Implication of endoplasmic reticulum stress and mitochondrial perturbations in remote liver injury after renal ischemia/reperfusion in rats: potential protective role of azilsartan

ABSTRACT

Objectives: Distant liver injury is a complication of renal ischemia-reperfusion (I/R) injury, which imposes mortality and economic burden. This study aimed to elucidate the cross-talk of endoplasmic reticulum (ER) stress and mitochondrial perturbations in renal I/R-induced liver injury, and the potential hepatoprotective effect of azilsartan (AZL).

Methods: Male albino Wister rats were pre-treated with AZL (3 mg/kg/day, PO) for 7 days then a bilateral renal I/R or sham procedure was performed. Activities of liver enzymes were assessed in plasma. The structure and ultra-structure of hepatocytes were assessed by light and electron microscopy. Markers of ER stress, mitochondrial biogenesis and apoptosis were analyzed in livers of rats.

Results: Renal ischemic rats showed higher plasma levels of liver enzymes than sham-operated rats, coupled with histological and ultra-structural alterations in hepatocytes. Mechanistically, there was up-regulation of ER stress markers and suppression of mitochondrial biogenesis-related proteins and enhanced apoptosis in livers of renal ischemic rats. These abnormalities were almost abrogated by AZL pretreatment.

Discussion: Our findings uncovered the involvement of mitochondrial perturbations, ER stress and apoptosis in liver injury following renal I/R, and suggested AZL as a preconditioning strategy to ameliorate remote liver injury in patients susceptible to renal I/R after adequate clinical testing.

KEYWORDS:

• Azilsartan
• Renal ischemia/reperfusion
• Remote liver injury
• Endoplasmic reticulum stress
• CHOP
• Mitochondrial biogenesis
• PGC-1α
• Apoptosis

1. Introduction

Renal ischemia reperfusion (I/R) is a leading cause of acute kidney injury (AKI) that frequently occurs in patients during hospitalization or some clinical settings such as trauma, sepsis, kidney transplantation and cardiac surgical procedures. Many lines of evidence have demonstrated complex organ crosstalk in AKI between the kidney and distant organs such as the brain, heart, lung, spleen and liver, which may cause multiple organ dysfunctions and failure. A potential association between renal I/R and remote liver dysfunction has been emphasized in clinical research that contributes to increasing mortality in affected patients [1]. Several factors trigger remote liver injury following renal I/R, but the precise pathological mechanisms are still poorly understood. It has been postulated that renal I/R can enhance oxidative stress, deplete the antioxidant defense system, promote a pro-inflammatory cascade and/or exacerbate apoptotic pathways in the liver as a remote organ, resulting in the accumulation of stressors in liver tissue which deteriorates proper liver functions [2].

The endoplasmic reticulum (ER) is a specialized organelle for modification, processing, folding and assembly of proteins in eukaryotic cells. The presence of chaperones in the ER lumen renders it an ideal compartment for proper protein-folding processes [3]. The ER homeostasis can be disrupted under a wide array of stressors such as hypoxia, glucose deprivation, environmental toxins, Ca ^+ 2 perturbations or oxidative injury causing ER stress. This consequently activates the unfolded protein response (UPR), to rescue the cell and reestablish ER homeostasis [4]. The UPR is mediated by three sensors: protein kinase RNA-like ER kinase (PERK), inositol requiring enzyme 1 (IRE1) and activating transcription factor 6 (ATF6). In normal cells, the sensor proteins are inactivated via binding with glucose-regulated protein 78 (GRP78). Whereas, in the case of ER stress, GRP78 dissociates allowing activation of three branches of URP [5]. The UPR can alleviate mild ER stress. In contrast, during severe or persistent ER stress, the UPR can not restore the normal ER, which eventually sensitizes the cells to apoptosis. Thus, the outcome of ER stress determines whether the cell will survive or undergo apoptosis [6].

The liver is the major site for the production of proteins that experience folding and modification processes inside the ER. Therefore, it is not surprising that ER stress is critically implicated in the pathogenesis of liver diseases including non-alcoholic fatty liver [7], viral hepatitis [8], hepatic ischemia [9] and liver fibrosis [10] but the contribution of ER stress and the related perturbations in remote liver injury has not been fully investigated.

