Taurine Supplementation Alleviates Puromycin Aminonucleoside Damage by Modulating Endoplasmic Reticulum Stress and Mitochondrial-Related Apoptosis in Rat Kidney

Taurine (TAU) is a sulfur-containing beta amino acid that is not involved in protein composition and anabolism, conditionally essential in mammals provided through diet. Growing evidence supports a protective role of TAU supply in osmoregulation, calcium flux, and reduction of inflammation and oxidant damage in renal diseases like diabetes. Endoplasmic reticulum (ER) stress, due to abnormal proteostasis, is a contributor to nephrotic syndrome and related renal damage. Here, we investigated the effect of dietary TAU (1.5% in drinking water for 15 days) in an established rat model that mimics human minimal change nephrosis, consisting of a single puromycin aminonucleoside (PAN) injection (intraperitoneally 15 mg/100 g body weight), with sacrifice after eight days. TAU limited proteinuria and podocytes foot processes effacement, and balanced slit diaphragm nephrin and glomerular claudin 1 expressions. In cortical proximal tubules, TAU improved lysosomal density, ER perimeter, restored proper ER-mitochondria tethering and mitochondrial cristae, and decreased inflammation. Remarkably, TAU downregulated glomerular ER stress markers (GRP78, GRP94), pro-apoptotic C/EBP homologous protein, activated caspase 3, tubular caspase1, and mitochondrial chaperone GRP75, but maintained anti-apoptotic HSP25. In conclusion, TAU, by targeting upstream ER stress separate from mitochondria dysfunctions at crucial renal sites, might be a promising dietary supplement in the treatment of the drug-resistant nephrotic syndrome.


Introduction
Nephrotic syndrome, which is one of the most widespread renal disorder in humans [1], includes a wide spectrum of diseases, with an idiopathic, toxic, or immunologic etiology, which is classified into four histological variants: minimal change disease (MCD), focal segmental glomerulosclerosis like heart failure, diabetes, and muscle and inflammatory diseases, in rodent models but also in humans, has been recently reported by Schaffer and Kim [43]. TAU has also been rediscovered for the treatment of renal infections, nephrolithiasis and diabetes [44][45][46][47], but to the best of our knowledge, its role in acute PAN nephrosis is limited to a biochemical study by Venkatesan et al. [48].
Therefore, we completed this study to explore if ER stress in parallel to mitochondrial damage in podocytes and epithelial tubular cells can be potential targets for TAU in experimental PAN nephrosis in rats.
Our results outlined a novel therapeutic role for TAU dietary supplementation against ER stress and for mitochondria-driven events, like apoptosis and inflammation, in experimental PAN nephrotic syndrome coupled to massive proteinuria.

Animal Treatment
Twenty two-month-old male Sprague Dawley (SD) rats were purchased from Harlan Laboratories Srl (San Pietro al Natisone, Udine, Italy). Animal care was performed according to the Italian Ministry of Health to comply with the commonly-accepted international guidelines known as "3Rs" European Community Council Directive 86/609/EEC. SD rats were divided into the following four groups: (1) n = 4 SD rats fed standard rodent diet and tap water (control group); (2) n = 4 SD rats fed standard diet but drinking taurine (TAU) dissolved 1.5% in tap water for 15 days (TAU group) [49]; (3) n = 6 SD rats fed normal standard diet and injected with PAN (intraperitoneally injected; 15 mg/hg) (PAN group) [50] and (4) n = 6 SD rats fed standard diet injected with PAN, but drinking TAU one week before and another week after PAN injection (PAN plus TAU group). The rationale for two weeks of TAU supplementation was obtained from the study of Moloney et al. [51], where this duration of treatment reversed early diabetic damage in patients. The power analysis to define the number of animal for each group was evaluated by the Student t test with a significance of 0.05. All animals were sacrificed by cervical dislocation on day 8 after PAN injection, and the kidneys were removed for histopathology, immunohistochemical, and electron microscopy analysis. A similar approach was successful to best analyze stress response in the kidney in a previous study [52]. Proteinuria was measured at day 8 after PAN injection in 16 h collected urine samples during overnight placement in metabolic cages. Urinary parameters were assessed using Multistix 10SG Reagent Strips and the animals were considered to have a severe proteinuria following three consecutive readings >300 mg/dL [53].

