A potential strategy for improving offspring behavior in maternal immune activation: Amantadine‐mediated suppression of neuroinflammation

Abstract

Background

Maternal viral infection during pregnancy can lead to maternal immune activation (MIA), increasing the risk of neurodevelopmental disorders in offspring. Amantadine (AMA) exhibits antiviral activity and is widely employed in the management of neurologic conditions. Nevertheless, the efficacy of AMA in treating MIA is currently not established.

Methods

MIA was induced by polyinosinic acid–polycytidylic acid (poly(I:C)); AMA was administered from embryonic (E) day 11.5 for 3 days. BV‐2 cells were stimulated using poly(I:C) and treated with AMA. Behavior was assessed via open field test, elevated plus maze test, three‐chamber sociability test, and marble burying test. Neuronal morphology was vizualized using Nissl stain; apoptosis via TUNEL (terminal deoxynucleotidyl transferase dUTP nick‐end labeling) stain; protein expression (Iba1, NeuN, CD68, TNF‐α [tumor necrosis factor‐alpha], IL‐1β [interleukin‐1β]) using immunofluorescence (IF); interleukin‐6 (IL‐6) levels using enzyme‐linked immunosorbent assay; reactive oxygen species using staining; Iba1, NeuN, Bcl‐2, Bax, and cleaved caspase 3 using Western blot; and gene expression changes using RNA‐seq.

Results

AMA treatment reduced the levels of IL‐6 in maternal blood, improved autism‐like behaviors in MIA offspring, and effectively prevented neuronal damage and neuroinflammation. In vitro cellular studies have demonstrated that AMA effectively downregulates the expression levels of pro‐inflammatory cytokines, including IL‐6, TNF‐α, and IL‐1β. RNA‐seq analysis indicated that AMA mitigates abnormal activation of microglia by modulating inflammatory pathways associated with IL‐6.

Conclusion

AMA can prevent the development of neuropsychiatric disorders in MIA offspring. This effect may be related to its ability to attenuate neuronal damage, reduce neuronal apoptosis, and inhibit neuroinflammation, indicating that the antiviral drug AMA may be a potential treatment for MIA.

Amantadine (AMA) regulates neurocyte apoptosis and inhibits neurocyte inflammation to prevent maternal immune activation (MIA) from affecting offspring behavior.

1. INTRODUCTION

In recent years, some viruses, including Zika virus, avian influenza virus, and coronavirus (SARS‐CoV‐2), have wreaked havoc globally, posing significant challenges to health‐care systems. ¹ These viruses pose significant risks to susceptible populations, including the elderly, pregnant women, and fetuses. ² , ³ Maternal infection during pregnancy is highly likely to disrupt the maternal immune environment, leading to irreversible harm to the developing fetus, such as miscarriage, fetal growth abnormalities, and neurodevelopmental disorders in offspring. ⁴ , ⁵ Epidemiological studies and animal experiments have confirmed that neurodevelopmental disorders increase the risk of autism, depression, and schizophrenia in offspring. ⁶ , ⁷ Therefore, the development of maternal immune activation (MIA) models for studying the occurrence, pathogenic mechanisms, and prevention of psychiatric disorders in offspring has become a research topic in neuroscience. ⁸

MIA animal models have been successfully established, including mice, rats, and nonhuman primates. ⁴ , ⁹ , ¹⁰ The most commonly used modeling drugs are the virus mimic polyinosinic acid–polycytidylic acid (poly(I:C)) and lipopolysaccharides (LPS), which, respectively, activate the toll‐like receptor 3 and 4 (TLR3 and TLR4) pathways to mimic viral and bacterial infections in pregnant mothers, thus activating the maternal immune system. ⁴ , ¹¹ Offspring of MIA are prone to behavioral defects, including impaired social abilities, spatial learning impairments, increased anxiety‐like behaviors, and deficits in prepulse inhibition (PPI), all indicating the potential development of autism or schizophrenia in offspring. ¹² , ¹³ Studies have found diverse neurological damages in MIA offspring, such as decreased activity of dopamine neurons, increased dendritic spine density in the hippocampus, and abnormal development of astrocytes in the amygdala. ¹⁴ , ¹⁵ Many studies have focused on abnormal changes occurring in the offspring's cerebral cortex and hippocampus due to MIA, primarily characterized by increased inflammatory factors in these brain regions. ¹³ , ¹⁴ , ¹⁶ Interleukin‐6 (IL‐6) is so crucial in MIA, and maternal administration of IL‐6 can induce abnormal behavior in offspring. ¹⁷ Simultaneously, abnormal activation of microglia is also a prominent feature in MIA offspring. ¹⁸ , ¹⁹ Microglia play crucial regulatory roles in synaptic connections and neuroinflammation. ²⁰ , ²¹ Research has shown that MIA alters both the number and status of microglia in offspring, thus detecting microglial status becoming an important indicator for evaluating the efficacy of MIA treatments. ²²

Amantadine (AMA, C10H17N) is a compound used to treat influenza virus and Parkinson's disease and has beneficial effects in treating traumatic brain injury and neurodegenerative diseases, mainly due to its resistance to inflammation and protection of neurons. ²³ , ²⁴ Studies have shown that AMA promotes dopamine release; it protects neuron activity by inhibiting calpain‐dependent phosphodiesterase 1; and it prevents neuroinflammation by increasing the release of glial cell‐derived neurotrophic factor (GDNF). ²⁵ , ²⁶ , ²⁷ AMA effectively lowers circulating levels of pro‐inflammatory cytokines in depression. ²⁸ Microglial activation is a critical feature of neuroinflammation, and experiments have demonstrated that AMA suppresses the inflammatory activation of microglia. ²⁹

Due to the broad‐spectrum antiviral activity and neuroprotective properties of AMAs, and recognizing viral infection as a substantial trigger for MIA and subsequent neurodevelopmental disorders in offspring, our objective is to explore the effects of AMA on the poly(I:C)‐induced MIA model. We aim to assess AMA's efficacy for autism in MIA offspring and explore new therapeutic strategies for MIA, including (1) MIA offspring behavior, (2) IL‐6 levels, (3) neuronal damage and apoptosis, and (4) microglial inflammation.