The renin-angiotensin system (RAS) has emerged as a key player in liver diseases. Substantial evidence has reported the activation of systemic and local RAS during hepatitis, nonalcoholic fatty liver disease, and liver cirrhosis [11, 12]. Angiotensin II (Ang II), via activating Ang II type 1 receptors (AT₁R), triggers pro-inflammatory, fibrogenic and proliferative actions in hepatic tissues [13]. Therefore, we hypothesized that RAS blockade may exhibit beneficial effects against remote liver injury following renal I/R. Azilsartan (AZL) is a potent antihypertensive drug acting as an AT₁R blocker (ARB), but with a higher affinity for AT₁R, compared to other ARBs [14]. Furthermore, AZL has marked renoprotective [15], cardioprotective [16], and neuroprotective [17] effects in numerous experimental animal models. More importantly, AZL exerts a favorable role against non-alcoholic fatty liver disease in rats [18]. However, the effect of AZL on liver injury associated with renal I/R is still elusive. Therefore, the fundamental goal of this study was to characterize the implication of ER stress/mitochondrial biogenesis perturbations crosstalk in renal I/R-induced liver injury and to investigate the impact of AZL pretreatment.

2. Materials and methods

2.1. Animals

40 male Albino Wister rats (8 weeks old, weighing 200–250 g) were purchased from the animal house of Zagazig University. The animals were maintained at 24 ± 2°C with a light/dark cycle of 12 h. and had free access to standard food and water during the whole study. Ethical approval for the experimental procedures was granted from the institutional committee of laboratory animal use (approval # ZU-IACUC/3/F/175/2023). During the study, the NIH guidelines for animal research were strictly followed. All possible efforts were made to minimize the pain and suffering of animals.

2.2. Experimental design and surgical procedures

After acclimatization for one week, rats received 3 mg/kg/day AZL (Sanofi, Paris, France) suspended in water using 1% w/v sodium carboxymethylcellulose or matched amounts of drug vehicle by oral gavage for 7 days then subjected to sham or renal I/R procedure. The rats were randomly allocated into five groups (n = 8 each): i- the vehicle-treated normal control group (NC); ii- the vehicle-treated sham group (sham); iii- the AZL-treated sham group (AZL + sham); iv- the vehicle-treated renal I/R group (Renal I/R); v- the AZL-treated renal I/R group (AZL + renal I/R). AZL was used in a clinically relevant dose according to a published report [19].

The bilateral renal I/R procedure was conducted in rats, as previously reported [20]. Briefly, rats were anaesthetized with an xylazine/ketamine mixture (8/75 mg/kg, IP) and kept on a 37°C electrically-heated pad. The right and left renal pedicles of rats were exposed through a midline incision and then clamped by vascular clips for 1 h to induce renal ischemia. 1 h later, the clips were removed gently and the kidneys were checked to make sure that perfusion was resumed. The incision was closed with the appropriate silk suture and the rats were allowed to recover. Sham-operated rats were exposed to the same procedure but clips were not applied.

24 h after the reperfusion, rats were anesthetized by 75 mg/kg pentobarbital sodium. Blood samples were obtained from cardiac puncture and processed for plasma preparation for liver enzyme assays. Rats were sacrificed and liver specimens were excised and snap frozen in liquid nitrogen. Thereafter, frozen liver tissue was homogenized in 10 ml of phosphate-buffered saline for each gram on ice, using a glass tissue homogenizer then centrifuged at 5000xg for 5 min. The supernatant was stored at −20°c and used for assaying tissue parameters. Part of the liver tissue was processed for histological examination, as mentioned below.

2.3. Liver enzymes analysis

Levels of AST, ALT and ALP were measured in plasma aliquots using commercially available kits (Spinreact, Spain).

2.4. Measurement of hepatic oxidative stress

Levels of malondialdehyde (MDA) and reduced glutathione (GSH) in hepatic tissue homogenates were determined using commercially available kits (Biodiagnostic, Giza, Egypt), following the manufacturer’s instructions. Results were expressed as µmol/g tissue.

2.5. ELISA analysis

Liver tissue homogenate was used to measure levels of PERK (LifeSpan BioSciences, Seattle, WA; USA); TNF-α, GRP78, ATF6, IRE1, eIF2α, CHOP, PGC-1α and NRF-1 (MyBioSource, San Diego, CA, USA) using the commercially available rat ELISA kits, following the manufacturers’ instructions.

2.6. Determination of caspase-3 activity

Caspase-3 activity was analyzed in liver specimen lysate using the caspase-3 colorimetric assay kit (Sigma-Aldrich, MI, USA; #CASP-3-C), adopting the suppliers’ instructions.

2.7. Histological assessment

Specimens of liver tissue were fixed in 10% neutral buffered formalin and then embedded in paraffin. After being deparaffinized, 5-μm-thick slices were processed for hematoxylin & eosin (H&E) staining and then examined under the light microscope (Primo star, ZEISS, China) for assessment of the general liver architecture. The level of tissue damage in H&E-stained liver sections was quantified, as previously reported [21]. The area percentage of hepatic tissue damage manifested as hepatocyte vacuolation and mononuclear cellular infiltration as well as central and portal veins dilatation and congestion, was analyzed in six randomly chosen microscopic fields of H&E-stained slides using Leica Qwin image analyzer system (Cambridge, UK).

