Paracetamol, silymarin (MilliporeSigma), gentamicin (PT. Bernofarm Pharmaceutical Company), 0.9% sodium chloride (PT. Widatra Bhakti), diethyl ether (PT. Brataco), 10% formalin solution, xylene, paraffin, 70% ethanol, diethyl ether, hematoxylin-eosin stains, pulvis gummi arabicum and potassium chloride (EMSURE^®; Merck KGaA) were of analytical grade. Kits for the estimation of total albumin (TA), aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), total bilirubin (TB), total cholesterol (TC), total protein (TP), serum creatinine (SCr), serum urea (SU), uric acid (UA) were from PT. Wacana Indo Mitra, tumor necrosis factor alpha (TNF-α), interferon gamma (IFN-γ) were from PT. Biolab Science Universal) and superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPx) and glutathione (GSH) were from Sigma-Aldrich (Merck KGaA).

A total of 10 kg of fresh C. costata leaves were bought from the Pancur Batu traditional market in North Sumatra, Indonesia in March 2022. The plant was identified at the Herbarium Medanense, Universitas Sumatera Utara, Indonesia (voucher number: 183/MEDA/2022). The cleaned C. costata leaves were brought to Pharmacognosy Laboratory, Universitas Buana Perjuangan Karawang for the extraction procedure. A total of 5 kg of C. costata powder was macerated in 70% ethanol three times in 24 h. The liquid extract was gathered and concentrated at 50˚C using a rotary evaporator (32).

Potassium bromide (KBr) pellets were mixed with CcE and the results were evaluated with a Shimadzu IRPrestige-21 FT-IR Spectrophotometer (Shimadzu Corporation). At a resolution of 4 cm^-1, the spectra were collected in the 400-4,000 cm^-1 range.

For randomization, an identification number was first assigned to each rat and then randomization was performed, which generated random numbers and allocated rats to study groups. Randomization was performed using online software (https://www.graphpad.com/quickcalcs/randomize1/). Meanwhile, in blinding during the experiment, alphanumeric codes were used to identify vials and syringes and each rat was given a number. Then, each sample code was placed in a sealed envelope and revealed at the end of the experiment.

A total of 44 male Wistar rats in good health, weighing 150-250 g and 8-12 weeks old, were employed in the hepatoprotective and nephroprotective research. Rats were acquired from CV. Mitra Putra Animal. The rats were kept at a 12-h light/dark cycle in the Pharmacology Laboratory at Department of Pharmacology and Clinical Pharmacy, Faculty of Pharmacy, Universitas Buana Perjuangan Karawang with a temperature range of 20-26˚C and 30-70% humidity. In addition, the experimental animals had unrestricted access to drinking water and normal pellets. The human endpoints established for this study were deteriorating body condition, weight loss, the inability to rise or ambulate and the presence of labored respiration. No animal reached this stage.

A hepatotoxicity model produced by paracetamol was used to investigate hepatoprotective activity. The experimental rats were housed in six groups of four rats each. Group I, II and III served as normal, negative and positive control given 1% w/v pulvis gummi arabicum (PGA) suspension, paracetamol at a dose of 1,000 mg/kg and silymarin at a dose of 50 mg/kg, respectively. Groups IV, V and VI were each given CcE at doses of 100, 200 and 400 mg/kg orally. Determination of the CcE doses in this study refers to previously published research (21,23). The experimental treatments and group designs were:

Group I (normal control): For 21 days, rats were given 1% w/v PGA suspension orally (10 ml/kg/day).

Group II (negative control): For 7 days, rats received a dose of 1,000 mg/kg of paracetamol.

Group III (positive control): For 21 days, rats received 50 mg/kg of silymarin.

Group IV (CcE 100): CcE was administered to rats for 21 days at a dose of 100 mg/kg.

Group V (CcE 200): CcE was administered to rats for 21 days at a dose of 200 mg/kg.

Group VI (CcE 400): CcE was administered to rats for 21 days at a dose of 400 mg/kg.

Rats in groups II-VI were given paracetamol induction from days 15 to 21 at a dose of 1,000 mg/kg orally (33). Meanwhile, on the 22nd day, the treatment groups were anesthetized with diethyl ether at a dose of 4 ml. Diethyl ether was administered to rats by simple ‘open-drop’ methods using an ether-impregnated cotton ball in a bell jar for induction followed by inhalation via a simple face cone. The parameters monitored to ensure the animals were anesthetized after diethyl ether administration were ataxic, recumbent, with a steady, slow respiratory rate, immobile and loss of palpebral blink reflex. After a cardiac puncture, 2 ml of blood was extracted and placed in a tube holding heparin. The rats were euthanized by cervical dislocation. The liver was immediately removed and washed with cold 0.9% NaCl solution to remove the blood before weighing. For histological analysis, a portion of the liver's median lobe was preserved in a 10% formalin solution (fixation was carried out for 24 h at room temperature 20-22˚C) (34). Using a motor-driven Teflon pestle, homogenate was prepared for the liver antioxidant enzyme level test by combining one gram of wet tissue with 9 ml of 1.25% KCl. The homogenate was centrifuged for 10 min at 4^˚C at 2,737 x g to extract the supernatant, which was then used to measure the levels of SOD, CAT, GPx and GSH (35).

