A total of 32 male rabbits (age, 4–5 months; weight, 1.80–2.20 kg), were provided by the Department of Experimental Animals in University of South China (Hengyang, China). The rabbits were housed at room temperature in 50–60% humidity with one rabbit per cage in a 12 h dark/light cycle. The rabbits were provided with water and commercial rabbit pellets ad libitum. The animals were divided at random into control, sham, sepsis and PAG groups (n=8 rabbits/group). All animal experiments were approved by the Ethics Committee of The Second Affiliated Hospital of Nanhua University (Hengyang, China).

The rabbit model of urine-derived sepsis was established as described previously (21). Briefly, a vertical incision was made along the left abdominal rectus muscle, and the middle section of ureter was freed and ligated. A suspension of Escherichia coli [E. coli; 10⁸ cells/ml; ATCC, 25922; American Type Culture Collection, Manassas, VA, USA;] was then injected into the ureter at the proximal end of the ligation site at a dose of 0.5 ml/kg body weight. The incision was subsequently sutured and rabbits were fed a normal diet. Food was given at the indicated time points but the rabbits had free access to water. The urine-derived sepsis model was additionally established in rabbits in the PAG group, with the addition of a single intraperitoneal injection of 37.5 mg/kg body weight PAG immediately following the surgical operation (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). The same surgical procedure was performed in rabbits in the sham group, excluding the injection of the bacteria. Rabbits in the control group were fed on normal diet without any additional treatments.

For each rabbit, the rectal temperature (using a rectal thermometer), heart rate (using a watch and stethoscope) and respiratory rate (by counting the nose hair swing frequency) were recorded prior to surgery, and at 12, 24, 36 and 48 h following surgery.

A total of 1–2 ml peripheral blood samples were collected from the ear vein prior to surgery, and at 12, 24, 36 and 48 h following surgery. White blood cells were counted using a hemocytometer.

Peripheral blood samples were collected prior to surgery, and at 12, 24, 36 and 48 h following surgery. The levels of creatinine and urea nitrogen were measured using an automatic biochemical analyzer (AU800; Olympus Corporation, Tokyo, Japan).

At 48 h following surgery, the rabbits were anesthetized and their left kidneys were removed. Kidney tissues were subsequently fixed with 4% formaldehyde at room temperature for 24–48 h, embedded in paraffin and cut into 5-µm-thick sections. Tissues were then subject to H&E staining at 60°C for 30 min, and visualized under a light microscope at magnification, ×400. In total, 10 fields of view were randomly selected and observed from each section.

At 48 h following surgery, the rabbits were anesthetized, and approximately 5–7 ml venous blood from the left kidney was collected and centrifuged at 1,500 × g at room temperature for 10 min. A total of 0.5 ml zinc acetate solution (1%) was subsequently added to 0.1 ml test plasma, followed by 0.5 ml dimethyl-p-phenylenediamine sulfate/HCl solution (20 mM/7.2 M) and 0.4 ml ferric chloride/HCl solution (30 mM/1.2 M). The solution was incubated at room temperature for 20 min, and 1 ml trichloroacetic acid (10%) and 2.5 ml ddH₂O were subsequently added. Samples were then centrifuged at 2,500 × g at room temperature for 8 min, the supernatant was collected and the absorbance at 679 nm was read using a spectrophotometer. The H₂S concentration was calculated according to a standard curve.

