Cu (II) oxide nanopowder purchased from Sigma, St. Louis, MO, USA (Catalog No. 544868) was used for the treatment. Based on the manufacturers' specifications, size of the nanopowder was determined as < 50 nm particle size by TEM analysis and the surface area was 29 m²/g. The purity of Cu nanopowder was specified as 99.5% based on the analysis report. Physiochemical properties of Cu-NPs such as pH and alkalinity were analyzed by dissolving the Cu-NPs in water before the treatment. For Cu-NPs stock preparation, 100 μg/ml of Cu-NPs was weighed and mixed in distilled water. The solution was sonicated up to 6 h and then used for the treatment. The effect of particles before and after sonication was observed through Field emission scanning electron microscopy (FESEM) and TEM. X-Ray Diffraction (XRD) and particle size distribution of Cu-NPs were also done. The percentage of Cu in CuSO₄ was calculated and the equivalent concentration of Cu load was measured and chosen for the treatment.

C. batrachus (L.) of similar age group and size (~120 g of 21 months old; 25 ± 3.5 cm) reared in the laboratory were taken for the experiment (n = 5). Each group (control, Cu-NPs-treated and CuSO₄-treated) were acclimated in glass tank (one fish/tank) contains 50 L of water and maintained at ambient light and dark cycle (20 ± 2°C; 12 h L/12 h D). During treatment, fish were daily fed commercially available pelleted fish feed, ad libitum. Catfish were separated into three groups (n = 5) such as control, CuSO₄ and Cu-NPs. Adult catfish (n = 5) were exposed to Cu-NPs (100 μg/L) and CuSO₄ (equivalent conc. of Cu load) for 21 days. Treatment was given every day once after the replenishment of water was done. Since the replenishment of water was done with Cu in both forms every day, the exposure of these correlates was authenticated without any measurement of dissolved Cu. Further, the histology and TEM analysis in the results section demonstrate the desired exposure of Cu in any form. The dosage selected for the experimental study was eco-relevant and also less than LC₅₀ value reported previously, which is why a single dose Cu-NPs treatment was done with equivalent amount of Cu separately. Further, the aim of this study is not for performing dose-dependency in in vivo study, however, the same was done in in vitro. Fish behavior and water temperature were monitored for all groups and no mortality was seen during the treatment period. After 21 days, blood was drawn from dorsal vein for androgen quantification while testes were dissected out for real-time PCR (qPCR), histology and TEM analysis. All the animal (fish) experiments was performed by following the general guidelines of the Institutional Animal Ethics Committee, University of Hyderabad.

Quantification of certain mRNA transcripts in control and treated samples were done using qPCR by SYBR® Green detection method. Total RNAs were prepared from testis of control, CuSO₄ and Cu-NPs treated batches using TRI-reagent® (Sigma) as per the protocol and quantified by a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). First strand synthesis was carried out for 500 ng of total RNA extracted from testis of three groups using verso cDNA synthesis kit (Thermo Scientific Inc., Waltham, MA, USA) based on the protocol followed. PCR amplification was performed for all the samples using internal control, 18S rRNA to check the efficiency of the reaction. There is no significant difference between the expression level of 18S rRNA between control and treated testis and hence it was used as a reference gene for the analysis. All qPCR primers were designed with at least one primer should spanned the junction of two exons by giving a single PCR product where there is no possibility for genomic DNA amplification. All the genes analyzed in this study were previously cloned from catfish in our laboratory and the primer sequences were listed in Supplementary Table 1. The reaction was setup in triplicates for each sample using SYBR green master mix and done on a 7500 Fast Real-Time PCR system (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's recommendations. Melting-curve analysis was performed to check the efficiency of PCR product amplification while zero amplification was observed in no template control. Cycle threshold (Ct) values were calculated from the exponential phase of PCR amplification and the target gene expression was normalized against the expression of 18S rRNA to generate a ΔCt value (Ct of target gene–Ct of endogenous control). Relative expression of the target gene was then calculated according to the equation 2^−ΔΔCt.

