2.1. Microscopy Studies
Most of the work on tissue distribution of nanoparticles has been restricted to rodents [16–19]. To date, the only studies about the toxicity of silver [20] and titanium dioxide nanoparticles have been performed on fish [21]. The present study, as far as we know, is the first report describing silicon-based QD distributions in the fish kidney. Previous studies on mice have indicated that silica nanoparticles mostly accumulate in the organs of the reticuloendothelial system, such as the liver and spleen, and also in the kidney [17–19,22].
The IP injected QDs were taken up by blood circulation of C. auratus gibelio and spread in tissues such as the white muscle [14] and kidney.
Due to intrinsic photoluminescence under ultraviolet excitation, silicon-based QDs have been detected in tissue sections by the emission of red light (Figure 1A–D). The fluorescence was not detectable for the control (non-injected and injected with NaCl 7‰) animals (Figure 1A). At 24 h after IP injection, silicon-based QDs were detected mainly in the epithelial cells cytoplasm of the proximal tubule (Figure 1B). At 3 (Figure 1C) and 7 days (Figure 1D) after IP injection cells from proximal tubule, distal tubule and collecting tubule were loaded with silicon-based QDs. The density of silicon-based QDs in the cytoplasm of cells from different nephron segments was higher at 3 and 7 days after IP injection. In the interstitial tissue, fluorescent silicon-based QDs were detected in the macrophages (Figure 1E).
To investigate the toxicity, histological assessment was performed to determine whether silicon-based quantum dots cause kidney damage. The crucian carp trunk kidney is composed of nephrons and collecting ducts surrounded by interstitial tissue (Figure 2A). In the kidney of fish injected with silicon-based QDs alterations were observed, including detached epithelial cells from basal lamina (Figure 2B), dilated tubules and debris in the lumen of tubules (Figure 2C). Besides degenerative processes, at 7 days after IP injection, kidney regeneration was also observed in Carassius auratus gibelio, indicating that silicon-based QDs do not impede the regeneration of tissue. Nephroneogenesis was characterized by the appearance of small, intensely basophilic clusters of cells, frequently near collecting ducts (Figure 2D).
By contrast to marine fish species, the glomerulus of freshwater fish is well vascularized and the glomerular filtration rate is high. Freshwater fish produce large amounts of dilute urine, maintaining the acid-base and electrolyte equilibria and regulating the blood pressure.
However, at the level of the renal corpuscle, in the absence of the glomerular basement membrane and podocytes, the sole dividing barrier between the mesangium and glomerulus is the fenestrated endothelium. Due to the high rate of glomerular filtration, nanoparticles smaller than 10 nm in diameter diffuse and accumulate in the mesangium after they depart from the glomerulus through these pores [23].
Histological examination revealed that the kidney of the crucian carp was able to regenerate after silicon-based QD injury. The formation of clusters of basophilic cells represents the first phase of de novo nephroneogenesis in fish kidney, followed by the development of a cavity which grows to form the renal vesicle and later on glomerulus and Bowman’s vesicle [24]. Previous studies have indicated that chemically-induced nephrotoxicosis in fish is followed by new nephron development, and tubular regeneration can be a good sign of adaptation and recovery [25]. This process was observed in the kidney of many fish species, including goldfish [26], rainbow trout [27], zebrafish [24], tilapia [11] and medaka [28].
The histological studies revealed that in addition to degenerative processes visible at 7 days after IP injection, kidney regeneration could also be observed in Carassius auratus gibelio, indicating that silicon-based QDs do not impede tissue regeneration, i.e., nephroneogenesis.
The detection of QDs in the macrophages might explain the increased CAT and SOD activities after 3 and 7 day exposure; this could be explained by the fact that these cells, when activated, produce oxygen and nitrogen reactive species during phagocytosis [29]. In theory, the contact with silicon QDs could generate the ROS cascade which ensues, either due to the activity of the macrophage NADPH oxidase (NOX) complex, or due to the activation by mechanical stress of NADPH oxidase expressed by renal cells. Taking into account that the greatest quantity of QDs is present in the renal tubular epithelial cells it seems that these could contribute to a larger extent to superoxide anion generation. In the kidney, NOX1 and NOX4 are present [30], but only NOX4 is highly expressed [31].
