Estrogenic Responsiveness of Brown Trout Primary Hepatocyte Spheroids to Environmental Levels of 17α-Ethinylestradiol

Three-dimensional (3D) fish hepatocyte cultures are promising alternative models for replicating in vivo data. Few studies have attempted to characterise the structure and function of fish 3D liver models and illustrate their applicability. This study aimed to further characterise a previously established spheroid model obtained from juvenile brown trout (Salmo trutta) primary hepatocytes under estrogenic stimulation. The spheroids were exposed for six days to environmentally relevant concentrations of 17α-ethinylestradiol—EE2 (1–100 ng/L). The mRNA levels of peroxisome (catalase—Cat and urate oxidase—Uox), lipid metabolism (acyl-CoA long chain synthetase 1—Acsl1, apolipoprotein AI—ApoAI, and fatty acid binding protein 1—Fabp1), and estrogen-related (estrogen receptor α—ERα, estrogen receptor β—ERβ, vitellogenin A—VtgA, zona pellucida glycoprotein 2.5—ZP2.5, and zona pellucida glycoprotein 3a.2—ZP3a.2) target genes were evaluated by quantitative real-time polymerase chain reaction. Immunohistochemistry was used to assess Vtg and ZP protein expressions. At the highest EE2 concentration, VtgA and ZP2.5 genes were significantly upregulated. The remaining target genes were not significantly altered by EE2. Vtg and ZP immunostaining was consistently increased in spheroids exposed to 50 and 100 ng/L of EE2, whereas lower EE2 levels resulted in a weaker signal. EE2 did not induce significant changes in the spheroids’ viability and morphological parameters. This study identified EE2 effects at environmentally relevant doses in trout liver spheroids, indicating its usefulness as a proxy for in vivo impacts of xenoestrogens.


Introduction
A wide range of negative impacts on aquatic species has been claimed by endocrinedisrupting chemicals (EDCs), namely estrogenic compounds [1].In fish, estrogens may cause changes in lipid and steroid hormone profiles [2], disruptions in the biotransformation metabolism [2,3]), and even behavioural and body phenotype alterations [4].In salmonids, the liver effects associated with the classic endocrine disruptor 17α-ethinylestradiol (EE2) have been intensely discussed and investigated using both in vivo [5][6][7] and in vitro [8][9][10][11] models.Transcriptomics has often been explored to study the impacts of EE2, as reviewed by Martyniuk et al. [12], unveiling estrogen-regulated genes.As an example, according to Hultman et al. [9], the microarray analysis of rainbow trout (Oncorhynchus mykiss) primary hepatocytes after EE2 exposure revealed gene expression alterations in distinct biological pathways related to estrogen receptor (ER) regulation, biotransformation, lipid metabolism, and cellular growth.In accordance, the EE2 treatment of liver slices of Atlantic cod (Gadus morhua) upregulated vitellogenin (Vtg) and zona pellucida (ZP) genes [13], both recognized as estrogenic biomarkers [14,15].
In mammals, three-dimensional (3D) hepatic models are increasingly being used as alternatives to in vivo assays for investigating metabolism [16,17], hepatic diseases [18], and hepatotoxicity [19], and even for testing new drug candidates [20].
While research on 3D systems in fish is not as advanced as that in mammals, studies in recent years have contributed to the establishment of experimental approaches for the maintenance of viable and functional hepatocytes from a variety of fish species [21][22][23][24][25].There is an enormous potential for using 3D liver models in the context of environmental toxicology.In this line, spheroids from a clearfin livebearer (Poeciliopsis lucida) hepatocellular carcinoma (PLHC-1) cell line were used to test the effects of a mixture of plastic additives (1 to 50 µM) and revealed a liver-like lipid profile phenotype [23].Further, PLHC-1 spheroids exposed to benzo(a)pyrene (1 nM to 5 µM) showed Cyp1a induction and cellular alterations, evidencing its high potential for screening the induction of xenobiotic metabolism and tissue damage [26].
Estrogenic compounds were already tested in fish hepatocyte spheroids.For example, Park et al. [25] found upregulations of Vtg5 mRNA expressions in zebrafish liver (ZFL) cell line spheroids stimulated with 1 nM of 17β-estradiol (E2) for 24 h.Additionally, an increase in ER and Vtg mRNA levels was observed in rainbow trout hepatocyte spheroids from day 1 to day 8 in culture after 24 h exposure to 1 µM of E2 [27].
Primary hepatocyte brown trout spheroids were previously characterized in terms of viability, morphological parameters, histomorphology, and basal gene expression [21,22].The present study aimed to test the applicability of those spheroids by exposing them to environmentally relevant concentrations of EE2 (1 to 100 ng/L).Gene expression analyses of a selection of target genes and immunohistochemistry were used to assess effects.To our knowledge, this is the first study using 3D cultures of fish primary hepatocytes under environmentally relevant concentrations of EE2.It uncovers innovative findings and sets baselines to enable further advances for implementing 3D cultures in fish toxicity assays.

