Effects of Zinc Status on Expression of Zinc Transporters, Redox-Related Enzymes and Insulin-like Growth Factor in Asian Sea Bass Cells

Simple Summary This article is of special interest for understanding the detriments of zinc deficiency and the benefits of zinc supplementation for aquacultures of Asian sea bass. We found that altering extracellular zinc conditions significantly affected the expression of certain zinc transporters and zinc-binding proteins as well as intracellular free zinc levels in cells from Asian sea bass (Lates calcarifer, SF cells). Interestingly, we detected no difference in the effects of organic compared to inorganic zinc supplements, when the same molar concentrations were added to the cultures. Moreover, changes in extracellular zinc conditions impacted on the expression of genes related to the redox balance and to growth hormone metabolism. Our data indicate that zinc deficiency induces a stress response in cells from Asian sea bass, which may be avoided when sufficient amounts of zinc are provided to the cells. We identified SF cells as a suitable model for studies to optimize zinc supplementation in the aquaculture of Asian sea bass. Abstract Since Asian sea bass is one of the economically most important fish, aquaculture conditions are constantly optimized. Evidence from feeding studies combined with the current understanding of the importance of zinc for growth and immune defense suggest that zinc supplementation may be a possible approach to optimize aquacultures of Asian sea bass. To investigate the effects of zinc deficiency and zinc supplementation, cells from Asian sea bass were incubated in culture medium with different zinc contents. The expression of genes, important for zinc homeostasis, redox metabolism, and growth hormones was analyzed using RT-PCR. Zinc deficiency induced the expression of certain zinc transporters (ZIP14, ZIP10, ZIP6, ZIP4, ZnT4, ZnT9) as well as of SOD1, IGF I and IGF II, while expression of ZnT1 and metallothionein (MT) was reduced. Zinc supplementation decreased the expression of ZIP10, while expression of ZnT1 and MT were elevated. No differences in the effects of zinc supplementation with zinc sulfate compared to supplementation with zinc amino acid complexes were observed. Thus, extracellular zinc conditions may govern the cellular zinc homeostasis, the redox metabolism and growth hormone expression in cells from Asian sea bass as reported for other fish species. Our data indicate that supplementing aquacultures with zinc may be recommended to avoid detriments of zinc deficiency.


Introduction
Fish and its products play a major role in human nutrition, providing at least 20% of the protein intake for a third of the world's population, and the dependence is highest in developing countries [1]. Asian sea bass (Lates calcarifer) is one of the economically most important fish in many countries. It is a popular species for aquaculture, because of its

Measurement of Free Intracellular Zinc
The uptake of ZnSO 4 and ZnAA into SF cells was analyzed using the zinc-specific probe FluoZin-3AM (Invitrogen, Karlsruhe, Germany) according to [24]. Cells (ZA) were washed, trypsinized, rewashed, mixed with an equal volume of fresh medium and seeded into 96 well plates (eight wells per treatment) on the day before the experiment. The cells reached 80% confluence at the time of analysis. On the day of the experiment, the adherent cells were washed once with PBS at room temperature. Detachment of the cells was controlled using microscopy. Incubation buffer (100 µL) made 1 µM with FluoZin3-AM was added to each well. Cells were then incubated at 28 • C for 30 min. After incubation, cells were washed once with PBS. Thereafter, 100 µL of measurement buffer were added to the wells. To determine the minimum (F min ) and maximum fluorescence (F max ), wells for F min were incubated with 100 µM N, N, N', N'-tetrakis (-) [2-pyridylmethyl] ethylenediamine (TPEN), while wells for F max were incubated with 100 µM ZnSO 4 and 50 µM pyrithione. Then, cells were incubated at 28 • C for 15 min and the baseline was recorded. The resulting fluorescence was measured with a Tecan 340 fluorescence multi-well plate reader (Tecan, Crailsheim, Germany) using excitation and emission wavelengths of 485 and 535 nm, respectively. Subsequently, the remaining wells were adjusted to contain 50 µM ZnSO4 or ZnAA by adding stock solutions (100 mM) and a kinetic was recorded. The zinc concentration was then calculated by the formula (KD = 8.9 nM):

