Ripe and healthy pomegranate fruits were collected from the northern parts of Khyber Pakhtunkhwa and authenticated by the taxonomist Dr. Muhammad Ilyas at the Department of Botany, University of Swabi, Pakistan, with the voucher number (PG/HA-1), deposited in the herbarium.

Pomegranate fruits were collected, washed thoroughly with tap water, and then peeled. The peels were shade-dried for one month. The dried peels were powdered in an electric grinder. The accurately weighed powdered sample (1 kg)/was mixed in 2 L of methanol to yield 100 g extract (Met-PPx). After removal of the solvent, the concentrated methanolic extract was suspended in water for further solubilization to get the aqueous fraction (15 g). Then n-hexane was added to the mixture to remove the fatty material, and the defatted fraction was then dissolved in ethylacetate and butanol to finally yield n-hexane (12 g), ethylacetate (25 g), and butanol (9 g) fractions, respectively. The extract and fractions were kept at 4°C until further use. The extract preparation scheme is given below in Figure 1.

Male Wistar rats (150–180 g) were housed in individual cages for a week at ambient temperature under 12-h light/dark cycles, with access to chow and tap water ad libitum according to standard laboratory protocol. All animals received humane care, and all protocols concerning the animals were in agreement with the guidelines approved by the Institutional Ethics Committee approval letter (CUI/Bio/ERB/4-21/19) of COMSATS University, Islamabad.

Acute hepatotoxicity studies were performed using male Wistar rats. The experimental protocol for the current study was based on previous studies (Ali et al., 2014). All the experimental animals were divided into different groups, and six rats were included in each group as described here: Group 1 (normal control) were injected with vehicle only (1 ml/kg body weight olive oil), Group 2 (hepatotoxicity group) were treated with i.p. injection of CCl₄ (1 ml/kg) with 1:1 olive oil, Group 3 (positive control) were injected intraperitoneally with CCl₄ (1 ml/kg) with 1:1 olive oil and also administered orally with silymarin (100 mg/kg) 3 days before treatment and 2 days after treatment, and Groups 4–8 were treated with i.p. injection of CCl₄ (1 ml/kg) with 1:1 olive oil but also received methanolic extract, aqueous, n-hexane, ethyl acetate, and butanol fractions, respectively, at a dose of 100 mg/kg body weight for 3 days before treatment and 2 days after treatment.

All experimental animals were terminated 48 h after the administration of CCl₄ under sodium pentothal anesthesia. From animals, the blood was collected by cardiac puncture using a 5-ml sterile syringe and with a slight possible pressure to avoid hemolysis. Blood was centrifuged for 15 min at 2000 rpm to extract serum. The serum was used for the determination of liver damage by analyzing biological parameters (ALT, AST, and ALP) using a dry chemistry analyzer (Roche Diagnostics, Mannheim, Germany).

For identification of oxidative stress, the livers were excised and homogenized (Polytron homogenizer) in 50 mM phosphate buffer saline with a pH value of 7.4 (Kinematica, Lucerne, Switzerland). The homogenates were centrifuged at 15 000 g for 20 min at 4°C using a Beckman L7-65 Ultracentrifuge (Beckman, Fullerton, United States), and the supernatants were used to determine the Cu/Zn SOD activity, MDA, and GSH content. The quantification of protein in liver tissues was assessed using Bradford’s method (Bradford, 1976). The activity of Cu/Zn SOD was measured at 550 nm by the reduction in cytochrome c by superoxide radicals using a UV–VIS spectrophotometer (Domitrović et al., 2011). Similarly, the MDA and GSH contents were determined according to a reported study (Anderson, 1985). The supernatant was deproteinized with 1.2 M metaphosphoric acid and centrifugation at 4500 g for 8 min (Rotina 420R, Andreas Hettich GmbH, Tuttlingen, Germany). Consequently, 700 μL of 0.3 mM NADPH in phosphate buffer saline and 25 μL of the deproteinized sample along with water and 100 μL of 6 mM DTNB were thoroughly mixed in a cuvette to get a final volume of 1.0 ml. Finally, 10 μL of glutathione reductase (50U ml⁻¹) was added to the mixture, and absorbance was observed for 30 min at 405 nm. From the series of dilution of stock solution of glutathione, their content was identified from the standard curve.

After dissection, the liver was identified in the right upper quadrant and sliced up. The liver was kept in formalin for histopathological examination. Formalin-fixed liver tissues were desiccated through a series of graded alcohol, followed by their entrenchment in paraffin, and cut into 6-μm-thick sections. The liver sections were transferred onto the slides, dewaxed, and rehydrated through xylene and graded alcohol in the reverse sequence from dehydration. The deparaffinized tissues were stained with hematoxylin–eosin (H&E) and examined under a bright field microscope. The necrotic area of each group was measured in at least 30 different tissue sections of the liver using NIS-elements software from Nikon, Japan, and was expressed as the percentage of the entire area of the section.

