The mouse breast cancer cells 4T1 (AW-CCM376) were obtained from Abiowell (Changsha, China) and maintained in Dulbecco’s Modified Eagle Medium containing 10% fetal bovine serum and 1% penicillin-streptomycin. The 4T1 cells were treated with 25, 50, and 100 µM kaempferol (N1719; APExBIO Technology LLC, Houston, TX, USA) to investigate the effects of kaempferol [17]. BCRD was mimicked in vitro, as previously described [18]. Briefly, 4T1 cells were stimulated with 10 µg/mL lipopolysaccharide (Sigma, St. Louis, MO, USA) for 6 h, and the supernatant was collected after centrifugation. The cell supernatant and 200 µM corticosterone (Cort, 50-22-6; Shanghai Aladdin Biochemical Technology Co., Ltd., Shanghai, China) were added to mouse hippocampal neuronal cells HT-22 (AW-CNM116, Abiowell) and cultured for 6 h to obtain BCRD cells. Then the BCRD cells were treated with kaempferol. Cells were divided into Control, BCRD, and Kaempferol groups. 4T1 cells and HT-22 cells were tested for mycoplasma. Through short tandem repeat (STR) detection, it was verified that the cell line we used was 4T1 and HT-22 cells, and there was no human-source pollution. The 4T1 (https://www.cellosaurus.org/CVCL_0125) and HT-22 cells (https://www.cellosaurus.org/CVCL_0321) used in this study were checked and found no cross-contamination.

The overexpression (oe) plasmids, oe-NC and oe-COX-2(HG-MO011198) (both from Honor Gene Co. Ltd., Changsha, China), were transfected into BCRD cells using Lipofectamine 2000 reagent (11668019; Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The sequence of COX-2 was obtained from the National Center for Biotechnology Information, and primers were designed to isolate the target gene. The target fragment was ligated into the pcDNA3.1(+) vector. After being stimulated with 25 µM kaempferol for 48 h, oe-NC and oe-COX-2 were transfected into BCRD cells. Cells were divided into BCRD, oe-NC, oe-COX-2, Kaempferol, Kaempferol+oe-NC, and Kaempferol+oe-COX-2 groups.

To assess cell viability, we used the Cell Counting Kit-8 (CCK8) assay. In brief, after pancreatic enzyme digestion, cells in the logarithmic growth phase were seeded in a 96-well plate at a density of 5 $\times$ 10${}^3$ cells/well in a volume of 100 µL per well. After drug treatment, 10 µL CCK-8 reagent (AWC0114a, Abiowell) was added to each well, thoroughly mixed, and incubated for 4 h. A microplate reader (MB-530, Shenzhen Heales Technology Development Co., Ltd., Shenzhen, China) was used to measure the absorbance at 450 nm. Each group had at least three replicates.

The wound healing assay was performed to assess cell migration. After digestion with trypsin, 4T1 cells were seeded in a 6-well plate at a density of 5 $\times$ 10${}^5$ cells/well in a volume of 1 mL per well. Subsequently, a sterile pipette tip was used to streak vertically. The scratched cells were washed away with phosphate-buffered saline (PBS). Then, the complete medium was replaced with a serum-free medium to limit the effects of cell proliferation [19], and cells were cultured at 37 °C for 48 h. Images were taken at 0 h, 24 h, and 48 h, and the width value of the area not covered by cells (width value, µm) was recorded.

The colony formation assay was used to analyze cell proliferation. After kaempferol treatment, 4T1 cells (200 cells/mL) were seeded in 6-well plates and dispersed evenly. Cells were allowed to grow for the next 14 d to form colonies. After terminating the culture, the cells were fixed in 4% paraformaldehyde for 15 min, stained with crystal violet dye (AWC0333; Abiowell), and counted.

Six-week-old female BALB/c mice were purchased from Hunan SJA Laboratory Animal Co., Ltd. (Changsha, China). As previously described, 4T1 cells (1 $\times$ 10${}^6$ cells/100 µL) were injected into the right mammalian fat pads of mice to establish a BC model [20]. Tumor formation was observed after 7 days. Subsequently, a mouse model of BCRD was established by administering 30 mg/kg Cort daily through subcutaneous injection in the back of BC mice for 21 days. The experimental groups included Control, BCRD, and Kaempferol groups (n = 5). The Kaempferol group additionally received 20 mg/kg/d kaempferol (HY-14590; MedChemExpress, Monmouth Junction, NJ, USA) by intraperitoneal injection for 21 d [14, 17]. The Control group was treated with the same dose of normal saline. The survival rate of mice remained at 100% throughout the modeling process.

