Prenatal valproate exposure alters midbrain and striatal neuronal morphology along with dopamine levels
Barbora Bugar Bodorova, Tomas Havranek, Denisa Mihalj, Kristina Kupkova, Ema Bogyova, Hisham El Falougy, Zuzana Bacova, Jan Bakos
TL;DR
Prenatal exposure to valproate changes brain neuron structure and dopamine levels in ways linked to autism-like symptoms.
Contribution
This study reveals region-specific neuronal and dopaminergic changes in midbrain and striatum due to prenatal valproate exposure.
Findings
VTA neurons from VPA-exposed rats had fewer and shorter neurites.
Midbrain dopamine levels were significantly higher in VPA-exposed rats.
Dopamine receptor type 4 gene expression increased in the VTA of male VPA-exposed rats.
Abstract
Alterations in dopaminergic pathways are implicated in the pathogenesis of neuropsychiatric disorders, including autism spectrum disorder (ASD). However, changes in morphology and neurite outgrowth within the dopaminergic system under autism-relevant conditions remain poorly understood. Prenatal valproate (VPA) administration is a well-established animal model for studying autism-related brain changes in offspring. Therefore, the aim of this study was to examine the effects of prenatal VPA exposure on (1) the morphology of primary neurons isolated from dopaminergic brain regions, and (2) the expression of molecular motors and neurite outgrowth-related proteins, along with (3) the levels of dopamine and dopamine receptors in the midbrain and striatum, analysed at postnatal day 30 (P30). Neurons from the ventral tegmental area (VTA) of rats prenatally exposed to VPA showed a significant…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6- —Comenius University in Bratislava
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAutism Spectrum Disorder Research · Pharmacological Effects and Toxicity Studies · Parkinson's Disease Mechanisms and Treatments
Introduction
Most dopaminergic neurons in the midbrain project to the striatum and cerebral cortex, playing a crucial role in regulating social behaviors, particularly those related to the perception of social rewards and motivation for social interaction (Bissonette and Roesch 2016; Manduca et al. 2021). Several studies suggest that disruption of dopaminergic pathways during brain development is associated with social dysfunction and altered ability to perceive socially relevant stimuli (Kosillo and Bateup 2021; Ijomone et al. 2025) which have been implicated in the etiology of neuropsychiatric disorders such as autism spectrum disorder (ASD, Pavăl and Miclutia 2021; Lu et al. 2024). Particularly, the mesocorticolimbic system including the ventral tegmental area (VTA), nucleus accumbens, ventral pallidum, olfactory tubercle, and prefrontal cortex, is important in understanding the etiology of ASD due to its key role in reward processing and social adaptation (Baumeister et al. 2023). Abnormalities in dopamine neurotransmission have been observed both in individuals with autism and in animal models exhibiting autistic-like behaviors (Ijomone et al. 2025; Barbier et al. 2025).
In our recent study, we observed reduced neurite outgrowth in primary dopaminergic neurons isolated from the midbrain and striatum of a transgenic animal model exhibiting autism-like characteristics (Bacova et al. 2025). Other studies have also shown alterations in neural circuits originating from the VTA of the midbrain, connecting its impaired function to social behavior deficits seen in autism (Bariselli et al. 2016). A well-established experimental model used to study autistic symptomatology involves the administration of valproate (VPA) to pregnant rodent females, followed by observation of the consequences in their offspring (Schneider and Przewołski 2005). Although recent studies confirm that prenatal administration of VPA induces autism-like symptoms in adult rats and mice and impacts the dopaminergic system (László et al. 2022; Cezar et al. 2025), the development of dopaminergic pathways and the possible changes that occur during early and juvenile stages are not yet fully clear. Nevertheless, previous studies have characterized dopaminergic innervation from the ventral midbrain to the prefrontal cortex and ventral striatum, as well as its developmental changes from embryonic stages through postnatal day 90, including in the context of ASD etiology (Bayer et al. 1995; Blaess et al. 2011; Kalsbeek et al. 1988; Kosillo and Bateup 2021).