Fish
A stock of sexually immature brown trout (Salmo trutta) of 12 (±2) months-mean (±SD)-was obtained from the Aquaculture Station of Torno (Amarante, Portugal).The animals were randomly distributed (maximum of 5 fish per tank) in 6 independent 100 L fiberglass tanks under a continuous dechlorinated water recirculation system and acclimatized for at least two weeks before use in diverse experiments.For this study, 6 animals were sampled from the different tanks.The fish were maintained under a photoperiod of 12 h light/12 h dark and fed daily ad libitum (Trout Plus 4, AquaSoja), except before isolations.The fish were visually inspected daily and did not have lack of appetite or any common stress behaviours.Finally, no gross lesions were noted, either when living or after sacrifice.Fish used in this study had a mean weight of 49.0 (±26.1)g and a mean total length of 16.3 (±2.9) cm.

Hepatocyte Isolation
Six fish were independently used for primary hepatocyte isolation.The fish were euthanized with an aqueous solution (0.6 mL/L) of ethylene glycol monophenyl ether (Merck KGaA, Darmstadt, Germany).All experimental procedures followed the Portuguese Decree-Law No. 113/2013, implementing EU Directive No. 2010/63 on animal protection for scientific purposes and respecting animal handling.The blood (±1 mL) was collected using insulin syringes through the caudal vein in order to minimize the amount of blood in the liver.After the liver was sampled, primary hepatocytes were isolated by ex situ collagenase perfusion using Hanks' Balanced Salt Solution (HBSS), as previously described in detail [11].Cell viability was checked in all experiments using an automatic cell counter (Invitrogen TM , Countess TM Automated Cell Counter) based on the trypan blue exclusion assay; cell viability was 76% (±7.0).

Spheroids Morphological Parameters
For each experiment on the 18th day post-isolation, brown trout PHS were photographed (n = 30 spheroids/condition) under a 10× objective lens in an Olympus CKX41 light microscope connected to an M5C-CYL-PL-D685CU Pixelink ® camera (Barrington, NJ, USA).Spheroid photos were analysed using AnaSP software version 2.0 to obtain the area, equivalent diameter, and sphericity [28].

Lactate Dehydrogenase (LDH) Assay
An LDH Cytotoxicity WST Assay Kit (ENZ-KIT157, Enzo Life Sciences, New York, NY, USA) was used to determine the LDH leakage in cell culture supernatants derived from each well housing the spheroids, across various experimental conditions.For each experiment on the 18th day post-isolation, 100 µL of cell culture supernatants per well (total of 4 replicates per fish, 1 from each plate/condition), previously centrifuged at 1500 rpm (0.2 rcf) for 5 min, were transferred to 96-well microplates (351172, non-tissue culture treated sterile plate with a flat bottom and low evaporation lid, Falcon, Corning, New York, NY, USA).Background controls were also included by adding 100 µL of fresh culture medium (n = 8 wells).Then, all wells received 100 µL of working solution, and the plate was incubated for 30 min at room temperature (±20 • C), protected from light.Absorbances were measured at 490 nm after stopping the reaction in a Multiskan TM GO microplate spectrophotometer (Thermo Scientific, Vantaa, Finland).Background subtraction was performed, and data were plotted for each condition.

AlamarBlue™ HS Cell Viability Reagent Assay
A total of 8 spheroids (total of 4 replicates per fish, 1 from each plate per condition) were transferred individually to a 96-well microplate (351172, non-tissue culture treated sterile plate with a flat bottom and low evaporation lid, Falcon, Corning, New York, NY, USA).After transfer, 90 µL of the fresh culture medium was added to each well, followed by 10 µL of the AlamarBlue™ HS Cell Viability Reagent (A50101, Invitrogen, Thermo Fisher Scientific, Life Technologies Corporation, Eugene, OR, USA).Blanks (n = 8 wells) were performed using the same amount of the corresponding medium but without spheroids.The plates were incubated at 18 • C for 24 h, protected from light, and at constant agitation (~100 rpm) (IKA ® MTS 2/4 digital microtiter shaker).The fluorescence was read at 550 nm and 588 nm (excitation and emission lengths, respectively) in a microplate Biotek Synergy™ HTX multimode reader (Agilent, Santa Clara, CA, USA) with the software Gen5 version 3.05.11(Agilent, Santa Clara, CA, USA).The fluorescence values of each sample were adjusted by blank subtraction and plotted for each condition.