Flame Atomic Absorption Spectrophotometry (AAS)
The total zinc content of all media (data not shown) and cells was analyzed by AAS (AAnalyst 800, Perkin Elmer, Baesweiler, Germany). SF cells were washed with PBS, trypsinized, rewashed with wash buffer (0.9% sodium chloride (AppliChem, Darmstadt, Germany), 10 mM ethylenediaminetetraacetic acid (Sigma-Aldrich) and 10 mM N- piperazine-N'-[2-ethanesulfonic acid] (HEPES, AppliChem)), counted and the pellet of 5 × 10 6 cells was afterwards lysed in water. Lysate (1 µL) was used to determine the protein content of the samples, using a Bradford assay (Biorad, Germany). The remaining lysate was digested in nitric acid (AppliChem) for 2 h at 80 • C and diluted with water, followed by subsequent zinc measurements with an AAS. Zinc concentrations for cells were normalized to the protein content of the sample.

Real-Time PCR
Isolation of total RNA and reverse transcription into cDNA were performed, according to the manufacturer's instructions, using the NucleoSpin RNA II kit (Macherey-Nagel, Dueren, Germany) and the qScript cDNA synthesis kit (QuantaBioscience, Beverly, MA, USA), respectively. Quantitative real-time PCR was performed using SYBR ® green reagent and a StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, Darmstadt, Germany).
The relative expression of target genes was determined using a standard curve, established by a mixture of stimulated and unstimulated SF cells. The internal reference control gene was elongation factor-1 (EF-1)A [2] because of its consistent expression level in ZA, ZnSO4, ZnAA or CHELEX cultures. The qPCR primers were as previously published or designed according to the CDS of reference sequences (RefSeq) for the gene of interest including the isoforms or transcript variants found on NCBI [25,26]. Primers were blasted to exclude possible unspecific PCR products. All PCRs were run in duplicate. The qPCR processes were performed as follows: the first step was performed for 15 min at 95 • C, followed by 40 cycles of 15 s at 95 • C for denaturation, and 30-60 s at 50-61 • C for annealing. For Zip10, 30 s at 72 • C for extension were added and water in the PCR mix was replaced by Q Solution (Qiagen, Germany). The third step for the melting curve analysis was performed for 15 s at 95 • C and then 60 s at 60 • C. Product sizes were verified using gel electrophoresis. The primer sequences are summarized in Table 1. For some genes multiple isoforms were available. In that case, all NCBI-RefSeq for isoforms covered by the primers are indicated.

MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) Test
SF cells were washed, trypsinized, rewashed with PBS, taken up in ZD, ZA or ZS (50 µM ZnSO 4 ) medium, diluted 1:30 in the respective medium and seeded on 96 well plates (8 wells per treatment). After incubating cells for 7 days, MTT-Solution was added until needle-like formazan aggregates were visible. The supernatants were discarded, and dimethyl sulfoxide was added. The plates were shaken for 10 min, and absorbance was measured at 570 nm using 690 nm as reference wavelength. Means of the eight wells were calculated.

Statistics
Outliers were excluded by Grubb's outlier test and normal distribution of the data was assessed by Shapiro-Wilk normality test. The statistical significance of the results was analyzed by GraphPad Prism software version 9 (GraphPad Software, La Jolla, CA, USA). For comparing two data sets (Figure 1a), multiple paired t-tests were performed, * p < 0.05. For comparing more than two data sets, one-way ANOVA with Tukey's post hoc test was used (p < 0.05).

