Renal oxidative stress and renal CD8⁺ T-cell infiltration in mercuric chloride-induced nephropathy in rats: role of angiotensin II

Abstract

Mercuric chloride (HgCl₂) induces kidney damage, in part, through oxidative stress. A role for angiotensin II (Ang II) in pro-inflammatory events in a model of acute HgCl₂-induced nephropathy was reported. Ang II is a potent oxidative stress inducer; however, its role in oxidative/anti-oxidative events in HgCl₂-induced nephropathy remains unknown. The aim of this study was to determine the role of Ang II in the oxidative stress and renal infiltration of CD8⁺ T-cells after an acute HgCl₂ intoxication. Three groups of Sprague Dawley rats were treated with a single subcutaneous dose of 2.5 mg/kg HgCl₂: for 3 days prior to and for 4 days after that injection, rats in one group received Losartan (30 mg/kg), in another group Enalapril (30 mg/kg) or normal saline in the last group. Two other groups of drug-treated rats received saline in place of HgCl₂. A final group of rats received saline in place of HgCl₂ and the test drugs. All treatments were via gastric gavage. At 96 h after the vehicle/HgCl₂ injection, blood and kidney samples were harvested. Renal sections were homogenized for measures of malondialdehyde (MDA), reduced glutathione (GSH) and catalase activity. Frozen sections were studied for the presence of superoxide anion () and CD8⁺ T-cells. HgCl₂-treated rats had increased interstitial and tubular expression of , high levels of MDA, normal catalase activity and GSH content, increased levels of interstitial CD8⁺ T-cells and an increased percentage of necrotic tubules. Anti-Ang II treatments diminished the HgCl₂-induced increases in interstitial , CD8⁺ T-cells and tubular damage and increased catalase and GSH expression above that due to HgCl₂ alone; the HgCl₂-induced high MDA levels were unaffected by the drugs. These data provide new information regarding the potential role of Ang II in the oxidative stress and renal CD8⁺ T-cell infiltration that occur during HgCl₂ nephropathy.

Keywords:

• Angiotensin II
• Enalapril
• kidney
• Losartan
• mercuric chloride
• oxidative stress

Introduction

Heavy metals like mercury are potential threats to humans (Hu 2000). Mercury is excreted almost exclusively via the kidney; thus, renal exposure to mercury is unavoidable and subsequent toxicity commonly evolves in the kidney (Madden & Fowler 2000). Renal mercury chloride (HgCl₂)-induced damage has been studied using experimental models of nephropathy, showing that the toxic effects of HgCl₂ in the kidney are related to inflammatory events and/or oxidative stress (Ghielli et al. 2000; Aleo et al. 2002; Shimojo et al. 2002; de Greef et al. 2003; Stacchiotti et al. 2003).

One of the major mechanisms associated with mercury toxicity has been attributed to generation of reactive species that, in turn, leads to an imbalance between pro- and anti-oxidant (reducing) elements in the body. This shift in balance toward the pro-oxidants is termed oxidative stress. The oxidative stress-mediated toxicity induced by mercury is based upon a strong affinity for nucleophiles (like -SH and -SeH moieties; Farina et al. 2013), increased generation of oxygen and nitrogen free radicals and alterations in enzymatic (catalase, glutathione peroxidase and superoxide dismutase) and non-enzymatic (e.g. carotenoids, flavonoids, glutathione, Vitamin C, Vitamin E, etc.) anti-oxidants, each leading to an increased incidence in lipid peroxidation and subsequent damage to DNA, lipids and proteins (Sharma et al., 2014).

Previous reports have suggested that angiotensin II (Ang II) plays an important role in triggering inflammatory and oxidative renal injury and, further, that anti-Ang II treatments reduced those events in several experimental models of renal failure (Benigni et al. 2010; van der Meer et al. 2010). The association of Ang II with the pathogenesis of HgCl₂-induced nephronpathy has been previously documented (Yanagisawa et al. 1998; Peña et al. 2012). In this regard, a vasopressor role for glomerular Ang II during the early phase of HgCl₂-induced nephropathy was reported (Yanagisawa et al. 1998). Increased expressions of intercellular adhesion molecule-1 (ICAM-1), ED-1 (monocytes/macrophages) and inducible nitric oxide synthase (iNOS) mediated by Ang II have also been documented (Peña et al. 2012). However, the role of this hormone in the oxidative stress and in the infiltration of cytotoxic CD8⁺ T-cells (an oxidative radical producer; Chou et al. 2014) during HgCl₂-induced nephropathy remains unknown.