To assess anxiety, the open field test (OFT) was conducted. Briefly, mice were quickly placed in the center of a square experimental box (50 $\times$ 50 $\times$ 40 cm), and the mice were allowed to move freely. During 15 min of training, the total distance traveled by the mice was recorded [21].

To assess depressive-like states, the forced swimming test (FST) was performed. Briefly, mice were placed in clear glass vials that were 21 cm high and 16.5 cm in diameter. Water about 13 cm deep was added to the bottle, and the mice were forced to swim for 6 min. The behavior of the mice was observed for 4 min, and the immobility time was recorded. Mice were considered stationary when their head was above the water’s surface with no apparent struggle [22].

For the tail suspension test (TST), mice (1 cm from the tail tip) were suspended in the air with adhesive tape, about 60 cm above the ground. Mice were separated from each other. During the 6 min test period, the activity of the mice was observed, and the immobility time within the next 4 min was recorded [23].

Hematoxylin and eosin (HE) staining was used to assess the degree of pathological damage. Tumors from BCRD mice were collected and then subjected to fixing, embedding, and sectioning (4 µm). Sections were placed in xylene for 20 min and repeated three times. Then, the sections were hydrated through different gradients of ethanol (75–100%) and washed with distilled water. Hematoxylin (AWI0009; Abiowell) and eosin (AWI0020, Abiowell) were used for staining. Sections were dehydrated in graded alcohol (95%–100%) and placed in xylene. The morphological changes were observed and photographed using a microscope (BA210T; Motic Microscopes, Xiamen, China).

Total RNAs from cells were extracted with TRIzol (15596026; Thermo Fisher Scientific, Waltham, MA, USA) and reversely transcribed to prepare cDNAs. The relative expression of targets was performed by the UltraSYBR Mixture kit (CW2601; ConWin Biosciences, Taizhou, China) and normalized to $\beta$-actin. The relative mRNA enrichment of targets was determined with the 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ method. The primers are listed in Table 1.

Total protein from mice hippocampal tissues or cell lysates was extracted with RIPA (AWB0136, Abiowell). After protein quantification with a BCA kit (AWB0104, Abiowell), proteins were resolved by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then electrotransferred to nitrocellulose membranes. After blocking in 5% skimmed milk for 90 min, the membranes were incubated overnight at 4 ℃ with the following primary antibodies: COX-2 (1:2000, 12375-1-AP; Proteintech Group Inc., Rosemont, IL, USA), PGE2 (1:300, ab217966; Abcam, Cambridge, MA, UK), interleukin 1 beta (IL-1$\beta$) (1:2000, 16806-1-AP; Proteintech), IL-6 (1:1000, ab233706; Abcam), transforming growth factor beta 1 (TGF-$\beta$1) (1:1000, ab31013; Abcam), IL-10 (1:1000, 20850-1-AP; Proteintech), and $\beta$-actin (1:5000, 66009-1-Ig; Proteintech). Then the membranes were washed with PBS with Tween 20 (PBST), followed by incubation with anti-mouse (1:5000, SA00001-1; Proteintech) and anti-rabbit (1:6000, SA00001-2; Proteintech) secondary antibodies for 90 min and washing with PBST. ECL Plus Substrate (AWB0005; Aiowell) and a gel imaging system (ChemiScope 6100; Clinx, Shanghai, China) were used to visualize the proteins. Protein expression levels were quantified by scanning densitometry using Quantity one 4.6 software (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

According to the manufacturer’s instructions, the levels of serotonin (5-HT, CSB-E08365m), dopamine (DA, CSB-E08661m), norepinephrine (NE, CSB-E07870m), IL-1$\beta$ (CSB-E08054m), IL-6 (CSB-E04639m), TGF-$\beta$1 (CSB-E04726m), IL-10 (CSB-E04594m), COX-2 (CSB-E12910m), and PGE2 (CSB-E07966m) were measured. All kits were from Cusabio Biotech Co., Ltd. (Wuhan, China).