Spheroids Morphology
For each experiment on the 18th day, spheroids (n = 6 per condition per fish) were transferred individually into 1.5 mL microtubes.The fixation was conducted using 500 µL of 10% buffered formalin (Epredia, Breda, The Netherlands) at room temperature.After 24 h, the fixative was changed to 70% ethanol.Richard-Allan Scientific HistoGel (HG-4000, Epredia, Breda, The Netherlands) was used to embed the spheroids.Dehydration, clearing, and paraffin impregnation were carried out over 12 h in an automatic processor (TP 1020, Leica Biosystems, Wetzlar, Germany).Posteriorly, the samples were embedded in paraffin (Histoplast, Epredia, Breda, The Netherlands) using an embedding station (EG1140C, Leica Biosystems, Wetzlar, Germany).After embedding, 1 spheroid/fish per exposure condition was sectioned with a thickness of 3 µm in a fully automated rotary microtome (RM2255, Leica Biosystems, Wetzlar, Germany).Sections were stained with hematoxylin and eosin (H&E), and their visualization and photographs were obtained using a light microscope (BX50, Olympus, Tokyo, Japan) coupled with a digital camera (EP50, Olympus, Tokyo, Japan).

RNA Extraction and cDNA Synthesis
One pool of spheroids per condition was obtained on the 18th day post-isolation for each independent experiment.The minimum number of brown trout PHS required to obtain the pellet was previously optimized [21].Spheroids were transferred into 1.5 mL microtubes, centrifuged at 1500 rpm for 5 min, and the pellets were snap-frozen in liquid nitrogen and stored at -80 • C. Total RNA was extracted with an illustra TM RNAspin Mini RNA Isolation Kit (GE Healthcare, Chicago, IL, USA), which involves a step of treating with DNase I to prevent contamination from genomic DNA.RNA quantification and purity were determined using a Multiskan TM GO microplate spectrophotometer (Thermo Scientific, Vantaa, Finland) with a µDrop™ Plate and a SkanIt Microplate Reader software version 4.1.The cDNA synthesis was performed using the iScript™ Reverse Transcription Supermix kit (Bio-Rad, Hercules, CA, USA), using 300 ng of total RNA for a total volume of 20 µL.

Quantitative Real-Time Polymerase Chain Reaction (RT-qPCR)
The CFX Connect real-time PCR detection system and CFX Manager software version 3.1 (Bio-Rad, Hercules, CA, USA) were used for RT-qPCR.Reactions included 5 µL of cDNA (diluted 1:5), 10 µL of iQ™ SYBR ® Green Supermix (Bio-Rad, Hercules, CA, USA), and 200 nM of each primer (total volume of 20 µL).Duplicates of cDNA samples and notemplate controls were always included in each analysis.A melt curve was used to assess the product's specificity.

Statistical Analyses
Descriptive and inferential statistics and graphs were created using Past 3 software version 3.25 [36] and the GraphPad Prism version 8.0.1, respectively.Shapiro-Wilk and Levene's tests checked the assumptions of normality and homogeneity of variance of data sets, respectively.A one-way analysis of variance (ANOVA) was followed by Tuckey's pairwise comparison post-hoc test.The non-parametric Kruskal-Wallis ANOVA and the Mann-Whitney pairwise comparison post-hoc test with sequential Bonferroni corrections were used in a few cases where the mentioned assumptions were not verified, even after data transformation.The differences were considered significant for p < 0.05.

Morphological Parameters
Bright-field images showed compact spheroids in all conditions with well-defined limits (Figure 1).For all experimental conditions, the area, equivalent diameter, and sphericity of the spheroids were not significantly influenced by EE2 exposures (Figure 1).

Viability-LDH and Resazurin Assays
The LDH and resazurin assays did not evidence significant differences betwe exposure conditions, as shown in Figure 2.

Viability-LDH and Resazurin Assays
The LDH and resazurin assays did not evidence significant differences between exposure conditions, as shown in Figure 2.

Viability-LDH and Resazurin Assays
The LDH and resazurin assays did not evidence significant differences between exposure conditions, as shown in Figure 2.

Morphology
No evident alterations were noted in the general structure of PHS between exposure conditions (Figure 3).Overall, spheroids showed a spherical/elliptical shape in all conditions.Hepatocytes had well-defined cellular limits and an intact nucleus (Figure 3).