The previously dewaxed and rehydrated liver sections (6 μm thick) were used for immunohistochemical assessment. All selected liver sections were incubated with blocking solution in phosphate buffered solution (PBS, Roti-immunoblock, Carl Roth, Karlsruhe, Germany) for 25 min at normal temperature. Subsequently, these tissues were incubated with primary monoclonal antibody for CD68 (clone ED1, abcam) and iNOS (ab3523) at dilution (1:100) for 45 min at 37°C. The liver tissues were thoroughly washed with PBS and incubated with secondary antibodies, FITC/Texas Red-conjugated goat anti-mouse IgG (1:100), for 40 min. After briefly rinsing twice with PBS, the nuclei were stained with DAPI for around a minute and then rinsed with PBS and mounted in mounting media. The immunostained tissues were analyzed by fluorescence microscopy, and their images were acquired using a Nikon DXM 1200 C camera using image analysis software AR 3.0 (Nikon, Japan). All the acquiesced images were processed using Adobe Photoshop software. The number of Kupffer cells was counted in different portions surrounding the injured area from their elongated nuclei stained with DAPI using NIS-elements software.

Results are expressed as mean ± SD. Total variation present in a set of data was estimated by one-way analysis of variance (ANOVA), followed by Dunnett’s post hoc test. p < 0.05 and p < 0.01 were considered to be significant.

Liver sections of the normal control group exhibited typical lobular architecture, with a central vein fenced by the hepatic cord of cells with clear spaces lined by elongated endothelial and Kupffer cells (Table 1; Figures 2A,B). In contrast, the CCl₄-treated group revealed loss of hepatic architecture with clear pale areas having degenerated and necrotic hepatocytes and the presence of inflammatory cells around the central vein with distinct nuclei (Table 1). This pointed toward the fact that the injurious free radical CCl₃ ⁻ is particularly generated only in the hepatocytes. Normal-shaped hepatocytes (Figures 2C,D) could also be perceived around the necrotic area, which could simply be identified because of their darker staining characteristic. Cells around the pale necrotic area exhibited hydrophic degeneration and ballooning hepatocytes (Table 1). Moreover, histopathological analysis (Figure 3B) had shown ∼47% damage in the CCl₄-treated group. Liver samples acquired from the CCl₄ ⁺ silymarin–treated group showed a substantially attenuated (p < 0.001) necrotic area around the central vein and a great decrease (Figures 2E,F) compared to the CCl₄ group (Table 1). Moreover, minimal hepatic damage (∼10%) was observed by histopathological analysis (Figure 3B). Interestingly, the CCl₄ ⁺ Met-PPx completely prevented liver necrosis (Table. 1; Figures 2G,H) with 0% damage (Figure 3B). Therefore, the Met-PPx at an optimum dose of 100 mg/kg presented restoration of injury and hence showed hepatoprotective activity (p < 0.001) compared to the positive control silymarin (100 mg/kg).

Figure 4 indicates the effect of different fractions of the Met-PPx on the histology of CCl₄-induced liver toxicity. The CCl₄ ⁺ EtOAc group had shown a similar protective effect to that of the Met-PPx, with restoration of the hepatic architecture with no sign of necrosis (Table. 1; Figures 4A,B), and presented 0% damage (Figure 3B). Even the liver architecture was in such a way that it was difficult to distinguish it from the untreated group. Likewise, the CCl₄ ⁺ aqueous group also completely abolished and showed normal arrangement of hepatocytes with no sign of necrosis (Figures 4C,D) and 0% damage (Table 1; Figure 3B). On contrary, more necrosis and massive inflammatory recruitment (Table 1) around the central vein was found in the CCl₄ ⁺ butanol group than in the positive control silymarin (Figures 4E,F). Histopathological investigation revealed ∼29% damage (Figure 3) which is less than that in the CCl₄-treated group but more than that in the silymarin control. The CCl₄ ⁺ n-hexane group also indicated necrotic areas (Table 1) around the injurious place of the central vein with congestion of RBCs (Figures 4G,H).

Several enzymes coming from the liver in the serum were used as useful biochemical markers for investigating early hepatotoxicity. After 48 h of CCl₄ intoxication, rats demonstrated a substantial increase in the levels of serum ALT, AST, and ALP as compared to the normal control (Figure 3A). The positive control silymarin at a dose of 100 mg/kg was not as effective in the reduction of hepatocellular damage as 100 mg/kg Met-PPx, which reduced (p < 0.001) the rise in the serum level of hepatic enzymes to a remarkable extent. However, it is interesting and noteworthy that in the silymarin-treated group, the level of enzymes was not completely halted down to the control, which represents some damage to the hepatocytes. Interestingly, the Met-PPx substantially stopped (p < 0.001) the rise in the level of serum enzymes and brought it down to control levels, indicating no damage to the parenchyma cells regardless of the CCl₄ treatment. The EtOAc fraction abrogated (p < 0.001) the elevation in serum enzymes. Similarly, the aqueous fraction also inhibited (p < 0.001) the increase in serum enzymes. However, the butanol and the n-hexane fractions did not considerably reduce the hepatic enzymes. These results strongly suggest that the hepatoprotective and free radical–quenching potential of the PP exists in the EtOAc and aqueous fraction due to the existence of bioactive compounds.