The two-dimensional (2D) structure of COX-2 was downloaded from the Protein Data Bank (https://www.rcsb.org/), and the 3D structure of kaempferol was obtained from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/compound/5280863). Autodock Vina software was performed to analyze the interaction of kaempferol and COX-2, and Discovery Studio was used for analyzing and viewing the 2D and 3D images.

Mouse brain tissue underwent embedding and sectioning. After deparaffinization and hydration, slices were placed in an EDTA buffer and heated for thermal antigen retrieval. After cooling, the slices were washed with PBS. Then sections were sequentially incubated in sodium borohydride, 75% ethanol, and Sudan black solution. Slices were blocked in 5% bovine serum albumin for 1 h, then incubated with 5-Bromo-2${}^{\prime }$-deoxyuridine (BrdU) (1:50, ab8152; Abcam) and neuronal nuclei (NeuN) (1:50, 26975-1-AP; Proteintech) primary antibodies overnight at 4 °C. Subsequently, the slices were incubated with anti-rabbit (1:500, SA00013-2; Proteintech) or anti-mouse (1:500, SA00013-3; Proteintech) secondary antibodies at 37 °C for 1 h. Slices were counterstained with DAPI for 10 min and then imaged and photographed under a fluorescent microscope.

Statistical analysis was conducted using GraphPad Prism 9.0 software (GraphPad Software Inc., San Diego, CA, USA). Data are presented as the mean $\pm$ standard deviation. Data from two samples were compared using the Student’s t-test. Analysis of variance (ANOVA) was used to compare data from multiple groups. All experiments included three biological replicates. p 0.05 was considered statistically significant.

To assess the effects of kaempferol on BC, 4T1 cells were stimulated with different concentrations of kaempferol (25, 50, and 100 µM). The results showed that after treatment for 24 and 48 h, kaempferol significantly reduced the viability of 4T1 cells, and the inhibitory ability was gradually increased in a dose-dependent manner (Fig. 1A). The wound healing assay showed that after 24 h and 48 h, kaempferol alleviated the migration of 4T1 cells in a dose-dependent manner (Fig. 1B,C). In addition, the number of cloned 4T1 cells was reduced after kaempferol intervention, and the effect of 100 µM kaempferol was the most prominent (Fig. 1D). These results showed that kaempferol inhibited 4T1 cell proliferation, migration, and colony formation.

Fig. 1.

Kaempferol reduced 4T1 cell viability, migration, and colony formation. (A) 4T1 cell viability was detected by the cell counting kit-8 (CCK8) assay at 24 h and 48 h. (B,C) The wound healing assay was performed to measure 4T1 cell migration at 24 h and 48 h. Areas not covered by 4T1 cells were measured and counted. (D) Cell numbers were counted to assess 4T1 cell clonality. n = 3. * p 0.05 vs. Control.

Next, the effects of kaempferol on inflammation in neuronal cells were examined. The non-toxic dose of kaempferol was determined by the CCK-8 assay. Compared with the Control group, BCRD cells showed decreased cell viability, whereas the cell viability was partially restored after stimulation with different concentrations of kaempferol (10 µM, 25 µM, and 50 µM). Notably, in BCRD cells, when the dose of kaempferol reached 50 µM, the cell viability showed a downward trend (Fig. 2A). Therefore, we applied 25 µM kaempferol for experiments. Subsequently, the levels of inflammatory factors were detected. BCRD cells showed increased levels of pro-inflammatory cytokines (IL-1$\beta$ and IL-6), and kaempferol reversed this trend (Fig. 2B,C). Meanwhile, kaempferol increased the levels of anti-inflammatory factors (TGF-$\beta$1 and IL-10) levels in BCRD cells (Fig. 2D,E). In addition, kaempferol inhibited the protein accumulation of COX-2 and PGE2 in BCRD cells (Fig. 2F). Taken together, these results showed that kaempferol attenuated inflammatory manifestations effects and downregulated COX-2 and PGE2 in BCRD cells.