Early adolescence in rats, corresponding to postnatal day 30 (P30), is a period of significant dopamine circuit remodeling (Reynolds and Flores 2021; Gilman et al. 2019). Moreover, some studies show that P30 is characterized by increased dopaminergic terminal density and enlargement of midbrain dopaminergic neurons (Park et al. 2000). However, it is still not clear whether prenatal VPA exposure affects dopaminergic brain regions during adolescence or changes the growth and shape of dopaminergic neurons early in development, which are both important for making synaptic connections with other brain areas.
Changes in neuronal morphology, including neurite (representing dendritic precursors) outgrowth and dendrite arborisation, are regulated by the expression levels of proteins responsible for anterograde transport of cargo such as molecular motors kinesins and dyneins (Lestanova et al. 2016; Guedes-Dias and Holzbaur 2019). It is known that alterations in expression of kinesin I family protein in early developmental stages result in severe defects in dendritic spine morphogenesis and synaptic plasticity (Zhao et al. 2020). Although limited information exists on changes in the expression of molecular motors during development in dopaminergic pathways, some evidence from other brain regions suggests they may play a role in the pathogenesis of autism (Gromova et al. 2018).
Therefore, the aim of this study was to examine the effects of prenatal VPA exposure on (1) the morphology of primary neurons isolated from dopaminergic brain regions, and (2) the expression of molecular motors and neurite outgrowth-related proteins, along with (3) the levels of dopamine and dopamine receptors in the midbrain and striatum, analysed at P30.
Materials and methods
Animals
A VPA autism rat model was established based on previous studies (László et al. 2022). Gestating Wistar rats were purchased from the Charles River Laboratories (Germany), housed in an air-conditioned room (22 ± 2 °C, 55 ± 10% humidity), and maintained on a 12 h light/dark cycle with food and water available ad libitum. On embryonic day 12.5–13, pregnant dams received a single intraperitoneal injection of 450 mg/kg VPA (sodium valproate, P4543, Merck, Germany) or saline solution (n = 6–8/group). Pups of mixed sexes from different dams were sacrificed by decapitation on postnatal day 0 (P0) to isolate primary neuronal cells (Fig. 1). On postnatal day 30 (P30) a separate cohort of male and female rats prenatally exposed to saline (control- CTRL) or VPA (n = 15/group) were sacrificed by decapitation, and their brain tissue was collected (Fig. 1). Dopaminergic brain regions were dissected and further analysed as described below. All experimental procedures used in this study were in accordance with the international standards (European Community Council Directive, 86/609/EEC, 1986, 2010 ). The State Veterinary and Food Administration of the Slovak Republic also approved all experimental procedures (5467-3/2023-220).
Fig. 1. Timeline of experimental procedures. Offspring from mothers prenatally treated with valproate (VPA) or saline were used at postnatal day 0 (P0) to isolate primary neuronal cells from the ventral midbrain (tegmentum) and ventral striatum. At postnatal day 30 (P30), a separate cohort of male and female offspring underwent dissection of the ventral tegmental area (VTA) and ventral striatum for gene expression analysis (qPCR), dopamine quantification (ELISA), and tyrosine hydroxylase measurement (Western blot). DIV days in vitro, ICC immunocytochemistry
Isolation of primary neurons
Neuronal cell cultures were prepared from CTRL and VPA prenatally treated animals as previously described (Reichova et al. 2021). Brains were placed in ice-cold sterile Hank’s balanced salt solution, further supplemented with 100 U/ml penicillin and 100 U/ml streptomycin, 0.3 M Hepes (Sigma-Aldrich, Germany). Specific brain regions (ventral striatum, midbrain) were individually collected (adapted according to Khazipov atlas (Khazipov et al. 2015) for P0 and Paxinos and Watson 1997 for P30), specifically, P0 the ventral striatum (ML ± 1.5 mm, AP 0.60 to 1.90 mm, DV −6.5 to −9 mm), ventral midbrain (tegmentum, ML ± 0.5 mm, AP −2.8 to −3.8 mm, DV −3 to −3.5 mm), P30 the ventral striatum (ML ± 2 mm, AP 0.80 to 2.60 mm, DV −6 to −9 mm), VTA (ML ± 1 mm, AP −5.5 to −5.6 mm, DV −8 to −9 mm). For the isolation of primary neurons (P0) from the midbrain, we collected tissue specifically from the ventral medial region according to above coordinates to avoid the substantia nigra, which is located more laterally. We refer to these cells as tegmental neurons rather than directly as VTA neurons, since precise neuroanatomical boundaries of the VTA cannot be reliably determined in neonatal rats. Afterwards, tissues were enzymatically dissociated for 20 min at 37 °C (HBSS, 0.1% Trypsin, 0.1 mg/ml DNAse I). Cells from pooled dissociated tissues (n = 6–8) were plated into 24-well plates with individual 12-mm round poly-d-lysin pre-coated coverslips (10 µg/ml; Sigma-Aldrich, Germany) and initially incubated for 3 h in RPMI medium containing 10% foetal bovine serum (both Sigma-Aldrich, Germany) under standard condition (37 °C and 5% CO_2_). After pre-incubation, RPMI medium was removed and exchanged for neuron-selective-growth medium (Neurobasal A; 100 U/ml penicillin; 100 U/ml streptomycin; 2 mM l-glutamine (all Gibco, USA) enriched with 2% B27 supplement (Invitrogen, USA) for 5 days. Subsequently, on the 5th and 7th day in vitro (DIV5, DIV7), 50% of neuron-selective-growth medium volume was exchanged.