2.8. Immunohistochemical staining:

The expression of Bcl-2 and Bax in liver tissue was assessed by immunohistochemical staining using mouse monoclonal antibodies (dilution 1:100; Santa Cruz Biotechnology Inc, Texas, USA; # sc-509 and sc-7480, respectively). Briefly, the slides were deparaffinized followed by antigen retrieval. To avoid the effect of endogenous peroxidases, 3% H₂O₂ in methanol was used for 15 min and then washed with Tris-buffered saline (TBS). Slides were then incubated with normal goat serum at room temperature for 30 min to block non-specific binding of immunoglobulin. Primary antibodies were applied overnight at 4°c then washed with TBS. Then, slides were embedded with the proper biotinylated secondary antibodies for 1 h at room temperature. After washing with TBS, sections were treated with Avidin–biotin-peroxidase kit for 30 min then embedded with diaminobenzidene (DAB) substrate. Slides were counter-stained by hematoxylin. For negative control slides, all the previous steps were performed without the addition of the primary antibody. The numbers of Bcl-2- and Bax-positive cells were determined in ten randomly selected high-power microscopic fields of each slide. Four liver slides from each group were examined blindly.

2.9. Transmission electron microscopy (TEM)

Fresh liver samples were fixed instantly in 3% glutaraldehyde for 2 h at 4°C, washed with PBS and then fixed in 2% perosmic acid for 2 h. Following dehydration in gradient ethanol concentrations, the slides were implanted in resin then sectioned into 50 nm per slice and stained with lead citrate and uranyl acetate. The stained slides were examined under TEM (H-7650, Hitachi, Japan).

2.10. Statistical analysis

Data were presented as means ± standard deviation (SD). GraphPad Prism version 5.0 (GraphPad Software, Inc. San Diego, USA) was used for conducting the statistical analysis. Arithmetic means of the different groups were compared using one-way ANOVA and Tukey's post hoc tests. A parametric Pearson correlation was applied for correlation analysis using the pooled data from all experimental groups. The statistical significance was reached when p < 0.05.

3. Results

3.1. Effects of renal I/R and AZL pretreatment on plasma levels of liver enzymes

As shown in Table 1, induction of bilateral renal ischemia for 1 h and then reperfusion for 24 h caused a marked elevation of plasma levels of AST, ALT and ALP, compared to NC and both sham groups (p < 0.001), indicating liver injury. These data suggested that renal I/R impaired liver functions. AZL pretreatment significantly attenuated ALT (p < 0.05), AST and ALP (p < 0.001) levels compared to the renal I/R group, underscoring the potential of AZL to rescue the hepatocytes integrity and liver function.

Table 1. The effect of azilsartan (AZL, 3 mg/kg/day, PO for 7 days) on liver enzymes of sham and renal ischemic (I/R) rats.

	ALT (U/l)	AST (U/l)	ALP (U/l)
NC	54.19 ± 5.47	76.50 ± 6.73	450.7 ± 41.62
Sham	56.17 ± 7.47	73.50 ± 8.43	476.7 ± 40.82
AZL + Sham	56.15 ± 5.64	70.17 ± 8.75	466.7 ± 95.01
Renal I/R	81.83 ± 8.52^ΘΘ	124.80 ± 7.19^ΘΘ	896.7 ± 50.86^ΘΘ
AZL + Renal I/R	67.67 ± 7.61^Ψ	89.50 ± 12.2^Θ,ΨΨ	585.0 ± 51.28^Θ,^ΨΨ

Values represent mean ± SD (n = 6/group); ^ϴp < 0.05, ^ϴϴp < 0.001 vs. NC and sham control groups; ^ΨP < 0.05, ^ΨΨP < 0.001 vs. renal I/R group; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; ALP: Alkaline phosphatase.

3.2. Effects of renal I/R and AZL pretreatment on liver architecture

As illustrated in Figure 1, H&E-stained liver sections of the NC group displayed normal hepatic architecture. Hepatocytes appeared with acidophilic cytoplasm and vesicular nucleus, arranged in cords and separated by blood sinusoids radiating from the central vein. Liver sections of sham-operated groups appeared with normal histological structure as that of the NC group. In contrast, examination of liver sections from the renal I/R group denoted disturbed liver architecture. The central vein appeared dilated and congested with dilated liver sinusoids. Some hepatocytes appeared with highly vacuolated cytoplasm, others appeared with dark pyknotic nuclei. The portal area revealed a dilated portal vein surrounded by inflammatory cell infiltrate. AZL pretreatment markedly alleviated the histological lesions of liver tissue that appeared more or less similar to the NC group. Quantitative analysis of damage/degeneration in stained liver sections showed a dramatic increase in the area percentage of hepatic tissue damage in the renal I/R group when compared to NC and sham-operated groups (p < 0.0001), suggesting the remote adverse effect of renal I/R on liver architecture. On the contrary, there was a marked decrease in the area percentage of tissue damage in liver sections from AZL + Renal I/R, compared with the renal I/R group (p < 0.001)