At 48 h following surgery, rabbits were anesthetized and the left kidneys were removed. A total of ~200 mg tissue was mixed with 10% (w/v) 50 mM/l potassium phosphate buffer (pH 6.8), and then homogenized on ice. Following centrifugation at 4°C at 2,500 × g at room temperature for 10 min, the supernatant was collected, and 80 µl L-cysteine (0.5 M) plus 3.52 ml 5-pyridoxal phosphate (0.5%) was added to 400 µl tissue homogenate. A total of 0.5 ml zinc acetate (1%) was then added to the central chamber; in which folded filter paper was placed. The reaction system and the central chamber were placed in a 25-ml conical flask filled with nitrogen, and incubated in a water bath at 37°C for 90 min. The reaction was terminated by adding 0.5 ml trichloroacetic acid (50%), and the flask was sealed with a paraffin membrane. Following a further 60 min incubation at 37°C, the mixture in the central chamber was transferred to a fresh tube containing 3.5 ml ddH₂O. A total of 0.5 ml dimethyl-p-phenylenediamine sulfate/HCl solution (20 mM/7.2 M) and 0.4 ml ferric chloride/HCl solution (30 mM/1.2 M) was added, and the solution was incubated at 37°C for 20 min prior to centrifugation at 1,500 × g at 37°C for 5 min. The absorbance at 670 nm was read using a spectrophotometer, and the H₂S concentration was calculated according to a standard curve. CSE activity (nM/min/mg) was set as the concentration of H₂S produced during 1 min in 1 mg kidney tissue.

The expression levels of NF-κB, IL-6, and TGF-β1 in kidney tissues were determined by western blot analysis. The kidney tissue (harvested 48 h post surgery) was lysed with a PBS lysis buffer (0.24 g potassium dihydrogen phosphate, 1.44 g disodium hydrogen phosphate, 0.2 g potassium chloride, 8 g sodium chloride in 1,000 ml solution; pH 7.4) and total protein was extracted according to the manufacturer's protocol. Sample protein concentrations were determined using the bicinchoninic acid assay method (Beyotime Institute of Biotechnology, Shanghai, China). A total of 50 µg protein sample was loaded per lane and separated by 10% SDS-PAGE, and then electronically transferred onto a polyvinylidene difluoride membrane. The membranes were subsequently blocked with 5% non-fat milk at 37°C for 1 h prior to incubation with primary antibodies against NF-κB (bsm-33117M; 1:500 dilution), IL-6 (bs-4587M; 1:500 dilution), TGF-β1 (bsm-33287M; 1:500 dilution), or β-actin (bsm-33036M; 1:5,000 dilution) (all purchased from BIOSS, Beijing, China) at 4°C overnight. Following washing with Tris-buffered saline and Tween-20 solution, the membrane was incubated with a goat anti-mouse immunoglobulin M/horseradish peroxidase antibodies (bs-0368G-HRP; 1:10,000; BIOSS) at room temperature for 2 h. Protein bands were visualized using an enhanced chemiluminescence method (Beyotime Institute of Biotechnology), according to the manufacturer's protocol and band densities were analyzed using the GSD8000 image analysis system (UVP, Cambridge, UK).

The results are expressed as the mean ± standard deviation. The SPSS software program (version, 18.0; SPSS, Inc., Chicago, IL, USA) was used to perform statistical analyses. The results were analyzed by one-way analysis of variance followed by a post hoc least significant difference test. P<0.05 was considered to indicate a statistically significant difference.

Monitoring of vital signs, routine blood examinations and kidney function tests were performed in rabbits from the control, sham, sepsis and PAG groups prior to surgery, and at 12, 24, 36 and 48 h following surgery. Based on the monitoring of vital signs, no significant differences in rectal temperature, heart rate and respiratory rate among these four groups was observed prior to surgery (Fig. 1A-C). By contrast, the rectal temperature, heart rate and respiratory rate were significantly elevated at 12 h following surgery in the sepsis and PAG groups compared with the sham group (P<0.05); however, a gradual plateauing in the level of these factors was observed at 48 h following surgery (Fig. 1A-C). In the sepsis group, the rectal temperature, heart rate and respiratory rate of rabbits were significantly elevated at 12 h following surgery compared with the sham group (P<0.05), and continued to increase at 24, 36 and 48 h following surgery. In addition, the rectal temperature, heart rate, and respiratory rate of rabbits in the PAG group were significantly elevated following surgery when compared with the sham group and the sepsis group (P<0.05; Fig. 1A-C). There was no significant difference between the control group and the sham group.