Histopathology and Immunohistochemistry
For histopathology, the kidneys were fixed for immersion in 4% phosphate buffered paraformaldehyde for 48 h and were embedded in paraffin wax, as previously reported [54]. Paraffin sections (5 µm-thick) were prepared for PAS staining and immunohistochemistry. For quantitative analysis of glomerular tuft area, PAS stained renal sections were observed at a final magnification of 400× using a light microscope (Olympus, Hamburg, Germany) and at least 25 glomeruli for four different slides were randomly analyzed for each experimental group. The glomerular area was manually traced and estimated using imaging software (Image Pro Plus, Milan, Italy) by two investigators in a blinded manner. On the same section, also renal tubular-interstitial damage was also assessed on 10 non-overlapping fields at 200× and four different slides for each experimental group were randomly chosen. Normal cortical proximal tubules presented an epithelium with a single layer of cubic cells showing a purple PAS-positive brush border (consisting of microvilli during ultrastructural analysis) and a central nucleus, but these findings were lost in damaged tubules. Tubular atrophy, epithelial disruption, and protein casts were semi-quantitatively graded, as follows: grade 0, normal; grade 1, lesion area <25% of the field; grade 2, lesion area between 25% and 50% of the field; grade 3, lesion area between 51% and 75% of the field; and, grade 4, >75% lesion area. Tubular injury score represented the average of the grades that are assigned to the fields analyzed. The main changes examined include tubular atrophy, tubular casts, sloughing of tubular epithelial cells, and the presence of inflammatory cells, as previously reported by He et al. [55]. Serial paraffin sections were used for immunohistochemistry, dewaxed in xylene, rehydrated and subjected to antigen retrieval in 0.01 M citrate buffer (pH 6.0) in a microwave oven, and then treated in 3% hydrogen peroxide to block the endogenous peroxidase background. Slides were treated with normal serum of the same species producing the secondary antibody (diluted 1:5) for 1 h at room temperature in the dark and subsequently incubated, overnight at 4 • C in humidified chambers, with: rabbit polyclonal anti-GRP75 (diluted 1:200); rat monoclonal anti-GRP94 antibody (diluted 1:200); goat polyclonal anti-CHOP, anti-HSP25 and anti-nephrin antibodies (diluted 1:50-1:100-1:50 respectively), rabbit polyclonal anti-GRP78, anti-caspase3 and anti-caspase 1-p10 antibodies (diluted 1:250 and 1:50 respectively), mouse monoclonal anti-CD68/ED1 antibody (diluted 1:200); and, rabbit polyclonal 4HNE(diluted 1:400). Viewing was performed using the avidin-biotin-peroxidase complex (ABC-peroxidase kit, Vector Labs, Burlingame, CA, USA), according to manufacturer's instructions. The peroxidase reaction was developed using 3 -3 -diminobenzidine tetrahydrochloride (Sigma Aldrich, St. Louis, MO, USA) as the substrate and hydrogen peroxide as the catalyst for 7 min in the dark. To assess the specificity of each immunoreaction, we omitted primary antibodies and substituted those with tris buffered solution. All sections were counterstained by hematoxylin, dehydrated and mounted for light microscopy analysis (Olympus, Hamburg, Germany). Then, an immunopositive area was quantified at 400× using computerized image analysis software (Image Pro Plus, Milan, Italy), as follows: brown peroxidase-positive area for each field was manually circled, then automatically converted into a gray-scale pixel to obtain a total value as arbitrary value (AU). Twenty randomly chosen non-overlapping fields for each immunostaining were estimated. In particular, for the quantification of nuclear markers, nuclei that were further stained with haematoxylin appeared blue-violet and were not estimated by the software set on brown color.

Transmission Electron Microscopy
The second kidney was fragmented in small pieces and was fixed for immersion in 2.5% glutaraldehyde in cacodylate buffer 0.1 M for 3 h then post-fixed in 2% osmium tetroxide in the same buffer for 1 h at 4 • C. Dehydration in progressive ethanol concentrations, propylene oxide, and embedding in Epon 812 mixture were performed, as previously described [56].
Semithin sections (1 µm thick) obtained with an UltraCut E ultra-microtome were then stained by methylene blue-azure II, and then observed under a light microscope to verify the presence of lysosomes at 100×. Subsequently, thin sections (80 nm thick) collected on cupper grids were observed under a transmission electron microscope Tecnai G2 Spirit (FEI Company, Eindhoven, The Netherland) at 80 kV. Morphometric computerized analysis of foot processes of podocytes was performed at 15,000×, but the ER perimeter was estimated at 26,000×, and ER-mitochondria distance at 135,000× by two observers blinded of the treatment, as previously described [57].

Statistical Analysis
All data are expressed as mean ± standard deviation and analysis of the statistical significance was obtained using one-way analysis of variance (ANOVA) corrected by Bonferroni test to compare the variability of a group with all the other experimental groups and p < 0.05 being considered to be significant. All of the experiments were performed in triplicate, and data were collected and analyzed by Origin Pro 9.1 software (Origin Lab Corporation, Northampton, MA, USA).