Testis of control, CuSO₄ and Cu-NPs treated adult catfish (n = 5) were dissected out and fixed in Bouin's fixative (15: 5: 1 of saturated picric acid: formaldehyde: glacial acetic acid). The tissues were dehydrated using graded ethanol series, followed by xylene and finally embedded in paraplast (Sigma). On a rotatory microtome (Leitz, Wetzlar, Germany), sections of about 5 μm thickness were cut, spread and dried in a hot plate (Yorko Sales Pvt. Ltd., New Delhi, India). The tissue slides were treated with xylene for deparaffinization and rehydrated using graded ethanol series. Then, the sections were stained with hematoxylin and eosin, followed by dehydration to mount using DPX mountant (SRL, Mumbai, India). All the microphotographs were taken using Olympus CX41 bright field light microscope attached with a Micropublisher (Model-MP3.3) cooled CCD camera (Q-imaging, BC, Canada).

For TEM analysis, control and treated (CuSO₄ and Cu-NPs) testis (n = 5) were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffered (PB) with pH 7.2 for 24 h at 4°C. On following day, the samples were washed for 45 min each with PB-saline (PBS) thrice. Further, it is post-fixed in 1% aqueous osmium textroxide for 2 h and rinsed with deionized water for 6 times each 45 min, dehydrated in series of graded alcohol, infiltrated and embedded in araldite 6005 resin (Spurr, 1969). The samples were then incubated at 80°C for 72 h for complete polymerization. Ultra-thin sections of about 1 μm were made using a glass knife on Ultramicrotome (Leica Ultra cut UCT-GA-D/E-1/00), mounted on Cu grids and stained with saturated aqueous Urenyl acetate and counter stained with Reynolds' lead citrate. TEM microphotographs were obtained with a FEI Technai G2 S-Twin instrument at an acceleration voltage of 200 kV. All the TEM sample processing was carried out at RUSKA Labs, Department of Veterinary Pathology, Acharya N G Ranga Agricultural University, Hyderabad.

T and 11-KT levels in serum were estimated by EIA kit (Cayman, Ann Arbor, MI, USA) as per the manufacturer's protocol. Blood was drawn from dorsal vein of catfish (n = 5) using a disposable syringe to obtain serum by centrifugation at 5000 g for 10 min at 4°C. The supernatant (serum) was carefully pipetted out by leaving the blood pellet and stored in −80°C for hormone analysis. The sensitivity limit for T and 11-KT measurement is 6 and 1.3 pg/ml, respectively as per the kit procedure and the assay validation was done as per Swapna et al. (2006).

Catfish primary testicular cell culture (mixed) was prepared based on the protocol followed by Song and Gutzeit (2003). Adult male catfish was anesthetized using 100 mg/dl MS222 (Sigma) and disinfected with 70% ethanol. Testes were dissected out and washed with sterile PBS. Later, the testes were kept in L-15 medium and chopped into small pieces, and then wash again with sterile PBS for 5 times. The testicular pieces were then incubated with 0.25% collagenase type I and 0.005% DNase at room temperature for 15 min. The mixture was poured into a sterile petriplate and gently pressed with blunt end of syringe to get cell suspension. The suspension was filtered through a 40 μm cell strainer and collected in a centrifuge tube. After centrifugation by 100 g at 4°C, the pellet was resuspended in Dulbecco's modified Eagle's medium (DMEM) and washed three times using DMEM. The viability of cells was assessed using trypan blue staining and about 1 × 10⁶ cells were plated in each well of 24 well culture plate containing DMEM, 10% FBS, 1 × Glutamax and antibiotic and antimycotic and incubated at 30°C in 5% CO₂. The culture was maintained for 24 h and then used for experiment.