A nephrotoxicity model produced by gentamicin was used to perform the nephroprotective activity test. Random selection was used to choose five groups of four rats each from the experimental animals. Groups I and II as normal and negative controls were given a 1% w/v PGA suspension and gentamicin at a dose of 80 mg/kg, respectively. Meanwhile, groups III, IV and V were given treatment using CcE at 100, 200 and 400 mg/kg, respectively. The experimental treatments and group designs were as follows:

Group I (normal control): Rats were given 1% w/v PGA suspension orally (10 ml/kg/day) for 8 days.

Group II (negative control): Gentamicin was administered to rats for 5 days at a dose of 80 mg/kg.

Group III (CcE 100): For 8 days, rats received a 100 mg/kg dosage of CcE.

Group IV (CcE 200): For 8 days, rats received a 200 mg/kg dosage of CcE.

Group V (CcE 400): For 8 days, rats received a 400 mg/kg dosage of CcE.

Gentamicin induction was administered intraperitoneally to rats in groups II-V at a dose of 80 mg/kg from days 4-8(36). On the ninth day, the rats in each treatment group were anesthetized using diethyl ether at a dose of 4 ml. Then 2 ml of blood was quickly collected into a heparin tube through a cardiac puncture and rats were euthanized by cervical dislocation. The kidney was cleaned with a cold 0.9% NaCl solution to remove blood and foreign tissue. This was followed by weighing and preserving the organs in 10% formalin solution for histopathological examination (fixation was carried out for 24 h at 20-22˚C) (37).

Fresh rat blood samples were centrifuged for 20 min at 503 x g and at 22˚C to produce blood serum. The serum was put in an Eppendorf tube and its levels of ALT, AST, TB, ALP, TC, TA, TP, SCr, SU and UA were promptly measured. In this procedure, commercial kits were used in accordance with the manufacturer's instructions [cat. nos. : ALT (32941-05121), AST (31335-05121), TB (3417012999-AL2-175423984), ALP (32918-05121), TC (3417012020-LAH-176618380), TA (3417012020-LAH-176587091), TP (3417012020-LAH-176626657), SCr (3417012020-LAH-176623909), SU (3417012020-BSS-211916981), UA (3417012999-LAB-205299812)] and a HumaLyzer 2000 photometer was used for measurement (PT. Sali Polapa Bersama).

TNF-α and IFN-γ levels were measured in the present study using the ELISA technique. The collected serum was immediately analyzed using a commercially available ELISA kit [cat. nos. : TNF-α (MBS2707992) and IFN-γ (MBS2708210, PT. Biolab Science Universal] containing a microtiter plate coated with specific antibodies against TNF-α and IFN-γ standards as well as a washing buffer and horseradish peroxidase (HRP) conjugate. Meanwhile, an automatic microplate reader recorded optical density at 450 nm (ELx50; BioTek; Agilent Technologies, Inc.).

After being cleaned during the autopsy, the water content in liver and kidney tissue samples is removed using an alcohol dehydration process. Next, clearing was performed using xylene to remove alcohol and make the tissue transparent. Then, paraffin penetration was performed to make the tissue harden at room temperature and make it easier to cut using a microtome. Paraffin blocks were sectioned at 3.4-4.6 µm and the slides were deparaffinized in xylene, followed by H&E staining (at 30˚C: hematoxylin ~10 min, eosin 2 min). A 100x objective lens on a light microscope (BX-51; Olympus Corporation) with a connected camera (Olympus Q Color-5; Olympus Corporation) and computer connection was used to view the slides at a total magnification of x1,000. A pathologist assessed and rated the liver and kidney sections based on the degree of damage, somewhat modified from Zakaria et al (34).

The experimental results were shown using the mean ± standard error of the mean. One-way analysis of variance was used to examine the variations in the means of the variables that were measured. This was followed by Tukey's post hoc test using GraphPad Prism version 8 (Dotmatics). P<0.05 was considered to indicate a statistically significant difference. Sample size determination was based on Federer calculation formula, which is (t-1) (n-1) ≥15; where t is the number of the groups and n is the experimental animal per group. (6-1) (n-1) ≥15 -> n≥4, for the testing of hepatoprotective and (5-1) (n-1) ≥15 -> n≥4.75 for the testing nephroprotective properties. According to this calculation, the minimum sample size was four experimental animals in each treatment and control group.

FT-IR revealed that there were several distinct functional groups by identifying 27 peaks for CcE. There were obvious peaks at 1,201.51, 1,444.98 and 1,515.13 cm^-1, showing C-O bending mode. This finding demonstrated the presence of a number of chemicals, including ethers, alcohols, esters and carboxylic acids. Furthermore, amines (N-H stretching), alcohol (O-H stretching), alkanes (C-H stretching), alkynes (C≡C stretching), carboxylic acid (C=O stretching), alkenes (C=C stretching) and imines (C=N) were among the functional groups found in a range of peaks that extended from 3,333.13 to 1,606.32 cm^-1. Fig. 1 shows the results of FT-IR analysis of CcE.