Immunocytochemistry
On the DIV9, stabilized culture of the primary neurons was inspected and fixed with 4% paraformaldehyde, pH 7.4 for 20 min at room temperature (RT). DIV9 was selected due to the intensive neurite outgrowth and synaptogenesis that occurs at this stage (Grabrucker et al. 2009; Fletcher et al. 1994). Coverslips were gently washed two times with ice-cold PBS and blocked in normal goat serum (3% NGS in PBS; PCN5000, Gibco, USA) with the presence of 0.1% Triton X-100 for 30 min at RT. Proteins were detected with specific antibodies: anti-microtubule-associated protein 2 (MAP2; mouse; M4403; Sigma-Aldrich, Germany; 1:2000), anti-Tyrosine Hydroxylase (TH; rabbit; ab112; Abcam; UK; 1:500), anti-mouse (Alexa Fluor 488; A11001; Invitrogen; USA; 1:500), and anti-rabbit (Alexa Fluor 555; A21428; Invitrogen; USA; 1:500). Nuclei were stained with 300 nM 4′,6-diamidino-2-phenylindole (DAPI; D1306, Invitrogen, USA) diluted in PBS for 1 min at RT. The coverslips were mounted using Fluoromount-G (00–4958-02, Invitrogen, USA).
Imaging and analysis
Immunofluorescence images were captured using a fluorescence Olympus BX63 (2904 × 1815 pixels). Three coverslips per group were used, with at least 15 randomly selected images per coverslip and 50–90 neurons per group. Image analysis was performed using the ImageJ software with the SNT plugin (Arshadi et al. 2021). The arborization of dendritic trees was evaluated by adapting the technique from our previous work in other autism-like animal model and assessed using Sholl analysis, while neurite outgrowth was measured by the length of the longest neurite (Bacova et al. 2025).The Sholl intersection profiles were obtained by counting the number of neurite branches in the interval from the soma to 150 μm, with the distance increasing by 1 μm. The length of the longest neurite was quantified from the edge of the nucleus to the apical end.
Quantitative real-time PCR
Total RNA was extracted from the ventral striatum and VTA in rats that were prenatally exposed to either saline (control, CTRL) or valproic acid (VPA), utilizing the phenol-chloroform extraction method with TRIzol Reagent (15596026, Invitrogen, USA). The concentration and purity of the isolated RNA were determined by UV absorbance using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA). cDNA samples were generated using a High-Capacity cDNA Reverse Transcription Kit (4368814, Applied Biosystems, USA) and stored at −20 °C until subsequent analysis. Reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) assays were performed using specific primers designed with NCBI Primer-BLAST (sequences listed in Table 1) and Power SYBR^®^ Green PCR Master Mix (4367659, Applied Biosystems, USA) on a QuantStudio5 Real-Time PCR System (Applied Biosystems, USA). The relative expression levels of the selected genes were calculated using the 2^−ΔΔCt^ method (Livak and Schmittgen 2001), in which the threshold cycle (Ct) values were first normalized to the reference gene Gapdh in the same sample and then to the average Ct value of the control samples.