Metabolical and Histopathological Effects of TAU in Nephrotic Rats
Similar metabolic and histological data were recorded in the control and TAU only treated rats, so they are labelled the "control group" in the next described evaluation. Metabolic and urinary parameters are summarized in Table 1. PAN-treated rats developed massive proteinuria (1200 ± 130 mg/dL) compared with the control group in which proteinuria was undetectable. However, in the TAU supplemented PAN group, proteinuria significantly decreased to a low range (80 ± 22 mg/dL). Urinary volume tripled in PAN nephritic rats, but significantly decreased in the TAU supplemented PAN group. Urinary leukocytes were absent in the control group, abundant in PAN group (125/dL), and decreased after PAN plus TAU treatment (70/dL) (all p < 0.05). To further corroborate the role of TAU preconditioning in the PAN model, the renal expression of the lipid peroxidation marker 4HNE was analyzed and quantified ( Figure S1). PAS staining did not reveal any damage to glomeruli, and similarly no differences in glomerular area were observed in experimental groups (data not shown). These findings confirmed the hallmark of "minimal change nephrosis" (MCD), i.e., absence of glomerular damage under light microscopy analysis. So, we analyzed the filtration barrier by transmission electron microscopy, the elective method that best defines proteinuria-induced changes in MCD patients [58]. In control rats, regular podocytes' foot processes adhered to the basal membrane and well-structured slight diaphragms were visualized ( Figure 1A,E). In contrast, in PAN-injected rats, prominent fusion of pedicels and tight junctions between pedicels, together with limited slight diaphragms, were observed ( Figure 1B,F). In contrast, after TAU pre-treatment, more regular thin foot processes and recovered filtration slits in the glomerular barrier were evident ( Figure 1C,G). Estimation of foot processes width on electron micrographs demonstrated a significant enlargement in PAN-injected rats (750 nm) when compared to the control group (200 nm), but a decreased width in the PAN supplemented with TAU group ( Figure 1D). These data indicated that TAU intake partially improved podocytes effacement in the PAN-treated group. Slight diaphragms quantification significantly decreased by 42% in the PAN group (p < 0.05), but increased by 21% in PAN plus TAU supplemented rats ( Figure 1H). These values indicated that TAU pre-treatment positively influenced the distribution of glomerular filtration sites.
Nutrients 2018, 10, x FOR PEER REVIEW 6 of 17 rats (750 nm) when compared to the control group (200 nm), but a decreased width in the PAN supplemented with TAU group ( Figure 1D). These data indicated that TAU intake partially improved podocytes effacement in the PAN-treated group. Slight diaphragms quantification significantly decreased by 42% in the PAN group (p < 0.05), but increased by 21% in PAN plus TAU supplemented rats ( Figure 1H). These values indicated that TAU pre-treatment positively influenced the distribution of glomerular filtration sites.

TAU Changes Podocytes Markers Distributions in Nephrotic Rats
To confirm podocyte impairment using transmission electron microscopy, we explored the glomerular localization of the cytoskeletal markers nephrin and claudin 1 to highlight the extent of PAN nephrosis in different groups. Nephrin, which is a structural marker of podocyte foot processes and the main component of a normal slight diaphragm, was highly expressed in control rats (Figure 2A), diminished in PAN nephrotic rats ( Figure 2B), was partially restored in the PAN group receiving TAU supplementation ( Figure 2C). Estimation of nephrin staining intensity, as arbitrary units, was high in the control group, lower in the PAN group (p < 0.05), and then increased in the PAN group receiving TAU ( Figure 2D).
In contrast, claudin 1, which is a tight junctional marker that is localized in the external epithelial layer of Bowman capsule, was barely detectable in the control group ( Figure 2E), markedly expressed in PAN group ( Figure 2F), and significantly reduced at the glomerular level in the PAN supplemented TAU group ( Figure 2G). The immunoreaction was also quantified and the relative immunopositivity evaluation is summarized in Figure 2H.

TAU Changes Podocytes Markers Distributions in Nephrotic Rats
To confirm podocyte impairment using transmission electron microscopy, we explored the glomerular localization of the cytoskeletal markers nephrin and claudin 1 to highlight the extent of PAN nephrosis in different groups. Nephrin, which is a structural marker of podocyte foot processes and the main component of a normal slight diaphragm, was highly expressed in control rats (Figure 2A), diminished in PAN nephrotic rats ( Figure 2B), was partially restored in the PAN group receiving TAU supplementation ( Figure 2C). Estimation of nephrin staining intensity, as arbitrary units, was high in the control group, lower in the PAN group (p < 0.05), and then increased in the PAN group receiving TAU ( Figure 2D).
In contrast, claudin 1, which is a tight junctional marker that is localized in the external epithelial layer of Bowman capsule, was barely detectable in the control group ( Figure 2E), markedly expressed in PAN group ( Figure 2F), and significantly reduced at the glomerular level in the PAN supplemented TAU group ( Figure 2G). The immunoreaction was also quantified and the relative immunopositivity evaluation is summarized in Figure 2H.