2. MATERIALS AND METHODS

2.1. Experimental animals

Three‐month‐old female C57BL/6 mice (weighing 22–28 g) were purchased from Shanghai Slack Laboratory Animal Co., Ltd. The animal housing facility was maintained at a temperature of 24 ± 1℃, a relative humidity of 45%–65%, and a 12‐h light–dark cycle.

2.2. MIA and treatment

Poly(I:C), a viral RNA analog, is a commonly used reagent for establishing the MIA mouse model. ¹² At 6:00 p.m. the day before the procedure, male and female mice were placed in breeding cages in the ratio 1:1. The following morning at 8:00 a.m., female mice were separated, and vaginal plugs were checked; the vaginal plug was considered as embryonic day (E) 0.5 of pregnancy. The significant weight gain of pregnant mice was observed at E12.5.

Pregnant mice were divided into three groups (control, poly(I:C) + AMA, and poly(I:C) groups). Pregnant mice in the control group were intraperitoneally injected with 1 mL/kg of saline at E12.5. Mice in the poly(I:C) + AMA group were intraperitoneally administered 25 mg/kg of amantadine hydrochloride once daily for three consecutive days starting from E11.5. Additionally, they received an intraperitoneal injection of 20 mg/kg of poly(I:C) at E12.5. The poly(I:C) group received only a 20‐mg/kg poly(I:C) injection at E12.5. The body weight changes were recorded.

After birth, the offspring weight was recorded every 2 days. The offspring were weaned and separated into individual cages at postnatal (P) day 21. Because male offspring from MIA models tend to exhibit a higher tendency toward depressive‐like behaviors, only male offspring were used for subsequent experiments. ³⁰ , ³¹ The male offspring were assigned to cages within each group, with four to six mice per cage. Behavioral tests were conducted when the offspring reached 2 months of age. The experimental process is shown in Figure S1.

2.3. Animal behavioral tests

Behavioral tests were conducted on 2‐month‐old offspring. Before each test, the mice were transported to the laboratory for 30 min. To eliminate scent cues, 75% ethanol was used to clear the apparatus. ³² , ³³ All behavioral data were analyzed using Noldus Ethovision XT 15 software (Noldus, the Netherlands).

The open field box consisted of a 40 × 40 × 40‐cm cube with a 20 × 20‐cm central area. Mice were gently placed in the central area, and their movements were recorded for 5 min.

The elevated plus maze consisted of two open arms (30 × 8 cm), two closed arms (30 × 8 × 15 cm), and a central platform (8 × 8 cm). Each mouse was gently placed on the central platform facing one of the open arms. The number of entries into the open and closed arms, as well as the time spent in each arm, was recorded over 5‐min period.

The three‐chamber apparatus consisted of three identical compartments (40 × 20 × 22 cm) connected by two small doors (8 × 5 cm) (Figure 2). The test mouse was free to explore the central chamber for 5 min. Social interaction phase: a mouse (stranger 1) was placed in one chamber, whereas an inanimate object (a cylindrical container) was placed in the other chamber. Test phase: the test mouse was allowed to explore for 5 min Then, the inanimate object was replaced with a new mouse (stranger 2), and the test mouse's activity was recorded again. The social preference index was calculated in the two phases.

FIGURE 2.

Effects of poly(I:C) (polyinosinic acid–polycytidylic acid) stimulation on the behavior of offspring. (A) Representative graphs of the open field experiment. (B) Total distance in the open field. (C) Frequency of entered central area of open field. (D) Residence time in the central area of open field. (E) Representative graphs of elevated plus maze experiment. (F) Residence time in the open arm of the elevated plus maze. (G) Frequency of entered in the open arm of the elevated plus maze. (H) Results of marble bead burial experiment. (I) Experimental design of the first phase of the three‐chamber social test. (J) Residence time in different compartments during the first phase of social test. (K) Preference index of first phase of the social test. (L) Experimental design of the second phase of the three‐chamber social test. (M) Residence time in different compartments during the second phase of social test. (N) Preference index of second phase of social test; in the open field and elevated plus maze experiments, control group (n = 12), poly (I:C) + AMA (amantadine) group (n = 10), and poly (I:C) group (n = 12); in the social test, each group (n = 8). Data are presented as mean ± SEM (standard error of the mean). *p < 0.05, **p < 0.01, and ***p < 0.001.

The marble burying test was conducted in a 40 × 40 × 40‐cm box. A 5‐cm‐thick layer of corncob bedding was evenly spread on the floor, and 20 black marbles (4 × 5 grid) were uniformly placed on top. Each mouse was free to explore for 30 min. The marbles buried to a depth of at least two thirds were counted for statistical analysis.

2.4. Nissl staining

Nissl staining was utilized to visualize Nissl bodies within neurons, serving as a crucial method for assessing neuronal damage. ³⁴ After completion of behavioral tests, the mice were killed, and they were perfused with phosphate buffered saline (PBS) to ensure thorough tissue preservation. Subsequently, brain tissues were dissected, fixed in 4% paraformaldehyde, and dehydrated in 30% sucrose solution. After dehydration, the brain tissue with olfactory bulbs and brainstem removed was cryo‐embedded in optimal cutting temperature (catalog no. 4583, Sakura) compound. Coronal hippocampal sections (5 μm thick) spanning coordinates from bregma −2.03 to −2.27 mm were prepared using bregma and lambda as anatomical landmarks. Histological staining was performed using a Nissl staining kit (catalog no. G1432, Solarbio). After staining, slides were coverslipped using neutral resin for preservation. Nissl staining patterns were observed under an inverted microscope (Ti2‐u, Nikon, Japan), and the number of Nissl‐positive cells was quantified using Image J software.