Table 1. List of primer sequences used in this studyNamePrimersGene bankReferences Dr1 Fw: CCAAGAATTGCCAGACCACRv: GTCTCCTCAGAGCCACAGAANM_012546.3 Noori-Daloii et al. 2012 Dr2 Fw: TCAATGGGTCAGAAGGGAAGRv: CAGAAGATCAGCCACAGCAANM_012547.3 Noori-Daloii et al. 2012 Dr3 Fw: AGGTGACAGGTGGAGTCTGGRv: CCGTGCTGATAGTGAACTGGNM_017140.2 Noori-Daloii et al. 2012 Dr4 Fw: CCACCCTCGGAGTAGACAAARv: ATGGTGTTGGCAGGGAACTNM_012944.2 Noori-Daloii et al. 2012 Dr5 Fw: CTAGTGTGTGCTGCCATCGTRv: ACCCAGATGTCGCAGAATGNM_012768.2 Noori-Daloii et al. 2012 Dynein Fw: TTCTGGCGTAGTCCTATTRv: ACACCACATCTCAAGTCTNM_019234.2 Baptista et al. 2013 Gapdh Fw: TGCACCACCAACTGCTTAGRv: GGATGCAGGGATGATGTTCNM_017008.4 Horii et al. 2002 Kif1a Fw: CATTAGTTAGTGGCGTTGARv: TACCTGGAGGCATTAGAAANM_001408926.1 Baptista et al. 2013 Kif5b Fw: GTGATGATTGCGTCCAAGRv: CTTCTTTGCACAATCGTTGNM_057202.2 Baptista et al. 2013 Wnt5a Fw: GCCACTTGTATCAGGACCACARv: GGCATTTACCACTCCAGCAGNM_022631.4 Sklepkiewicz et al. 2011Fw forward, Rv reverse; DR dopamine receptor, Gapdh glyceraldehyde 3-phosphate dehydrogenase, Kif kinesin-like protein, Wnt wingless/integrated
Western blot analysis
Tissues from the ventral striatum and VTA of the midbrain of CTRL or prenatally VPA-exposed animals were subsequently processed by ultrasonic homogenization in ice-cold phosphate-buffered saline (PBS) containing Protease Inhibitor Cocktail (Sigma-Aldrich, Germany). Homogenates were then centrifuged for 5 min at 5000 × g and the protein concentrations were quantified using a BCA kit (Thermo Fisher Scientific, Slovakia) with bovine serum albumin (BSA) as a standard. Total proteins were diluted with ultra-pure water and a sample loading buffer (1:2 solution; 10% w/v SDS, 0.02% bromophenol blue, and 25% glycerol in 0.5 M Tris- HCl, pH 6.8) and separated 12% SDS-PAGE running gels (8 µg protein/lane) with 5% stacking gels at a constant current of 150 mA as a running parameter. The proteins were transferred to a low fluorescent polyvinylidene difluoride membrane Immobilon-FL (0.45 μm pore size; Merck-Millipore, Czech Republic) by a Mini Trans-Blot^®^ cell system (Bio-Rad, Slovakia) for 2 h in wet conditions at 300 mA. After the blocking with 4% w/v BSA (Sigma-Aldrich, Germany) for 1 h at RT, membranes were incubated overnight at + 4 °C with primary antibodies anti-Tyrosine Hydroxylase (TH; rabbit; ab112; Abcam; UK; 1:1000) and anti- GAPDH (mouse; G8795; Sigma-Aldrich, Germany, 1:1000). The GAPDH protein was used as a reference. Blots were washed three times for 10 min in Tris Buffered Saline solution with Tween^®^ and then treated with corresponding fluorescent secondary antibodies: Anti-rabbit CF™ 770 (goat, SAB4600215; Sigma-Aldrich, Germany, 1:5000) and Anti-mouse CF™ 680R (donkey, SAB4600207; Sigma-Aldrich, Germany, 1:5000) for 1 h at RT. The Odyssey Infrared Imaging System (LI-COR, USA) was used to detect and quantify fluorescent signals. Odyssey 2.0 analytical software (LI-COR, USA) was used to manually signpost bands of interest and quantify the density of each blot band. Results are presented as a percentage of the control group normalized to GAPDH, reference protein.