Because Ang II can be chemotactic for T-cells (Silva-Filho et al., 2011) and this hormone can mediate oxidative stress (Toda et al. 2007; Garrido & Griendling 2009; Harrison & Gongora 2009; Wysocki et al. 2014), it could be expected that Ang II would modulate those parameters during HgCl₂-induced nephropathy. Ang II acts through two receptor sub-types, AT1 and AT2 receptors. Most of the classic physiological effects of Ang II are mediated by the AT1 receptor, including induction of oxidative stress (Abdi & Johns 1997; Garrido & Griendling 2009; Benigni et al. 2010). In addition, increased production of Ang II by the effect of Ang I-converting enzyme (ACE) is related to enhanced Ang II-related actions (Abdi & Johns 1997; Benigni et al. 2010).

With the advent of drugs such as Losartan that specifically blocks the Ang II AT-1 receptor and Enalapril that blocks ACE activity, it is possible to dissect the physiological roles of the renin-angiotensin system (Abdi & Johns 1997; Donmez et al. 2002; Benigni et al. 2010; Muñoz et al. 2011; Peña et al., 2013). Therefore, the aim of this study was to determine the role of Ang II in the renal oxidative stress and CD8 cell infiltration in rats treated with HgCl₂ using the angiotensin II receptor (AT-1) blocker Losartan and the ACE inhibitor Enalapril.

Materials and methods

Animals and experimental design

Sprague-Dawley rats (male, 200–250 g) were obtained from the Instituto de Investigaciones Cientificas (Caracas, Venezuela) and housed at the Instituto de Investigaciones Clinicas “Dr. Americo Negrette”, Zulia University Care Services Center according to the guidelines by the Institutional Animal Care and Use Committee. The facilities were maintained at 25 °C with a 12-h light:dark cycle. All rats had ad libitum access to standard chow (Ratarina, Valencia, Venezuela) and filtered water throughout the study. The Committee of Bioethics and Biosecurity of FONACIT (Caracas) and of Bioethics of the Medical School (Universidad del Zulia, Maracaibo, Venezuela) approved all the studies here. Since hormone fluctuations had a potential to influence the experiments, only male rats were used.

For these studies, the rats were randomly allocated into six groups (n = 10/group). These groups were designated as Group I (HgCl₂ only-treated), Group II (HgCl₂ and Losartan-treated), Group III (HgCl₂ and Enalapril-treated), Group IV (Losartan only-treated), Group V (Enalapril only-treated) and Group VI (untreated). Rats in Groups I, II and III were to be injected subcutaneously with a single 500 μl dose (2.5 mg/kg) of HgCl₂ (99.5% purity; Sigma, St. Louis, MO) in normal (0.9%) saline. Rats in Groups II and IV were to be treated daily by gastric gavage with 30 mg Losartan/kg (Cozaar, Merck & Co., Whitehouse Station, NJ) in normal saline (Abdi & Johns 1997). Rats in Groups III and V were to be treated daily by gavage with 30 mg Enalapril/kg (Renitec, Merck & Co.) in normal saline (Guo et al. 2008). Rats with no HgCl₂ treatment (Groups IV, V and VI) received, respectively, Enalapril, Losartan or normal saline only daily by gavage (500 μl). The Anti-Ang II doses used here were previously shown to be capable of diminishing pro-inflammatory events in different inflammatory models (Vargas et al. 2012; Muñoz et al. 2011; Peña et al. 2012).

Each of the drug/water treatments occurred for 3 days prior to the HgCl₂ injection and thereafter (up to 4 days post-HgCl₂ injection). This particular timing of treatments was necessary because mercury enters cells rapidly after administration (Hultman & Enestrom 1986) and renal damage can be detected early. Therefore, when the anti-Ang II treatment is given 3 days after HgCl₂ injection, it may be acting after the initiation of renal cellular injury. Muñoz et al. (2011) showed that use of this anti-Ang II pre-treatment (3 days before and then throughout the experimental period) protocol diminished the expression of Ang II and pro-inflammatory molecules in an Adriamycin-induced nephrosis model. The specific dose of HgCl₂ (2.5 mg/kg) and evolution time (96 h) used here were selected based on the fact that: (1) doses ranging from 1.0–3.5 mg HgCl₂/kg induce dose-dependent alterations in the cytoplasm and nucleus of proximal tubular cells in rats (Stacchiotti et al. 2003); and (2) important inflammatory events are seen 2–6 days post-exposure (Ghielli et al. 2000). Accordingly, kidneys of the HgCl₂-treated animals here were analyzed on Day 4 (i.e. 96 h) post-HgCl₂ injection.