2.5. Immunofluorescence

Similar procedures were followed to obtain 10‐μm‐thick brain tissue sections for immunofluorescence (IF) experiments. The brain tissue sections were permeabilized using 0.5% Triton X‐100 for 20 min and then washed thrice with PBS. The sections were blocked using 5% bovine serum albumin–PBS solution for 30 min. The sections were incubated with primary antibodies Iba1 (17198, 1:200, Cell Signaling Technology [CST]) and NeuN (A6NFN3, 1:200, CST) at 4℃ overnight. The sections were washed thrice with PBS and incubated with secondary antibodies for 2 h, including goat anti‐rabbit IgG Brilliant 647 (catalog no. bs‐0295G‐BF647, Bioss) and goat anti‐mouse IgG Alexa 488 (catalog no. bs‐0296G‐AF488, Bioss). The sections were stained with 4′,6‐diamidino‐2‐phenylindole at room temperature for 15 min and then washed thrice with PBS. In addition, IF of CD68 (97778, 1:800, CST), tumor necrosis factor‐alpha (TNF‐α, 29084, 1:100, Signalway Antibody LLC), and interleukin‐1β (IL‐1β, ab254360, 1:100, Abcam) was assessed in BV‐2 cells. Slides were coverslipped, and images were obtained using a confocal microscope (C2, Nikon). Fluorescence intensity was quantitatively analyzed using ImageJ software.

2.6. TUNEL staining

Following protocols from previous studies and the instructions provided in the TUNEL (terminal deoxynucleotidyl transferase dUTP nick‐end labeling) staining kit (P‐CA‐301, Procell), 10‐μm‐thick brain tissue sections were then subjected to sequential procedures, including permeabilization, equilibration, labeling, nuclear staining, and coverslipping. Neuronal apoptosis was observed under a confocal microscope, and the apoptotic cells in various brain regions were quantitatively analyzed using ImageJ software.

2.7. Cell culture

BV‐2 cells (iCell‐m011), from C57BL/6 mouse microglia, were obtained from Shanghai iCell Bioscience Inc. The cells were cultured in Dulbecco's modified Eagle's medium (C11995500BT, Gibco) supplemented with 10% (v/v) fetal bovine serum (13011‐8611, Zhejiang Tianhang Biotechnology) and 1% penicillin/streptomycin (15140122, Gibco). Cells were cultured at 37.0℃ in a humidified atmosphere of 95% air and 5% CO₂. The complete medium was replaced every 2 days during normal cell culture. Once cells reached approximately 70% confluence, drugs were added to the medium for treatment. The control group received no extra treatment, the poly(I:C) group was treated with 5 μg/mL of poly(I:C) for 24 h, and the poly(I:C) + AMA groups received different concentrations of AMA and then 5 μg/mL of poly(I:C). The optimal treatment concentration of AMA was screened using t CCK‐8.

2.8. CCK‐8 assay

Cell viability was assessed using the Cell Counting Kit‐8 [CCK‐8] kit (AC11L054, Life‐iLab); 3000 BV‐2 cells were seeded into a 96‐well plate and inoculated for 24 h. Drugs were added to the wells according to the experimental groups: no drugs for the control group, 5 μg/mL of poly(I:C) for the poly(I:C) group, and varying concentrations (0, 1, 5, 10, 20, 40, and 80 μM) of AMA followed by 5 μg/mL of poly(I:C) for the poly(I:C) + AMA groups. After 24 h of incubation with the drugs, the 100‐μL medium containing 10% CCK‐8 was added to each well, and the cells were incubated for an additional 2 h. The Optical Density at 450 nm (OD₄₅₀) value was used to calculate cell viability.

2.9. Enzyme‐linked immunosorbent assay

Poly(I:C) can induce an increase in plasma IL‐6 levels in MIA, reaching a peak 2–3 h after intraperitoneal injection in rats, which subsequently affects offspring behavior and neurodevelopment. ⁴ IL‐6 concentrations in mouse plasma and cell culture supernatants were determined using the enzyme‐linked immunosorbent assay (ELISA) kit (JL20268, Jianglai). Briefly, serum or supernatant was centrifuged at 1000 g for 20 min and then incubated with biotinylated antibodies at 37°C for 1 h. After incubation, the reaction was terminated, and OD₄₅₀ values were measured using a microplate reader. The standard curve is y = 204.82x + 9.0596 (R ² = 0.998). The coefficients of variation were less than 10%.

2.10. Reactive oxygen species assay

Reactive oxygen species (ROS) staining of BV‐2 cells was performed using the ROS assay kit (E‐BC‐K138‐F, Elabscience). Briefly, cells were incubated with 2′‐7′‐dichlorofluorescein diacetate at 37°C for 20 min. The cells were washed thrice with PBS. Fluorescence images were obtained using a confocal microscope, and intensity was quantitatively analyzed using ImageJ software.