Immunohistochemistry
Overall, for both antibodies used, the C and SC conditions showed no significant immunostaining (Figure 5), whereas EE2 exposure resulted in positive labelling, which was classified into three levels of intensity (in particular for Vtg): low, moderate, and strong.Vtg and ZP immunostaining were both cytoplasmic and varied between diffuse and a granular pattern homogeneously distributed throughout the spheroids.Despite interanimal variability, the overall pattern showed weaker immunostaining signals in spheroids exposed to EE2 concentrations of 1 ng/L and 10 ng/L (Figure 5).Moderate to strong immunolabelling was noticed for EE2 at 50 ng/L and 100 ng/L (Figure 5), respectively.The pattern obtained with both antibodies was well-defined, especially for the 100 ng/L condition.Negative controls did not show immunostaining.

Morphology
No evident alterations were noted in the general structure of PHS between exposure conditions (Figure 3).Overall, spheroids showed a spherical/elliptical shape in all conditions.Hepatocytes had well-defined cellular limits and an intact nucleus (Figure 3).The lipid metabolism-and peroxisome-related genes were not significantly influenced by EE2 exposure (Figure 4).Regarding the direct estrogen-related genes, the ZP3a.2,ERα, and ERβ showed stable mRNA levels, while VtgA and ZP2.5 expressions were dose-responsive and significantly upregulated at the highest concentration (100 ng/L) (Figure 4).and a granular pattern homogeneously distributed throughout the spheroids.Despite inter-animal variability, the overall pattern showed weaker immunostaining signals in spheroids exposed to EE2 concentrations of 1 ng/L and 10 ng/L (Figure 5).Moderate to strong immunolabelling was noticed for EE2 at 50 ng/L and 100 ng/L (Figure 5), respec tively.The pattern obtained with both antibodies was well-defined, especially for the 100 ng/L condition.Negative controls did not show immunostaining.