TAU Reduces Glomerular ER Chaperones and Apoptosis
ER stress response was assessed by the immunohistochemistry analysis of the glomerular expression of ER resident chaperones in all of the experimental groups. With this approach, the master ER chaperone GRP78, was weakly or moderately detected in glomeruli and surrounding cortical distal tubules in control rats ( Figure 3A). In contrast, in PAN-treated nephrotic rats, a marked signal was detected in glomeruli and proximal tubules also at the nuclear level ( Figure 3B), but decreased in the proximal tubules in TAU pre-treatment ( Figure 3C). Estimation of the GRP78 signal doubled in the PAN nephrotic group and significantly decreased by 60% in PAN-injected rats supplemented with TAU ( Figure 3D). To confirm the evidence of ER stress, GRP94 localization was further assessed. GRP94 immunostaining, barely detectable in control rats ( Figure 3E), was strongly expressed in glomeruli and proximal tubules of PAN-treated rats ( Figure 3F), but decreased in the PAN group that was supplemented with TAU ( Figure 3G). Quantification of GRP94 immunoreaction showed a decrease in PAN nephrotic rats that were supplemented with TAU ( Figure 3H). Furthermore, to better examine the influence of ER stress on apoptosis, we examined the glomerular expression of CHOP. Remarkably, CHOP was undetectable in the control group ( Figure 3I), evident in podocyte nuclei in the PAN-injected rats ( Figure 3L), and decreased after TAU supplementation ( Figure 3M). Quantitative analysis of CHOP positive nuclei, expressed as percentage, was 4% in the control group, about 28% in the PAN nephrotic rats, and 16% in the PAN group that was treated with TAU ( Figure 3N). These data clearly demonstrated that TAU limited glomerular ER stress and the pro-apoptotic cascade.

TAU Reduces Glomerular ER Chaperones and Apoptosis
ER stress response was assessed by the immunohistochemistry analysis of the glomerular expression of ER resident chaperones in all of the experimental groups. With this approach, the master ER chaperone GRP78, was weakly or moderately detected in glomeruli and surrounding cortical distal tubules in control rats ( Figure 3A). In contrast, in PAN-treated nephrotic rats, a marked signal was detected in glomeruli and proximal tubules also at the nuclear level ( Figure 3B), but decreased in the proximal tubules in TAU pre-treatment ( Figure 3C). Estimation of the GRP78 signal doubled in the PAN nephrotic group and significantly decreased by 60% in PAN-injected rats supplemented with TAU ( Figure 3D). To confirm the evidence of ER stress, GRP94 localization was further assessed. GRP94 immunostaining, barely detectable in control rats ( Figure 3E), was strongly expressed in glomeruli and proximal tubules of PAN-treated rats ( Figure 3F), but decreased in the PAN group that was supplemented with TAU ( Figure 3G). Quantification of GRP94 immunoreaction showed a decrease in PAN nephrotic rats that were supplemented with TAU ( Figure 3H). Furthermore, to better examine the influence of ER stress on apoptosis, we examined the glomerular expression of CHOP. Remarkably, CHOP was undetectable in the control group ( Figure 3I), evident in podocyte nuclei in the PAN-injected rats ( Figure 3L), and decreased after TAU supplementation ( Figure 3M). Quantitative analysis of CHOP positive nuclei, expressed as percentage, was 4% in the control group, about 28% in the PAN nephrotic rats, and 16% in the PAN group that was treated with TAU ( Figure 3N). These data clearly demonstrated that TAU limited glomerular ER stress and the pro-apoptotic cascade.

TAU Decreases Tubular Apoptosis and Oxidative Damage
While considering the crucial role of the first tract of the nephron in the reabsorption of excessive urinary proteins, we extended immunohistochemical analysis of ER-driven apoptosis to proximal cortical tubules. Occasional pro-apoptotic CHOP in the nuclei of the proximal tubules of control rats ( Figure 4A) was clearly expressed in disrupted proximal tubules in PAN-injected rats on day eight ( Figure 4B) and decreased in the PAN plus TAU group ( Figure 4C). Accordingly, the percentage of nuclear positivity shifted from 8% in control rats to 36% in the PAN group and to 18% in TAU supplemented group ( Figure 4D).
Caspase 3 is the crucial marker of the final step in the apoptotic cascade that must be activated by proteolysis [59]. So, we evaluated the presence of cleaved caspase 3 in cortical renal tubules in our model. Control rats did not express cleaved caspase 3 in the kidney ( Figure 4E); in contrast, in the PAN group, an intense nuclear staining was observed in the proximal tubules, but only occasionally in the glomeruli on day eight ( Figure 4F). After TAU pre-treatment, cleaved caspase 3 immunostaining was limited to a few nuclei in the proximal tubules ( Figure 4G). Quantitative analysis showed that nuclear positivity for the cleaved activated caspase 3 was very low in control rats, 26 AU in the PAN group, but decreased in rats that were receiving PAN and TAU (p < 0.05) ( Figure 4H).
To best correlate the extent of apoptosis to oxidative damage, we visualized the renal distribution of HSP25, an inducible actin chaperone able to oppose ATP-depletion in the kidney. Nuclear HSP25 expression was weak in the proximal tubules of the control rats ( Figure 4I), and intense in the PAN-treated group localized in the proximal and distal tubules ( Figure 4L). In contrast, in PAN nephrotic rats supplemented with TAU, the HSP25 signal shifted to the brush border in proximal tubules and to the cytoplasm of the distal tubules ( Figure 4M). Quantitative