2.11. Protein extraction and Western blot

The protein in the hippocampal region was extracted using radioimmunoprecipitation assay buffer. BCA kit (SK1070‐500, Coolaber) was used to detect the protein concentration. The separated proteins were then transferred onto a PVDF membrane. The PVDF membrane was blocked with 5% skim milk for 1 h and incubated at 4°C in primary antibodies overnight, including Iba1 (17 198, 1:1000, CST), NeuN (A6NFN3, 1:1000, CST), GAPDH (10494‐1‐AP, 1:5000, Proteintech), Bcl‐2 (ab182858, 1:2000, Abcam), Bax (AF0120, 1:1000, Affinity), and cleaved caspase 3 (AF7022, 1:1000, Affinity). The next day, the membrane was washed thrice with TBST (C05‐07003, Bioss) and incubated for 2 h at room temperature in secondary antibodies such as goat anti‐rabbit IgG H&L (catalog no. bs‐0295G‐SA, Bioss) and goat anti‐mouse IgG H&L (catalog no. bs‐0296G‐SA, Bioss). After incubation, protein bands were visualized using an ECL kit (E412, Vazyme), and images were obtained using a Vilber Fusion Fx6 imaging system. Protein bands were analyzed using ImageJ software.

2.12. RNA‐seq

The cells from the poly(I:C) and the poly(I:C) + AMA (10 μM) groups were used for RNA‐seq. Transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China).

2.13. Statistical analysis

Data were analyzed using SPSS 20.0 and expressed as mean ± standard error of the mean. Graphical presentation was performed using GraphPad Prism 8.02. Data normality and homogeneity of variances were assessed using Kolmogorov–Smirnov and Levene's tests, respectively. For comparisons involving three or more groups, one‐way/multi‐way analysis of variance (ANOVA) was performed, and then Tukey's honestly significant difference test was performed for post hoc pairwise comparisons. Predefined comparisons between the two groups were analyzed using two‐tailed unpaired Student's t‐test, with Welch's correction for unequal variances. Categorical variables were evaluated using χ ² test with Yates's continuity correction. Multiple comparisons were adjusted using Tukey's (ANOVA) or Bonferroni (t‐test) methods to maintain a family‐wise error rate < 0.05. Statistical significance was defined as *p < 0.05, **p < 0.01, and ***p < 0.001; N.S. denotes nonsignificance (p ≥ 0.05).

3. RESULTS

3.1. Impact of AMA treatment on maternal and offspring weight

Based on previous studies, the MIA animal model using poly(I:C) was established successfully. ¹² The weight of pregnant mice at 12.5 days was significantly higher than that of nonpregnant mice (Figure S2). However, the injection of poly(I:C) significantly decreased body weight, and early treatment with AMA failed to reverse the decrease (Figure 1A). The release of IL‐6 in mothers reaches a peak within 3 h after the injection of poly(I:C). ⁴ , ³⁵ The IL‐6 levels at the 3‐h time point after the injection of poly(I:C) significantly increased in this study. But AMA treatment significantly decreased the level of IL‐6 (Figure 1B). To evaluate the effect of AMA treatment on offspring growth and development, we further analyzed the weight changes in all mice. The results showed signs of increased weight recovery in the offspring of the AMA treatment group, but the difference was not significant (Figure 1C).

FIGURE 1.

Effects of poly(I:C) (polyinosinic acid–polycytidylic acid) stimulation on maternal weight, inflammation, and offspring growth. (A) Changes in maternal mouse weight after poly(I:C) stimulation. The maternal mice were weighed after 48 h of injection. Each study has six parallel samples (n = 6). (B) Changes in maternal serum IL‐6 (interleukin‐6) after poly(I:C) stimulation. The concentration of IL‐6 in serum was detected after 3 h of injection. Study consists of 12 parallel samples in the control group (n = 12), 21 parallel samples in the poly(I:C) + AMA (amantadine) group (n = 21), and 18 parallel samples in the poly(I:C) group (n = 18). (C) Growth curves of the offspring in three groups. Each study has 12 parallel samples (n = 12). Data are presented as mean ± SEM (standard error of the mean). *p < 0.05, **p < 0.01, and ***p < 0.001.

3.2. AMA treatment effectively improves MIA offspring autistic‐like behaviors

Based on previous studies, male offspring were selected for experiment. ¹⁵ When the offspring reached 60 days of age, the offspring were subjected to a series of animal behavior tests to assess whether AMA treatment would affect offspring behavior. In the open field test, AMA treatment significantly increased the offspring's movement distance and frequency of entry into the central area (Figure 2A–C), but the difference in time spent in the central area was not significant (Figure 2D). In the elevated plus maze test, AMA treatment significantly increased the time and frequency of offspring entering into the open arms (Figure 2E–G). In the marble burying test, AMA treatment significantly reduced the number of buried marbles (Figure 2H).

The three‐chamber social test is commonly used to assess offspring's social ability when encountering unfamiliar conspecifics. ³⁶ The three‐chamber social test was used to evaluate the effect of AMA treatment on offspring social ability. In the sociability stage, MIA offspring spend more time interacting with objects than with conspecifics, and AMA treatment significantly increased the offspring's interaction time with conspecifics rather than objects, as confirmed by the preference index (Figure 2I–K). In the social novelty stage, AMA treatment did not significantly increase the time spent by the offspring with stranger 2 mice, but an increasing trend in time existed, which was well supported by the preference index. AMA treatment significantly increased the offspring's interaction with unfamiliar mice (Figure 2L–N). The three core behavioral impairments of autism are social deficits, communication impairments, and repetitive behaviors. The aforementioned behavioral studies confirm that AMA treatment can effectively improve autistic‐like behaviors induced by poly(I:C) in offspring.

3.3. AMA treatment reduces neuronal damage in MIA offspring

Impaired social ability in mice is closely related to neuronal damage in the hippocampal region, and Nissl staining is an important method for detecting neuronal damage. ³⁴ Behavioral tests confirmed that the poly(I:C)‐induced MIA caused behavioral impairments in offspring, and AMA treatment partially altered this outcome. Therefore, Nissl staining was used to evaluate the neuronal status of offspring. The MIA offspring exhibited more severe neuronal nuclear condensation and fragmentation in the hippocampal CA1 region, CA3 region, and cortex (Cx), and AMA treatment effectively prevented damage, leading to neuronal status in the three brain regions being closer to that in the blank control group (Figure 3A,B). Poly(I:C) induced the decrease in neurons in the hippocampal CA1 region (Figure 3C), CA3 region (Figure 3D), and Cx (Figure 3E), and AMA treatment prevented the reduction in neurons in the CA1 region and Cx (Figure 3C–E).

