Adult male Sprague–Dawley rats weighing 225–300 g were purchased from the Shanghai SLAC Laboratory Animal Co., Ltd., Shanghai, China. The rats were maintained in a facility with specific pathogen-free and controlled temperature (20–22°C), humidity (45–55%), and light (12-h light/12-h dark cycle) conditions, with standard rat chow and water made available ad libitum, except for 1 day of fasting before and after the operation. The experimental procedures were approved by the Institutional Animal Committee of Wenzhou Medical University. All rats were well-cared for throughout the experiment in accordance with the Guide for the Care and Use of Laboratory Animals.

Ginsenoside Rb1 (GRb1) (purity > 98%) was purchased from Shanghai Tauto Biotech Co., Ltd. (Shanghai, China). All reagents and chemicals were procured from local suppliers and were of analytical grade.

After an adaptation period of 1 week, rats were randomly divided into three groups—control group (CG), POFS model group (PG), and GRb1-treated POFS model group (GG)—into four cohorts: post-operative days 1, 3, 5, and 7.

The POFS rat model was induced by major small intestinal resection as described in a previous study (Zhang et al., 2011). After opening the abdomen, the length of the small intestine was measured. We removed 70% of the small intestine starting from 10 cm below the ligament of Treitz and ending at the terminal ileum. End-to-end intestinal anastomosis can then be conducted while the mesentery is repaired (Figure 1B). Before closing the abdomen, the anastomotic leak was detected by squeezing the intestinal contents. Rats in PG and GG received surgery as mentioned above, while rats in the CG went through the same procedure, without any small intestinal resection. All rats received 24-h fasting but water free before surgery.

Rats in the CG and PG were intraperitoneally administered saline at a dose of 4 ml/kg for 3 days before surgery and once daily after surgery until sacrifice. Rats in GG were administered GRb1 dissolved in saline at a dose of 15 mg/kg (4 ml/kg) using the same method as in the other two groups.

Four cohorts of rats including those in the PG, CG, and GG received the open field test (OFT) on days 1, 3, 5, and 7 after surgery, respectively. Once the OFT finished, the rats were sacrificed and samples of hippocampal tissues were prepared after left ventricular perfusion by saline (Figure 1A2). A special cohort of rats was used for the sucrose preference test (SPT). Rats were deprived of water for 12 h water on the day of surgery, and food and water were suspended for 12 h before every subsequent SPT day (post-operative days 3, 5, and 7) (Figure 1A1).

The open field arena was a square black plastic box (100 × 100 × 50 cm³) and used an automated video tracking system (ANY-maze, Shanghai, China). We divided the bottom into 16 grids and defined 4 central grids as the center arena. Each rat was placed individually at the corner of the arena and its behavior monitored for 5 min. The journey (the number of grids the rats walked through), stationary time, and time in the center area of each rat were recorded.

After food and water deprivation for 12 h, each rat was exposed to the sucrose solution (1% w/v) and pure water for 12 h. The positions of the two bottles were exchanged every 6 h to prevent place preference. Fluid consumption was measured by weighing the bottles before and after the test. The sucrose preference percentage (SPP) was calculated using the following formula: SPP (%) = sucrose solution consumption (g)/(water consumption + sucrose solution consumption) × 100%. The same process was performed 7 days before surgery as a habituation.

After the OFT, blood was collected from the postcava and centrifuged at 5,000 rpm for 15 min after allowing to stand for 4 h. Then, the supernatant was collected and placed in a refrigerator at −80°C until subsequent testing. According to the manufacturer’s instructions, the concentrations of interleukin-1β (IL-1β), IL-6, and tumor necrosis factor alpha (TNF-α) in the serum were determined by ELISA kits (PI303, PI328, and PT516; Beyotime, Shanghai, China).

Real-time quantitative polymerase chain reaction (PCR) assays were performed to quantitate the expressions of the inflammatory cytokines IL-6, IL-1β, and TNF-α (n = 8 for the CG, PG, and GG). The β-actin gene, which is a housekeeping gene, was chosen as the internal gene control. Total RNA was extracted from the isolated hippocampus of rats using the TRIzol reagent (Invitrogen, Carlsbad, CA, United States). RNA purity and concentration were assessed with UV spectrophotometry (2.0 > A₂₆₀/A₂₈₀ > 1.8). Complementary DNA (cDNA) was synthesized using RT kit (Toyobo, Osaka, Japan). The PCR reaction mixture was prepared using the SYBR Green Real-time PCR Master Mix-Plus (Toyobo, Osaka, Japan). The primer sequences were designed by the Shanghai Sangon Biological Engineering Technology and Services Co., Ltd. (Shanghai, China). Amplification of cDNA was performed with a 7500 Real-Time Quantitative PCR instrument (Applied Biosystems, Carlsbad, CA, United States). The reaction parameters were incubated at 95°C for 1 min, then 40 cycles at 95°C for 15 s, 60°C for 15 s, and 72°C for 1 min. All reactions were performed in triplicate and normalized using β-actin as an endogenous control gene. Relative data quantitation was performed by the 2^–ΔΔCT method.

