Effect of WiFi signal exposure in utero and early life on neurodevelopment and behaviors of rats

The aim of this study is to examine the long-term effects of prenatal and early-life WIFI signal exposure on neurodevelopment and behaviors as well as biochemical alterations of Wistar rats. On the first day of pregnancy (E0), expectant rats were allocated into two groups: the control group (n = 12) and the WiFi-exposed group (WiFi group, n = 12). WiFi group was exposed to turn on WiFi for 24 h/day from E0 to postnatal day (PND) 42. The control group was exposed to turn-off WiFi at the same time. On PND7-42, we evaluated the development and behavior of the offspring, including body weight, pain threshold, and swimming ability, spatial learning, and memory among others. Also, levels of proteins involved in apoptosis were analyzed histologically in the hippocampus in response to oxidative stress. The results showed that WiFi signal exposure in utero and early life (1) increased the body weight of WiFi + M (WiFi + male) group; (2) no change in neuro-behavioral development was observed in WiFi group; (3) increased learning and memory function in WiFi + M group; (4) enhanced comparative levels of BDNF and p-CREB proteins in the hippocampus of WiFi + M group; (5) no neuronal loss or degeneration was detected, and neuronal numbers in hippocampal CA1 were no evidently differences in each group; (6) no change in the apoptosis-related proteins (caspase-3 and Bax) levels; and (7) no difference in GSH-PX and SOD activities in the hippocampus. Prenatal WiFi exposure has no effects on hippocampal CA1 neurons, oxidative equilibrium in brain, and neurodevelopment of rats. Some effects of prenatal WiFi exposure are sex dependent. Prenatal WiFi exposure increased the body weight, improved the spatial memory and learning function, and induced behavioral hyperactivity of male rats.


Introduction
As the economy and society develops, Wireless Fidelity (WiFi) communication services are extensively used in household, industrial, military, medical, and scientific setting in recent years. The increase in exposure to the WiFi wireless communication signal has posed major concerns with regard to its effects on human health (Othman et al. 2017a;At-Assa et al. 2013).
Potentially harmful effects of WiFi exposure have been studied on various tissues and body systems (Redmayne et al. 2013). Studies have suggested that continuous RF radiation exposure could have an effect on human health and lead to disorders like headaches, cancer, anemia, among other health problems (Othman et al. 2017b;Redmayne et al. 2013;Pallarés et al. 2013;Aït-Aïssa et al. 2012). One research revealed that in rats, WiFi radiation (2.45 GHz) can cause a reduction in sperm parameters and an upsurge in apoptosis-positive Responsible Editor: Mohamed M. Abdel-Daim Hongmei Wu and Dongyu Min equally contributed to this paper.

Highlights
• WiFi signal exposure in utero and early life increased the body weight of male rat. • No change in neuro-behavioral development was observed in WiFi signal exposure group. • WiFi signal exposure increased learning and memory function of male rat. • WiFi signal exposure enhanced relative levels of BDNF and p-CREB proteins in the hippocampus. • No neuronal loss and degeneration were observed in WiFi signal exposure group.
• No change of apoptosis-related proteins (Bax and caspase-3) and activities of SOD and GSH-PX in hippocampus were observed in WiFi signal exposure group.
Extended author information available on the last page of the article cells in the seminiferous tubules (Mortazavi et al. 2013;Le Quément et al. 2012) Saili et al. demonstrated that acute WiFi signal exposure (2.45 GHz) can affect the cardiovascular system (heart rhythm and blood pressure) in mature male rabbits. Review studies raised a degree of scientific uncertainty of the risk of radiofrequency (RF) transmissions to human health and suggested taking precautions, especially in children (Castaño-Vinyals et al. 2022;Kheifets et al. 2005). Among several possible biological targets, the effect of WiFi signal on the nervous system has received a special focus because of its immense cellular diversity, electrical nature, and organizational complexity (Shokri et al. 2015). The memory performance for a year was inversely related to the collective duration of wireless phone use and more significantly to RF-EMF dose (Schoeni et al. 2015). The implication of this is that RF-EMF exposure influences memory function (Zhijian et al. 2013). Other studies have reported some beneficiary effects of microwave irradiation (Bayat et al. 2023). They announced beneficial cognitive effects of RF radiation in Alzheimer's disease (Mortazavi et al. 2013;Banaceur et al. 2013), but the evidence for this remains indistinct.
Currently, the effects of 2.4 GHz WiFi signal exposure on the nervous system have mostly been studied in adult animals, and few in utero and postnatal exposure studies have been carried out. The embryonic period is considered a critical phase in the development and the nervous system and immature brain plasticity toward environmental factors recognized them as possible RF field targets. However, the neurological effects of WiFi exposure early in life, particularly during pregnancy, have not been extensively studied (Çelik et al 2016;Altunkaynak et al. 2016). There is limited data with regard to the long-term WiFi effects on the physiology of the brain. Whether WiFi exposure has positive, negative, or no effects on neurodevelopment and behaviors is not clear.
Given the progressive increase in the 2.45 GHz wireless networks, concerns should be raised concerning the effects of continuous and long-term whole-body exposure to these high WiFi network frequencies (Shokri et al. 2015). Therefore, we explored the long-term effects of prenatal and early life exposure to 2.54 GHz WiFi signals on neurodevelopment and behaviors as well as biochemical alterations of Wistar rats.