Figure 1. Photomicrographs of H&E-stained liver sections from sham and renal ischemic rats. NC group (a): showing normal liver architecture. Cords of hepatocytes (thick arrow) radiate from the central vein (CV) with blood sinusoids in between (thin arrow). (b): showing portal tract containing; a branch of the portal vein (Pv), bile duct (Bd) and hepatic artery (arrowhead). Hepatocytes (thin arrow) appear with acidophilic cytoplasm, a pale vesicular nucleus with a prominent nucleolus. A binucleated hepatocyte is noticed (curved arrow). Sham (c&d) and AZL + Sham (e&f) groups: displaying the same as the NC group. Renal I/R group (g): displaying apparent dilated and congested central vein (CV). Hepatocytes appear with highly vacuolated cytoplasm (thick arrow). Congested blood sinusoid (thin arrow) is noticed. (h): showing dilated and congested portal vein (CV) with mononuclear cell infiltration (star). Hepatocytes appear with highly vacuolated cytoplasm (thin arrow). (i): showing a congested central vein (CV). Most hepatocytes appear with vacuolated cytoplasm (thin arrow), and some appear swollen with small, dark and pyknotic nuclei. AZL + Renal I/R group (j): displaying restoration of normal histological structure of liver more or less as NC group. (k): showing portal tract containing a branch of the portal vein (Pv), bile duct (Bd) and hepatic artery (arrowhead). Cytoplasm of some hepatocytes is still vacuolated (thin arrow) [H&E; (a, c, e, g, h & j) magnification x200, scale bar = 100μm; (b, d, f, I & k) magnification x400, scale bar = 50μm]. (l): quantitative analysis of the area % of hepatic tissue damage in H&E-stained slides using a computer-assisted automated image analyzer. Six microscopic fields of H&E-stained slides were randomly chosen, analyzed and averaged for each slide. Each bar represents the mean ± SD of four rats per group. ^ϴp < 0.0001 vs. NC and sham control groups; ^ΨP < 0.001 vs. Renal I/R group.

3.3. Effects of renal I/R and AZL pretreatment on hepatocytes ultra-structure

As shown in Figure 2, the ultra-structural changes in liver cells were evaluated by TEM. The hepatocytes of the NC group had normal morphology with a euchromatic nucleus. The cytoplasm contains numerous mitochondria and rough ER. Bile canaliculi in between adjacent liver cells were obvious with a junctional complex close to it and attached to the adjacent hepatocytes. Long microvilli in the hepatocytes membrane were observed facing the blood sinusoid. The ultra-structure of hepatocytes from sham-operated groups was similar to that of the NC group. Conversely, renal I/R surgery resulted in marked changes in hepatocytes ultra-structure. Nucleus appeared shrunken and irregular with more peripheral heterochromatin. A marked cytoplasmic vacuolation was observed, along with dilated rough ER with structural disruption. The mitochondria were distorted, swollen and had uneven ridges. AZL pretreatment restored the normal ultra-structure of hepatocytes and alleviated the distortion in both organelles.

Figure 2. Transmission electron photomicrographs of liver sections from sham and renal ischemic rats. NC group (a): showing part of two adjacent liver cells with a euchromatic nucleus (N), numerous mitochondria (M), rough endoplasmic reticulum (rER), long microvilli (red arrow) facing blood sinusoid (BS), junctional complex (arrowhead) close to bile canaliculus (bc) between two adjacent cells. Cell membrane between two hepatocytes can be noticed (thick arrow). Sham group (b) and AZL + Sham group (c): showing the same as NC group. Renal I/R group (d): showing distorted swollen mitochondria (M), dilated rough endoplasmic reticulum (rER) and multiple cytoplasmic vacuolation (stars). (e): showing shrunken, irregular nucleus (N) with peripheral heterochromatin and marked destructed mitochondria (M). AZL + Renal I/R group (f): showing a liver cell with euchromatic nucleus (N), numerous mitochondria (M), rough endoplasmic reticulum (rER), more or less as NC group (Uranyl acetate & lead citrate; magnification x5000; scale bar = 2 µm).