Western blotting was performed on proteins from hippocampal tissue homogenates (n = 8 for the CG, PG, and GG). The hippocampal cytoplasm and nuclear protein were extracted using a nuclear extraction kit (Pierce, Rockford, IL, United States) according to the manufacturer’s instructions. Total protein was also extracted from the hippocampus using a protein extraction kit (Beyotime, Nanjing, China). The protein concentrations were determined using a Micro BCA Protein Assay kit (Beyotime, Nanjing, China). Thirty micrograms of protein per specimen was separated on a sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and blotted onto a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA, United States). The membranes were blocked for 1.5 h with Tris-buffered saline with Tween (TBST) containing 5% non-fat dry milk at room temperature and incubated overnight with the primary antibodies at 4°C, washed, and then incubated with the appropriate secondary antibodies [horseradish peroxidase (HRP)-conjugated anti-IgG antibody, 1:5,000; Biosharp, Hefei, China] for 1 h at room temperature. After washing, the protein bands were visualized using enhanced chemiluminescence (ECL) and exposed to X-ray film. After scanning the X-ray film, the pixel density was calculated with Quantity One software ver. 4.6.2 (Bio-Rad, Hercules, CA, United States). The primary antibodies were: anti-IDO (1:200, sc-53978; Santa Cruz Biotechnology, Dallas, TX, United States), anti-p38MAPK (1:1,000, 8960s; Cell Signaling, Danvers, MA, United States), anti-phospho-p38 (p-p38) (1:250, 4511s; Cell Signaling), anti-NF-kappaB/p65 (NF-κB/p65) (1:1,000, 8242s; Cell Signaling), anti NF-kappaB/phosphor-p65 (NF-κB/pp65) (1:250, 3033s; Cell Signaling), and anti-GAPDH (1:1,000; Hangzhou Goodhere Biological Technology Co., Ltd., Hangzhou, China), anti-β-actin (1:1,000; Hangzhou Goodhere Biological Technology Co., Ltd.), or anti-lamin-B1 (1:500; BioWORLD, Dublin, OH, United States) polyclonal antibody used as a loading control.

The hippocampus was dissected out, imbedded in Tissue-Tek (Sakura, Torrance, CA, United States), and frozen in liquid nitrogen (n = 8 for the CG, PG, and GG). Sections of 10 μm thick were cut using a frozen section machine. After fixation in paraformaldehyde (4%), the sections were incubated with the primary antibodies overnight, rabbit anti-NF-κB/p65 (1:100, 8242s; Cell Signaling, Danvers, MA, United States), followed by incubation with the secondary antibody Dylight 488 anti-rabbit (1:500; Abcam, Cambridge, United Kingdom). Immunoreactivity was visualized at the appropriate wavelengths with a light and an epifluorescence microscope (Nikon 80i, Tokyo, Japan).

Samples of hippocampal tissue were weighed and dissected before homogenizing at 4°C with 0.2 M perchloric acid including 100 μm EDTA-2Na in a Teflon/glass homogenizer (n = 8 for the CG, PG, and GG). The homogenate was centrifuged at 4°C for 15 min at 20,000 × g. The supernatant was collected and filtered through a 0.45-μm centrifuge tube filter before injection into the high-performance liquid chromatography (HPLC) system. Determination of the concentrations of tryptophan, kynurenine, and serotonin in hippocampal tissue homogenates was performed using a Hypersil C18 column (250 mm × 4.6 mm i.d., 5 mm particle size; Elite Analytical Instrument Co., Ltd., Dalian, China). The concentrations of kynurenine and tryptophan were determined according to the method described by Wang and Tang (2006) and Herve et al. (1996). Briefly, the mobile phase for fluorescent light detection consisted of 220 mmol/L zinc acetate/acetic acid solution containing 5% (v/v) acetonitrile, pH 6.0. The flow rate was set at 1.5 ml/min. The injected sample volume was 20 μl. The excitation wavelength was 365 nm and the emission wavelength was 480 nm for kynurenine; for tryptophan, these were 254 and 404 nm, respectively. Determination of serotonin concentration was according to the method recorded by Szeitz and Bandiera (2018). Briefly, the mobile phase consisted of 100 mmol/L sodium acetate/acetic acid solution containing 10% (v/v) methanol, pH 5.0. The flow rate was set at 1.0 ml/min. The injected sample volume was 20 μl. The excitation and emission wavelengths were 330 and 250 nm, respectively.

All values were expressed as the mean ± standard deviation (SD). Statistical analysis was performed using two-way ANOVA, followed by post-hoc tests (using the least significant difference test, LSD-t) for multigroup comparisons. A value of p < 0.05 was considered statistically significant.

We used the OFT and SPT to judge whether large abdominal surgery will induce fatigue syndrome and depression-like behaviors and whether GRb1 will reverse them. The results of the OFT showed that, in the early stage after surgery (post-operative day 1), rats in the PG had fewer journeys and longer stationary times on the first day after surgery during the 5-min test (Figures 1D,F). In addition, rats in the PG showed significantly less exposure in the center of the open field arena on post-operative days 1 and 3, which could be improved by GRb1 (Figures 1C,E).