Zinc Homeostasis in Zinc Deficient and Zinc Supplemented SF Cells
It has been hypothesized that the uptake of zinc into cells strongly depends on the chemical nature of the zinc compound used and that zinc in complex with sulfate, gluconate, aspartate or other amino acids is taken up by cells with differing efficiency. Thus, as a first step, the uptake of zinc sulfate (ZnSO4) and a zinc-amino acid complex (ZnAA) into SF cells was analyzed using the zinc-specific probe FluoZin3 AM. As depicted in Figure 1a, ZnSO4 and ZnAA were taken up by the cells. Maximum intracellular zinc levels were reached 30 min after adding the supplement. However, the amount of zinc taken up by the cells during this time was significantly higher when ZnAA was added compared to ZnSO4 (Figure 1a).
To evaluate if this difference was limited to short incubation times, total cellular zinc was measured 7 days after supplementation was started. As in the short-term incubation, both ZnSO4 and ZnAA incubation resulted in a significant increase in total cellular zinc. However, interestingly, no difference was observed at 7 days when total cellular zinc was measured by atomic absorption spectrometry ( Figure 1b). In line with this, the expression of metallothionein, the most important intracellular zinc-binding protein, was significantly increased by ZnSO4 and ZnAA in the same magnitude ( Figure 1c). When cells were cultured in medium where zinc had been depleted using CHELEX resin, total cellular zinc was significantly decreased compared to cells cultured in zinc-adequate conditions (Figure 1b) as was also observed for the expression of metallothionein (Figure 1c).

Zinc Transporter Expression in SF Cells
Aiming at a more detailed characterization of zinc homeostasis and at finding an explanation for the changes in the zinc status of SF cells, the expression of zinc importing transporters was analyzed. No data on the expression of zinc transporters in cells from

Zinc Homeostasis in Zinc Deficient and Zinc Supplemented SF Cells
It has been hypothesized that the uptake of zinc into cells strongly depends on the chemical nature of the zinc compound used and that zinc in complex with sulfate, gluconate, aspartate or other amino acids is taken up by cells with differing efficiency. Thus, as a first step, the uptake of zinc sulfate (ZnSO 4 ) and a zinc-amino acid complex (ZnAA) into SF cells was analyzed using the zinc-specific probe FluoZin3 AM. As depicted in Figure 1a, ZnSO 4 and ZnAA were taken up by the cells. Maximum intracellular zinc levels were reached 30 min after adding the supplement. However, the amount of zinc taken up by the cells during this time was significantly higher when ZnAA was added compared to ZnSO 4 ( Figure 1a).
To evaluate if this difference was limited to short incubation times, total cellular zinc was measured 7 days after supplementation was started. As in the short-term incubation, both ZnSO 4 and ZnAA incubation resulted in a significant increase in total cellular zinc. However, interestingly, no difference was observed at 7 days when total cellular zinc was measured by atomic absorption spectrometry ( Figure 1b). In line with this, the expression of metallothionein, the most important intracellular zinc-binding protein, was significantly increased by ZnSO 4 and ZnAA in the same magnitude ( Figure 1c). When cells were cultured in medium where zinc had been depleted using CHELEX resin, total cellular zinc was significantly decreased compared to cells cultured in zinc-adequate conditions ( Figure 1b) as was also observed for the expression of metallothionein (Figure 1c).