Animals
Adult Wistar rats (both sexes) were procured from Harbin Medical University (Harbin, China). Both female (weight 240-280 g) and male (weight 280-320 g) rats were used for the experiments. The feeding method and first day of pregnancy (E0) determination were conducted as described previously. Approval for animal procedures was provided by the Experimental Animal Ethics Committee, Daqing Campus of Harbin Medical University.

Experimental design
On E0, the gravid rats were separated into 2 groups: the control group (n = 12) and the WiFi-exposed group (WiFi group, n = 12). Separate housing was provided for every pregnant rat. WiFi group was exposed to turn on WiFi for 24 h/day for 9 weeks. The control group was exposed to turn-off WiFi for the same time. WiFi device was (802-16e 2005 WiMAX Indoor CPE antennae, model number: WIXFMM-130, China) with a frequency of 2450 MHz (2.45 GHz). The duration of radiation was 24 h/day in a 30-cm distance from the antenna to the cages. We test the average electric field intensity is 2.1 V/m, the average power density is 82.32 mV/ m 2 , average magnetic field intensity is 14.31 mA/m, and there are no differences between the inside and outside of the plastic cage.
The litters born from the control and WiFi group rats were included in the control and WiFi groups, correspondingly. On the PND21, the offspring were weaned then separated into various cages based on sex. We carried out the following experiments on the offspring.

Neuro-behavioral development
The pups were tested for behavioral development as described by Schneider and Przewłocki (2005). This test was performed on each group containing eight animals.
Swimming performance: Swimming test was used to evaluate motor development and integration of a coordinated series of reflex responses on PNDs 8, 10, 12, and 14.
Pain threshold: Pain threshold was determined by the hot plate method on PNDs 9, 11, 13, and 15.

Self-grooming test
Self-grooming test was conducted to assess the repetitive and stereotyped behavior of each rat. The tests were performed as previously described (Kim et al. 2017). Briefly, the rats were individually put into standardized cages under light (40 lx). The cages were empty to avoid digging in the bedding, which is considered a competing behavior. The rats were familiarized to the cages for 5 min, then timed for 10 min. An experienced investigator scored the total time spent in grooming during the period.

Open field test
In the present study, open field test was conducted on PND 30 to evaluate the rats' locomotor activity. The test was performed as previously described. Prior to the test, the rats were familiarized with the test box for 5 min. Subsequently, the cumulative distance traveled by each rat plus the movement duration (within a 10-min session) was recorded using the auto-tracking camera system (YH-OF, YiHong, China).

Morris water maze
This was carried out to determine the spatial learning and memory of rats, as detailed before (Wu et al. 2017). The rats were first trained for 4 consecutive days (i.e., from PND 36 to 39). After the training, an 8-cm-diameter platform was put at the center of the third quadrant of the water maze, then hidden 1.5 cm underneath the surface of the water. Each rat was given four trials (60 s each) daily to locate the platform. The time taken to locate the platform (escape latency) was regarded as index of performance. On day 5, the platform was eliminated, and the number of times that a rat passed through the circular area (diameter, 10 cm) that previously contained the platform within 60 s was recorded. And this was taken as the index of spatial memory.