The SPT was used to assess the rats for loss of pleasure. From the results, we found that the SPP was significantly decreased after the large abdominal surgery procedure in the PG rats. Compared with those in the CG, rats in the PG showed obviously lower levels of SPP on post-operative days 1–5, and in the first 3 days after surgery; SPP decreased to below 65%, which reached the criterion for anhedonia. After intervention with GRb1, rats in the GG recovered markedly on post-operative day 3 (Figure 1G).

To investigate post-operative inflammation in both peripheral blood and the CNS and to explore the link between them, we performed ELISA and real-time PCR to assess the levels of inflammatory cytokines in serum and the hippocampus, respectively. Serum cytokine concentrations showed obvious and lasting increases in rats in the PG; conversely, GRb1 decreased the concentrations of IL-1β, IL-6, and TNF-α. Compared with rats in the CG, the serum IL-1β, IL-6, and TNF-α levels in the PG were significantly increased from day 1 to day 7 after surgery, while rats in the GG showed obvious attenuation than those in the PG at different post-operative time points (Figures 2A–C).

The expressions of inflammatory cytokines in the hippocampus revealed the same increasing trend as those in serum, but only showed significant difference in the early period after surgery. On post-operative days 1 and 3, the expressions of IL-6, IL-1β, and TNF-α messenger RNA (mRNA) in the PG rats were significantly upregulated compared with those of rats in the CG. In addition, GRb1 downregulated the mRNA expressions of IL-1β on post-operative day 1 and of IL-6 and TNF-α on post-operative day 3 (Figures 2D–F). These results suggest that 70% of major small intestinal resection will result in systemic inflammation. Dramatic and long-lasting increases of peripheral blood cytokines may act as the link between surgery and inflammation in the hippocampus. In addition, hippocampal inflammation was serious on post-operative days 1 and 3, then gradually recovered to the CG level; GRb1 played an alleviating role in the first 3 days after surgery.

To further explore the mechanism of activation of the inflammation pathway in the hippocampus of POFS rats, we examined the protein expressions of phospho-p38MAPK and NF-κB/phospho-p65. In addition, NF-κB/p65 nuclear translocation was detected by both immunofluorescence and nuclear protein extraction followed by Western blot. Rats in the PG, which had large abdominal operation, showed significantly increased phospho-p38MAPK expression than did rats in the CG on post-operative days 1, 3, and 5. Compared with rats in the PG, those in the GG obviously downregulated the phospho-p38MAPK protein expression on days 1 and 3 after surgery (Figures 3A,B).

The expression of NF-κB/phospho-p65 had a similar change to phospho-p38MAPK. Compared with rats in the CG, those in the PG revealed significantly upregulated expressions on post-operative days 1 and 3, while the GG rats showed a reversion effect on day 5 after surgery (Figures 3C,F). The ratio of NF-κB/p65 protein expression in the nucleus and cytoplasm illustrates the activation of the NF-κB/p65 pathway from another direction. The PG rats showed significantly higher ratios than did the CG rats on post-operative days 1, 3, and 5, while GRb1 obviously inhibited the nuclear translocation of NF-κB/p65 on post-operative days 1 and 3 (Figures 3D,G). In immunofluorescence, an obvious upregulation of the translocation of NF-κB/p65 from the cytoplasm into the nucleus in the PG has been detected, accompanied with a modest upregulation in the GG and less translocation in the GG (Figure 3E). These results revealed that large abdominal surgery and sequenced neuroinflammation will result in p38MAPK and NF-κB/p65 phosphorylation and promote NF-κB/p65 to translocate from the cytoplasm into the nucleus.

Based on the effect of the inflammatory response on tryptophan metabolism, we examined the expression of the key enzyme IDO in the tryptophan pathway and its metabolism productions: kynurenine and serotonin. The protein expression of IDO was detected by Western blot in the hippocampus. The results showed that large abdominal operation increased the expression of IDO significantly on days 1–5 after surgery. Meanwhile, GRb1 obviously downregulated IDO expression on post-operative day 3 (Figures 4A,B).

Indoleamine 2,3-dioxygenase (IDO) is the rate-limiting enzyme in the tryptophan–kynurenine metabolic pathway. HPLC was used to detect the levels of tryptophan, kynurenine, and serotonin. Firstly, tryptophan concentration in the hippocampus of the PG rats was shown to be significantly increased on post-operative days 1 and 3, while rats in the GG showed a decreasing trend on the first day after surgery (Figure 4C). Secondly, the level of kynurenine was enhanced significantly in the PG compared to the CG on days 1 and 3 after surgery and was attenuated at the same time points in rats in the GG (Figure 4D). Serotonin in the hippocampus showed a relatively stable level and only had an obvious increase on post-operative day 1 in the PG rats (Figure 4E). Furthermore, although tryptophan, the metabolism material in the IDO-mediated pathway, had been upregulated in the early stage after surgery, the ratio of kynurenine/tryptophan still showed a dramatic increase on post-operative days 1 and 3; the ratio in the GG rats was significantly lower than that in the PG rats at these time points (Figure 4F).