Zinc Transporter Expression in SF Cells
Aiming at a more detailed characterization of zinc homeostasis and at finding an explanation for the changes in the zinc status of SF cells, the expression of zinc importing transporters was analyzed. No data on the expression of zinc transporters in cells from the Asian sea bass were available to date. Figure 2 summarizes the results for transcription of those transporters that were found to be expressed by SF cells. However, the expressions of Zip1, Zip7, Zip9, Zip11, Zip12 and Zip13 were below the detection limit and no homologous sequence was found in NCBI for Zip5 (Table 2). Similar to the observations for the transporters of the Zip family, no significant difference in the effects of ZnSO4 compared to ZnAA on the expression of ZnTs was found (data not shown). Thus, data were pooled and denoted as ZS as well. Zinc supplementation increased the expression of ZnT1 compared to ZD and ZA cells (Figure 3a), whereas zinc deficiency increased the expression of ZnT4 and ZnT9 compared to ZA and ZS cells (Figure 3b,d).   No significant difference was found between the expression induced by ZnSO 4 compared to ZnAA, as exemplified for Zip14 (Figure 2a). Thus, data for ZnSO 4 and ZnAA supplementation were averaged and are subsequently summarized as zinc supplementation (ZS). Interestingly, ZD cells showed higher expression levels for all the Zip mRNAs detected than in ZA or ZS cells, with some Zips, e.g., Zip14, Zip10, Zip6, and Zip4 being more strongly affected than others (Figure 2b,c,e,f). Zinc supplementation significantly decreased the expression of Zip10 in comparison to ZA controls. There was no statistically significant difference between the ZA or ZS cells for any of the other detected Zips (Figure 2b-f). However, some of the transporters, including Zip2, Zip4 and Zip14 showed some increase in expression in zinc supplemented compared to zinc adequate cells.
Similar to the observations for the transporters of the Zip family, no significant difference in the effects of ZnSO 4 compared to ZnAA on the expression of ZnTs was found (data not shown). Thus, data were pooled and denoted as ZS as well. Zinc supplementation increased the expression of ZnT1 compared to ZD and ZA cells (Figure 3a), whereas zinc deficiency increased the expression of ZnT4 and ZnT9 compared to ZA and ZS cells (Figure 3b,d). Similar to the observations for the transporters of the Zip family, no significant difference in the effects of ZnSO4 compared to ZnAA on the expression of ZnTs was found (data not shown). Thus, data were pooled and denoted as ZS as well. Zinc supplementation increased the expression of ZnT1 compared to ZD and ZA cells (Figure 3a), whereas zinc deficiency increased the expression of ZnT4 and ZnT9 compared to ZA and ZS cells (Figure 3b,d).  Only negligible zinc-dependent changes were detected for ZnT5 expression. No homologous sequence was found for ZnT3 and ZnT10. The mRNA expression of ZnT2, ZnT6, ZnT7 and ZnT8 was below the detection limit (Table 2).

Effect of Zinc Status on the Expression of Redox-Related Mediators
Zinc is generally known for its anti-oxidant function, which has so far not been investigated for the Asian sea bass. While no significant effects of zinc supplementation (ZS) with either ZnSO 4 or ZnAA on the expression of redox-related enzymes were found in comparison to ZA cells, zinc deficiency caused a significant increase in SOD1 transcription in comparison to ZA and ZS cells (Figure 4a). Catalase expression showed the same trends but the differences between the treatments were not statistically significant (Figure 4b). Glutathione peroxidases 2, 3 and 7 were tested as well, but no expression of those genes was detected in SF cells (data not shown).

Expression of Growth-Related Factors Is Affected by Zinc Deficiency
Similar to what was observed for genes involved in regulating the redox metabolism, zinc supplementation did not significantly alter the expression of IGF-I or IGF-II in comparison to ZA cells. However, zinc deficiency significantly induced the expression of both IGF-I and IGF-II compared to zinc adequate control cells (Figure 5a,b).
with either ZnSO4 or ZnAA on the expression of redox-related enzymes were found in comparison to ZA cells, zinc deficiency caused a significant increase in SOD1 transcription in comparison to ZA and ZS cells (Figure 4a). Catalase expression showed the same trends but the differences between the treatments were not statistically significant (Figure 4b). Glutathione peroxidases 2, 3 and 7 were tested as well, but no expression of those genes was detected in SF cells (data not shown).

Expression of Growth-Related Factors Is Affected by Zinc Deficiency
Similar to what was observed for genes involved in regulating the redox metabolism, zinc supplementation did not significantly alter the expression of IGF-I or IGF-II in comparison to ZA cells. However, zinc deficiency significantly induced the expression of both IGF-I and IGF-II compared to zinc adequate control cells (Figure 5a,b).

