Tilapia skin was donated sincerely by Guolian (Zhanjiang, China). CP was provided via Sigma-Aldrich (St. Louis, MO, United States). Superoxide dismutase (SOD) kit, malondialdehyde (MDA) kit, glutathione peroxidase (GSH-Px) kit, and bicinchoninic acid (BCA) kit were purchased from Jiancheng Institute (Nanjing, China). The tumor necrosis factor-alpha (TNF-α) and interleukin 1 beta (IL-1β) commercially mouse ELISA kit were bought from Invitrogen (Carlsbad, CA, United States). The in situ TUNEL kit was provided by Roche Group (Basel, Switzerland). Rabbit antimouse ionized calcium-binding adaptor molecule-1 (Iba-1) primary antibody was purchased from Wako (Tokyo, Japan). Mouse antimouse B-cell lymphoma-2 (Bcl-2), caspase-3, and Bcl-2-Associated X (BAX) primary antibodies were donated via Santa Cruz (Dallas, TX, United States). Rabbit antimouse kelch-like ECH-associated protein 1 (Keap1) and nuclear factor E2-related factor (Nrf2) primary antibodies were bought from Boster (Wuhan, China). Rabbit antimouse doublecortin (DCX) primary antibody was obtained from Abcam (Cambridge, United Kingdom). Rabbit antimouse heme oxygenase-1 (HO-1), brain-derived neurotrophic factor (BDNF), cAMP response element-binding protein (CREB), and phosphorylated CREB (p-CREB) primary antibodies were bought from Cell Signaling Technology (Danvers, MA, United States).

The preparation of TSP follows our previously published papers, which are briefly described as follows (17): tilapia skins were hydrolyzed by neutrase (Ryon Biological, Shanghai, China) at 50°C for 90 min with 0.3% under a pH 7.0, and were hydrolyzed by alcalase (Ryon Biological, Shanghai, China) at 55°C, pH 9.0 for 90 min with 0.3%. Hereafter, the hydrolysis process was interrupted via heating at 90°C for 10 min. After passing through a 10000 Da ultrafiltration membrane, TSP was obtained by freeze-drying. The amino acid composition of TSP is as follows (g per 100 g): glycine (22%), proline (10.10%), arginine (8.23%), alanine (7.86%), glutamic acid (6.83%), lysine (3.45%), leucine (3.10%), serine (3.01%), threonine (2.98%), aspartic acid (2.96%), valine (2.16%), phenylalanine (2.10%), isoleucine (1.06%), tyrosine (0.86%), histidine (0.77%), methionine (0.35%) (17). The content of 180–1000 Da molecular weight of TSP was 71.86%. Among them, 34.05% were 500–1000 Da, 37.78% were 180–500 Da, and 5.75% were below 180 Da (17).

Male C57BL/6 mice (aged 8 weeks) were bought from Guangdong Medical Laboratory Animal Center (Guangzhou, China). Five animals were housed per cage in an environment with a temperature of 23 ± 2°C and 40–70% humidity. The mice receive 12 h of light and 12 h of the dark cycle with adequate food and water. Mice were randomly allocated to five groups (n = 15 per group): the normal control group (marked as “Control”), CP model group (marked as “Model”), 250 mg/kg/d TSP group (marked as “Model + TSP250”), 500 mg/kg/d TSP group (marked as “Model + TSP500”), and 1000 mg/kg/d TSP group (marked as “Model + TSP1000”). After 1 week of the adaptation period, CP was dissolved in saline solution and administered intraperitoneally (ip) for 2 weeks (10 mg/kg/d) to mice in Model and three doses of TSP groups. The mice in the Control group were received normal saline (1 mL/100 g, ip) for 2 weeks. Three doses of TSP were dissolved in distilled water and treated intragastrically (ig) to mice in three doses of TSP groups for 30 days, respectively. Moreover, the mice in the Control and Model groups were received deionized water (1 mL/100 g, ig) for 30 days. After the behavioral experiments, the hippocampus was collected for testing under pentobarbital sodium anesthesia. Three mice per group were randomly selected and transcardially perfused with PBS and 10% paraformaldehyde during anesthesia. The brain was dehydrated by a sucrose gradient and fixed with the optimal cutting temperature (OCT) compound for analysis. The schematic graph of the experimental process is exhibited in Figure 1A.