Tissue preparation
Histological tissue preparation was conducted according to the methods described by Wu et al. (2018). Briefly, on PND 43, rat tissues from each group were perfused in ice-cold saline (0.9%) then in 4%-paraformaldehyde. Subsequently, the brains were removed, then post-fixed using 4% paraformaldehyde, then cryopreserved at 4 °C in a 30% w/v sucrose solution. Next, frozen coronal sections (50 μm thick) were generated on slides for staining and immunohistochemistry. All the sections were preserved at − 80 °C for further studies.

Nissl staining
Hippocampal neuron apoptosis was evaluated based on the results of Nissl staining in the brain section. For Nissl staining, the normal neurons were round, and the nuclei appeared as pallid blue (Guan et al. 2019). The sections were washed with distilled water and immersed for 15 min in crystal violet (0.1%) at room temperature (RT), then washed again with distilled water. Subsequently, the sections were dehydrated, then sealed with neutral balsam before taking photomicrographs using a light microscope.

NeuN immunohistochemistry
NeuN immunohistochemistry was carried out to assess neuron apoptosis as described by our previous report (Wu et al. 2017;Wang et al. 2017). Briefly, the sections were incubated with the primary antibody mouse anti-NeuN (1:50, Chemicon International, USA) overnight at 4 °C. After rinsing, the sections were incubated with suitable biotinylated antibodies (1:200) for 15 min at 37 °C then subjected to 3.3-diaminobenzidine (DAB) exposure for about 2 min. Sections were subsequently put into tap water to stop the DAB reaction. Following several washing and dehydration steps, the sections were put in coverslips then visualized using a light microscope and photographed. The control sections were incubated with 10% normal goat plasma as opposed to the primary antibody. All the subsequent sections were incubated as illustrated above and were observed under a light microscope. No positive immunoreaction was observed.

Measurement of superoxide dismutase (SOD), malondialdehyde (MDA)
The contents and enzymatic activity of SOD in the brain of rats from every group were assessed using various commercial assay kits as per the methods described by the manufacturer (Nanjing Jian Cheng Bioengineering Institute, Nanjing, China). The level of MDA was expressed in nmol/mg protein. The SOD activity was identified as U/mg protein.

Western blot assay
This assay was conducted as per the methods described by Wu et al. (2017). The experiment involved using the hippocampal tissues from different groups. An aliquot of hippocampal tissue was homogenized in ice-cold RIPA lysis buffer. Then, 30-μg protein samples were resolved using SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel) electrophoresis and transferred to nitrocellulose membranes. After blocking for 1 h using 5% skim milk, the membranes were incubated overnight at 4 °C together with the following antibodies: anti-BDNF (1:1000; Cell Signal, USA), anti-CREB (1:500; Cell Signal, USA), anti-Bcl-2 (1:1000; Cell Signal, USA), anti-Caspase3 (1:500; Cell Signal, USA), and mouse anti-GAPDH (1:1000, Zhongshan Jinqiao Biotechnology, China). The next day, the membranes were incubated for 1 h at RT together with the HRP-labeled goat anti-rabbit/anti-mouse secondary antibody (1:2000, Santa Cruz, USA). HRP signals were detected using a chemiluminescence imaging machine (Image Quant LAS 4000 minis; Applygen Technologies Inc., China). A scanner was used to digitized the images before they were analyzed by Quantity One software (Bio-Rad Laboratories, USA).

Statistical analysis
All data were shown as the mean ± SEM. The analyses were completed in SPSS version 18.0 (Beijing Stats Data Mining Co., Ltd.). Statistical analysis was conducted using a two-way analysis of variance (ANOVA) to compare two levels of prenatal status (WiFi versus control) and sex (males versus females), and the means were separated by Bonferroni's post-tests.