Expression of Growth-Related Factors Is Affected by Zinc Deficiency
Similar to what was observed for genes involved in regulating the redox metabolism, zinc supplementation did not significantly alter the expression of IGF-I or IGF-II in comparison to ZA cells. However, zinc deficiency significantly induced the expression of both IGF-I and IGF-II compared to zinc adequate control cells (Figure 5a,b).

Metabolism Is Affected by the Zinc Status
To obtain a first impression on the effects of the extracellular zinc availability on the metabolism of the cells, the MTT test was performed. As illustrated in Figure 6, cells cultured in zinc deficient medium tended to have a reduced metabolic activity, but this was not significantly different from ZA cells, whereas zinc supplementation significantly increased metabolic activity in comparison to ZA cells.

Metabolism Is Affected by the Zinc Status
To obtain a first impression on the effects of the extracellular zinc availability on the metabolism of the cells, the MTT test was performed. As illustrated in Figure 6, cells cultured in zinc deficient medium tended to have a reduced metabolic activity, but this was not significantly different from ZA cells, whereas zinc supplementation significantly increased metabolic activity in comparison to ZA cells.

Discussion
The uptake of zinc by cells may depend not only on the nature of the zinc compound used, but also on the species and cell type investigated as well as the overall composition of the diet [27]. While zinc transporters are largely responsible for the transport of zinc across cellular membranes in mammalian cells, data on zinc transport into fish cells is rather scarce and mostly limited to investigations on zebrafish and catfish [18,20]. Our general knowledge on the effects of alterations in zinc availability on the metabolism of fish cells at the molecular level is also limited but would be helpful to improve growth