TAU Decreases Tubular Apoptosis and Oxidative Damage
While considering the crucial role of the first tract of the nephron in the reabsorption of excessive urinary proteins, we extended immunohistochemical analysis of ER-driven apoptosis to proximal cortical tubules. Occasional pro-apoptotic CHOP in the nuclei of the proximal tubules of control rats ( Figure 4A) was clearly expressed in disrupted proximal tubules in PAN-injected rats on day eight ( Figure 4B) and decreased in the PAN plus TAU group ( Figure 4C). Accordingly, the percentage of nuclear positivity shifted from 8% in control rats to 36% in the PAN group and to 18% in TAU supplemented group ( Figure 4D). Caspase 3 is the crucial marker of the final step in the apoptotic cascade that must be activated by proteolysis [59]. So, we evaluated the presence of cleaved caspase 3 in cortical renal tubules in our model. Control rats did not express cleaved caspase 3 in the kidney ( Figure 4E); in contrast, in the PAN group, an intense nuclear staining was observed in the proximal tubules, but only occasionally in the glomeruli on day eight ( Figure 4F). After TAU pre-treatment, cleaved caspase 3 immunostaining was limited to a few nuclei in the proximal tubules ( Figure 4G). Quantitative analysis showed that nuclear positivity for the cleaved activated caspase 3 was very low in control rats, 26 AU in the PAN group, but decreased in rats that were receiving PAN and TAU (p < 0.05) ( Figure 4H).
To best correlate the extent of apoptosis to oxidative damage, we visualized the renal distribution of HSP25, an inducible actin chaperone able to oppose ATP-depletion in the kidney. Nuclear HSP25 expression was weak in the proximal tubules of the control rats ( Figure 4I), and intense in the PAN-treated group localized in the proximal and distal tubules ( Figure 4L). In contrast, in PAN nephrotic rats supplemented with TAU, the HSP25 signal shifted to the brush border in proximal Nutrients 2018, 10, 689 9 of 17 tubules and to the cytoplasm of the distal tubules ( Figure 4M). Quantitative analysis of HSP25 abundance confirmed the highest value in PAN-injected rats, which decreased by 50% after TAU supplementation to a value that was not that different from the control (p < 0.05) ( Figure 4N). These results indicated that dietary TAU supplementation reduced tubular apoptosis and oxidative damage in proximal tubules.
Nutrients 2018, 10, x FOR PEER REVIEW 9 of 17 analysis of HSP25 abundance confirmed the highest value in PAN-injected rats, which decreased by 50% after TAU supplementation to a value that was not that different from the control (p < 0.05) ( Figure 4N). These results indicated that dietary TAU supplementation reduced tubular apoptosis and oxidative damage in proximal tubules.

TAU Improves Tubular Inflammation and Mitochondrial Stress in PAN Nephrosis
In addition to apoptosis and ER stress, tubular inflammation and the interstitial recruitment of macrophages from circulation are also involved in the renal lesions that are induced by acute proteinuria.
In our nephritic model, the histological analysis of PAS stained cortical tubules revealed a regular purple brush border in the control rats ( Figure 5A), PAS stained proteinaceous casts in the lumen of the PAN nephrotic group (Figure 5B), and enlarged dilated proximal tubules in the PAN group receiving TAU ( Figure 5C). Semi-quantitative estimation of proximal tubular damage, expressed as a "tubular score", was 0.5 in control rats, significantly increased to 3.5 in nephrotic rats, and decreased to 2 AU in PAN group pre-treated with TAU ( Figure 5D).
Moreover, we decided to examine the tubular presence of activated caspase 1, a component of NLRP3 inflammasome, which is the multiprotein complex sensitive to ER damage. The staining of activated caspase 1, i.e., the subunit caspase 1-p10, was barely detectable in the proximal tubules in control rats ( Figure 5E), intense in the proximal tubules in the PAN-injected group (Figure 5F), and weaker in the PAN rats that received TAU ( Figure 5G). Quantitative analysis revealed that caspase1 p10 intensity was very low in the control group, dramatically increased 14 fold in PAN, but significantly decreased in the PAN plus TAU group, p < 0.05 ( Figure 5H). Figure S1 shows the evaluation of the CD68/ED1 macrophage infiltration in the tubule interstitial area.