3.4. Effects of renal I/R and AZL pretreatment on hepatic redox and inflammatory status

As shown in Table 2, renal I/R surgery enhanced the oxidative injury and inflammatory burden in the liver tissue of rats, as manifested by increased levels of MDA and TNF-α, compared to NC and sham-operated control groups (p < 0.001). There was also a decrease in GSH content in livers of rats subjected to renal I/R (p < 0.001). Preconditioning with AZL significantly blunted the derangements of redox and inflammatory status of liver tissue, compared to vehicle-treated renal I/R exposed rats (p < 0.001). These findings highlighted the anti-oxidant and anti-inflammatory properties of AZL.

Table 2. The effect of azilsartan (AZL, 3 mg/kg/day, PO for 7 days) on oxidative stress and inflammatory markers in hepatic tissue of sham and renal ischemic (I/R) rats.

	MDA (µmol/g tissue)	GSH (µmol/g tissue)	TNF-α (pg/g tissue)
NC	24.30 ± 4.05	7.11 ± 1.56	110.9 ± 13.71
Sham	26.50 ± 6.35	6.20 ± 0.56	120.7 ± 15.78
AZL + Sham	27.00 ± 6.03	6.17 ± 0.42	122.8 ± 10.53
Renal I/R	67.67 ± 8.38^ΘΘ	2.57 ± 0.24^ΘΘ	542.5 ± 65.56^ΘΘ
AZL + Renal I/R	44.33 ± 7.15^Θ,Ψ	4.75 ± 0.34^ΘΘ,Ψ	227.8 ± 44.27^Θ,Ψ

Values represent mean ± SD (n = 6/group); ^ϴp < 0.01, ^ϴϴp < 0.001 vs. NC and sham control groups; ^ΨP < 0.001 vs. Renal I/R group.

3.5. Effects of renal I/R and AZL pretreatment on ER stress markers in hepatic tissue of rats

As indicated in Figure 3, GRP78 was evaluated as a prominent ER-resident chaperone. PERK is considered the central regulator of ER stress which determines the ultimate fate of the cell. Eukaryotic Initiation Factor 2 alpha (eIF-2α) is the principal down-stream effector of activated PERK, acting as a key player in the global translation process. Rats that underwent renal I/R experienced significantly higher levels of GRP78, PERK and eIF-2α than normal and sham-operated control groups (p < 0.01). Other sensors of UPR were also assessed. There were significant increases in hepatic expression of IRE1 and ATF6 in renal ischemic rats, compared with normal and sham controls (p < 0.01). These findings suggested the promotion of UPR as a pro-adaptive response to overcome the ER stress in hepatocytes as a consequence of renal I/R.

Figure 3. The effect of azilsartan (AZL, 3 mg/kg/day, PO for 7 days) on endoplasmic reticulum (ER) stress markers in liver tissue of sham and renal ischemic (I/R) rats. Each bar represents the mean ± SD of six rats per group; ^ϴp < 0.01 vs. NC and sham control groups; ^ΨP < 0.001 vs. Renal I/R group. GRP78: glucose-regulated protein 78 kDa; PERK: protein kinase RNA–like endoplasmic reticulum kinase; ATF6: activating transcription factor 6; IRE1: inositol requiring enzyme 1; eIF2α: eukaryotic initiation factor 2 alpha; CHOP: C/EBP homologous protein.

CHOP is regarded a well-established ER stress-associated pro-apoptotic factor. There is marked up-regulation in the hepatic expression of CHOP in renal ischemic rats, compared to normal and sham control groups (p < 0.01), implying that the pro-adaptive responses failed and apoptotic cell death proceeded. AZL pretreatment almost restored the GRP78 levels in liver tissue of rats subjected to I/R, to values of control groups. Meanwhile, AZL pretreatment significantly minimized the expression of PERK, eIF-2α, ATF6, IRE1 and CHOP, relative to the renal I/R group (p < 0.001). These findings unveiled the potential of AZL to offset the ER stress.

3.6. Effects of renal I/R and AZL pretreatment on mitochondrial biogenesis in hepatic tissue of rats

As illustrated in Figure 4, PGC-1α was examined as a hallmark of mitochondrial biogenesis and function by induction of NRF-1 and other down-stream proteins. Renal ischemic rats displayed a marked decline in hepatic levels of PGC-1α and NRF-1, compared to NC and sham-operated rats, (p < 0.001), suggesting that renal I/R could impair mitochondrial biogenesis in distant liver organs. On the contrary, the hepatic expressions of both markers were enhanced by AZL preconditioning (p < 0.001). These data suggested that AZL promoted mitochondrial biogenesis in hepatocytes of rats undergoing renal I/R. Interestingly, Pearson correlation analysis demonstrated negative correlations between CHOP and both PGC-1α (r = −0.9270, p < 0.0001) and NRF-1 (r = −0.9045, p < 0.0001) which implied the antagonistic crosstalk between CHOP and mitochondrial biogenesis-related proteins.