The hippocampal tissue was homogenized in phosphate-buffered saline (PBS, 10% w/v) and centrifuged (10000 g, 12 min) to obtain the supernatant. Additionally, the protein concentration was tested via the BCA commercial kit. The resulting supernatant was measured the SOD activity, MDA content, TNF-α level, and IL-1β level using the commercial kit, and the operating procedures followed the instruction manual.

The immunofluorescence experiment was followed the previous pattern with minor modifications (25). The sections (20 μm thick) were mounted on glass slides and washed for 15 min in PBS containing 0.2% Tween-20 (PBST). The sections were blocked in PBST containing 10% normal horse serum for 1 h and reacted with the primary antibody in blocking buffer overnight. The sections were rewashed for 15 min in PBST to remove unbound antibodies. What is more, the sections were incubated with FITC-conjugated secondary antibodies for 2 h, and the sections were rewashed for 15 min in PBST. The slices were reacted with 4, 6-diamidino-2-phenylindole (DAPI) for 2 min, and the sections were washed for 5 min in PBST. After mounting in glycerol, imaging was performed using a DM2500 Fluorescence Microscope (Leica, Wetzlar, Germany). The mean fluorescence intensity of positive cells was counted via the Image-Pro Plus 6.0 software.

Some DNA structure-specific endonucleases are activated when neurons undergo apoptosis, and the internucleosomal genomic DNA is cut via these endonucleases. The exposed 3’-OH of broken DNA can be catalyzed by terminal deoxynucleotidyl transferase (TdT) to bind fluorescein-labeled deoxyuridine triphosphate (dUTP), which an optical microscope can detect. The three brain slides of each sample were randomly selected. The brain slices (20 μm thick) were placed for 20 min at 25°C, fixed in PBS containing 4% paraformaldehyde solution for 25 min. Next, the slices were washed with PBS for 10 min and reacted with proteinase K solution (20 mg/ml) for 15 min, and rinsed with PBS for 10 min. The TdT labeling buffer was dropped to the brain slices, incubated for 60 min at 25°C, and washed with PBS for 30 min. The tissue was reacted with DAPI for 2 min and detected via a DM2500 Fluorescence Microscope (Leica, Wetzlar, Germany). All operations were implemented in a dark and wet environment. The number of TUNEL-positive neurons was calculated by the Image-Pro Plus 6.0 software.

The Western blotting experiment referred to previous studies and modified it appropriately (26, 27). Trivially, hippocampal tissues were homogenized in 300 μL of radio-immunoprecipitation assay (RIPA) buffer containing phosphatase inhibitor and centrifuged at 10000 g for 12 min. The protein concentration was tested via the BCA kit. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred immediately to the polyvinylidene fluoride (PVDF) film, and placed with blocking 5% nonfat dry milk in Tris-buffered saline containing 0.2% Tween-20 (TBST) for 1.5 h. Moreover, The PVDF film was reacted with the primary antibody for 12 h. After TBST washing, the PVDF film was incubated with a horseradish peroxidase-conjugated secondary antibody for 1.5 h. After the rewash, ECL was used for the Western blot experiment. All details were performed at 4°C. The image was processed by the ChemiDoc™ XRS + Imaging System (Bio-Rad, Hercules, CA, United States). Protein bands were normalized relative to the β-actin.

The results were exhibited using mean ± standard error of measurement (SEM). Experimental data were performed using one-way ANOVA followed by Bonferroni’s post-hoc analysis and analyzed via the Graphpad Prism 8.0 (GraphPad Software, San Diego, CA, United States). P-value < 0.05 was considered statistically significant.