TAU Improves Tubular Inflammation and Mitochondrial Stress in PAN Nephrosis
In addition to apoptosis and ER stress, tubular inflammation and the interstitial recruitment of macrophages from circulation are also involved in the renal lesions that are induced by acute proteinuria. In our nephritic model, the histological analysis of PAS stained cortical tubules revealed a regular purple brush border in the control rats ( Figure 5A), PAS stained proteinaceous casts in the lumen of the PAN nephrotic group (Figure 5B), and enlarged dilated proximal tubules in the PAN group receiving TAU ( Figure 5C). Semi-quantitative estimation of proximal tubular damage, expressed as a "tubular score", was 0.5 in control rats, significantly increased to 3.5 in nephrotic rats, and decreased to 2 AU in PAN group pre-treated with TAU ( Figure 5D).
Moreover, we decided to examine the tubular presence of activated caspase 1, a component of NLRP3 inflammasome, which is the multiprotein complex sensitive to ER damage. The staining of activated caspase 1, i.e., the subunit caspase 1-p10, was barely detectable in the proximal tubules in control rats ( Figure 5E), intense in the proximal tubules in the PAN-injected group (Figure 5F), and weaker in the PAN rats that received TAU ( Figure 5G). Quantitative analysis revealed that caspase1 p10 intensity was very low in the control group, dramatically increased 14 fold in PAN, but significantly decreased in the PAN plus TAU group, p < 0.05 ( Figure 5H). Figure S1 shows the evaluation of the CD68/ED1 macrophage infiltration in the tubule interstitial area.
Finally, to better understand the effect of different treatments on mitochondrial homeostasis, we analyzed the expression of the mitochondrial chaperone GRP75.
In control rats, GRP75 immunoreaction was faint in the cytoplasm of the proximal tubules ( Figure 5I), strong in the damaged proximal tubules of the PAN nephrotic rats (Figure 5L), and limited to few epithelial cells in the PAN rats supplemented with TAU ( Figure 5M). Estimation of GRP75 immunostaining tripled in the PAN group as compared with the control group, but halved in PAN rats that were supplemented with TAU (value 18 AU; p < 0.05) ( Figure 5N). Therefore, all of these data indicated that TAU pre-treatment was effective against "sterile" inflammation and oxidative damage by modulating caspase 1 activation and GRP75 abundance in proximal tubules. Finally, to better understand the effect of different treatments on mitochondrial homeostasis, we analyzed the expression of the mitochondrial chaperone GRP75. In control rats, GRP75 immunoreaction was faint in the cytoplasm of the proximal tubules ( Figure 5I), strong in the damaged proximal tubules of the PAN nephrotic rats (Figure 5L), and limited to few epithelial cells in the PAN rats supplemented with TAU ( Figure 5M). Estimation of GRP75 immunostaining tripled in the PAN group as compared with the control group, but halved in PAN rats that were supplemented with TAU (value 18 AU; p < 0.05) ( Figure 5N). Therefore, all of these data indicated that TAU pre-treatment was effective against "sterile" inflammation and oxidative damage by modulating caspase 1 activation and GRP75 abundance in proximal tubules.

Effect of TAU on Mitochondria and Rough ER in Nephrotic Proximal Tubules
To understand whether TAU alleviation of tubular apoptosis and altered oxidative status that we observed via immunohistochemistry may be correlated to the morphology of the ER, mitochondria, or other organelles, we examined the proximal tubules using electron microscopy. At low magnification, we observed scattered dense mitochondria and scarce lysosomes in the control group ( Figure 6A), excessive perinuclear lysosomes, and multiple rough ER (RER) cisternae in the PAN group ( Figure 6B), and elongated mitochondria in the PAN plus TAU supplemented group ( Figure 6C). At high magnification, we focused on mitochondria and RER features. In the control proximal tubules, mitochondria displayed parallel cristae and a regular distance from RER ( Figure  6D). In contrast, in the PAN nephrotic group, the mitochondria contained a few short cristae, a disrupted outer membrane, and were fused in multiple points with RER ( Figure 6E). Remarkably, in the PAN rats that were supplemented with TAU, mitochondria presented restored cristae, well-defined outer membrane, and regular RER distance that was similar to the control rats ( Figure  6F).