Figure 4. A&B: The effect of azilsartan (AZL, 3 mg/kg/day, PO for 7 days) on mitochondrial biogenesis-related proteins in liver tissues of sham and renal ischemic rats. Each bar represents the mean ± SD of six rats per group; ^ϴp < 0.001 vs. NC and sham control groups; ^ΨP < 0.001 vs. Renal I/R group. C&D: Parametric Pearson correlation between CHOP and mitochondrial biogenesis-related proteins. PGC-1α: Peroxisome proliferator-activated receptor gamma coactivator 1-alpha; NRF-1: Nuclear respiratory factor-1.

3.7. Effects of renal I/R and AZL pretreatment on apoptotic markers in hepatic tissue of rats

As indicated in Figure 5, Bax (pro-apoptotic marker) and Bcl-2 (anti-apoptotic protein) yielded brown cytoplasmic reactions upon immunohistochemical analysis. Renal I/R considerably increased the number of Bax immunopositive cells in liver specimens, in comparison to NC and sham-operated control groups (p < 0.001). This was markedly reversed by AZL pretreatment, as shown in the AZL + Renal I/R group, in which the hepatic tissue showed an apparent decrease in the number of Bax immunopositive cells (p < 0.001). On the other hand, the number of Bcl-2 immunopositive cells in the liver tissue of the renal I/R group markedly declined in comparison to control groups (p < 0.001). AZL pretreatment effectively prevented the decreased Bcl-2 expression in liver tissue (p < 0.001).

Figure 5. Photomicrographs of immunohistochemical staining of Bax and Bcl-2 in liver sections from sham and renal ischemic rats. (a, b & c): NC, Sham and AZL + Sham groups showing few Bax immunopositive liver cells. (d): Renal I/R group showing extensive Bax immunopositive liver cells. (e): AZL + Renal I/R group showing that some liver cells are BAX positive cells. (f, g & h): NC, Sham and AZL + Sham groups showing abundant Bcl-2 immunoreactions in most hepatocytes. (i): Renal I/R group revealing an apparent decline in Bcl-2 immunoreactions in most of the hepatocytes. (j): AZL + Renal I/R group demonstrating extensive Bcl-2 immunoreactivity in liver cells more or less as control groups (Avidine biotin peroxidase stain with Hx counter stain; Magnification x400, scale bar = 50 µm). (k & l): Quantification of Bax and Bcl-2 immunoexpressions in liver tissue of rats, respectively. Each bar represents the mean ± SD of four rats per group; ^ϴp < 0.001 vs. NC and sham control groups; ^ΨP < 0.001 vs. Renal I/R group.

Caspase-3 serves as a key executioner of apoptosis. As shown in Figure 6, renal I/R surgery significantly promoted the activity of caspase-3 in the liver tissue of rats compared to NC and sham-operated controls (p < 0.001). AZL pretreatment markedly reduced caspase-3 activity in hepatic tissue (p < 0.001), ensuring the anti-apoptotic properties of AZL. Taken together, renal I/R enhanced the apoptosis of hepatocytes, while AZL pretreatment impeded apoptotic death of liver cells by enhancing the Bcl-2 expression and suppressing the apoptotic proteins.

Figure 6. The effect of azilsartan (AZL, 3 mg/kg/day, PO for 7 days) on caspase-3 activity in livers of sham and renal ischemic rats. Each bar represents the mean ± SD of five rats per group; ^ϴp < 0.05, ^ϴϴp < 0.001 vs. NC and sham control groups; ^ΨP < 0.001 vs. Renal I/R group.

4. Discussion

Remote or distant liver injury is still an unresolved clinical challenge, emphasizing the urgent need to develop new management approaches. The current study highlighted the implication of ER stress and mitochondrial biogenesis perturbations in hepatic injury associated with renal I/R. Pretreatment with AZL protected against remote liver injury, likely via restoring ER homeostasis, enhancing mitochondrial biogenesis and inhibiting hepatocyte apoptosis.

Herein, we found that induction of renal I/R imposed a detrimental impact on rat liver, as manifested by increased plasma levels of liver enzymes which indicated the injury of hepatocytes and leakage of intracellular enzymes. Such findings were further supported by the histopathological assessment that revealed marked lesions in liver specimens of renal ischemic rats. Our results were closely aligned with previous studies [20, 22].