TM3 Leydig cells (Mus musculus) purchased from NCCS, Pune, India were used for the assay. The culture was maintained as per the method described above. For cytotoxicity studies, Cu-NPs and CuSO₄ were dissolved at a final concentration of 10, 25, 50, and 75 μg/ml.

Mitochondrial function was analyzed using the CellTiter 96® AQ_ueous One Solution Cell Proliferation Assay (Promega, Alexandria NSW) based on the procedure followed by Braydich-Stolle et al. (2005). After 24 h maintenance, increasing concentration of CuSO₄ and Cu-NPs were added to 96 well cell culture plate. The cultures were kept for 48 h with supplement of 5% CO₂ at 30°C and further 20 μl of Aqueous One Solution reagent was directly added to the culture wells. After 3 h incubation, the color developed was directly measured at 490 nm using a standard microplate reader (Bio-Rad, Ireland). The absorbance measured at 490 nm is the quantity of formation of formazan product which is directly proportional to the number of live cells in culture. The experiment was repeated in triplicate, independently (n = 5). The relative cell viability (%) related to control wells was expressed in percentage relative to untreated cells.

Determination of cellular apoptosis or necrosis level was done using Vybrant Apoptosis Assay kit #4 (Molecular Probes, Eugene, OR) as per the protocol described by manufacturer. The kit supplied with YO-PRO®-1/propidium iodide, green fluorescent YO-PRO®-1 binds specifically to apoptotic cells while red-fluorescent propidium iodide stains necrotic cells. After 48 h treatment with Cu, the cells were washed with PBS and incubated with 2 ml of PBS contains 2 μl of YO-PRO®-1 and propidium iodide and further incubated for 30 min in ice. Following incubation, the cells were analyzed using BD LSRFortessa™ cell analyzer (BD Biosciences, Franklin Lakes, NJ, USA).

The results represented for qPCR and EIA were expressed as mean ± SEM (Standard Error of the Mean). All statistical analyses were done using SigmaPlot 11.0 software (Systat Software Inc., Chicago, USA). Data passed homogeneity and normality tests and compared by one-way analysis of variance (ANOVA) followed by Student-Newman-Keuls' post-hoc test. Statistical analysis was considered significant at the level of P < 0.05.

Characterization of Cu-NPs were done through FESEM, TEM and XRD, and distribution of Cu-NPs were provided (Figures 1A–G). Before sonication, the Cu-NPs had large size to meet the desired properties, hence sonication was performed without interruption for longer period of time (6 h) to reach the desired size and shape of NPs for biological response. Before sonication, the FESEM image of Cu-NPs showed large particle size, ~40–60 nm and the size was not uniform. Agglomerated with particle to particle and in the background tiny particle with island structure was formed which is not enough to reach the goal (Figure 1A). Thus, the phase and spherical morphology formed during sonication at 6 h of time with the same speed and the particles were well controlled in size (15–20 nm), spherical shape and particle uniformity (Figure 1B). Sonication process led to the NP size, agglomeration reduction due to dispersion of coarse agglomeration. The key observation was that the 6 h sonication time is an optimum duration to initiate the controlled size, stability and refine the particle size. TEM images before and after sonication were taken (Figures 1C–E), the particle size reduction was also evident in TEM images. Thus, the toxicity or biological response was experimented using the sample after sonication. In addition, pH and alkalinity of the water sample was unchanged after the addition of Cu-NPs into 50 L water tank during the treatment. FESEM, TEM images are conforming the shape (spherical NP), average particle size of the Cu and conformed as 40–50 nm. X-ray diffraction is one of the important techniques to determine the size of the unit cell of material. X-ray diffraction pattern of the prepared Cu-NP, the sharp and broad crystalline peaks clearly conforming the Cu-NP nature (Figure 1F). The mean Cu-NP size was measured from the full width half maxima of the most intensive peak (111) of Cu. The 2θ values of 43.29, 50.42, 74.12, and 89.92 corresponding to (111), (200), (220), and (311) planes of the face centered cubic (fcc) crystalline cubic structure of metallic Cu. The intensities of all reflections are well matched with standard data (JCPDF File No. 04-0836) and conforming that Cu-NP has high purity and other impurities was not observed. The particle size of Cu NPs were quantitatively evaluated according to Debye-Scherer formula from the broadening of intensive peak (111).