All rats (experimental and controls) at 72 h post-HgCl₂ injection were placed in individual metabolic cages for 24 h to collect urine samples and to determine urinary protein concentration (using sulfosalicylic acid method; Tietz 1970). Then, 24 h later, the rats were anesthetized (ether) and blood samples collected via abdominal aortic puncture to determine blood creatinine by a direct colorimetric method in an Astra 4 Model analyzer (Beckman Instruments Inc., Fullerton, CA). At necropsy, the kidneys were perfused with saline solution. A section of the left kidney was removed, embedded in Tissue Tek (Miles Inc., Kankakee, IL), frozen in acetone-dry ice solution and stored at −70 °C until used in immunofluorescence studies and for determination of superoxide anion content. The remaining kidney was homogenized for measures of local oxidative stress.

Oxidative stress parameters

Thiobarbituric reactive substances (TBARS)

Kidney malondialdehyde (MDA) content was determined using a thiobarbituric acid assay (Ohkawa et al. 1979). In brief, perfused kidney sections were homogenized in a buffer containing 100 mM KCl and 3 mM EDTA. After centrifugation (600 × g, 15 min) a volume of supernatant was removed for measure of protein content (using bicinchoninic acid [BCA]; Wiehelman et al. 1988). From the remaining supernatant, 400 μl was added to 200 μl of reaction solution (8.1% SDS, 20% acetic acid and 0.8% thiobarbituric acid [pH 3.5]) (Sigma). The mixture was then heated to 95 °C for 60 min and the chromogenic substrate then extracted into an organic phase with butanol/pyridine (15:1, v/v). The absorbance of the organic layer was then assessed at 532 nm in the Astra 4 Model analyzer spectrophotometer (Beckman Instruments). TBARS levels in samples were extrapolated from a standard curve generated in parallel using bis-dimethyl acetal standard (Sigma). Results were expressed as nmol TBARS/mg protein.

Catalase activity

Catalase activity in the homogenates prepared above was determined using the method described by Aebi (1982). Catalyzed decomposition of the H₂O₂ would be followed by monitoring decreases in absorbance at 240 nm (A₂₄₀; ɛ = 0.0039 mmol⁻¹•mm⁻¹); the change in absorbance (ΔA₂₄₀)/unit time was then used to reflect catalase activity. In the reaction, 25 μl homogenate was combined with 725 μl of a 7.7 mM hydrogen peroxide (H₂O₂) in 50 mM potassium phosphate buffer (pH 7.0) solution and the reaction then continuously monitored at 240 nm in the spectrophotometer. Due to abnormal kinetics inherent to catalase, it was not possible to define catalase units (U) according to standard recommendations. Thus, a different definition of activity, i.e. one based on a rate constant of a first-order reaction (k) as recommended by Aebi (1982) was applied. All results were then expressed in terms of k/mg homogenate protein.

Reduced glutathione (GSH)

The GSH content in the homogenates prepared above was determined as described by Beutler et al. (1963). In this case, aliquots of the isolated supernatants were treated with m-phosphoric acid to precipitate proteins and then centrifuged at 3000 x g for 10 min. The GSH in each supernatant was then measured by adding 500 μl to 2 mL of a solution of 0.04% v) 5,5/v) 5,5-dithio-bis-2-nitrobenzoic acid (DTNB; Sigma) in 0.2 M phosphate buffer (pH 7.4). The absorbance of each sample at 410 nm was then measured in the spectrophotometer. Levels of GSH in samples were extrapolated from a standard curve generated in parallel using purified GSH. All results were expressed as nmol GSH/mg renal protein.

Histological identification of superoxide anion

Superoxide anion () production at the cellular level in frozen renal sections was determined using a method described in Briggs et al. (1986). In brief, 10-μm thick sections from each rat were incubated for 60 min at 37 °C in a solution containing 0.05 M Tris-HCl buffer (pH 7.6) containing 0.05 M 3,3’-diaminobenzidine (DAB), 8% (w/v) NiCl, 0.5 M MnCl₂ and 10% (w/v) NaN₃ (Sigma). The sections were then fixed in 10% formalin and counterstained with 1% methyl green solution. Using a light microscope, positive interstitial or tubular cells were identified and counted; levels were expressed as number of cells/mm² in 10 randomly-selected fields.

Histopathological studies

The left kidney was processed for light microscopy. Tissue samples were fixed in 10% formalin and embedded in paraffin; subsequently, 4-μm thick sections were prepared and then stained with periodic acid-Schiff base (PAS) and hematoxylin. Each section was then examined using an Axioscope light microscope (Zeiss, Oberkochen, Germany) and pathological alterations were quantified. Analyses were performed blinded to sample identity. The extension of tubular injury was assessed as focal tubular necrosis and complete tubular necrosis. At least 100 S1–S2 and S3 proximal tubular segments were analyzed and reported as a percentage of tubular segments showing given histological damage.