Fig. 2.

Kaempferol regulated the expression of inflammatory cytokines and cyclo-oxygenase (COX)-2/PGE2. (A) The effect of kaempferol (10 µM, 25 µM, and 50 µM) on cell viability was assessed at 24 h and 48 h. The mRNA levels of IL-1$\beta$ (B), IL-6 (C), TGF-$\beta$1 (D), and IL-10 (E) were measured. (F) The abundance of COX-2 and PGE2 was examined. n = 3. * p 0.05 vs. Control, # p 0.05 vs. breast cancer-related depression (BCRD).

The role of COX-2/PGE2 in the anti-inflammatory effect of kaempferol was further explored. To confirm the kaempferol-COX-2 interaction, molecular docking was selected for analysis. Kaempferol and COX-2 had a binding energy of –9.1 kcal/mol, which indicated a strong and stable interaction between kaempferol and COX-2. In the visualized images, kaempferol bound to specific residues within the active pocket of COX-2, including PRO127, TYR2373, ARG376, GLN374, GLN2374, PHE142, PRO2538, GLY2227, VAL2228, GLY2533, ASN2537, TRP139, GLY2536, ASN-2375, GLY2225, ARG2376, and LEU2145. More specifically, kaempferol formed a Pi-Sigma conjugation at the GLN2374 residue. Kaempferol formed an Amide-Pi stacked at residue GLY2536. Kaempferol formed an unfavorable Donor-Donor at the ASN2537 residue. Kaempferol formed Pi-Alkyl and carbon-hydrogen bonds at the PRO2538 residue. Kaempferol formed Pi-Donor hydrogen and conventional hydrogen bonds at the ASN2375 residue. Kaempferol formed a conventional hydrogen bond at residues GLN374, VAL2228, and ARG2376. Furthermore, kaempferol formed van der Waals at residues ARG376, TYR2373, PRO127, PHE142, GLY2227, GLY2533, TRP139, GLY2225, LEU2145. These results suggested that kaempferol can bind to COX-2 (Fig. 3A).

Fig. 3.

Kaempferol inhibited the COX-2/PGE2 pathway to relieve neuroinflammation. (A) Molecular docking revealed the binding of kaempferol to COX-2. (B) The mRNA and protein levels of COX-2 were measured. (C) The mRNA and protein levels of PGE2 were measured. (D) The effect of COX-2 on cell viability was detected. (E) The effects of COX-2 on the levels of IL-1$\beta$, IL-6, TGF-$\beta$1, and IL-10 were examined. n = 3. & p 0.05 vs. oe-NC, * p 0.05 vs. BCRD, # p 0.05 vs. Kaempferol+oe-NC.

Next, BCRD cells were transfected with the overexpression plasmid oe-COX-2. Our results showed that compared with the oe-NC group, oe-COX-2 increased the expression of both COX-2 and PGE2 in BCRD cells. By contrast, treatment with kaempferol reduced the expression of COX-2 and PGE2 in BCRD cells compared with the BCRD group. Compared with Kaempferol+oe-NC, oe-COX-2 elevated the expression of COX2 and PGE2 in BCRD cells after kaempferol intervention. Compared with the oe-COX-2 group, kaempferol inhibited the expression of COX2 and PGE2 (Fig. 3B,C). In addition, oe-COX-2 decreased BCRD cell viability compared to the oe-NC group, whereas Kaempferol enhanced BCRD cell viability. The Kaempferol+oe-COX-2 group exhibited reduced BCRD cell viability compared to Kaempferol+oe-NC. Kaempferol decreased cell viability compared with the oe-COX-2 group (Fig. 3D). Compared with the oe-NC group, oe-COX-2 increased IL-1$\beta$ and IL-6 levels and decreased TGF-$\beta$1 and IL-10 levels. Compared with the BCRD group, kaempferol downregulated IL-1$\beta$ and IL-6 and upregulated TGF-$\beta$1 and IL-10. oe-COX-2 increased the kaempferol-decreased IL-1$\beta$ and IL-6 levels and attenuated the kaempferol-increased TGF-$\beta$1 and IL-10 levels compared to Kaempferol+oe-NC. Compared with the oe-COX-2 group, the Kaempferol+oe-COX-2 group showed decreased IL-1$\beta$ and IL-6 levels and increased TGF-$\beta$1 and IL-10 levels (Fig. 3E). These results suggested that COX-2 overexpression inhibited BCRD cell viability and regulated the expression of inflammatory factors. Thus, kaempferol may have a therapeutic role in neuroinflammation, at least in part, by targeting COX-2/PGE2.

