To study the effects of MGF on PPD, we established a hormone-simulated pregnancy (HSP) mouse model combined with ovariectomy (OVX). As shown in Figure 1, behavioral tests began 10 days after progesterone (P4) withdrawal. Two doses of MGF (20 and 60 mg/kg) were orally administered once per day. Moreover, the novelty-suppressed feeding (NSF) test was used to evaluate exploration and anhedonia behaviors, while the forced swim test (FST) and tail-suspension test (TST) were utilized to assess depression-like behaviors. We found that mice in the PPD model group showed increased immobility time in the NSF test, FST, and TST (Figures 2A–F), indicating impaired emotional functions. Interestingly, administration of MGF significantly decreased the immobility time in the NSF test in a dose-dependent manner compared with the PPD group (p < 0.001) (Figures 2A,B). Consistently, administration of MGF significantly decreased the immobility time in the FST and TST, suggesting alleviated depression-like behaviors (Figures 2C–F). Furthermore, we compared the PPD/MGF groups with the control groups through behavioral tests and found that PPD/MGF groups reduced the immobility time of PPD mice in NSF, which was still higher than that in the control groups. However, there was no difference in immobility time between high doses of the MGF and the control group in TST and FST, indicating a protective effect of MGF. Collectively, these results suggest that the administration of MGF could alleviate HSP-induced depression-like behavior in mice.

To further study the mechanism underlying the protective effect of MGF, we examined the expression of synaptic plasticity–related protein 95 (PSD95) and brain-derived neurotrophic factor (BDNF) in the hippocampus. However, no significant differences were observed between the MGF-treated groups and the PPD model groups (Figures 3A–C). Multiple studies have suggested that neuroinflammation is involved in the development of depression (Engler et al., 2017; Moisan et al., 2021). To determine whether neuroinflammation is involved in this process, we first examined the protein levels of IBA1 and GFAP in the mouse brain. We found that the expression level of IBA1 was increased in the PPD group compared to that in the control group. MGF treatment inhibited this increase in a dose-dependent manner. There were no significant changes in the protein levels of GFAP among the four groups (Figures 3D–F). Moreover, we found that the levels of inflammatory cytokines TNF-α, IL-6, and IL-1β were significantly increased in the PPD group mice. Interestingly, treatment with MGF significantly inhibited the increase in the levels of these cytokines (Figures 3G–I). Together, these results show that MGF treatment inhibited inflammatory cytokine levels in the PPD mouse brain.

Next, we investigated whether microglial activation is involved in this process. To address this, we performed an IBA1 immunofluorescence staining assay, which showed that a higher number of microglia existed in the hippocampus of PDD mice (Figures 4A,B). Using Image-Pro Plus software, we analyzed the number of microglia in the CA1 and DG areas of the hippocampus in these four groups of mice. The number of microglia was significantly increased in the CA1 and DG areas of the hippocampus in the PPD group mice (p < 0.001 and p < 0.01, respectively), while treatment with a high concentration of MGF significantly inhibited this increase, with a decreasing trend seen in the low concentration of MGF treatment groups (Figures 4C,D). Thus, these results suggest that microglia were activated in the PPD model mouse brain and that MGF treatment could significantly inhibit microglial activation.

To find the potential molecular targets of MGF, bioinformatic analysis of 3D similarity searching, ranking, and superposition was performed using ChemMapper (http://www.lilab-ecust.cn/chemmapper/index.html). Among the predicted targets (MAP kinase–activated protein kinase 2, amine oxidase [flavin-containing] A, sialidase, fatty acid synthase, and transcription factor p65), MAP kinase–activated protein kinase 2 (MAPK) was ranked first, with a 3D similarity score of 1.0 (Figures 5A,B). Next, to study changes in MAPK signaling in the hippocampus of the mouse brain, the levels of p-JNK, p-p38, and p-ERK were investigated. As shown in Figure 5C, increased levels of these three markers were observed in the PPD group compared to the control group. Notably, administration of MGF inhibited the increase in p-JNK, p-p38, and p-ERK levels, suggesting downregulation of MAPK signaling in the mouse brain. To further confirm the effects of MGF on microglia, we cultured microglial BV2 cells and studied the effect of MGF on LPS-induced MAPK signaling activation in vitro (Figure 5D). As shown in Figure 5E, LPS treatment increased the protein levels of iNOS, p-JNK, and p-p38, whereas pretreatment with MGF largely inhibited increased levels. Consistently, the levels of the downstream inflammatory cytokines TNF-α, IL-6, and IL-1β were significantly inhibited in the MGF treatment group (Figures 5F–H). Collectively, these results show that MGF inhibits microglia-mediated inflammation by targeting MAPK signaling.

In summary, our results show that treatment with MGF significantly alleviated PPD-like behaviors in mice. Mechanistically, we found that MGF inhibited microglial activation by targeting MAPK signaling in vivo and in vitro (Figure 6), providing a potential therapeutic strategy for PPD treatment.

MGF (purity ≥98%) was purchased from Chengdu Desite Biotechnology (Chengdu, China). β-estradiol (E8875), dimethyl sulfoxide (DMSO), and LPS were purchased from Sigma-Aldrich (St. Louis, MO, United States). Progesterone was obtained from VETEC (V900699). The antibodies used for western blotting were as follows: Iba1/AIF-1 (E4O4W) (#17198), GFAP (E4L7M) (#80788), PSD95 (D27E11) (#3450), BDNF (#47808), iNOS (D6B6S) (#13120), anti-p-ERK1/2 (Thr202/Tyr204) (#9101), anti-ERK1/2 (#9102), anti-p-p38 MAPK (Thr180/Tyr182) (#4511), anti-p38 MAPK (#9212), and anti-p-JNK (Thr183/Tyr185) (#9251) were purchased from Cell Signaling Technology (Beverly, MA, United States). β-tubulin (#CW0098A) and β-actin (#CW0096M) were procured from CWBiotech (Beijing, China).