Immunofluorescence studies

Perfused kidneys were used to determine CD8 expression in acetone-fixed 4-μm frozen sections by indirect immunofluorescence. For the study here, a murine monoclonal antibody against rat CD8 (Chemicon International, Inc, Temecula, CA) was used at manufacturer-recommended dilutions. Isotype-matched monoclonal antibody was used as a negative control. A secondary antibody, FITC conjugated anti-mouse IgG (Fab′₂) (Pierce Biotechnology, Inc., Rockford, IL) was then used to localize the mouse monoclonal antibody in the tissues. After washing away unbound antibodies, the sections were then mounted in a solution of p-phenylene-diamine (in PBS-glycerol) and observed using an Axioskop microscope-epifluorescent system (Zeiss, Oberkochen, Germany). Sections of rat lymphoid nodes were processed as positive controls. Levels of positive cells in the tissues were expressed as number of interstitial cells/mm² or cells/glomerular cross-section (gcs) in 10 randomly-selected fields. All microscopic analyses were done in a blinded manner.

Statistical analysis

Data were expressed as mean ± SD. Data analysis was performed using a non-parametric analysis of variance (ANOVA: Kruskal-Wallis Test) and a Dunn’s multiple comparisons test. Pearson’s correlation was used to analyze associations between variables. All data were analyzed using GraphPad Instat (GraphPad Instat software Inc. La Jolla, CA). A p value < 0.05 was considered significant.

Results

Superoxide anion production and lipid peroxidation

Oxygen species (such as superoxide anion, ) and lipid peroxidation (MDA) were examined in the kidney 4 days after HgCl₂ intoxication with/without Losartan and Enalapril co-treatments. The data showed that HgCl₂ increased interstitial and tubular expressions of relative to levels in the naïve control rats (Figures 1 and 2). This change was reflected as interstitial and tubular cells with a strong dark precipitate in the cytoplasm (Figure 2). Treatment with Losartan or Enalapril diminished the significant increase in interstitial expression (Figure 1a), but failed to diminish the significant tubular expression (Figure 1b). The HgCl₂ also caused significant renal tissue lipid peroxidation as evidenced by increases in MDA content relative to levels in the naïve control rats; neither anti-Ang II treatment decreased the induced increase in MDA levels (Figure 3).

Figure 1. Losartan or Enalapril effects on renal superoxide anion () production in HgCl₂-induced nephropathy. Increased expression of (a) interstitial and (b) tubular in tissues from all HgCl₂-treated rats was observed. Losartan and Enalapril treatments diminished only the induced increase in interstitial expression (to levels non-significantly higher than in Group VI control rat tissues: untreated, saline gavaged). In (a) *value is significantly different from those of all other groups of rats shown (p < 0.05). In (b) *of groups shown, value significantly different from that of Group VI control rats only (p < 0.05). Values from rats that received only Losartan- or Enalapril-only were not significantly different from the Group VI rats and so are not presented. Values shown are mean ± SD (n = 10/group). Data analysis was done using a non-parametric ANOVA (Kruskal-Wallis Test) and a Dunn’s multiple comparisons test.

Figure 2. Expression of cellular in renal tissues from HgCl₂-treated rats. (a) High numbers of -positive cells are seen in tubules (arrows) and interstitium (cross arrow) of HgCl₂ only-treated rats. (b) Losartan treatment diminished the number of -positive cells (arrows) relative to that in HgCl₂ only-treated rats. Rats in Groups IV, V and VI presented with normal tissues and so micrographs are not shown. Representative micrographs are presented. Original magnification = 400×.

Figure 3. Effects of Losartan or Enalapril on renal thiobarbituric reactive substance (TBARS) production during HgCl₂-induced nephropathy. Increased renal levels of TBARS were found in all HgCl_2-treated rats relative to in Group VI (untreated, saline gavaged) control rat tissues. Losartan and Enalapril were not capable of reducing the increase in peroxidation. *Value significantly different from Group VI control rats (p < 0.05). Values from rats that received only Losartan- or Enalapril-only were not significantly different from the Group VI rats and so are not presented. Values shown are mean ± SD (n = 10/group). Data analysis was performed using a non-parametric ANOVA (Kruskal-Wallis Test) and a Dunn’s multiple comparisons test.