Effect of TAU on Mitochondria and Rough ER in Nephrotic Proximal Tubules
To understand whether TAU alleviation of tubular apoptosis and altered oxidative status that we observed via immunohistochemistry may be correlated to the morphology of the ER, mitochondria, or other organelles, we examined the proximal tubules using electron microscopy. At low magnification, we observed scattered dense mitochondria and scarce lysosomes in the control group ( Figure 6A), excessive perinuclear lysosomes, and multiple rough ER (RER) cisternae in the PAN group ( Figure 6B), and elongated mitochondria in the PAN plus TAU supplemented group ( Figure 6C). At high magnification, we focused on mitochondria and RER features. In the control proximal tubules, mitochondria displayed parallel cristae and a regular distance from RER ( Figure 6D). In contrast, in the PAN nephrotic group, the mitochondria contained a few short cristae, a disrupted outer membrane, and were fused in multiple points with RER ( Figure 6E). Remarkably, in the PAN rats that were supplemented with TAU, mitochondria presented restored cristae, well-defined outer membrane, and regular RER distance that was similar to the control rats ( Figure 6F). PAN-treated rats, ER perimeter significantly increased up to 120 µm/field in proximal tubules when compared to the control group (40 µm/field). After TAU pre-treatment, the ER perimeter was similar to that of the control rats (44 µm/field) ( Figure 6H). Furthermore, when considering the crucial role of ER and mitochondria juxtaposition in regulating ER signaling [60], we estimated ER-mitochondria tethering in our experimental model. The distance between organelles in control rats was less than 50 nm, significantly decreased to 26 nm in the PAN-injected group, but was almost similar to the control value in TAU supplemented PAN rats (43 nm) (all p < 0.05) ( Figure 6I). These results indicated that TAU inhibited mitochondrial alterations and it contributed to the maintenance of a proper distance between mitochondria and RER in proximal tubules.

Discussion
Proteinuria, which is the presence of proteins in urine, is not only a diagnostic marker of glomerular damage but also a strong contributor to tubular decline and the progression of chronic kidney disease [61,62]. ER stress and UPR are fundamental mechanisms that are involved in the pathogenesis of proteinuric renal failure [25,63]. However, ER modulation by drugs may be of great To best corroborate the above findings, lysosomal density, RER perimeter, and the distance between ER and the outer mitochondrial membrane were measured. In proximal tubules of control rats, lysosomal density was 45 (expressed as number per 100 µm 2 ), almost doubled (80) in PAN injected rats, but decreased to the control level after TAU supplementation (48) (Figure 6G). In PAN-treated rats, ER perimeter significantly increased up to 120 µm/field in proximal tubules when compared to the control group (40 µm/field). After TAU pre-treatment, the ER perimeter was similar to that of the control rats (44 µm/field) ( Figure 6H). Furthermore, when considering the crucial role of ER and mitochondria juxtaposition in regulating ER signaling [60], we estimated ER-mitochondria tethering in our experimental model. The distance between organelles in control rats was less than 50 nm, significantly decreased to 26 nm in the PAN-injected group, but was almost similar to the control value in TAU supplemented PAN rats (43 nm) (all p < 0.05) ( Figure 6I). These results indicated that TAU inhibited mitochondrial alterations and it contributed to the maintenance of a proper distance between mitochondria and RER in proximal tubules.

