Autism spectrum disorder (ASD) is a complex neurodevelopmental condition that affects social communication and behavior (Hodges et al., 2020). While the precise cause of ASD remains unknown, it is believed that both genetic and environmental factors contribute to its development (Almandil et al., 2019).

Initially introduced as an antiepileptic drug, valproic acid (VPA) is now also used to treat neurological disorders such as bipolar disorder and migraines (Romoli et al., 2019). However, it has been identified as a teratogenic drug, associated with risks such as congenital malformations, developmental delays, reduced cognitive function, and an increased risk of ASD when taken during pregnancy (Christensen et al., 2013; Juliandi et al., 2015). Evidence also shows that VPA exposure during pregnancy in humans increases the risk of autism in children (Roullet et al., 2013; Taleb et al., 2021). Also, is commonly used to induce autistic-like behavior in rodent models for research purposes (Nicolini and Fahnestock, 2018).

ASD affects various brain regions, with the hippocampus being a crucial part impacted by the disorder (Banker et al., 2021). Studies have shown that prenatal exposure to VPA leads to oxidative stress in the hippocampus (Jian et al., 2022), potentially affecting neural structures, especially dendrites and their spines, which are essential for proper hippocampal function (Martínez-Cerdeño et al., 2016). Oxidative stress may also influence beta-secretase 1 (BACE1), which has been linked to cognitive decline (Mouton-Liger et al., 2012; Zuliani et al., 2021). Inflammation has also been reported to play an important role in the pathophysiology of ASD, with markers like TNFα increased in both ASD patients and VPA-exposed rodents (Deckmann et al., 2018). One consequence of these changes may be a reduction in the density of dendritic spines in areas such as the hippocampus, eventually leading to behavioral alterations such as core autism behaviors and cognitive declines. Additionally, there may be an increased chance of apoptosis in the hippocampus, as seen in other studies (Zohny et al., 2023).

Transcranial magnetic stimulation (TMS) is known for its ability to stimulate the brain cortex and induce changes in activity (Afshari et al., 2023). Repeated TMS (rTMS) protocols involve multiple magnetic stimulations at known frequencies over multiple sessions to produce long-term changes in brain cortex activity (Marder et al., 2022). High-frequency rTMS (HF-rTMS), typically defined as frequencies ≥ 5 Hz, can increase activity, while low-frequency rTMS (LF-rTMS) inhibits brain cortex activity (Wang J. et al., 2020). Several studies have reported the antioxidant activity of rTMS, leading to neuroprotective effects, and its ability to increase brain-derived neurotrophic factor (BDNF) (Shang et al., 2016; Medina-Fernández et al., 2018), potentially influencing dendrite growth and spine maturation. It has also been shown that TMS, specifically LF-rTMS, can have beneficial effects on models of ASD induced by both VPA and maternal separation (Tan et al., 2018; Xu et al., 2024). Also, In our previous study, we observed beneficial effects of LF-rTMS on ASD conditions, consistent with the findings of the abovementioned studies (Afshari et al., 2024). There are reports indicating that HF-rTMS does not have a treatment effect on social dysfunction in FMR^–/– mice (Hou et al., 2021) However, several studies report beneficial effects in different rodent models (Choung et al., 2021; Kim et al., 2024; Zhang et al., 2024). As noted, the effects of LF-rTMS and HF-rTMS on the brain differ and have been reported with varying outcomes, making it necessary to investigate both frequencies to determine the most effective method.

As mentioned earlier, there are studies that have focused on the effect of LF-rTMS on ASD, but the effects of HF-rTMS on this condition have not been thoroughly investigated. Additionally, the exact mechanisms by which TMS affects ASD models remain unknown. Therefore, the primary objective of our study is to utilize HF-rTMS treatment with a focus on the hippocampus to investigate the potential treatment effects of this method on autistic behavior and cognitive impairment in VPA-induced ASD, as well as the possible mechanisms underlying the effects of HF-rTMS.