We developed a mouse model of BCRD to confirm the therapeutic effect of kaempferol in vivo. Compared with the BCRD group, kaempferol promoted the growth of mice (Fig. 4A). Compared with the BCRD group, kaempferol significantly inhibited the volume and weight of BC tumors (Fig. 4B,C), suggesting that kaempferol had certain anti-BC effects on BCRD mice. Next, the tumor structure of BC was observed. As shown in Fig. 4D, the tumors of kaempferol-treated BCRD mice showed reduced inflammatory cell infiltration, clear cell structure, and tight arrangement, showing that kaempferol protected the pathological tumor structure of BCRD mice.

Fig. 4.

Kaempferol inhibited the growth of BC and neuroinflammation in BCRD mice. (A) Within 28 days of modeling, the mouse weight in each group (n = 5) was recorded. (B) Within 28 days of modeling, the tumor growth in each group was recorded. (C) On the 28th day, the tumors of mice in each group were photographed and weighed. (D) Representative images of the pathological phenotype in each group were presented. n = 3. * p 0.05 vs. BCRD.

The effect of kaempferol on the depressive behavior of mice in the BCRD group was assessed by behavioral studies. The results of the FST showed that the immobility time of BCRD mice increased, and kaempferol effectively reduced the immobility time of BCRD mice (Fig. 5A). The OFT showed that the total distance of mice in the BCRD group was significantly reduced, but kaempferol reversed this trend (Fig. 5B). Kaempferol inhibited the stress-induced increase in immobility time during the TST (Fig. 5C). Major neurotransmitters (NTs) in depression include (5-hydroxy tryptamine, 5-HT), dopamine (DA) and norepinephrine (NE). Compared with the Control group, 5-HT, DA, and NE levels were downregulated in the BCRD group, while kaempferol partially restored their levels (Fig. 5D). Taken together, these findings demonstrated that kaempferol contributed to reducing depression-like behaviors in BRCD mice.

Fig. 5.

Kaempferol improved depression in BCRD mice. (A) Forced swimming test (FST). (B) Open field test (OFT). (C) Tail suspension test (TST). (D) The contents of 5-hydroxy tryptamine (5-HT), dopamine (DA), and norepinephrine (NE) were examined by ELISA. n = 3. * p 0.05 vs. Control, # p 0.05 vs. BCRD.

IL-1$\beta$ and IL-6 levels in the peripheral blood of mice in the BCRD group were increased, and TGF-$\beta$1 and IL-10 levels were decreased, whereas kaempferol reversed the levels of these cytokines (Fig. 6A). The protein abundance of these cytokines in the hippocampus of BCRD mice was also restored by kaempferol (Fig. 6B). Moreover, peripheral blood and hippocampal tissues of BCRD mice showed increased COX-2 and PGE2 expression, which were effectively interfered with by kaempferol (Fig. 6C,D). BrdU/NeuN staining revealed that kaempferol reversed the reduction of BrdU/NeuN neurons in BCRD mice (Fig. 6E). Our results suggested that kaempferol improved the development of BCRD at least in part by inhibiting COX-2/PGE2.

Fig. 6.

Kaempferol improved neuronal damage in BCRD mice, at least in part through COX-2/PGE2 signaling. The levels of IL-1$\beta$, IL-6, TGF-$\beta$1, and IL-10 in the peripheral blood (A) and hippocampal tissues (B) of mice in each group were examined. The levels of COX-2 and PGE2 in peripheral blood (C) and hippocampal tissues (D) of mice in each group were measured. (E) Immunohistochemistry staining was applied for the detection of hippocampal neurogenesis. 5-hydroxy tryptamine (BrdU)/neuronal nuclei (NeuN) positive cells were calculated. n = 3. * p 0.05 vs. Control, # p 0.05 vs. BCRD.