MDA is one of the main end products of lipid peroxidation, and its elevated level could be used as an indicator of hepatotoxicity. CCl₄ intoxication significantly elevated the MDA content while reducing GSH and SOD levels in the liver tissues, compared to the normal control group (Figures 5A–C), indicating that the CCl₄ injection induced severe oxidative stress. Treatment with Met-PPx and its EtOAc and aqueous fractions significantly attenuated this and were more effective than silymarin. Hepatic antioxidant enzyme levels were substantially increased (levels similar to those in the normal control) by Met-PPx and its EtOAc and aqueous fractions, which also significantly decreased MDA levels. In contrast, the butanol and the n-hexane fractions did not considerably increase the antioxidant activity and also showed high levels of MDA.

Inflammation of the liver is linked with activation of Kupffer cells (KCs) and their migration into the hepatic cords where they secrete TNF- and IL-6, proinflammatory cytokines (Kiso et al., 2012). A particular KC marker, CD68, was used to monitor KC activation in CCl₄-intoxicated rats. In the normal control liver (Figure 6), a small number (71 ± 10) of quiescent KCs are restricted to the liver sinusoids only (Figures 7A–C). The liver section of the control group did not show considerable CD68 immunopositivity (Figure 7A). In contrast, strong (p < 0.001) CD68 expression (190 ± 8) was observed around the central vein of the CCl₄-intoxicated group (Figures 7D–F). CD68 immunoreactivity was observed mainly around the injured areas, presenting mostly pericentral staining. CD68 immunopositivity increased in the livers of CCl₄-intoxicated rats. Mixed infiltration of inflammatory cells (Figure 7E) was identified from elongated nuclei with DAPI staining around the central vein in the CCl₄-treated group. It is obviously known from Figure 6 that treatment with silymarin has decreased (p < 0.001) the number of Kupffer cells (112 ± 10) compared to the intoxicated group. Moreover, an enhancement in the number of KCs surrounding the central vein and the pericentral region in the liver treated with silymarin was still higher (p < 0.001), relative to the control group (Figures 7G–I). Interestingly, the Met- PPx significantly reduced (p < 0.001) KCs (71 ± 9) around the central vein (Figure 6), similar to the normal control group. After CCl₄ intoxication, the activated KCs were recruited to the site of injury. The Met-PPx administration restricted the KCs to the sinusoidal spaces despite the CCl₄ treatment as supported by the immunohistochemistry (Figures 7J–L). The presence of polyphenolic compounds in the EtOAc fraction might significantly reduce (p < 0.001) CD68-positive (74 ± 8) and limited the KCs to the sinusoids, similar to the Met-PPx group as evident from immunohistochemistry (Figures 7A–C). Likewise, the aqueous fraction completely restricted (p < 0.001) the KCs (75 ± 9) (Figure 6) to the sinusoids (Figures 8D–F). Compared with normal control sections, the butanol group revealed a drastic increase in the number of KCs (153 ± 11) (Figure 6) around the central vein (Figures 8G–I). Similarly, the n-hexane–treated group showed a dramatic increase (141 ± 12) in the number of KCs (Figures 6, 7J–LJ–L7J–L).

To investigate the role of hepatic iNOS in vivo, a CCl₄-induced acute liver model of inflammation was used. In the control group, no apparent detection of iNOS expression was noticed by immunohistochemistry (Figures 9A,B), which was further confirmed from the merging of iNOS and DAPI as shown (Figure 9C). Moreover, quantification of the iNOS immunostained structure was identified and was measured under different conditions, showing a diminished number in contrast to the CCl₄ group (Figure 10). After 24 h of administration of CCl₄, the liver-associated necroinflammation is at its peak, especially ALT. In the CCl₄-intoxicated model group, prominent expression of iNOS was observed and localized to the cytoplasmic content surrounding the hepatocytes’ nuclei (Figures 9D, 10). By double fluorescent microscopy, the enhanced expression and colocalization of iNOS and DAPI were uniformly confined in the surrounding central vein, which represents necrosis of hepatocytes (Figures 9E,F, 10). The positive control silymarin group also presented the expression of iNOS in the central and pericentral hepatocytes (Figures 9G, 10). The iNOS positively stained cells were scattered all around the central vein with signatures of necroinflammation (Figures 9H,I). In contrast, treatment with the ethyl acetate fraction showed negligible expression of iNOS (Figures 9J, 10), whereas iNOS expression was further reduced in the at 100 mg kg⁻¹ in merged image (Figures 9L, 10). Moreover, hepatocyte and Kupffer cells’ nuclei were immune-negative for iNOS in the normal group like the ethyl acetate– and aqueous-treated groups.