FIGURE 3.

AMA (amantadine) reduced the offspring's Nissl body that was stimulated by poly(I:C) (polyinosinic acid–polycytidylic acid). (A) Map of location of CA1, CA3, and Cx (cortex) in mouse brain. (B) Nissl staining in CA1, CA3, and Cx areas. Scale bar: 200 μm. (C) Proportion of positive cells in the CA1 area. (D) Proportion of positive cells in the CA3 area. (E) Proportion of positive cells in the Cx area; each group (n = 5). Red arrow represents apoptotic cells. Data are presented as mean ± SEM (standard error of the mean). *p < 0.05 and **p < 0.01; N.S. denotes nonsignificance (p ≥ 0.05).

3.4. AMA effectively prevents neuronal apoptosis in MIA offspring

Neuronal damage is often closely associated with apoptosis, and TUNEL staining is an important method for detecting neuronal apoptosis. TUNEL staining was used to assess neuronal apoptosis in the hippocampal and Cx brain regions. Poly(I:C)‐induced MIA increased the TUNEL fluorescence intensity in the hippocampal CA1, CA3, and Cx brain regions of offspring, indicating an increase in the number of apoptotic cells within the brain regions of MIA offspring. Further statistical analysis revealed that AMA treatment effectively reduced the proportion of apoptotic cells in the CA1, CA3, and Cx regions (Figure 4A–D).

FIGURE 4.

AMA (amantadine) reduced cellular apoptosis that was stimulated by poly (I:C) (polyinosinic acid–polycytidylic acid) in the offspring brain. (A) TUNEL (terminal deoxynucleotidyl transferase dUTP nick‐end labeling) staining in CA1, CA3, and Cx (cortex) areas. Scale bar: 200 μm. (B) Proportion of TUNEL‐positive cells in the CA1 area. (C) Proportion of TUNEL‐positive cells in the CA3 area. (D) Proportion of TUNEL‐positive cells in the Cx area. (E) Western blotting plot of the apoptosis‐related protein in the mouse brain. (F) Relative expression of Bcl‐2 protein in the mouse brain. (G) Relative expression of Bax protein in the mouse brain. (H) Bcl‐2 to Bax ratio. (I) Relative expression of cleaved caspase 3 protein in the mouse brain. Each group (n = 5). Data are presented as mean ± SEM (standard error of the mean). *p < 0.05, **p < 0.01, and ***p < 0.001; N.S. denotes nonsignificance (p ≥ 0.05).

Previous studies have shown that the Bcl‐2 protein mainly inhibits the functions of Bax and caspase proteins, maintains mitochondrial membrane potential balance, and plays a crucial role in anti‐apoptosis. ³⁷ The Bcl‐2, Bax, and cleaved caspase 3 proteins in the offspring hippocampus were detected using Western blotting. The poly(I:C) injection decreased the expression of Bcl‐2 and increased the expression of cleaved caspase 3. AMA increased the expression of Bcl‐2 in the hippocampal region of MIA offspring and decreased the expression of cleaved caspase 3 (Figure 4E–I). AMA treatment effectively reduced the ratio of Bcl‐2 to Bax (Figure 4H). These studies confirm that AMA treatment can effectively reverse the increase in neuronal apoptosis in the hippocampal CA1, CA3, and Cx regions of MIA offspring, which is closely related to the Bcl‐2/Bax/caspase 3 signaling pathway.

3.5. AMA reduces the occurrence of neuroinflammation in MIA offspring

Studies have confirmed that poly(I:C)‐induced MIA led to changes in the specific markers of microglia Iba1 and neurons NeuN in offspring brain regions, inducing the occurrence of neuroinflammation. ³⁸ Behavioral tests showed that AMA treatment can improve autistic‐like behaviors in MIA offspring, and neuroinflammation is an important inducer of abnormal animal behavior. ³⁹ IF staining was used to evaluate whether AMA treatment has a beneficial effect on neuroinflammation in offspring. IF staining of Iba1 showed that microglia in the hippocampal CA1, CA3, and Cx brain regions of MIA offspring were highly activated, indicating that poly(I:C) induced neuroinflammation. AMA treatment partially alleviated the abnormal activation of microglia in the hippocampal CA1, CA3, and Cx brain regions (Figure 5A,B). On the contrary, the fluorescence intensity of NeuN showed an opposite trend to that of Iba1 (Figure 5C,D). The results of Western blotting of Iba1 and NeuN proteins in the hippocampal region were consistent with those of IF, showing that AMA treatment reduced the expression of Iba1 and increased the expression of NeuN (Figure 5E–G). These findings demonstrated that AMA treatment attenuated neuroinflammation in the hippocampal CA1, CA3, and Cx regions of MIA offspring.

FIGURE 5.

Effects of poly(I:C) (polyinosinic acid–polycytidylic acid) stimulation on the expression of Iba1 and NeuN proteins. (A) Iba1 immunofluorescence (IF) in CA1, CA3, and Cx (cortex) areas. Scale bar: 200 μm. (B) Iba1 fluorescence intensity in CA1, CA3, and Cx areas. (C) NeuN IF in CA1, CA3, and Cx areas. Scale bar: 200 μm. (D) Iba1 fluorescence intensity in CA1, CA3, and Cx areas. (E) Western blotting plot of Iba1 and NeuN proteins in the mouse brain. (F) Relative expression of Iba1 protein in the mouse brain. (G) Relative expression of NeuN protein in the mouse brain; each group (n = 5). Data are presented as mean ± SEM (standard error of the mean). *p < 0.05, **p < 0.01, and ***p < 0.001; N.S. denotes nonsignificance (p ≥ 0.05).