Discussion
Like in other research areas, the practice in ecotoxicology has been to limit the number of animals killed.In mammals, 3D hepatocyte cultures remain viable with morphological and functional capacities closer to in vivo liver and over extended periods more than the ones reported in two-dimensional (2D) cultures [18].
Hence, 3D cultures have recently been explored for developing and characterising new fish models [37][38][39].In this context, environmentally realistic conditions must be evaluated in 3D fish models before these can be adopted as accurate alternatives to in vivo studies.As such, the present study is a step toward determining the potential of using fish 3D models by testing environmental exposure concentrations.Using a previously developed spheroid model of primary brown trout hepatocytes, this study investigated the effects of EE2 at low, medium, and high environmental concentrations on estrogen-related targets.
EE2 environmental levels are variable, ranging from 0.002 ng/L [40] to 45 µg/L [41].The goal here was to test four concentrations of EE2, three of which were within the range of levels most frequently detected in the environment, as well as a concentration of 100 ng/L; although above the average, environmental monitoring also found such levels [41][42][43].
The morphological parameters (area, equivalent diameter, and sphericity) of 18-dayold brown trout PHS were not influenced by any EE2 concentration.Compact spheroids were observed in all conditions, which indicates that these EE2 levels did not compromise the cell-cell interactions and, consequently, the 3D structure of spheroids.Viable and metabolically competent brown trout PHS were observed in all conditions and resembled the biometric and morphological characteristics previously described for these spheroids with the same days in culture [22].
The genes selected here to evaluate the effect of EE2 were based on in vivo and in vitro assays, using liver or 2D hepatocytes isolated from brown trout, respectively [6,31,35].In these studies, the target genes associated with various pathways, including estrogenic, peroxisome, and lipid-related pathways, exhibited estrogen-specific responses.
Here, there were no significant differences in the mRNA levels of the selected peroxisomal and lipid metabolism-related genes after 6 days of exposure to EE2.Lipid-related pathways are frequently listed as estrogen-responsive in fish [3,44,45].Therefore, the absence of changes in the genes tested here may be related to the model, the doses, exposure time, and/or selected targets.For instance, in zebrafish liver, the expression of ApoAI was significantly downregulated in response to 10 ng/L of EE2 for 21 days [46].According to Madureira et al. [6], the same peroxisomal (Cat and Uox) and lipid (Fabp1 and ApoAI) target genes in brown trout liver showed a decrease or an increase (Acsl1) in gene expression after 28 days of in vivo exposure to EE2 (50 µg/L) [6].However, it should be noted that in the latter example, the selected concentration is approximately 5000-fold times higher than the usual environmental range (using a concentration of 10 ng/L as reference).A distinct profile was found regarding the expression of the VtgA and ZP2.5 genes, both widely accepted as estrogenic markers [47].In this study, the VtgA and ZP2.5 mRNA levels were significantly higher after exposure to EE2 at 100 ng/L compared to the other groups.An increasing trend was evident following exposure to 50 ng/L of EE2, although statistical significance was not proved.Data from in vivo EE2 experiments with different fish species support our results with brown trout PHS.For instance, increases in Vtg gene expression were observed in rainbow trout and male medaka (Oryzias latipes) after water exposure to 125 ng/L and 100 ng/L of EE2 over 61 days [7] and 4 weeks [48], respectively.Uren Webster et al. [33] found a significant upregulation of vtg1, ZP2.5, and ZP3a.2 genes after 4 days of water treatment with 34.4 ng/L of E2.Interestingly, ZP2.5 transcript was more responsive to E2 than ZP3a.2, which aligns with our results.Also, in brown trout, low doses of EE2 (3 ng/L) for 21 days caused a Vtg mRNA induction in vivo [5], but this did not happen in PHS, at least for the exposure time tested here.Evidence linked Vtg induction in trout to ER expression [49], but the ERα and ERβ mRNA levels did not change significantly in the present study.Mortensen and Arukwe [50] showed that liver ERα and ERβ mRNA levels significantly decreased, while Vtg and ZP increased at day 3 after EE2 exposures (50 ng/L) in juvenile Atlantic salmon.It was suggested by the authors that basal ER levels may be sufficient to trigger Vtg and ZP gene inductions, which may justify here the steady ER expression with increases of VtgA and ZP2.5 mRNA levels.Further, prior studies demonstrated a peak induction of ERα mRNA levels in less than 48 h after EE2 exposure in rainbow trout [7].As a result, we cannot rule out the possibility that we did not assess (at least) ERα expression at its maximum induction time.
In vitro studies with hepatocyte cell cultures also corroborate the VtgA and ZP2.5 gene expression patterns obtained in this study.In rainbow trout primary hepatocytes exposed to EE2 for 48 h, there was a dose-effect concentration on the upregulation of vtg1, zrp3, and zrp4 levels for concentrations above 88.9 ng/L (0.3 nM) [9].In fish hepatocyte spheroid cultures, estrogenic stimulation has also been tested but, so far, in unrealistic environmental concentrations [25,27,51,52].For instance, according to Pelissero et al. [52], EE2 highly increased the Vtg concentrations in the culture medium of rainbow trout hepatocyte aggregates at the 296,400-29,640,000 ng/L (1000-100,000 nM) range, although minimal induction was measured at 296.4 ng/L (1 nM).Flouriot et al. [27] described that the Vtg mRNA induction was caused by exposure to 272,380 ng/L (1 µM) of 17β-estradiol (E2) during 24 h in male rainbow trout hepatocyte spheroids.Also, in the zebrafish liver (ZFL) cell line, the Vtg5 mRNA expression was upregulated after exposure to E2 (1 nM = 272.38ng/L) [25].
Within the present study, EE2 also caused a specific induction of Vtg and ZP protein expression, as already evidenced in the liver of Atlantic salmon exposed to other estrogenic compounds [53,54].The positive control labelling for Vtg and ZP protein was cytoplasmic, as previously reported for both antibodies in brown trout hepatocytes [35].Here, immunohistochemistry corroborated gene expression, showing a consistent increase in Vtg and ZP proteins, particularly after exposure to 100 ng/L of EE2.Immunolabelling was weaker and variable across the fish for lower EE2 concentrations (1 and 10 ng/L).The effect concentration that caused a 10-fold Vtg protein induction in rainbow trout hepatocytes was 71.7 ng/L (2.42 × 10 −10 mol/L) [8], which aligns with the data obtained here.

Conclusions
The present study revealed the applicability, sensitivity, and specific responsiveness to EE2 of primary brown trout hepatocyte spheroids under distinct environmentally relevant levels.At 100 ng/L, remarkable effects were detected, particularly the stimulation of Vtg and ZP genes and protein expression.Lipid target genes did not change, even at the higher EE2 concentration, under the tested conditions.Given the published evidence from in vivo and 2D in vitro research, the model used here could serve as an alternate diagnostic tool for uncovering some effects arising from exposure to estrogen-disrupting compounds.However, further studies should be considered to disclose the EE2 effects at environmentally relevant levels in brown trout PHS and its applicability as an alternative model in ecotoxicology.