Discussion
The uptake of zinc by cells may depend not only on the nature of the zinc compound used, but also on the species and cell type investigated as well as the overall composition of the diet [27]. While zinc transporters are largely responsible for the transport of zinc across cellular membranes in mammalian cells, data on zinc transport into fish cells is rather scarce and mostly limited to investigations on zebrafish and catfish [18,20]. Our general knowledge on the effects of alterations in zinc availability on the metabolism of fish cells at the molecular level is also limited but would be helpful to improve growth and health of fish in aquaculture. In the current study, we thus addressed these open questions using epithelial SF cells derived from Asian sea bass. This fish type is frequently used for commercial fishing and aquaculture and is thus very important economically.
Intracellular free zinc levels were increased more rapidly and to a greater extent by ZnAA compared to ZnSO 4 , when analyzed in the short term within the first 30 min after supplementation had started (Figure 1a). However, in contrast, no differences between the two supplements were observed when total cellular zinc and the expression of MT were analyzed after supplementation for 7 days (Figure 1b,c). In agreement with the latter result for SF cells, no differences in final zinc levels after the uptake of ZnAA compared to ZnSO 4 by Asian sea bass where found in vivo [3]. Several studies indicate that FluoZin3-AM accumulates in certain cellular compartments depending on the cell type and that the probe does not necessarily remain in the cytoplasm [24,28,29]. Thus, although results show clear differences in the uptake of the two zinc compounds, this may be true for only distinct compartments, while in other parts of the cell, conditions may be different. Alternatively, uptake of inorganic zinc compounds may require de novo expression or intracellular re-distribution of zinc transporters or zinc-binding proteins and may thus be delayed compared to ZnAA sources [9]. However, and most importantly, supplementation with either ZnSO 4 or ZnAA was able to significantly elevate zinc levels within the cells (Figure 1), as was also described for RTgutGC (Rainbow Trout gut Guelph Canada) cells and cells derived from common carp [27,30]. As expected, intracellular zinc levels decreased in SF cells grown in zinc deficient medium.
Expression of some, but not of all, zinc transporters and zinc-binding proteins appear to be affected by changes in zinc homeostasis. Moreover, Zips and ZnTs exhibit tissuespecific expression and distribution [9,18]. In SF cells, we detected mRNA expression of MT and 11 zinc transporters. The expressions of MT, ZnT1, ZnT4, ZnT9, Zip4, Zip6, Zip10, Zip14 were affected by changes in zinc homeostasis (Figures 1c, 2 and 3). MT expression was increased by both zinc supplements and was decreased by zinc deficiency (Figure 1c), which is in line with reports from various studies in mammals and fish [31,32]. Interestingly, ZnT1 mRNA levels have been shown to be directly regulated by zinc across various cell types and species including zebrafish gill and zebrafish intestine [18,[30][31][32]. This indicates that ZnT1 may be a highly conserved transporter, exporting zinc to maintain zinc homeostasis in cells during high zinc conditions [32,33]. Down-regulation of ZnT1 expression during zinc deficiency has been described for zebrafish [18]. We also found a slight but not significant decrease in ZnT1 expression in zinc deficient SF cells (Figure 3a). On the other hand, zinc deficiency increased the expression of ZnT4 and ZnT9 in SF cells (Figure 3b,d). For ZnT4, up-and down-regulation by zinc deficiency were reported for rat jejunum and a human osteoblastic cell line, respectively [34,35]. ZnT9 mRNA was upregulated in zinc deficient human osteoblastic cells, as observed here [35] and the role of ZnT9 in maintaining zinc homeostasis is accepted [36]. Probably to facilitate or hinder zinc uptake, mRNA expression of Zip4, Zip6, Zip10 and Zip14 was upregulated in zinc deficient SF cells, while zinc supplementation reduced expression of Zip10. Zip10 expression was also upregulated in zinc deficient zebrafish together with Zip3 [37]. Zip3 expression was also slightly upregulated in zinc deficient SF cells (Figure 2g). Zinc supplementation affected the expressions of Zip1, Zip4 and Zip8 in zebrafish [38]. Regarding epithelial cells from yellow catfish, Zip1, Zip3, Zip6, Zip7, Zip8, Zip9, Zip10, Zip11, Zip13 and Zip14 appeared to be relevant for maintaining zinc homeostasis, depending on the duration and extent of changes in extracellular zinc [20,33]. Zip10 thus seems to be a common zinc-sensitive transporter in various species, while zinc-dependent regulation of other zinc transporters may strongly depend on cell type, zinc status, duration of zinc deficiency or supplementation and the species investigated.
Although only mRNA data are presented, our results indicate that zinc uptake, regulation of intracellular zinc levels and adjustment to extracellular zinc conditions by SF cells are similar to what has been described for mammals. However, it needs to be tested, whether localization and function of MT and the zinc transporters in SF cells and for example human epithelial cells are the same. Restricting our analysis to mRNA expression is a clear limitation of this study. Since the data shown for zinc transporters and MT presented here (Figures 1-3) match well with major findings regarding zinc metabolism in other species [9,19,31,39], the validity of our data is still probably high. Unfortunately, the available resources for studying zinc transporters and zinc-related processes in Asian sea bass on a molecular level are so far limited. For example, we were not able to find sea bass-specific antibodies for zinc transporters. However, we hope that our data will increase the attractivity of Asian sea bass for zinc-related research and thus the development of antibodies in the future.