Increasing evidence indicates crosstalk between ER stress and the pathogenesis of various diseases, including liver disorders [23]. We found that rats which underwent renal I/R surgery displayed dilated and distorted rough ER in TEM micrographs implying the accumulation of misfolded proteins in the ER lumen. On molecular levels, there were remarkable rises of PERK, ATF6 and IRE1 levels, coupled with up-regulation of GRP78 in hepatic tissue of renal I/R-exposed rats which indicated the stimulation of UPR as a survival attempt to restore the ER homeostasis. After being dissociated from GRP78, PERK is activated by self-phosphorylation. Activated PERK undergoes phosphorylation to its major down-stream effector, eIF-2α at serine 51 leading to repression of the global translation process, likely as a pro-adaptive endeavor to suppress the burden on ER [24]. Herein, renal ischemic rats exhibited a marked rise in eIF-2α levels, but regrettably, the phosphorylated eIF-2α was not assessed in the present study. However, the increased expressions of GRP78 and UPR sensors collectively suggested that renal I/R could trigger ER stress in rats’ hepatocytes. Hypoxia, reactive oxygen species, ATP/nutrient deprivation and/or calcium overload are highlighted as major contributing insults [23].

The disruption of hepatic ER homeostasis following renal I/R was also mirrored by the up-regulation of CHOP in the liver tissue of rats. CHOP is a crucial chaperone that drives the apoptotic impact of persistent ER stress. Upon unstressed physiological status, CHOP is slightly expressed, but its transcription is dramatically induced likely through PERK, ATF6 and IRE1, in response to ER stress [25]. Accordingly, the increased levels of CHOP in the livers of renal ischemic rats implied that the adaptive UPR was overwhelmed by severe or sustained ER stress. Thus, hepatic cells would switch from the pro-survival adaption to initiate signaling cascades that eventually contribute to apoptosis [6]. Therefore, targeting ER stress, particularly CHOP expression, might serve as a promising modality to combat distant liver injury.

Mitochondrial biogenesis entails the growth and auto-replication of existing mitochondria. PGC-1α is regarded as a co-transcriptional regulator of mitochondrial biogenesis and function through promoting different transcription factors, namely NRF-1, NRF-2, mitochondrial transcription factor A (mtTFA) and estrogen-related receptor-α, which interact together to induce transcription and replication of mitochondrial genome (mtDNA) [26]. Recent evidence demonstrates that ER stress and mitochondrial perturbations are closely interconnected in several pathological entities [27–30]. Severe or sustained ER stress can deteriorate mitochondrial biogenesis. CHOP down-regulates the transcription of PGC-1α expression [31]. Chen et al., have shown that in vivo deletion of Chop gene or in vitro knocking down of CHOP protein can promote the expression of PGC-1α [32]. These data exemplify that CHOP plays a critical role in linking ER stress to mitochondrial perturbations. We also found that CHOP expression was negatively correlated with both PGC-1α and NRF-1 levels. In light of these findings, it seems plausible to assume that the down-regulation of PGC-1α, and hence NRF-1, might be related, at least in part, to the increased CHOP in hepatic tissue of renal ischemic rats. Further mechanistic studies using CHOP inhibitors or genetically manipulated mice are needed to certify the causal relationship between CHOP and PGC-1α.

The decline of PGC-1α level in hepatic tissue might contribute to disruption of mitochondrial biogenesis and impairment of bioenergetic capacity. The ultra-structural abnormalities including swelling and uneven ridges of the mitochondria, further underpinned the dysregulation of mitochondrial biogenesis in livers of rats after renal I/R. Importantly, the decreased PGC-1α level has been associated with increased oxidant production and inflammatory response [33]. Meanwhile, the up-regulation of CHOP worsens the cellular redox state by promoting the expression of ER oxidoreductin 1 (ERO1α), resulting in the production of hydrogen peroxide and oxidative stress [34]. Consisted with these data, we found an enhanced lipid peroxidation and inflammatory response reflected by increased MDA and TNF-α levels in the livers of renal ischemic rats. We also found a notable depletion of their hepatic GSH content. The ratio of GSH to the oxidized glutathione (GSSG) is regarded as an important indicator for cellular redox potential, but unfortunately, the GSH/GSSG ratio was not assessed, which was a limitation of the present study.

An abundant body of evidence implicates CHOP as a fundamental mediator of ER stress-associated apoptosis [35]. CHOP can modulate the members of the Bcl-2 family by suppressing the anti-apoptotic factors like Bcl-2 while inducing the pro-apoptotic proteins like Bax and Bak. CHOP can also up-regulate the death receptors 4 and 5 which induce apoptosis through caspase 8 activation. Thus, CHOP can activate both intrinsic and extrinsic apoptotic pathways [36]. The pro-apoptotic properties of CHOP may be also owing to the induction of ERO1α, which triggers a hyper-oxidizing environment in the ER, leading to leakage of hydrogen peroxide and a cascade of inflammatory and apoptotic events [34]. Consistently, the decline of Bcl-2 expression, along with increased Bax expression and caspase-3 activity in hepatocytes of renal ischemic rats further reflected the enhanced apoptosis, and appeared to be secondary to CHOP up-regulation.