In order to research the effect of TSP on CP-induced anxiety- and depression-like behavior in mice, behavioral experiments were investigated. After CP treatment, sugar water preference and total travel distance were markedly reduced, and the immobility time of FST and TST was considerably prolonged (P < 0.05, P < 0.01, Figure 2) vs. that of the Control groups. These four behavioral conditions were dramatically improved via the treatment of different doses of TSP (250, 500, and 1000 mg/kg/d), showing a dose-dependently manner (P < 0.05, P < 0.01, Figure 2).

In order to research the possible mechanism of TSP on anxiety- and depression-like behavior following chemotherapy, mice were injected ip with CP, and a high-dose group was selected for in-depth studies. Compared with that of the Control groups, the SOD activity and GSH-Px level of the hippocampus from the Model group were significantly reduced (P < 0.05, P < 0.01, Figures 3A,C), and the MDA content was obviously increased (P < 0.05, Figure 3B). After treatment with the TSP, SOD activity and GSH-Px level were drastically increased (P < 0.05, Figures 3A,C), MDA level was considerably decreased (P < 0.05, Figure 3B), indicating that TSP prevents the balance of redox reactions broken by CP in the hippocampus.

Hippocampal neuroinflammation is closely related to depression. In order to explore the effects of TSP on hippocampal neuroinflammation, microglia labeled with Iba-1 (Figure 4A) were evaluated via immunofluorescence, TNF-α (Figure 4C), and IL-1β (Figure 4C) levels were assessed by ELISA kits in the hippocampus. Compared with that of the Model group, the expression of Iba-1, the levels of TNF-α, and IL-1β in the hippocampus of the mice treated with TSP were markedly reduced (P < 0.05, Figure 4). The data showed that TSP could rescue the inflammatory response in the hippocampus.

The nuclei of TUNEL-positive neurons (apoptotic neurons) in the hippocampus were stained (Figure 5A). The number of TUNEL-positive neurons in the hippocampus of CA1 and CA3 regions were considerably increased (P < 0.01, Figures 5B,C), indicating that CP has a damaging effect on neurons in the hippocampus. TSP-administered mice found that the number of TUNEL-positive cells was markedly reduced vs. that of the Model group (P < 0.05, P < 0.01, Figures 5B,C).

Accumulated literature has reported that hippocampal neurogenesis exerts an effect on the resistance to depression, so we tested the expression of the neurogenesis marker DCX (Figure 6A). In the Model group, the number of DCX positive cells in the hippocampus of the DG region was considerably lowered vs. that of the Control group (P < 0.01, Figure 6B). In contrast, TSP supplements significantly increased the number of DCX positive cells (P < 0.05, Figure 6B), compared with that of the Model group.

The Nrf2/HO-1 signaling pathway was investigated as a possible mechanism of oxidative stress (Figure 7). The results exhibited that the protein expression of Keap1 was obviously raised (P < 0.01, Figure 7B), and the protein expression of Nrf2 and HO-1 was markedly down-regulated (P < 0.05, Figures 7C,D), compared with that of the Control group. TSP-administered mice observed that the Nrf2/HO-1 signaling pathway was obviously activated in the hippocampus (P < 0.05, P < 0.01, Figures 7B–D).

The BDNF/TrkB/CREB signaling pathway was evaluated (Figure 8). The results showed that CP significantly suppressed BDNF expression (P < 0.05, Figure 8B) and p-CREB expression (P < 0.01, Figure 8C) in the hippocampus vs. that of the Control group. 1000 mg/kg-dose TSP administration restored these changes drastically (P < 0.05, Figures 8B,C).

The Bcl/Bax/caspase-3 apoptosis pathway was detected by the Western blotting experiment (Figure 9). Compared with that of the Control group, CP induced a considerable reduction in Bcl-2 expression (P < 0.01, Figure 9B) and a significant increase in Bax and caspase-3 (P < 0.01, Figures 9C,D). After 1000 mg/kg TSP administration, the apoptosis pathway-related protein markers were dramatically rescued in the hippocampus (P < 0.05, P < 0.01, Figures 9B–D).