Zinc is well known to have anti-oxidant properties and to support cell growth as well as proliferation [9]. However, our data revealed no significant effects of zinc supplementation on the mRNA expression of CAT, SOD, IGF-I or IGF-II (Figures 4 and 5). However, contrasting effects of zinc supplementation on expression and activity of all four factors have been described elsewhere for other species. Thus, zinc supplementation of coho salmon, olive flounders and other fish improved the activity of anti-oxidant enzymes, including SOD and CAT [10,40], and activities of SOD and CAT in serum, intestine, liver and muscle tissue of juvenile Jian carp increased after zinc supplementation [41]. Investigation of zinc-treated disk abalone revealed that the transcriptional responses of SOD genes were quite variable depending on the tissues and isoforms investigated [42]. Finally, zinc plus lead supplementation of carp increased SOD and CAT activity after 24 h of supplementation. In contrast, activities of both enzymes were decreased after several days of zinc and lead treatment [43]. Although zinc was combined with lead in that study, the time-dependency of the effects may explain some of the discrepancies between our data and existing studies. Similarly, effects of zinc supplementation on IGF-I and IGF-II expression strongly varied depending on species, dosage and duration of zinc treatment. Supplementation of zebrafish with moderate concentrations of zinc chloride (ZnCl 2 ) or zinc oxide nanoparticles did not affect growth or IGF-I mRNA expression, while treatment with high ZnCl 2 concentrations decreased both parameters in zebrafish [44,45]. Zinc supplementation of rainbow trout white muscle cells caused a decrease in IGF mRNA levels as well [7].
A lack of zinc induces a stress response in various cell types. Amongst others, production of reactive oxygen species is induced and growth retardation, maldevelopment and dwarfism in fish have been reported [11]. To counteract oxidative stress, anti-oxidatively active enzymes can be induced, as observed here in zinc deficient SF cells (Figure 4) [40]. However, several studies indicate that long-term feeding of a zinc deficient diet to animals such as young grass carp reduces the SOD and CAT expression and activity in various tissues [8,46]. In contrast, CAT activity was increased in the heart of zinc deficient rats and a slight increase in the activity of SOD was observed as well [47]. Embryos from zinc deficient ducks revealed higher expression and activities of CAT [39]. Moreover, thermal or nutritional stress significantly increased SOD mRNA and CAT levels in the tropical fish species Chanos chanos [48] and in plasma of Starry flounder [49]. This is in line with our data for zinc deficiency-induced stress. We were surprised that zinc deficiency significantly increased mRNA expression of IGF-I and IGF-II ( Figure 5). In fact, most studies in mammals revealed a decrease in IGF mRNA expression during zinc deficiency in various tissues [50]. On the other hand, no changes in IGF mRNA expression were found in zinc deficient rat liver cells in vitro. Furthermore, increased expression of IGF-I mRNA in the liver of zinc deficient weanling rats was described, while serum IGF-I levels remained constant [51]. Interestingly, effects of starvation and refeeding on IGF-I levels were slower and restricted to the liver in salmon compared to rats. This indicates that there may be large species-dependent differences in IGF-I expression as well as in the adaptation to food deprivation [52].
The available studies on effects of changes in zinc homeostasis on the expression and activity of anti-oxidant mediators and IGF are difficult to compare due to high heterogeneity. Large differences exist between the chosen (fish) species, supplementation in vivo or in vitro, the zinc dose, the zinc compound and the duration of supplementation. In addition, complete fish or organs were often analyzed and enzyme activity rather than mRNA expression of SOD and CAT was assessed. Thus, differences in experimental setups may explain the contrasting data. Furthermore, ectothermic animals such as Asian sea bass are known to generally have a slower metabolism and many fish are metabolically adapted to extended periods of food deprivation [52]. Thus, the zinc-deficient cells may still be capable of mounting an anti-oxidative and IGF response within the first 7 days of zinc deficiency. However, since metabolic activity was reduced in zinc-deficient SF cells (Figure 6), they may overall react to zinc deficiency with a stress response. The induction of anti-oxidative factors ( Figure 4) may overall not rescue the cells from zinc deficiency-induced apoptosis, In summary, our data indicate that zinc deficiency increases the expression of CAT, SOD, IGF-I and IGF-II in SF epithelial cells (Figures 4 and 5), which remains to be tested on the protein and enzyme activity level. Moreover, the activity of those factors needs to be assessed to test if our findings are relevant for the zinc-regulated anti-oxidative response of epithelial cells.

Conclusions
It can be concluded that our data provide essential basic information on mRNA levels indicating expression of genes relevant for zinc metabolism in cells from the Asian sea bass under different zinc supplementation regimes. Those data may enable future studies focusing on the molecular mechanisms underlying zinc's effect on the growth and the redox response of Asian sea bass. Parallels found in the expression of zinc transporters and MT between mammalian and Asian sea bass cells were found. Thus, Asian sea bass, or cultures of Asian sea bass cells, may provide valuable models to address zinc-related questions in fish in general.