Where D is the average particle size in nanometer, k is the Scherrer constant, β is the full-width half the maximum intensity (FWHM), λ is the X-ray wavelength (λ = 0.154 nm), θ is the Bragg angle. The average particle size was found as 50 nm according to Debye-Scherer formula using imageJ software based on the Cu morphology (Figure 1G). The size average size distribution was conformed as 40–60 nm which is well matched with XRD and TEM particle size analysis.

Expression level of several steroidogenic enzyme (3β-hsd, cyp11a1, 11β-hsd2, 11β-h, and star) and transcription factor genes (wt1, ad4bp/sf-1, dmrt1, sox9a, and gata4) were analyzed and quantified in control and treated (CuSO₄ and Cu-NPs) groups represented by fold increase. In general, expression level of all the steroidogenic enzyme genes analyzed showed significant upregulation in both the treated groups when compared to control (Figure 2A). The expression of 3β-hsd increased significantly to 2.8 and 2.6 (P < 0.01) fold in CuSO₄ and Cu-NPs treated groups, respectively when compared to control. Significant expression of cyp11a1 was found in treated batches of CuSO₄ and Cu-NPs with fold increase of 2.8 and 3.8 (P < 0.01), respectively in comparison with control fish. The expression of 11β-hsd2 also showed significant increase to 3.5 (P < 0.01) and 1.3 (P < 0.05) fold, respectively in CuSO₄ and Cu-NPs treated compared with control. Further, significant increase in the levels of 11β-h to 1.5 (P < 0.05) and 2.6 (P < 0.01) fold, respectively was observed in CuSO₄ and Cu-NPs treated groups in comparison with control. Significant (P < 0.01) increase of star expression to 2.2 and 3.0-fold was evident in CuSO₄ and Cu-NPs treated fish, respectively.

Similarly, the expression analysis of several transcription factor genes showed significant increase in expression when compared to control (Figure 2B). The expression of wt1 increased significantly (P < 0.05) to 1.9 and 1.7-fold in CuSO₄ and Cu-NPs treated groups, respectively when compared to control. Significant (P < 0.05) increase of ad4bp/sf-1 to 1.7 and 1.6-fold while dmrt1 showed significant (P < 0.01) increase to 2.7 and 2.2-fold, respectively in CuSO₄ and Cu-NPs treated fishes than control. Further, the expression of sox9a and gata4 levels were increased significantly (P < 0.05) to 2.0 and 1.6-fold and 1.5 and 1.6-fold in CuSO₄ and Cu-NPs treated fish, respectively when compared to controls.

Serum 11-KT and T levels estimated by EIA were compared between the control and treated groups (Figures 2C,D). The levels of 11-KT in treated groups showed significant increase (P < 0.05; Figure 2C) than the levels of control. Correspondingly, T levels in serum of treated fishes of C. batrachus (Figure 2D) were significantly elevated (P < 0.05) when compared to control.

Control testis of catfish stained with hematoxylin and eosin (Figures 3A–F) displayed normal basal lamina consisting of interstitial/Leydig cells (Figure 3D), spermatogonia with dark nuclei and spermatocytes with large oval nuclei (Figures 3C,E,F). The spermatids and/or sperm were detected in the lumen of testis in control (Figure 3C). In the treated groups, reduction in the size of testicular lumen with disrupted basal lamina was seen (Figures 3G–R). Spermatocytes were completely separated from the basal lamina integration with distinct shape were observed in each lumen of testis (Figures 3G,J,K,M,O,Q). Spermatogonia displayed abnormal structure with signs of disrupted nuclei (Figures 3K,Q) when compared to the control. The Cu-NPs treated group showed disorganization in testis lumen (Figures 3M,N) with irregular and enlarged spermatocytes (Figure 3Q). Interestingly, the interstitial cell layer is not affected as the testicular architecture was not altered, which is even evident at low magnification in the treated groups (Figures 3G,H,M,N). However, both the treated groups showed corroborative effects when exposed to Cu as a soluble metal and NPs.