An intriguing finding was that AZL preconditioning mitigated the renal I/R-induced increments of liver enzyme levels, suggesting that AZL preserved the hepatocellular membrane integrity. AZL effectively alleviated the histological and ultra-structural changes of hepatocytes. These findings underscored the hepato-protective potential of AZL against remote liver injury. Relevant to our findings, AZL exerts a hepato-protective effect in high-fat diet-fed rats [18]. We also found that AZL efficiently attenuated the expression of GRP-78 and the three sensors of UPR (PERK, ATF6 and IRE1) in the livers of renal ischemic rats. The ER expansion in hepatocytes was blunted by AZL pretreatment. Meanwhile, AZL markedly counteracted the up-regulation of CHOP in the livers of renal ischemic rats. These data collectively suggested that AZL could restore ER homeostasis in hepatic cells. As far as we know, this was the first report uncovering the ameliorative effect of AZL on hepatic ER stress and CHOP expression. The exact mechanism underlying this effect is still elusive but it might be related to AT₁R blockade. Supportively, Ang II induces ER stress in adipocytes [37]. Ang II also up-regulates ER stress markers [38] and increases the expression of CHOP in pancreatic β-cells [39]. Antagonism of AT₁R by losartan inhibits ER stress in the aorta of uremic mice and in vitro experiments [40]. Valsartan, another ARB, can protect rats from contrast-induced AKI by alleviating ER stress [41]. Intriguingly, olmesartan offsets CHO-induced apoptosis in kidneys [42] and hearts of diabetic mice [43], through an AT₁R blockade-dependent mechanism. Surely, additional investigations are required to confirm the postulated mechanism behind the repressive effect of AZL on ER stress and CHOP expression.

Numerous studies have indicated that gene deletion of Chop can inhibit apoptosis in neuronal cells [44], pancreatic β-cells [45], cardiomyocytes [46] and renal cells [47]. Accordingly, AZL-induced suppression of CHOP partly mediates the inhibitory effect of AZL on caspase-3 activity. These data illustrated that AZL could protect against hepatocellular apoptosis thereby promoting hepatocyte survival. Parallel to our findings, AZL mitigates apoptotic cell death of dopaminergic neurons in rats [48]. Given that apoptosis is an important pathogenic feature in liver dysfunction, regardless of the initiating insults; therefore, the anti-apoptotic properties of AZL might be involved in its hepato-protective effects.

The current study also unveiled the positive effect of AZL on mitochondrial biogenesis. AZL could promote hepatic expressions of PGC-1α as well as NRF-1 and alleviate mitochondrial swelling in hepatocytes of rats subjected to renal I/R. These findings suggested that AZL reversed mitochondrial perturbations in hepatocytes, in harmony with previous studies [17, 49]. This might be attributed, at least in part, to the RAS blockade. Supportively, RAS inhibition by enalapril or losartan induces up-regulation of PGC-1α and NRF-1 in the livers of aged rats [50]. Previous studies have demonstrated that PGC-1α can protect against alcoholic [51] and non-alcoholic fatty liver disease [52], viral hepatitis [53] and cholestasis [54], presumably via its antioxidant and anti-inflammatory effect. Therefore, we suggested that the enhancement of mitochondrial biogenesis by AZL might represent a possible mechanism underlying its hepato-protective effect. The hepatoprotective effect of AZL might also encompass its antioxidant and anti-inflammatory properties. AZL mitigated lipid peroxidation and inflammatory burden in hepatic tissue of renal ischemic rats, consistent with a previous report [15]. These actions might be attributed to the ability of AZL to suppress ER stress and mitochondrial biogenesis perturbations. Another postulated mechanism may be related to AZL-induced inhibition of NADPH oxidase enzyme, the major source of cellular ROS [55].

5. Conclusions

This study demonstrated new insights into the molecular mechanisms underlying renal I/R-induced remote liver injury through inducing ER stress, mitochondrial biogenesis perturbations and hepatocyte apoptosis. The current findings suggested that AZL could rescue the liver as a remote organ from renal I/R injury, and provided the first proof of the modulatory effect of AZL on ER stress and mitochondrial biogenesis. These results might extend the clinical applications of AZL to encompass its hepato-protective effect against remote liver injury in patients susceptible to renal I/R. Certainly, additional studies are essential to validate the clinical translation of these findings.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data are available upon reasonable request from the corresponding author.