Anti-oxidant activities

Any shift in host anti-oxidant response reflected through measures of catalase activity and GSH content was not observed in the mercury-treated rats, the values of catalase activity (Figure 4a) and GSH content (Figure 4b) in these hosts remained at normal levels. Total anti-oxidant levels were significantly increased with both Losartan and Enalapril treatment as compared to in hosts that only received HgCl₂ alone. While both drugs seemed to have similar effectiveness in increasing total anti-oxidant content (Figure 4), neither reduced the lipid peroxidation in the tissues that were noted above (re: Figure 3).

Figure 4. Losartan or Enalapril effects on renal catalase activity and reduced glutathione (GSH) content during HgCl₂-induced nephropathy. HgCl₂ itself did not modify (a) renal catalase activity or (b) GSH content relative to Group VI control rat tissue values (untreated, saline gavaged). Anti-Ang II treatments significantly increased both parameters in HgCl₂-treated hosts. *Value significantly different from Group VI control and from HgCl₂-only rats (p < 0.05). Values from rats that received only Losartan- or Enalapril-only were not significantly different from the Group VI rats and so are not presented. Values shown are mean ± SD (n = 10/group). Data analysis was performed using a non-parametric ANOVA (Kruskal-Wallis Test) and a Dunn’s multiple comparisons test.

Renal CD8 infiltration

To investigate whether HgCl₂ intoxication was capable of inducing renal infiltration of CD8⁺ T-cells, renal sections underwent immunofluorescence analyses using a monoclonal antibody against rat CD8. The data show there was a differential expression of CD8⁺ T-cells in the kidney. In this regard, a low number of CD8⁺ T-cells was seen in the glomerulus (Figure 5a); however, a high number of these cells was found in the interstitium (Figure 5b). Losartan or Enalapril treatments prevented (diminished with similar effectiveness) the enhanced CD8⁺ T-cell expression in both the interstitial and glomerular compartments (Figure 5). A representative photo of the immunofluorescence findings is presented in Figure 6.

Figure 5. Losartan or Enalapril effects on renal CD8⁺ cell infiltration during HgCl₂-induced nephropathy. (a) Significantly lower numbers of glomerular CD8⁺ cells in tissues from all HgCl₂-treated rats relative to in Group VI control rat tissues (untreated, saline gavaged) were found. (b) Significantly higher numbers of interstitial CD8⁺ cells in tissues from HgCl₂ only-treated rats relative to levels in Group VI control tissues. At both sites, use of Losartan and Enalapril reduced CD8⁺ cell levels relative to those in the HgCl₂ only rats. *Value significantly different from Group VI control rats (p < 0.05 in (a) and p < 0.01 in (b)). In (a) values from the HgCl₂-LOS and HgCl₂-ENA rats were significantly different from HgCl₂-only and from the control rats (p < 0.05). In (b) **value significantly different from HgCl₂-only rats (p < 0.05) only. Values from rats that received only Losartan- or Enalapril-only were not significantly different from the Group VI rats and so are not presented. All values shown are mean ± SD (n = 10/group). Data analysis was performed using a non-parametric ANOVA (Kruskal-Wallis Test) and a Dunn’s multiple comparisons test.

Figure 6. Indirect immunofluorescence analysis of renal CD8⁺ lymphocytes in HgCl₂-induced nephropathy. (a) An increased presence of interstitial CD8⁺ cells (arrows) in tissue from HgCl₂ only-treated rats. (b) Co-treatment with Losartan led to a decrease in numbers of these cells (arrows) relative to that in HgCl₂ only-treated rats. Rats in Groups IV, V and VI presented with normal tissues and so micrographs are not shown. Representative micrographs are presented. Original magnification = 400×.

Tubular damage

An increased percentage of necrotic tubules (partial or total necrosis) in HgCl₂-treated animals was observed. Losartan or Enalapril treatments reduced the percentages (mean ± SD) of necrotic tubules (i.e. Control: 1.86 ± 1.30; HgCl₂: 13.72 ± 4.55*; Losartan-HgCl₂: 4.15 ± 1.39; Enalapril-HgCl₂: 3.93 ± 1.20; *p < 0.01 vs all groups). Correlation analysis showed a significant association between CD8⁺ cell infiltration and the percentage of necrotic tubules in the Losartan- and Enalapril-treated animals (Figure 7). However, no association between tubular superoxide anion expression and tubular damage was observed.

Figure 7. Correlation between renal CD8⁺ T-cell infiltration and percentage of tubules with necrosis. Significant positive correlation between CD8⁺ cell infiltration and tubular necrosis was observed. Treatments with (a) Losartan or (b) Enalapril decreased levels of infiltrating cells and tubules with necrosis.

The findings in this study were not accompanied by altered renal functions, since serum creatinine and urinary protein values remained at normal levels (Figure 8). No significant differences between the non-HgCl₂-treated rats and controls were observed (Table 1).