2.2. Biochemical Studies
2.2.1. Lipid Peroxidation
The MDA level increased by 97% after one day of exposure, reaching the maximum level of 288% higher than the control after the third day, as indicated in Table 1. The doubling of MDA concentration was registered as a consequence of the cascade mechanism of lipid peroxidation. Taking into account that GPX activity decreased after three and seven days, whereas GST activity significantly increased during the same period, it would seem that GST was involved in the decrease of MDA after 7 days following its conjugation with GSH.
The decrease of MDA concentration after 7 days could also be explained by the significant increase of SOD and CAT activities, which are involved in ROS elimination.
2.2.2. GSH Level
GSH plays a central role in antioxidant defense and in the regulation of several signal transduction pathways and metabolic functions [32], and is considered the most important antioxidant in vivo. The GSH concentration (Table 1) decreased by 66%, 60% and 47% respectively after 1, 3 and 7 days of treatment, compared to controls. As these data indicated, the response of the kidney cells to QD exposure initially involved a decrease of the GSH level, as it was probably consumed by indirect chemical reactions, as well as GPX and GST catalyzed reactions that removed the deleterious compounds.
This tri-peptide is synthesized intracellularly from glutamate, cysteine and glycine in the proximal tubular cells and can remain in the cell or be exported in the extracellular space. In the cell, the most important part of GSH is present in the mitochondrial pool, the rest being located in the cytoplasm [33]. In our case, the GSH formation was not sustained by enzymatic reduction catalyzed by GR, but, possibly due to de novo synthesis, induced by lipid peroxidation products [34]. This could be the reason why the GSH concentration increased by 56% after the seventh day, compared to the first day of exposure. Our data indicate that in the first three days of exposure, the decrease of the GSH concentration is in accordance with the rise of MDA level; conversely, the decrease of MDA concentration after the seventh day can also be explained by the increase of the GSH concentration.
2.2.3. Oxidative Protein Alterations
ROS that elude the antioxidant system attack the proteins. Sulfur and aromatic-containing amino acid residues, such as tyrosine, cysteine and methionine are particularly susceptible to oxidative transformation [35]. Advanced oxidation protein products (AOPP) are formed in vivo as a consequence of the exposure of proteins to hypochlorous acid generated by myeloperoxidase, by oxidation of tyrosyl and histidyl residues, as well as by the deamination of lysyl residues [36]. These oxidative markers correlate with the levels of dityrosine [37] and could reflect the QD’s exposed macrophage involvement in the oxidative stress in fish kidney. AOPP concentration (Table 2) increased significantly after 3 days compared to control, whereas after 1 and 7 days, a non-significant difference of approximately 10% was observed.
In the case of cysteine, oxidation leads to the formation of disulfide bonds, mixed disulfides (e.g., with glutathione) and thiyl radicals [38]. The protein thiol level (Table 2) decreased significantly by 28%, 33% and 43% respectively after 1, 3 and 7 days of treatment, compared to control.
Our results suggest that the dityrosyl cross-links appeared due to the ROS attack on proteins could be lowered by the intervention of protein cysteine residues.
Carbonyl derivatives of proteins are formed by ROS-mediated oxidation of the side chains of some amino acid residues (especially proline, arginine, lysine, threonine), as well as by a reaction with glucoxidation and lipid peroxidation products [39]. Reactive carbonyl compounds derived from lipids and carbohydrates react with proteins resulting in a “carbonyl stress” which generates kidney pathologies [40]. The protein reactive carbonyl group level (Table 2) was significantly up-regulated after 3 and 7 days of exposure by 122% and 115% respectively compared to control levels. In our case, the excess MDA could react mainly with amino groups of lysine residues, producing MDA-modified protein adducts.
Protein oxidation has been associated with renal damage. The radicals generated on tyrosine could migrate to the cysteine residues, forming a thiyl radical on a local cysteine residue, which can generate a very reactive sulfenic acid; the latter might be converted back to the reduced state or, alternatively, further oxidized, depending on the presence of additional redox active molecules [41]. The downregulation of protein thiols could be due to intra-protein or inter-protein disulfide formation that affects the protein tridimensional structure and function. But, in some cases, these alterations do not cause long time changes in protein function, because tioredoxin or glutaredoxin rapidly catalyze the reduction of disulfide bonds.