Female BALA/c mice (8 weeks old, 20–25 g) were housed in the animal care facility of our institute. All animal experimental procedures were approved by the Biological and Medical Ethics Committee of Minzu University of China. All mice were maintained under conditions of a 12-h light/dark cycle at 23°C and were provided with food and water.

BV-2 microglial cell lines were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM, #11965-092, Life Technologies, Waltham, MA, United States) supplemented with 10% heat-inactivated fetal bovine serum (FBS, #04-001-1A, Biological Industries, Israel) and 1% penicillin-streptomycin solution (#03-031-1B, Biological Industries) at 37°C in a humidified atmosphere with 5% CO₂.

Two-month-old female mice were chosen, and hormone-induced pseudopregnancy (HSP)-induced PPD models were established as previously described (Li et al., 2018; Zhang et al., 2021). Mice were randomly divided into four groups (control, PPD, PPD/low MGF, and PPD/high MGF). OVX was performed under isoflurane anesthesia. After 7 days of recovery from OVX operation, mice in the PPD and PPD with MGF treatment groups were intraperitoneally injected with β-estradiol (E2, 0.5 g/day) and progesterone (P4, 0.8 mg/day) dissolved in 0.1 ml sesame oil daily for 16 days, resulting in a gradual increase in the concentration of E2 and P4 in mice to mimic the increases in hormone levels. Subsequently, mice were intraperitoneally injected with E2 (10 µg/day) alone for seven consecutive days to mimic high levels of E2 during pregnancy. Meanwhile, MGF was administered intragastrically at two different doses (20 and 60 mg/kg), as indicated in Figure 1.

The NST was performed as previously described, with minor modifications (Barbieri et al., 2021). Briefly, before the test, the mice were deprived of food but had free access to water for 24 h. Each mouse was positioned into the device with food placed on white paper in the same direction and allowed to freely explore for 5 min. The immobility time of each mouse was recorded.

One day before the test, mice were allowed to swim in water for 5 min. During the test, the mice were placed in a beaker (volume, 3 L) filled with water at 23–25°C. The total test time was 6 min, and the immobility time of the mice in the last 4 min was recorded.

Mice were placed in the test room 2 h before the test and hung on the instrument with a clip. Similar to the FST, the total experimental time was 6 min, and the immobility time of the mice in the last 4 min was recorded.

Total RNA was isolated from the hippocampus of mice in each group using a TRIzol reagent (Invitrogen, cat#15596018), and 1 μg of RNA was used to synthesize cDNA using a one-step first-strand cDNA synthesis kit (Transgen Biotech, cat#AT341). Quantitative real-time PCR was performed using a 2 × SYBR Green PCR master mix (Transgen Biotech, cat#AQ131) and an Agilent Mx3005P RT-PCR system. The expression levels of the tested genes were normalized to those of β-actin. The primers for mouse IL-1β, TNF-α, IL-6, and β-actin were as follows:

Mouse IL-1β: Forward: 5′-TGT​AAT​GAA​AGA​CGG​CAC​ACC-3′; Reverse: 5′-TCT​TCT​TTG​GGT​ATT​GCT​TGG-3′.

Mouse TNF-α: Forward: 5′-CAG​GCG​GTG​CCT​ATG​TCT​C-3’; Reverse: 5′-CGA​TCA​CCC​CGA​AGT​TCA​GTA G-3′.

Mouse IL-6: Forward: 5′-CTA​CCA​AAC​TGG​ATA​TAA​TCA​GGA-3′; Reverse: 5′-CCA​GGT​AGC​TAT​GGT​ACT​CCA​GAA-3′.

Mouse β-actin: Forward: 5′-GGCTGTATTCCC CTCCATCG-3′; Reverse: 5′-CCA​GTT​GGT​AAC​AAT​GCC​ATG T-3′.

The concentration of the extracted protein was determined using the BCA assay. Equal amounts of protein were separated by polyacrylamide gel electrophoresis (SDS-PAGE) and incubated with the primary antibody overnight at 4°C, followed by incubation with a secondary antibody (1:5,000) for 1 h at room temperature. An ECL luminescent solution was used for detection.

After anesthesia, the mice were perfused with normal saline, and then the whole brain was isolated and fixed with 4% paraformaldehyde for 24 h and dehydrated overnight in 30% sucrose solution. Whole brain tissue was embedded in OCT and sectioned using a freezing microtome (Leica CM3050S). Tissue sections were incubated with anti–goat IBA1 antibody (1:500, WAKO, Japan) overnight at 4°C with shaking. On the following day, tissue sections were incubated with secondary antibodies for 1 h at room temperature. Finally, images were captured using a laser scanning confocal microscope (Nikon, Tokyo, Japan).

All data are presented as mean ± SEM. The significance of the differences was determined by the t-test and one-way ANOVA using GraphPad Prism (GraphPad Software, San Diego, CA, United States). *p < 0.05, **p < 0.01, and ***p < 0.001 were considered as significant.