Similar results were observed for the routine blood examinations and kidney function tests in the control, sham, sepsis and PAG groups. Prior to surgery, no significant differences in the white cell count and the levels of creatinine and urea nitrogen among these four groups were observed (Fig. 1D-F). In the sepsis and PAG groups, the white cell counts and the levels of creatinine and urea nitrogen were significantly elevated following surgery when compared with the sham group at the corresponding time points (P<0.05), with increased levels of these factors observed in the PAG group compared with the sepsis group at all time points (Fig. 1D-F). No significant differences were observed between the control group and the sham group. These results indicated the successful establishment of a rabbit model of urine-derived sepsis, which was considered to be suitable for subsequent experiments. In addition, PAG treatment was demonstrated to exacerbate sepsis-associated pathophysiological alterations.

The effects of PAG on the concentration of serum H₂S and kidney CSE activity in a rabbit model of urine-derived sepsis were subsequently investigated. As demonstrated in Fig. 2A, no significant difference in the serum H₂S concentration between the control and sham groups at 48 h following surgery was observed. By contrast, the serum H₂S concentration was significantly reduced in the sepsis group at 48 h following surgery when compared with the sham group (P<0.05; Fig. 2A). In addition, a significant reduction the serum H₂S concentration was observed in the PAG group at 48 h after surgery when compared with the sham and sepsis groups (P<0.05; Fig 2A). Similar results were observed for the kidney CSE activity detection. When compared with the sham group, CSE activity in kidney tissues of rabbits in the sepsis group were significantly decreased at 48 h following surgery (P<0.05), with a greater reduction in kidney CSE activity in the PAG group compared with the sham and sepsis groups (P<0.05; Fig. 2B). These results suggest that PAG treatment further decreased the serum H₂S concentration and kidney CSE activity in a rabbit model of urine-derived sepsis, which suggests that PAG exacerbates disease progression.

In order to investigate the effects of PAG on the formation of histopathological kidney lesions in a rabbit model of urine-derived sepsis, H&E staining was performed. As indicated in Fig. 3, no obvious histopathological alterations in the kidney tissues of rabbits in the control and sham groups were observed (Fig. 3). However, in the sepsis group, H&E staining indicated obvious structural abnormalities in the kidney tissues, including an abnormal glomerulus morphology, glomus atrophy and deformation and kidney capsule expansion. In addition, edema and necrosis were observed in the kidney tubular epithelial cells, with detached necrotic cells blocking the lumen. Furthermore, the size of the renal capsule and renal tubular lumen were significantly expanded and significant kidney interstitium hyperemia and edema, and abundant inflammatory cell infiltration were observed in the kidney tissues of rabbits in the sepsis group (Fig. 3). When compared with the sepsis group, these histopathological alterations were significantly exacerbated in the PAG group (Fig. 3). These results provide evidence to suggest that treatment with PAG worsens the development of histopathological kidney lesions in a rabbit model of urine-derived sepsis.

To investigate the effects of PAG on the expression levels of NF-κB, IL-6 and TGF-β1 in the kidney tissues of rabbits with urine-derived sepsis, western blot analysis was performed. As indicated in Fig. 4, no significant difference in the protein expression levels of NF-κB, IL-6 and TGF-β1 was observed between the control and sham groups at 48 h following surgery. However, the expression levels of NF-κB and IL-6 in kidney tissues were significantly elevated in the sepsis group when compared with the sham group (P<0.05). The expression of NF-κB and IL-6 were significantly elevated in the PAG group when compared with sham and sepsis groups (P<0.05; Fig. 4). In addition, TGF-β1 protein expression was significantly elevated in the sepsis group compared with the sham group (P<0.05). However, compared with the sepsis group, the expression level of TGF-β1 was significantly reduced in the PAG group (P<0.05); however, its expression remained significantly higher than the sham group (P<0.05). No significant differences were observed between the control group and the sham group. These results suggest that PAG treatment may significantly increase the expression levels of NF-κB and IL-6, and decrease TGF-β1 expression in kidney tissues, in a rabbit model of urine-derived sepsis.