Effects of WIFI signals on bodyweight and neuro-behavioral development of offspring
The body weight of offspring was monitored for 6 weeks after birth. As showed in Fig. 1A, the mean body weight of male rats in the WiFi group (WiFi + M) significantly increased compared with male rats in the control group (Con + M) at the P28, P35, and P42 (p < 0.01). However, no considerable difference in the mean body weight was found during the entire exposure period between female rats in the WiFi group (WiFi + F) and female rats in the control group (Con + F). The result of the neuro-behavioral development test indicated no significant differences in the WiFi group relative to the control group in the following aspects: swimming performance (Fig. 1B), pain threshold (Fig. 1C), and negative geotaxis (Fig. 1D).

Effects of WiFi signals on the stereotyped and repetitive behaviors of offspring
Self-grooming was used to observe the stereotyped and repetitive behaviors of each group within 10 min. No remarkable changes in total times (Fig. 1E) and duration of self-grooming (Fig. 1F) were found between the WiFi group and the control group.
The results of open field tests showed that for the WiFi + M group, the distance spontaneously traveled was considerably longer than that of the Con + M group (p < 0.05), while the WiFi + F group did not differ significantly from the Con + F group (Fig. 1G). Furthermore, the results of the velocity of movement showed no significant changes between the groups (Fig. 1H).

Effects of WiFi signals on spatial learning and memory function of offspring
The spatial learning and reference memory of each group were assessed using the Morris water maze test.
In the place navigation test, the escape latency was remarkably shortened in the offspring of the WiFi + M group relative to other three group rats, especially during sessions 3 (p < 0.01) and 4 (p < 0.01) as showed in Fig. 2A and B.
Concerning the probe test, the frequency of passing (passing times) of rats in the WiFi + M group showed a marked elevation compared with other three groups (p < 0.01) ( Fig. 2C and D).
BDNF/CREB signaling pathway is considered to play significant functions in learning and memory processes. Thus, we evaluated the changes in BDNF protein levels and the ratio p-CREB/CREB in the hippocampal tissues of different groups of rats by Western blot analysis.
The BDNF expression in the rats' hippocampal tissues in each group showed that the expression of BDNF was remarkably elevated in WiFi + M group relative to the other three groups (p < 0.01) (Fig. 2E).
Results of p-CREB expression in the rats' hippocampal tissues showed that the ratio of p-CREB to total CERB was significantly increased in the WiFi + M group compared with other three groups (p < 0.01) (Fig. 2F).

Effects of WiFi signals on neurons in the hippocampus of offspring
We examined whether WiFi signals could affect neurons in the hippocampus of WiFi-exposed rats. We conducted a histological study through Nissl staining to examine neuronal changes in the CA1 region of the hippocampus. Regarding Nissl staining of the control group, a clear neuronal cell outline with a compact structure and abundant cytoplasm and a cell body was observed (Fig. 3A). Neuronal degeneration and loss were not detected in the WiFi group. No remarkable differences were detected with regard to the number of surviving cells in the CA1 region of the hippocampus between the WiFi group and the control group (Fig. 3B).
In neurological research, NeuN has been considered to be a reliable marker of mature neurons; the expression level of NeuN has been used to directly evaluate neuronal death or loss (Huttner et al. 2014;Soylemezoglu et al. 2003). Similarly, NeuN immunohistochemistry also revealed no differences in mature neurons in the pyramidal cells of the hippocampal CA1 region in the WiFi-treated rats (Fig. 3C). The number of positive cells (dark brown indicates NeuN-positive cells) in the hippocampal CA1 has no obvious change in the WiFi group relative to the control group (Fig. 3D).
Bax and caspase-3 are key proteins related to neuron damage. To examine the effect of WiFi on hippocampal neurons at the molecular level, the expressions of Bax and caspase-3 in the hippocampal tissues of different groups of rats were tested by Western blot analysis. The results showed no significant changes in the expression of caspase-3 and Bax proteins between the groups (Fig. 3E and F).

Effects of WiFi signals on brain oxidative stress response of offspring
To further study the effect of WiFi on the brain of offspring, MDA content SOD antioxidant enzyme activity in the rats' hippocampal tissues in each group was measured. MDA level is an important indicator of lipid peroxidation. No statistically significant alteration was found in MDA content in the hippocampal tissues of WiFi rats compared with the control group (Fig. 4B). Besides, we also tested the activity of the antioxidant enzyme. Figure 4A shows that compared with the Con group, SOD activity in the hippocampus of the WiFi group was not significantly changed.