Based on the present results, administration of paracetamol (1,000 mg/kg) to rats increased AST, ALT, ALP, TB and TC levels and also decreased TA and TP (P<0.001-<0.0001) when compared with normal controls. Pretreatment with CcE at all doses caused a significant decrease (P<0.05-<0.0001) in increasing AST, ALT, ALP, TB and TC, levels induced by paracetamol. Furthermore, the administration of CcE at all doses also caused a significant increase (P<0.01-<0.0001) in decreasing TA and TP levels induced by paracetamol. The pretreatment with silymarin (50 mg/kg) had an improved effect on changes in liver biochemical serum parameters compared with CcE. Additionally, compared with a normal control group, there was a statistically significant increase in liver weight (P<0.01) after the administration of paracetamol (1,000 mg/kg). Rats' liver weight significantly decreased (P<0.01) after receiving pretreatment with silymarin (50 mg/kg) or CcE at all dosages when compared with the paracetamol group. Table I shows the effects of pretreatment with CcE on liver function parameters and liver weight of rats.

The findings demonstrated that, in comparison with normal controls, the administration of paracetamol (1,000 mg/kg) significantly reduced the activities of SOD, CAT, GPx and GSH in liver tissue (P<0.001-<0.0001). The pretreatment with silymarin (50 mg/kg) and CcE at all doses showed a significant increase (P<0.05-<0.0001) in SOD, CAT, GPx and GSH activities when compared with paracetamol group. Therefore, CcE triggered hepatoprotective activity through the activation of endogenous enzymatic antioxidant systems. Fig. 2 shows the effects of the extracts on liver antioxidant enzymes.

Based on the present results, administration of paracetamol (1,000 mg/kg) induced a substantial increase in TNF-α and IFN-γ levels (P<0.0001) when compared with normal control. Pretreatment with CcE at all doses caused a significant decrease in increasing the levels of TNF-α (P<0.01-<0.0001) and IFN-γ (P<0.01-<0.001) induced by paracetamol. However, pretreatment with silymarin (50 mg/kg) had an improved effect on decreasing TNF-α levels (P<0.0001) and IFN-γ (P<0.001) than CcE. The effect of CcE on TNF-α and IFN-γ levels in rats with paracetamol-induced hepatotoxicity is depicted in Fig. 3.

Fig. 4A illustrated the cellular architecture with clear cells, sinusoidal gaps and central veins observed in the histopathological analysis of liver slices in the normal control group. However, the paracetamol group showed the most severe damage to cellular architecture, with centrilobular necrosis, hyperplasia, vascular and cellular degeneration, inflammation, polymorphonuclear aggregation, extensive lymphocyte infiltration and loss of cellular boundaries (Fig. 4B). Pretreatment with silymarin (50 mg/kg) showed complete improvement in cellular architecture, such as necrotic hepatocyte patches (Fig. 4C). In comparison with the paracetamol group, the pretreatment with CcE at all doses resulted in a lobular pattern that was nearly normal with mild degrees of necrosis and lymphocyte infiltration (Fig. 4D-F). Table II shows the histopathological scores of the changes.

In comparison with normal controls, the administration of gentamicin (80 mg/kg) to rats resulted in a significant rise (P<0.001) in the levels of SCr, SU and UA. Gentamicin-induced levels of SCr, SU and UA were significantly (P<0.05-<0.01) reduced after pretreatment with CcE at all dosages. Rat kidney weight increased significantly (P<0.01) after receiving gentamicin (80 mg/kg) in comparison with the normal control group. Furthermore, in comparison with the gentamicin group, the pretreatment with CcE at all dosages resulted in a significant (P<0.01) drop in the kidney weight of the rats. Table III shows the effects of CcE pretreatment on kidney function parameters and kidney weight of rats.

Compared with normal controls, the administration of 80 mg/kg of gentamicin resulted in a statistically significant rise (P<0.0001) in the levels of TNF-α and IFN-γ. The levels of TNF-α (P<0.01-<0.0001) and IFN-γ (P<0.01-<0.001) generated by gentamicin were significantly reduced following pretreatment with CcE at all dosages. Fig. 5 illustrates how CcE affects TNF-α and IFN-γ levels in rats that have gentamicin-induced nephrotoxicity.

When evaluated histopathologically, the kidney sections from the normal control group revealed normal tubules and glomeruli without any evident abnormalities (Fig. 6A). The gentamicin group showed severe acute glomerular and tubular necrosis, characterized by total obliteration of the tubular lumen, as well as intertubular hemorrhage and acute leukocyte infiltration (Fig. 6B). Meanwhile, pretreatment with CcE at all doses showed normal glomeruli, relatively normal tubular dilation, no interstitial edema and capillary congestion when compared with the gentamicin group (Fig. 6C-E). Table IV shows the histopathological scores of the changes.