3.6. Regulation of microglia by AMA

Our animal experiments have shown that AMA treatment can alleviate autistic‐like behaviors in MIA offspring, reduce neuronal damage, decrease neuronal apoptosis, and inhibit neuronal inflammation to some extent, revealing the important role of offspring microglia in the treatment process. Therefore, BV‐2 cells were used to explore the mechanism of action of poly(I:C) infection and AMA treatment. Based on previous studies, 5 μg/mL of poly(I:C) was used to stimulate BV‐2 cells and intervened with different concentrations of AMA. ²⁹ , ⁴⁰ The cell viability results showed that adding only poly(I:C) to the culture medium significantly increased the viability, whereas 10 μmol/L AMA significantly decreased the viability (Figure 6A). Additionally, ELISA results showed that treatment with 10 μmol/L AMA effectively reduced the IL‐6 concentration in the cell supernatant (Figure 6B). Therefore, subsequent cell experiments selected 10 μmol/L AMA as the optimal concentration for treatment.

FIGURE 6.

AMA (amantadine) reduced the impairment of poly(I:C) (polyinosinic acid–polycytidylic acid) stimulation on BV‐2 cells. (A) Cell viability of different concentrations of AMA upon poly(I:C) stimulation. The concentration of poly(I:C) was 5 μg/mL. (B) IL‐6 (interleukin‐6) concentration in the supernatant of the cells under poly(I:C) stimulation with different concentrations of AMA The concentration of poly(I:C) was 5 μg/mL. (C) The bright‐field image of cells in the different groups. Left image scale bar: 50 μm. Right image scale bar: 25 μm. (D) Cell ROS (reactive oxygen species) fluorescence image and mean fluorescence value statistics. Left image scale bar: 200 μm. Right image scale bar: 100 μm. (E) Immunofluorescence images of Iba1, CD68, TNF‐α (tumor necrosis factor‐alpha), and IL‐1β (interleukin‐1β) proteins in the BV‐2 cells. Scale bar: 100 μm; each group (n = 5). Data are presented as mean ± SEM (standard error of the mean). *p < 0.05, **p < 0.01, and ***p < 0.001; N.S. denotes nonsignificance (p ≥ 0.05).

Brightfield microscopy field showed that poly(I:C) stimulated the proliferation of BV‐2 cells. Meanwhile, poly(I:C)‐stimulated BV‐2 cells exhibited spindle‐shaped morphology, enlarged cell bodies, and increased processes. Treatment with AMA effectively prevented the morphological changes in BV‐2 cells (Figure 6C). AMA significantly decreased the ROS intensity in poly(I:C)‐stimulated BV‐2 cells (Figure 6D). Oxidative stress is a risk factor for neuroinflammation. Similar to animal experiments, the IF staining of Iba1 showed that poly(I:C) increased the fluorescence intensity in BV‐2 cells, whereas AMA treatment decreased the fluorescence intensity (Figure 6E). AMA decreased the expression of CD68 in poly(I:C)‐induced BV‐2 cells (Figure 6E). AMA decreased the expression of TNF‐α and IL‐1β in poly(I:C)‐induced BV‐2 cells (Figure 6E).

3.7. AMA treatment effectively regulates inflammation‐related pathways to inhibit microglial activation

RNA‐seq technology was used to investigate the mechanism that AMA improves poly(I:C)‐induced impairment in BV‐2 cells. Analysis of the RNA‐seq data identified 534 differentially expressed genes (DEG), as shown in Figure 7. The poly(I:C)‐induced group exhibited 521 significantly upregulated genes and 13 significantly downregulated genes compared to the AMA‐treated group (Figure 7A). KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment analysis showed that the DEGs were enriched in several pathways, including cytokine–cytokine receptor interaction, inflammatory bowel disease, and Th1 and Th2 cell differentiation pathways (Figure 7B). Interestingly, IL‐6 was associated with 7 pathways in the top 20 pathways.

FIGURE 7.

Transcriptome study of AMA (amantadine) reducing poly(I:C) (polyinosinic acid–polycytidylic acid)–induced BV‐2 cell damage. (A) Volcano plot presents DEGs (differentially expressed genes) in poly(I:C) versus AMA. The y‐axis represents the −log10 p‐value for each gene, whereas the x‐axis represents the log2 fold change. Horizontal dotted line indicates the cutoff criteria for DEGs with p‐value ≤0.05. Red circles indicate upregulated genes, whereas blue circles indicate downregulated genes. (B) KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis of the total DEGs. KEGG enrichment bubble plot of top 20 pathways. (C) Bar diagram shows GO (Gene Ontology) terms enriched from DEGs. The top 10 GO terms in the three categories ranked by −log10 p‐value. The red font denotes the signaling pathways associated with IL‐6 (interleukin‐6); N.S. denotes nonsignificance (p ≥ 0.05).

Furthermore, the results of Gene Ontology (GO) enrichment analysis showed that DEGs were enriched in biological processes related to cell adhesion, cell surface receptor signaling pathway, and angiogenesis. In the cellular component categories, DEGs were enriched in the external side of the plasma membrane, extracellular region, and plasma membrane. In the molecular function categories, DEGs were enriched in extracellular matrix structural constituent, transmembrane signaling receptor activity, and protein binding (Figure 7C).