In this study, we evaluated the potential therapeutic impact of HF-rTMS on ASD rats induced by prenatal VPA exposure. Our findings demonstrate that prenatal VPA administration leads to long-term effects on the nervous system, influencing various behavioral characteristics. Prolonged HF-rTMS treatment can effectively reverse cognitive dysfunction associated with ASD.

ASD affects multiple brain regions, with some identified as significant contributors to ASD symptoms, including the hippocampus. Primarily involved in learning and memory, the hippocampus is susceptible to damage from various conditions (Anand and Dhikav, 2012). Consequently, hippocampal abnormalities are implicated in cognitive impairments associated with ASD (Banker et al., 2021). Animal models of ASD also reveal abnormalities in synaptic functions within the hippocampus. Moreover, hippocampus connectivity with other brain areas, such as the striatum and amygdala, may have a role in repetitive behaviors, another core symptom of ASD, as well as in social dysfunction (Kim et al., 2016; Tan et al., 2018). Furthermore, this structure plays a role in anxiety and stress management (Ghasemi et al., 2022).

In our study, prenatal injection of VPA resulted in behavioral changes in offspring. In addition to core ASD behaviors such as reduced social interaction, increased grooming, and repetitive digging behavior, we observed anxiety, impaired spatial memory, reduced recognition memory, and decreased stress coping strategies in these rats. Other studies have also reported cognitive impairments in prenatal VPA-induced autism (Nicolini and Fahnestock, 2018).

The hippocampus of offspring rats exposed prenatally to VPA injection showed increased oxidative stress and inflammation in other studies (Cristiano et al., 2022; Saadat et al., 2023). Oxidative stress has the potential to damage neural cells in the hippocampus and increase apoptosis (Saadat et al., 2023). Damage caused by oxidative stress may also increase BACE1 activity, which is related to the accumulation of amyloid β in neural cells (Mouton-Liger et al., 2012). These events may potentially lead to cognitive impairments. MMP also plays a role as a marker for cell health (Sakamuru et al., 2016). Mitochondrial dysfunction can lead to elevated ROS levels and eventually increased oxidative stress (Lenaz, 2001). Additionally, microglial activation, observed in this model through triggering receptor expressed on myeloid cells 2 downregulation (Luo et al., 2023), which has been implicated in autistic individuals (Liao et al., 2020). In our study, VPA-induced autism led to increased oxidative factors, reduced antioxidants, elevated caspase 3, and increased MMP levels, ultimately leading to heightened BACE1 activity.

BDNF is a neurotrophic component involved in neural development, neural plasticity, and synaptogenesis (Wei et al., 2020). It may also increase antioxidant activity in the body and have a protective effect against oxidative damage (Wang D. et al., 2020). BDNF may also increase MAP2 expression in neural cells (Chen et al., 2017). MAP2 is crucial in the cytoskeleton of neural cells, playing a role in microtubule polymerization and ultimately being associated with axon and dendrite structure (Chen et al., 2017). It has also been reported that the CA1 area of the hippocampus is important for social recognition in ASD (Chai et al., 2021). Evidence also shows that the CA1 area is critical in terms of cognitive dysfunction when affected by related disorders like dementia (Langdon et al., 2013). In other words, in the VPA model, when it comes to cognitive impairment, this area is also important to consider, as other studies have shown reduced dendritic spine density in pyramidal neurons of this area and spatial memory impairment (Lima-Castañeda et al., 2023). In our experiment, VPA-exposed rats showed a decrease in BDNF and MAP2, which was also evident in histological analysis with a reduction in dendritic spines in the CA1 area of the hippocampus.