Discussion
Internet access is currently considered a must have in daily routines and therefore has been installed in almost all communication gadgets (Zhi et al. 2018;Nazıroğlu et al. 2013).
Consequently, continuous exposure to WiFi has become a very common risk factor for poor health (Foster and Moulder 2015;At-Assa et al. 2013;Aït-Aïssa et al. 2012).
The present study investigated the effects of a 2.54 GHz WIFI signal exposure during prenatal and early life (24 h/day for 9 consecutive weeks) on rat neurodevelopment, behaviors, and cognition as well as biochemical index alterations.
Our study found that WiFi exposure did not affect offspring physical and functional development. These results agree with a study by Poulletier et al. Behaviorally, exposed offspring exhibited no alteration in motor and emotional behavior. Contrarily, some studies have revealed that exposure to WiFi radio frequencies during pregnancy could affect neurological functions of offspring (Othman et al. 2017b, a). However, this could have been due to the high radiation dosage tested in these studies. Herein, we revealed that in most tests, the effect of WiFi treatment was dependent on the sex of the offspring, and this was consistent with the findings of Zhang et al.; however, its mechanism of action remains unclear.
EMF exposure has been shown to have contradictory effects on the cognitive functions of animals including humans. Dubreuil et al. revealed that RF exposure can reduce performance in rodents, particularly in tasks that require spatial memory (Dubreuil et al. 2003  900 MHz frequency for 5 weeks (i.e., 2 h a day, 5 days per week, SAR 3 W/kg) can significantly improve memory and learning abilities of young rats (Kumlin et al. 2007). Herein, we observed that prenatal WiFi exposure can enhance cognitive ability. Several studies have shown that protein synthesis occurs in neuronal dendrites and might be the cellular basis of memory and learning. Currently, it is not known whether microwave radiation affects protein No difference expressions of apoptosis protein in the hippocampus between different groups. E, F The expression of apoptosis-related proteins of hippocampus of offsprings in each group: E representative Western blots for Bcl-2 and F representative Western blots for caspase-3. All data are expressed as mean ± SEM (n = 6) Fig. 4 Effects of WiFi signals on brain oxidative stress response of offspring in different groups. A SOD activity in hippocampus. B MDA contents in hippocampus. All data are expressed as mean ± SEM (n = 6) synthesis, particularly in the brain. In this study, we found that prenatal WiFi exposure can increase the expression of BDNF and phosphorylated CREB proteins. However, the detailed mechanism still needs further study. The brain has been shown to be more prone to oxidative injury during development in early years of life (Çelik et al 2016). Besides, oxidative stress can be activated via several mechanisms, such as electromagnetic radiation, thus resulting in molecular impairment. The oxidative stress response due to exposure to WiFi signals has been previously investigated in an animal model. It is noteworthy that previous studies that investigated the harmful effects of RF-EMR have reported inconsistent findings. Ozben and Kamali et al. implicated microwave radiation (Shokri et al. 2015;Foster and Moulder 2015;At-Assa et al. 2013;Aït-Aïssa et al. 2012) in apoptosis through their ability to trigger lipid peroxidation of cell membranes and as a result yield apoptosis signal (Kamali et al. 2018;Ozben 2007;Dasdag et al. 2008). However, other studies have indicated that EMR has no considerable impact on the antioxidant defense system because of unaltered oxidative stress markers such as MDA (At-Assa et al. 2013). In the present study, WiFi exposure did not induce brain oxidative stress response in offspring, suggesting that the possible damaging impacts of such radiations could be dependent on the exposure duration, dose, age, and body posture (Peter and Richard 2010). Contradictory results of the mentioned research could be because of differences in study methods, especially in the duration of exposure and dose of WiFi signal.
In conclusion, the findings of this study indicate that prenatal WiFi exposure does not affect the offspring's hippocampal neurons, oxidative equilibrium in brain, and neurodevelopment and emotional responses. Notably, some effects of WiFi exposure are sex dependent. Prenatal WiFi exposure increased the body weight, improved the spatial memory, and learning function and induced behavioral hyperactivity of male rats. However, there is a need to conduct further studies, especially on biochemical and neuromolecular mechanisms underlying such effects.