4. DISCUSSION

Both epidemiological investigations and animal studies have demonstrated that pregnant mothers are highly susceptible to adverse external stimuli, leading to MIA. Abnormal MIA increases the risk of neurodevelopmental disorders and mental illnesses in offspring. The prevention and treatment of MIA offspring remain challenging in clinical medicine. ⁴¹ , ⁴² This study explores whether maternal administration of AMA during pregnancy has a positive effect on poly(I:C)‐infected MIA offspring.

During pregnancy, the embryo is closely connected to the mother, and signals of increased inflammatory factors due to maternal immune system activation may be transmitted through the placenta to the embryo, leading to neurodevelopmental disorders in offspring. ⁴³ The studies have found that poly(I:C) infection of the mother leads to a synchronized surge in IL‐6 in maternal blood, placenta, and fetal brain tissue. Additionally, injecting IL‐6 into pregnant mice can induce autism‐like behaviors in offspring, whereas using IL‐6 antibodies effectively prevents abnormal behaviors in offspring. ¹⁶ , ¹⁷ , ⁴⁴ Clinical studies have also confirmed that IL‐6 levels in the blood and cerebrospinal fluid of children with autism are significantly higher than those in healthy children. ⁴⁵ , ⁴⁶ These studies suggest that IL‐6 plays a crucial role in mediating neurodevelopmental disorders in MIA offspring. AMA treatment effectively reduced IL‐6 levels in the blood of MIA mice, indicating that AMA attenuates behavioral abnormalities and neurodevelopmental impairments in MIA offspring by reducing IL‐6 release. On the contrary, elevated IL‐6 can disrupt the placental immune balance, affect maternal nutrient transport efficiency, and directly impact progenitor cells, affecting neurogenesis and gliogenesis in the fetus. ⁴⁷ , ⁴⁸ , ⁴⁹ Furthermore, AMA treatment reduced the extent of weight loss in MIA mice and improved the developmental progression of offspring, indicating that AMA treatment mitigated the impact of poly(I:C) infection on both mothers and offspring.

Animal behavior represents the complex, multilayered, and dynamic final output of the central nervous system (CNS), controlling movement and emotions. Autism is primarily characterized by social communication deficits, increased repetitive behaviors, and cognitive impairments. ¹⁴ , ⁵⁰ Numerous studies have confirmed that MIA offspring induced by poly(I:C) exhibit prominent autism‐like behaviors, such as a significant increase in marble burying, ³⁵ , ⁴⁹ decreased PPI, ¹⁵ and lack of social ability. ⁴⁹ , ⁵¹ As expected, AMA effectively improved autism‐like behaviors in MIA offspring, which is consistent with previous results, showing that inhibiting maternal IL‐6 release improves autism‐like behaviors in offspring, ³⁵ further indicating that AMA may alter abnormal behaviors in MIA offspring by reducing IL‐6 expression.

The hippocampus is the primary brain region responsible for learning and memory, playing a crucial role in regulating animal behavior. ⁵² Previous research has shown that inflammatory factors induced by poly(I:C) can be transported from the mother to the embryo through the placenta, disrupting the proliferation and differentiation of early hippocampal neural progenitor cells and causing lasting harm to neural development in the hippocampal region of the embryo. ³⁵ , ⁴⁷ Studies have confirmed that MIA offspring exhibit significant neuronal damage in the hippocampus, with a decrease in Nissl‐positive cells and an increase in cells with pyknosis and chromatin dissolution, indicating impaired neuronal function in the hippocampus of autism offspring. ⁵³ , ⁵⁴ , ⁵⁵ Figure 3 shows that poly(I:C) infection leads to increased neuronal damage in the CA1, CA3, and Cx regions in offspring, with significant reduction in Nissl‐positive cells. AMA treatment effectively reduced neuronal damage, revealing its protective function for neurons in offspring. AMA can effectively inhibit the release of pro‐inflammatory factors from microglia while increasing the expression of GDNF. ⁵⁶ Clinical studies have also identified AMA as a potential drug for children with autism, as it can improve language skills and social interaction abilities in these children, ⁵⁷ , ⁵⁸ , ⁵⁹ supporting our findings that AMA treatment improves autism‐like behaviors in MIA offspring.

Neuronal apoptosis often follows neuronal damage. Previous studies have found an increase in neuronal apoptosis in the hippocampus and Cx regions of autism offspring, and hippocampal neuronal damage is an important factor inducing autism‐like behaviors. ³⁵ , ⁶⁰ We confirmed that poly(I:C) infection increases neuronal apoptosis in the hippocampus and Cx regions of offspring, whereas AMA treatment reduces the number of apoptotic neurons. Further examinations showed that Bcl‐2 was downregulated and cleaved caspase 3 was upregulated in the hippocampus of MIA offspring. AMA can reduce cell apoptosis, suggesting that AMA regulates microglial apoptosis through the Bcl‐2/Bax/caspase 3 signaling pathway in animal and cell experiments. Microglia play a key role in maintaining innate immunity in the brain, maintaining CNS homeostasis, and preventing neuroinflammatory pathology. ⁶¹ The poly(I:C)‐stimulated BV‐2 cells exhibited an increase in cell viability, with cells morphologically changing from round to spindle shaped, and an increase in the number of branches, along with a significant increase in ROS fluorescence intensity, indicating that poly(I:C) stimulation induces microglia to enter the M1 activation state. This explained that AMA can effectively prevent microglial activation. In previous studies, the BV‐2 cells can change shape to an amoeboid‐like or dense form when stimulated by poly(I:C) or LPS. Animal experiments also confirmed that poly(I:C) infection during pregnancy leads to significant activation of microglia in offspring, indicating that poly(I:C) induces neuroinflammation. ⁴¹ Poly(I:C) activates TLR3, initiating the TIR‐domain‐containing adapter‐inducing interferon‐β signaling pathway and increasing neuroinflammation levels. ¹¹ , ⁶²