TEM analysis of control testis showed normal testicular organization with basal lamina consists of interstitial/Leydig or myoid cells and regular shaped spermatocytes (Figures 4A,D,G). Spermatocyte of control testis displayed spherical/oval shaped nuclei with well-defined nuclear membrane which is attached to endoplasmic reticulum (Figure 4G). Cytoplasm of the spermatocyte consists of ribonucleic protein granules. Sertoli cell comprises of thick nucleolus which is present near basal lamina facing toward the testis lumen (Figure 4J). Erythrocyte of control testis showed normal nucleated uniform structure located inside the lumen of capillary blood vessel (Figure 4M).

TEM images from treated testis (both groups) exhibited degeneration of basal lamina with the presence of apoptotic cells (Figures 4B,C). Distinct shaped spermatocytes with degeneration of nuclear membrane were seen in both groups (Figures 4E,F,H,I). Sertoli cells of treated groups showed chromatin marginalization with disintegration of nuclear membrane and electron dense bodies (Figures 4K,L). Treated groups of testis displayed erythrocyte deformability inside the capillary. In addition, the capillary walls showed some damage after the treatment (Figures 4N,O).

Spermatocyte with vacuolated cytoplasm was observed in the treated group (Figure 5A). Interestingly, bio-distribution of Cu-NPs with apoptotic cells were observed in the treated testis (Figures 5B–F).

Relative cell viability decreased with the increasing concentrations of CuSO₄ and Cu-NPs exposure to cell cultures. For catfish primary testicular cells, the LC₅₀ value was calculated at 20.5 and 30.0 μg/ml for Cu-NPs and CuSO₄, respectively (Figure 6Ai). In the case of TM3 Leydig cell culture, the LC₅₀ value was noted at 37.5 and 25.3 μg/ml for Cu-NPs and CuSO₄, respectively (Figure 6Aii). In comparison, the cytotoxic effects were almost similar for both type of cells.

By examining the cell morphology, control groups of both the cultures showed normal adherent nature throughout the experiment period (Figures 6Bi,vi). Upon treatment, the cells exhibited dramatic change in the morphology. In 10 μg/ml treatment group, cells started to show signs of clumping in culture dishes (Figures 6Bii,iv,vii,ix). However, cell toxicity was not evident. Loss of adhesion, cell clumping and shrinkage were more in Cu-NPs and CuSO₄ of 75 μg/ml treated batches (Figures 6Biii,v,viii,x). However, similar results were observed in rest of the batches analyzed (photos not shown). Distinct morphological changes during treatment indicate the occurrence of apoptosis/necrosis.

To further investigate the Cu-induced cellular toxicity, control and treated cells of catfish primary testicular culture (Figure 7A) and TM3 Leydig cell (Figure 7B) were added with YO-PRO-1 and PI dyes and the difference in normal and apoptotic/necrotic cells were analyzed using flow cytometry. Results indicated that percentage of apoptotic/necrotic cells were increased with the increasing concentration of Cu-NPs and CuSO₄ exposure (Figures 7A,B) when compared to the control (low right quadrant). The percentage of cells appeared in the positively stained with YO-PRO-1 which represented early apoptotic cells (lower right quadrant) and cells stained with both YO-PRO-1 and PI (upper right quadrant) which denoted late apoptotic cells were calculated and compared with the control. However, the percentage of dead cell population was evaluated from the upper left quadrant. The percentage of population in each quadrant was given in Supplementary Table 2.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.