However, considering the mild histological modifications, we could conclude that silicon-based QDs induced the formation of oxidized proteins to a moderate degree, which could be degraded by the proteosomal system, as was proved by previous studies [42], avoiding the cross-linking aggregation and installment of possible pathological situations.
2.2.4. The Antioxidant Scavenging Enzymes
The SOD activity was unchanged in the first three days of exposure, but in the seventh day, an increase of 56.5% was observed, whereas CAT activities increased in a time-dependent manner, by 25% and 27% after 3 and 7 days of exposure, respectively (Figure 3).
2.2.5. Enzymes Involved in Glutathione Metabolism and in Generating Reducing Equivalents
After the first day of exposure, the increase of SOD and CAT activities suggests an insignificant increase of superoxide level and its downstream metabolite, hydrogen peroxide. When the concentration of hydrogen peroxide increased after the third and seventh days, catalase started to catalyze its degradation. After the seventh day, SOD activity increased significantly and the dismutation of superoxide occurred. Our results are in agreement with those of Oruc et al. (2004) who studied the SOD activity changes in kidneys of Oreochromis niloticus and Cyprinus carpio exposed to pesticides [43]. Nevertheless, in the presence of various doses of cadmium, a very toxic xenobiotic, this activity significantly decreased in the kidney of Clarias gariepinus [44]. Taking into account that catalase specific activity changed by a maximum of 27% after seven days of exposure, it seems that the generation of superoxide anions was at a moderate level.
The activities of GR and GPX increased after one day of exposure by 14% and 16% respectively, but decreased dramatically after 3 and 7 days by approximately 50% compared to control (Figure 4). The decrease of GR specific activity after the third day of exposure to QDs generated a decreased formation of reduced glutathione, and probably a lower level of protein glutathionylation. Additionally, the GST activity increased by 29.32%, 47.48% and 87.64% after one, 3 and 7 days of treatment, respectively (Figure 4).
The profile of GPX specific activity is inversely proportional with that of catalase. One possibility would be that, in the first day, the very low quantity of hydrogen peroxide was decomposed by GPX. At longer exposure times, this enzyme was not efficient in both hydrogen peroxide as well as lipid detoxification of peroxidation products. The variation of GST activity increased continuously, suggesting a significant role in the detoxification process in fish kidney, with some GST isoforms having a selenium independent GPx activity. This result is in agreement with previous findings [45].
The specific activity of G6PDH increased significantly by 49% and 94% after 3 and 7 days of treatment, respectively, compared to controls (Figure 4). This enzyme catalyzes the oxidative branch of the pentose phosphate pathway, generating NADPH, an electron donor in reductive biosynthesis, which is used for GSH regeneration [46]. According to our results, it appeared that NADPH was used for GSH regeneration at a reduced rate and only on the first day. It is also possible that this cofactor could interact directly with free radicals and have, after seven days, a pro-oxidant action being aimed at superoxide generation in the reactions catalyzed by NOXs.
2.3. Heat Shock Protein Hsp 70 Evaluation
Fish kidney tissue as well as liver and gill are sensitive to Hsp response [47]. The Hsp 70 family is highly conserved and generally acts under stress conditions [48]. Our results have shown that Hsp 70 protein expression in the kidney of IP injected fish with Si/SiO2 QDs increased in a time-dependent manner: 1.33-, 1.65- and 1.88-fold, after 1, 3 and 7 days, respectively, compared to controls (Figure 5).
These proteins assist, in an ATP-dependent manner, the folding of nascent proteins, the refolding of damaged proteins and mediate the degradation of irreversibly denatured proteins [49]; in addition, they also protect cells against caspase dependent and independent apoptosis [50]. It was also suggested that Hsp 70 forms mixed disulfides with other cytoplasmic proteins, conferring in this way a protection to redox sensitive proteins from potentially irreversible disulfide bond formation [51].
It seems that following the perturbation of the intracellular environment due to the exposure of kidney to silicon-based QDs, the resulting biochemical effects were not completely scavenged by detoxification processes and modifications at protein levels occurred. Hsp70 probably targeted the denatured proteins for degradation, which ensured kidney resistance, as previously thought [52].
The increase in Hsp 70 expression in fish exposed to different types of xenobiotics was also shown in other studies [53,54]. Our results are in agreement with recent literature [55] which revealed that IP injection of titanium dioxide nanoparticles in mice was associated with very light lesions in the kidneys.