Figure 8. Losartan and Enalapril effects on renal function in HgCl₂-induced nephropathy. Serum creatinine levels and urinary protein content remained similar in all HgCl₂-treated rats compared to values in Group VI control rats (untreated, saline gavaged). Values from rats that received Losartan or Enalapril only were not significantly different from the Group VI rats and so are not presented. Values shown are mean ± SD (n = 10/group). Data analysis was performed using a non-parametric ANOVA (Kruskal-Wallis Test) and a Dunn’s multiple comparisons test.

Table 1. Values from untreated and Losartan- or Enalapril-treated controls.

Parameters	Untreated	Losartan treated	Enalapril treated
Interstitial (cells/mm^2)	4.16 ± 1.82	4.90 ± 2.76	3.34 ± 2.30
Tubular (cells/mm^2)	0.64 ± 0.88	1.06 ± 0.86	0.82 ± 0.72
Glomerular CD8 (cells/gcs)	1.80 ± 1.64	1.02 ± 0.77	0.74 ± 0.54
Interstitial CD8 (cells/mm^2)	17.28 ± 8.90	14.89 ± 10.10	16.59 ± 9.20
Catalase (k/mg)	2.23 ± 0.67	3.03 ± 1.69	2.94 ± 0.92
TBARS (nmol/mg)	0.13 ± 0.09	0.15 ± 0.10	0.12 ± 0.05
GSH (nmol/mg)	49.72 ± 14.18	42.51 ± 20.15	44.45 ± 16.32

Controls represent non-HgCl₂-dosed animals treated or untreated with anti-Angiotensin II drugs. No significant differences between the controls were observed.

, superoxide anion; TBARS, thiobarbituric reactive substances; GSH, reduced glutathione; CD8, CD8⁺ T-lymphocyte.

Data analysis was performed using a non-parametric ANOVA (Kruskal-Wallis Test) and a Dunn’s multiple comparisons test.

In the analyses of CD8⁺ cell presence in the tissues analyzed, the values obtained were: Control glomerular CD8: 1.80 ± 1.64; Losartan control glomerular CD8: 1.02 ± 0.77; Enalapril control glomerular CD8: 0.74 ± 0.54 (all three groups in cells/gcs); Control interstitial CD8: 17.28 ± 8.90; Losartan control interstitial CD8: 14.89 ± 10.10; and Enalapril control interstitial CD8: 16. 59 ± 9.20 (all three groups in mean cell/mm² [± SD]).

With regard to the presence of catalase (k/mg), TBARS (nmol/mg) and GSH (nmol/mg) in the analyzed tissues, the values obtained here were (mean ± SD): Control catalase: 2.23 ± 0.67; Losartan control catalase: 3.03 ± 1.69; Enalapril control catalase: 2.94 ± 0.92; Control TBARS: 0.13 ± 0.09; Losartan control TBARS: 0.15 ± 0.10; Enalapril control TBARS: 0.12 ± 0.05; Control GSH: 49.72 ± 14.18; Losartan control GSH: 42.51 ± 20.15; and Enalapril control GSH: 44.45 ± 16.32.

Discussion

One of the major mechanisms associated with mercury toxicity is the generation of reactive species that develop, in part, due to an imbalance between pro-oxidant elements and anti-oxidant (reducing) elements in the body. In this process, a shift to the former is termed as oxidative stress. Mercuric chloride (HgCl₂) is nephrotoxic in rats and affects the pars recta (S3 segment) of the proximal tubule (Stacchiotti et al. 2003), inducing a necrosis that seems to be mediated by oxidative damage (Aleo et al. 2002; Shimojo et al. 2002). The oxidative stress induced by HgCl₂ is mediated by a strong affinity for nucleophiles (such as -SH and -SeH; Farina et al. 2013), by induction of oxygen/nitrogen free radicals and by alterations in the overall anti-oxidant response which involves enzymatic and non-enzymatic alterations and results in damage to DNA, proteins and lipids (i.e. malondialdehyde/lipid peroxidation) (Sharma et al., 2014).

Under the current experimental conditions (2.5 mg HgCl₂/kg and analysis at 96 h post-exposure), only increased renal and MDA levels (accompanied by normal expression of catalase and reduced glutathione) were observed, suggesting an oxidative stress that was probably mediated by an absent/poor anti-oxidant response. Despite the anti-oxidant effects of the anti-Ang II treatments, i.e. decreasing expression and increasing catalase activity and reduced glutathione levels, both Losartan and Enalapril were not capable of reducing lipid peroxidation (MDA content). At this point there is no clear explanation for this result; however, similar results have been previously reported. In this regard, increases in blood pressure, plasma creatinine and MDA were found in a 4/6 kidney ablation rat model (Krivosikova et al. 2001). In that study, Losartan reduced blood pressure and creatinine levels, but, MDA levels remained unaltered suggesting other mechanisms of lipid peroxidation different from those mediated by Ang II.