Discussion
Proteinuria, which is the presence of proteins in urine, is not only a diagnostic marker of glomerular damage but also a strong contributor to tubular decline and the progression of chronic kidney disease [61,62]. ER stress and UPR are fundamental mechanisms that are involved in the pathogenesis of proteinuric renal failure [25,63]. However, ER modulation by drugs may be of great value for a nephrologist to develop effective personalized ER-targeted treatments in unresponsive or recurrent nephrotic syndrome, instead of conventional glucocorticoids or immunosuppressive drugs [7,64].
One major point addressed in this study was that TAU alleviated experimental PAN nephrosis and proteinuria in rats by reducing ER stress and HSPs response.
Indeed, in rats, a single-dose PAN injection induced massive proteinuria on day eight, which was associated with ultrastructural glomerular and tubular alterations, mimicking human MCD [48]. In rats, TAU preconditioning, one week before and another after PAN injection, decreased proteinuria and renal damage. Through ultrastructural analysis, we clearly observed reduced podocyte effacement in the TAU supplemented nephrotic animals, but an enhanced slit diaphragm number and distribution. Furthermore, in agreement with morphological findings, TAU restored glomerular nephrin, which is the main slit diaphragm component, and minimized claudin 1 in our nephrotic model.
Notably, claudin 1 in nephrotic glomeruli was correlated with the presence of tight junctions, and it indicated a dysfunctional filtration barrier able to promote proteinuria in line with the findings that were reported by Gong et al. [16]. However, in this study, we demonstrated that TAU dietary supplementation restored proper expressions of these two markers in the PAN nephrotic model.
In vitro studies on malignant cells reported that PAN induced ER stress by translocon opening in the ER membrane, attracting proteins in the lumen and releasing calcium ions in the cytoplasm [65]. Moreover, also in PAN-treated podocytes, abnormal calcium flux, calpain, and calcineurin were described [18]. GRP78 and GRP94 expression was detected in our study in PAN nephrotic rats, which agrees with the findings that were reported by Inagi et al. [19]. However, TAU preconditioning inhibited ER stress in nephrotic glomeruli and proximal tubules in line with previous studies on cisplatin or starvation-triggered ER damage [66,67].
Another interesting novel finding of this study is the attenuation of ER stress driven by pro-apoptotic CHOP and HSP25 in glomeruli and proximal tubules in PAN nephrotic rats pre-treated with TAU. Although apoptosis and oxidative damage that was linked to ER stress in kidneys treated with PAN have been well established [68,69], the beneficial role of TAU in nephritic kidney is still unclear. However, TAU hampered ER and mitochondrial damage in neurodegeneration and stroke [70].
As podocytes are the main target of PAN toxin, proximal tubular cells may also be involved in the reabsorption of excessive urinary proteins from the ultrafiltrate by unique transporters in their apical site, called the brush border [38]. However, lysosomal cathepsins are also directly involved in acute kidney disease progression [71]. Therefore, we analyzed cortical proximal tubules in PAN nephrotic model while focusing on ER stress, apoptosis, and subcellular organelles changes in the lysosomes and mitochondria. Intriguingly, our findings on the maintenance of cortical tubular dilatations in PAN pre-treated with TAU mice agreed with those of Mozaffari et al. [72] who reported an apparent detrimental role of TAU in rat kidney after ischemia-reperfusion.
Notably, we demonstrated that TAU antagonized tubular ER stress and apoptosis in nephrotic animals. Moreover, TAU improved abnormal lysosomal accumulation, ER mitochondria tethering, and mitochondrial integrity reducing HSP25 and GRP75 expressions. These findings highlighted the preventive effect of TAU in subcellular sites essential for oxidative metabolism and calcium flux, and that HSPs may be therapeutic targets [73,74]. In addition, the recovery of proper ER-mitochondria tethering in TAU supplemented nephrotic rats suggested that TAU modulated bioenergetics [75] or worked as a mitochondrial buffer of calcium in the nervous system, according to El Idrissi [76]. Furthermore, in MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes), which is a human mitochondrial disease with stroke-like episodes, TAU supplementation recovered mitochondrial respiratory function revealing a promising therapeutic role [43]. Finally, we showed that oral TAU reduced tubular inflammation and caspase 1 activation in PAN nephrosis. Recent studies reported that NOD-like receptor protein 3 (NLRP3) inflammasome was overloaded with calreticulin in the ER in proximal tubular cells that were challenged with albumin [77]. ER stress triggered a sterile inflammatory cascade, activating NLRP3 and caspase 1 in proteinuric patients [78]. Consequently, we hypothesized that dietary TAU supplementation, improving ER homeostasis, may be able to limit the inflammatory reaction in proximal renal tubules, even if further experiments are needed to verify this hypothesis.

Conclusions
Taken together, our results suggest that dietary TAU significantly preserves ER and mitochondria features, promoting glomerular and tubular recovery and reducing experimental puromycin-induced damage in rat kidney. Intriguingly, TAU decreased ER stress, and in parallel, mitochondria-related apoptosis, drives mitochondria and ER expression of specific stress markers, hampering puromycin nephrotic damage (Figure 7). This study reveals that CHOP may be an upstream therapeutic TAU target able to drive downstream apoptosis triggered by ER and mitochondrial UPR in the puromycin model [79]. Finally, this translational study implies that TAU might act as a potential promising dietary additive in drug resistant-MCD patients. We only described the effective role of TAU in rat; new studies on this potentiality should be completed in the future. were challenged with albumin [77]. ER stress triggered a sterile inflammatory cascade, activating NLRP3 and caspase 1 in proteinuric patients [78]. Consequently, we hypothesized that dietary TAU supplementation, improving ER homeostasis, may be able to limit the inflammatory reaction in proximal renal tubules, even if further experiments are needed to verify this hypothesis.

Conclusions
Taken together, our results suggest that dietary TAU significantly preserves ER and mitochondria features, promoting glomerular and tubular recovery and reducing experimental puromycin-induced damage in rat kidney. Intriguingly, TAU decreased ER stress, and in parallel, mitochondria-related apoptosis, drives mitochondria and ER expression of specific stress markers, hampering puromycin nephrotic damage (Figure 7). This study reveals that CHOP may be an upstream therapeutic TAU target able to drive downstream apoptosis triggered by ER and mitochondrial UPR in the puromycin model [79]. Finally, this translational study implies that TAU might act as a potential promising dietary additive in drug resistant-MCD patients. We only described the effective role of TAU in rat; new studies on this potentiality should be completed in the future.