Several studies have used TMS to treat individuals with ASD, but despite all efforts, TMS therapy is still in its early stages (Savino et al., 2023). One study on rats showed that LF-rTMS can improve ASD induced by neonatal isolation through the regulation of GABA transmission (Tan et al., 2018). In another study, LF-rTMS on FMR-/- mice improved social dysfunction, but they also reported that HF-rTMS cannot affect social dysfunction (Hou et al., 2021). In our study, HF-rTMS improved the core symptoms of ASD in the prenatal VPA model. HF-rTMS in preclinical research has shown the ability to improve cognitive impairment (Ma et al., 2019). It also reduced immobility time in mice (Cambiaghi et al., 2022) and anxiety in radiation induced brain injury model (Qin et al., 2024), as observed in our study with VPA-exposed rats.

Several studies on autistic patients have reported that HF-rTMS may help relieve core symptoms associated with ASD. For example, a study by Yang et al. reported that HF-rTMS on the left inferior parietal lobule can improve core symptoms of ASD in children (Yang et al., 2019). Additionally, in another study by Enticott et al. (2014) 5 Hz rTMS on the dorsomedial prefrontal cortex reduced social impairment. On the other hand, our study focused on the hippocampus. HF-rTMS has been used in several preclinical studies to reverse hippocampal dysfunction. For instance, in one study on an Alzheimer’s mouse model, HF-rTMS showed therapeutic effects and activated the dopaminergic pathway (Choung et al., 2021). In another study, also in an Alzheimer’s model, HF-rTMS demonstrated neuroprotective effects (Kim et al., 2024). Additionally, in a study comparing both LF-rTMS and HF-rTMS on Alzheimer’s disease, it was reported that both frequencies had therapeutic effects. This suggests that despite different protocols, both LF and HF-rTMS can have beneficial outcomes (Zhang et al., 2024).

Another factor that may affect rat performance in behavioral tests is mobility. Although the distance traveled in the open field test is not enough to make it significant, it seems to have an increased effect on the mobility of the rats (evident in the excessive exploration frequency in the Barnes maze test compared to sham rats). Because the coil not only stimulates the desired areas but also stimulates other parts of the brain, this outcome in rat studies may be observed. As other studies have also reported, this increased activity may be related to striatal alteration (Arfani et al., 2017). Also, we did not observe any adverse effects with HF-rTMS on rats.

Studies have claimed that rTMS has an antioxidant effect (Medina-Fernández et al., 2018). This device has the ability to reduces neural cell death induced by oxidative stress (Kim et al., 2024). Conversely, it also affects inflammatory markers, leading to a reduction in ROS production (Medina-Fernández et al., 2018). However, the exact mechanism of the antioxidant activity of rTMS remains unclear. In our study, HF-rTMS over a 14-day period reduced oxidative markers and increased antioxidant activity. It also improved MMP and TNFα, which may further reduce oxidative damage and prevent apoptosis, as reported in our study. Reduction of oxidative stress may also play a role in reducing BACE1 and, therefore, improving cognitive processes. Our data also showed an increase in BDNF in the hippocampus under HF-rTMS treatment, which may also contribute to the reduction of oxidative stress, as mentioned earlier. Another study reported increased BDNF and SYN in the hippocampus after rTMS treatment (Shang et al., 2016). As mentioned before, BDNF may also increase MAP2 (Chen et al., 2017). The increased levels of MAP2, SYN, and eventually dendritic spines in the hippocampus lead to better behavioral outcomes with HF-rTMS treatment.

One limitation of our study is the lack of detailed histological analysis, such as immunohistochemistry and other methods, which are recommended for future studies. Additionally, investigating the density of each type of dendritic spine could provide a better understanding of the effectiveness of HF-rTMS treatment, and this is also recommended for future research. While our study did not aim to explore sex differences or the treatment effects of HF-rTMS on different genders, these are important considerations for future investigations.

In conclusion, our experiment demonstrated that HF-rTMS has the potential to improve ASD symptoms induced by VPA in rats. It showed efficacy in enhancing memory, reducing anxiety, and improving stress coping strategies. These improvements may be attributed to the antioxidant properties of HF-rTMS and its capacity to increase levels of BDNF, SYN, and the density of dendritic spines.