In the present study, tissue anti-oxidant responses (catalase activity, reduced glutathione content) were increased in the HgCl₂-treated rats that also received anti-Ang II drugs, suggesting a blocking effect of Ang II on the activity/production of those agents. In this regard, Ang II was previously seen to cause decreased expression and activity of catalase in adventitial fibroblasts, an outcome probably mediated by activation of ERK1/2; this effect was abolished by losartan (Yang et al. 2011). In addition, during experimental congestive heart failure after myocardial infarction, decreases in myocardial catalase activity have been reported; losartan treatment improved the activity of this enzyme (Khaper & Singal 2001). The observed increased levels of reduced glutathione after Ang II blocking could also be related to an inhibitory effect of Ang II on this molecule. In this regard, Losartan and Enalapril therapies have been shown to cause increases in levels of plasma glutathione in hypertension patients (Donmez et al. 2002).

As shown in the current experimental model, there is likely a role for Ang II in HgCl₂-induced oxidative stress. In a previous report, increased renal expression of Ang II in rats treated with HgCl₂ was noted (Peña et al. 2012). Increased levels of endogenous Ang II promote local tissue damage by several mechanisms, including oxidative stress (Wysocki et al. 2014), mainly through activation of NADPH oxidase and increased formation of (Toda et al. 2007; Garrido & Griendling 2009; Harrison & Gongora 2009; Wysocki et al. 2014). Since Losartan and Enalapril treatment diminished the high levels of renal found in the present study, a role for induced NADPH oxidase mediated by Ang II could be expected. Indeed, blocking of AT1 receptor by Losartan reduced NADPH oxidase activity in several experimental models (Toda et al. 2007; Wysocki et al. 2014). In conjunction with any enhanced production caused by Ang II, HgCl₂ could concurrently be decreasing local superoxide dismutase (SOD) activity (Jo et al. 2004), leading to overall increases in renal content. These data support the possibility of a dual mechanism involving actions of Ang II and HgCl₂ to explain the increases in production. In this regard, anti-Ang II treatments here failed to decrease oxidative stress in the tubules; however, they did have this effect in the interstitial areas. These data suggest two different oxidative stress inducers in the kidney, one acting on/in tubules and the other in the interstitium. Since anti-Ang II treatment failed to reduce tubular oxidative stress, an HgCl₂-mediated effect was expected. These data also suggested there were mechanisms unrelated (tubular) and related (interstitial) to Ang II; these differing pathways could help explain the differential findings in the interstitium and tubules in the present study.

Other potentially reactive species could also be being generated, i.e. via spontaneous dismutation or enzymatic catalysis, hydrogen peroxide (H₂O₂) (Rahal et al. 2014). Moreover, reactions between and nitric oxide (NO) during renal diseases induce formation of highly reactive peroxynitrites (Araujo & Welch 2006; Toda et al. 2007). In general, these inappropriate increases in reactive oxygen/nitrogen species can induce tissue damage (Nath et al. 1996).

The oxidative stress observed in this study was not accompanied by renal failure or glomerular permeability alteration. The literature is clear in that renal tubular damage induced by HgCl₂ is a function of dose (Stacchiotti et al. 2003). The data here were obtained using a dose of 2.5 mg/kg; this differed from reports showing renal functional alterations and high levels of tubular damage after hosts received a dose of 5 mg HgCl₂/kg (Gado & Aldahmash 2013; Velyka et al. 2014). The unaltered renal functions could also be related to a lack of inflammatory (CD8 infiltration) and oxidative stress ( -positive cells) in the glomeruli. Another explanation may be related to the low percentage of tubular necrosis (≈15%) noted when under the same protocol as used in this study (Peña et al. 2012). Accordingly, a low level of tubular necrosis was observed in the present study. In general, a high level of tubular necrosis induced by HgCl₂ can trigger acute renal failure by causing increases in afferent arteriole resistance and decreases in ultra-filtration coefficients (via mesangial cell contraction; Yanagisawa et al. 1998).