Poly(I:C) infection leads to significant activation of microglia in multiple brain regions of MIA offspring, and microglial activation is an important cause of autism‐like behaviors in offspring. ¹³ , ¹⁴ , ⁶³ Iba1 and CD68 are markers of M1 microglial activation; their IF staining is used to assess the occurrence of neuroinflammation. ¹⁵ , ³⁵ In animal experiments, we found the significant upregulation of Iba1 protein and downregulation of NeuN protein in the hippocampal CA1, CA3, and Cx regions of MIA offspring. AMA reduced Iba1 protein expression and increased NeuN expression, suggesting that AMA treatment can ameliorate neuroinflammation in the hippocampus and Cx of offspring, possibly related to reduced IL‐6 levels in maternal blood. Poly(I:C) stimulation for BV‐2 cells significantly upregulated Iba1 expression and CD68 expression, whereas AMA treatment reversed these effects, indicating a direct protective effect of AMA on microglia. To further investigate the protective mechanism and related pathways of AMA treatment, we conducted transcriptomic sequencing of BV‐2 cells. Bioinformatics analysis revealed that the therapeutic effects of AMA on BV‐2 cells might involve pathways such as cytokine–cytokine receptor interaction, inflammatory bowel disease, and Th1 and Th2 cell differentiation pathways, all closely related to IL‐6 expression. The studies have confirmed that these signaling pathways are altered in the brain tissue of autism patients. ⁶⁴ , ⁶⁵ , ⁶⁶

As shown in Figure 8, our results suggest that AMA may regulate the abnormal activation of microglia by controlling inflammation‐related pathways, particularly by inhibiting the release of IL‐6, returning microglia to a resting state, thereby protecting neurons. In this study, we administered AMA during the maternal infection window and found that it effectively reduced IL‐6 release and altered autism‐like behaviors in the offspring. AMA treatment mitigated neuronal damage and apoptosis in the hippocampal region of the offspring and attenuated neuroinflammation. Cellular assays confirmed that poly(I:C) induced M1‐type activation of microglia, leading to increased proliferation and ROS production, which were reversed by AMA treatment. RNA‐seq analysis further substantiates that AMA may exert protective effects on microglia by modulating inflammatory pathways. These indicate that AMA could be a potential therapeutic drug for treating MIA‐induced conditions.

FIGURE 8.

Amantadine (AMA) suppresses neuronal apoptosis and attenuates neuroinflammation, thereby mitigating maternal immune activation (MIA)–induced behavioral impairments in offspring.

5. CONCLUSIONS

This study found that AMA intervention treatment can effectively ameliorate autism‐like behaviors in MIA offspring, and its mechanism may be associated with the reduction in the inflammatory factors, as well as the prevention of the generation of MI‐type microglia.

AUTHOR CONTRIBUTIONS

Jianfei Wu: Conceptualization; data curation; formal analysis; investigation; resources; software; visualization; writing – original draft. daixu Wei: Validation; visualization; writing – original draft; writing – review and editing. Yu Liu: Conceptualization; resources; supervision; validation; visualization; writing – original draft; writing – review and editing. Binglong Wang: Conceptualization; project administration; resources. Yilin Wang: Formal analysis; software; supervision; validation. Bo Liu: Formal analysis; funding acquisition; writing – review and editing. Youguo Tan: Funding acquisition; project administration; writing – review and editing. Duanfang Cai: Project administration; resources; supervision. Kezhi Liu: Funding acquisition; resources. Daixu Wei: Funding acquisition; visualization; writing – original draft; writing – review and editing.

FUNDING INFORMATION

This work was supported by grants from the National Natural Science Foundation of China (31900950), Project Supported by the Natural Science Basic Research Plan in Shaanxi Province of China (2024JC‐YBMS‐706), the National Key Research and Development Program of China (2022YFC2009900), the Key Science and Technology Plan Projects in Zigong (2022ZCNKY07, 2023‐NKY‐01‐02, 2023‐NKY‐02‐13, and 2023‐NKY‐02‐14), the Collaborative Innovation Project of Zigong Medical Big Data and Artificial Intelligence Research Institute (2023‐YGY‐1‐02 and 2024‐YGY‐02‐04), the Zigong Science and Technology Program (2023YKY11), and the Scientific Research Project of Zigong Health Commission (22yb001 and 24zd008).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. They bear sole responsibility for the content and composition of this article.

ETHICS STATEMENT

The experimental design and use of animals in this study were approved by the Ethics Committee of the Zigong Mental Health Center (ethics approval no. ZPRC‐20230620), and complied with the guidelines provided by the National Research Council Guide for the Care and Use of Laboratory Animals in China.

Supporting information

FIGURE S1. An overview of the applied protocol for experiments. The control group was injected with saline after 12.5 days of pregnancy. The poly (I:C) (polyinosinic acid–polycytidylic acid) group was injected with poly (I:C) after 12.5 days of pregnancy. The poly (I:C) + AMA (amantadine) group was injected with poly (I:C) at 12.5 days after pregnancy and was also injected with AMA at 11.5, 12.5, and 13.5 days after pregnancy. The offspring was subjected to the behavioral experiment at 60 days after birth. The offspring was subjected to the molecular experiment at 75 days after birth.

FIGURE S2. The change in maternal body weight after 12.5 days of pregnancy. The maternal mice were not injected with any reagents; each group (n = 12). Data are presented as mean ± SEM (standard error of the mean). ***p < 0.001.

Wu J, Liu Y, Wang B, et al. A potential strategy for improving offspring behavior in maternal immune activation: Amantadine‐mediated suppression of neuroinflammation. Anim Models Exp Med. 2025;8:1836‐1850. doi: 10.1002/ame2.70059