Ingestion of inorganic mercury salts, mainly HgCl₂, results in corrosion of the alimentary tract and renal damage. The acute lethal dose for most inorganic mercury compounds for an adult is 14–57 mg/kg for a 70-kg person (US Environmental Protection Agency 1994; ATSDR 1999). Thus, the dose used in this study did not represent a lethal dose for humans. In addition, not one of the treated animals died during the experimental period (4 days), suggesting the dose used was not lethal to rats. However, in this study, 2.5 mg HgCl₂/kg was still capable of increasing renal tubular damage, CD8⁺ cell infiltration and inducing oxidative stress. This would be in keeping with earlier findings of increased renal pro-inflammatory events noted in rats after being treated with 2.5 mg HgCl₂/kg (Peña et al. 2012). Such events could, if untreated or mitigated, induce deleterious effects over time leading to host death.

Here, mercuric intoxication induced an increase in renal interstitial infiltration by CD8⁺ cells, a marker expressed on cytotoxic T-lymphocytes (Milstein et al. 2011). Since anti-Ang II treatments diminished the expression of these cells in renal tissues, a role for Ang II in leukocyte infiltration was suggested. This finding would be in accord with other reports that had shown that Ang II enhanced CD8⁺ T-cell infiltration into the myocardium during acute cardiac inflammation (Ma et al. 2011). To our knowledge, this is the first report regarding infiltration of CD8⁺ T-cells during the acute phase of HgCl₂ nephropathy and its potential modulation by Ang II. The current in vivo data also are in agreement with previous in vitro study results. In the latter, it was shown that Losartan or captopril was able to reduce adhesion/transmigration of CD8⁺ T-cells (Silva-Filho et al. 2011), again suggesting a role for Ang II in lymphocyte infiltration.

It is known that CD8⁺ T-cells are important intermediaries in the actions of Ang II in the kidney, by impacting upon renal infiltration, sodium and volume retention, renal cortical production, vascular hypertrophy and rarefaction (Chen et al., 2012; Trott et al. 2014). In this study, high levels of renal accompanied by lipid peroxidation were found. These data suggested that the oxidative stress mediated by Ang II involved, in part, increased production of by activated CD8⁺ T-cells. We, therefore, conclude that, in addition to effects on tissue infiltration, Ang II can also influence CD8⁺ T-cell activation. Activated CD8⁺ T-cells could be involved in tubular damage, i.e. the cytotoxic activity of these cells could induce deleterious effects on tubular cells. The high association seen here between interstitial CD8⁺ cell levels and the percentage of tubules with necrosis comports with this hypothesis.

Still, a direct effect of mercury on CD8⁺ T-cell activation here could also be important. In this regard, an increased number of CD8⁺ T-cells expressing pro-inflammatory genes in the thymus of mice treated with mercuric sulfide has been reported (Son et al. 2010). In addition, it is known that HgCl₂ exposure during immune system development can lead to persistent alterations in adult immune cell phenotypes and functions (Pilones et al. 2009). The effect on immune cells seems to be shared by other metals, as high levels of CD8⁺ T-cells and oxidative stress have been found in patients suffering from nickel or chromium intoxication (Ahmed et al. 2010).

Differential modulatory effects of HgCl₂ on CD8⁺ T-cell infiltration were observed here. Specifically, increased CD8⁺ cell infiltration in the interstitium and decreased infiltration in the glomeruli were found. While HgCl₂ has a potent nephrotoxic effect (mediated by Hg²⁺-glutathione complexes (Yanagisawa 1998), the effect is mainly in the proximal tubules, inducing acute tubular necrosis and interstitial inflammation (Yanagisawa 1998; Peña et al. 2012). Our earlier studies noted that increased expression of immunological markers of inflammation (e.g. ICAM-1, ED-1⁺ cell infiltration) was only noted in kidney tubular and interstitial areas and absent in glomerular regions during HgCl₂ acute intoxication in rats (Peña et al. 2012). Thus, necrotic and inflammatory effects in the interstitium could be modulated by differential degrees of infiltration of CD8⁺ cells, like that observed in the present study.

Conclusions

This study showed that inhibition of the AT1 receptor and angiotensin I-converting enzyme (by Losartan and Enalapril, respectively) decreased renal oxidative responses and CD8⁺ cell infiltration and increased anti-oxidant responses in a model of mercury-induced nephropathy. This suggested a new role for Ang II in mercuric intoxication. The study also showed that modulation of the rennin-angiotensin system, either by angiotensin-converting enzyme inhibition or by AT1 receptor blockade, was beneficial. However, the clinical relevance of the present findings requires performing additional therapeutic experiments.

Acknowledgements

We thank the Instituto de Investigaciones Clinicas “Dr. Americo Negrette”, Facultad de Medicina (Universidad del Zulia, Venezuela) that provided support for this study. The Institution was not involved in the study design, collection, analysis and interpretation of data, writing of the report or in the decision to submit the article for publication.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.