ELISA
The same protein extracts (50 µl) used for Western blot analysis were also employed to quantify dopamine concentrations using an ELISA kit (EU0392, FineTest, USA). This analysis is based on a competitive ELISA detection method. Dopamine levels in the samples were quantified by comparing their optical density values to a standard curve, measured using a spectrophotometric plate reader (Biotek, USA) at 450 nm. The dopamine concentration was calculated per microgram of total protein in each sample, obtained in above-described Western blot analysis.
Statistical analysis
Statistical analyses were performed using GraphPad Prism (version 10.0; GraphPad Software Inc., USA). Outliers were identified using the interquartile range (IQR) method. First, the sample data were tested for normal distribution using the Shapiro–Wilk test. If the data were normally distributed, the two-group means were analyzed using an unpaired two-tailed Student’s t-test. If the data distribution was not normal, the non-parametric Mann-Whitney U test was used. To evaluate differences in neurite arborization, two-way ANOVA was used. Figures were generated using GraphPad Prism version 10.0. The bar graphs and scatter plots represent mean ± standard error of the mean (SEM). Statistical significance was set at p < 0.05.
Results
Primary neuronal cell cultures were derived from the ventral tegmentum of the midbrain for the purpose of examining neuron morphology. Neurons isolated from prenatally VPA exposed rats had a significantly (two-way ANOVA, factor treatment F_(1,14700)_ = 474.9, p ≤ 0.001, factor distance F_(149,14700)_ = 114.9, p ≤ 0.001, interaction of factors F_(149,14700)_ = 2.212, p ≤ 0.001) reduced number of neurites compared to CTRL group that were also shorter in length (Fig. 2A). Post hoc Sidak’s test revealed a significantly reduced number of arborizations at distances of 2–6 μm (p ≤ 0.001) and 22–38 μm (p ≤ 0.001) from the nucleus. Furthermore, the length of the longest neurite was significantly shorter in neurons isolated from rats prenatally exposed to VPA compared to control samples (Mann-Whitney U = 739, p ≤ 0.001, (Fig. 2B).
Fig. 2. Dendritic arborization in primary neurons isolated from the ventral tegmentum of the midbrain in neonatal control (CTRL) and prenatally valproate (VPA)-exposed rats. Neurons were incubated in vitro for 9 days and then stained. The arborization of the dendritic tree was assessed using Sholl analysis (A). The neurite outgrowth was measured by the length of the longest neurite (B). Results are presented as mean ± SEM (n = 50 neurons per group). Statistical differences between groups were determined by two-way ANOVA for Sholl analysis (main effects in the graph) or Mann–Whitney test (***p < 0.001) for the longest neurite
To assess the morphology of dopaminergic neurons, TH-positive neurons from the midbrain tegmentum were stained (Fig. 3). Significant changes in neuronal morphology were observed (two-way ANOVA, factor treatment F_(1, 26878)_ = 6.769, p ≤ 0.01, factor distance F_(150, 26878)_ = 149.1, p ≤ 0.001, factor interaction F_(150, 26878)_ = 1.002, n.s.), post hoc analysis identified significant differences of dendritic arborisation at distances up to 5 μm. However, no significant differences were found in the length of the longest neurites between the groups (Fig. 3B).
Fig. 3. Dendritic arborization in tyrosine hydroxylase (TH) positive primary neurons isolated from the ventral midbrain (tegmentum) in neonatal control (CTRL) and prenatally valproate (VPA)-exposed rats. Neurons were incubated in vitro for 9 days and then stained. The arborization of the dendritic tree was assessed using Sholl analysis (A). The neurite outgrowth was measured by the length of the longest neurite (B). Results are presented as mean ± SEM (n = 90 neurons per group). Representative images of neurons isolated from CTRL (C) and prenatally VPA-treated (D) rats, immunostained for microtubule-associated protein 2 (MAP2), TH, and DAPI for nuclear visualization. Statistical differences between groups were determined by two-way ANOVA for Sholl analysis (main effects in the graph) or Mann–Whitney test for the longest neurite
Similar changes, although to a lesser extent, were also observed in primary neurons isolated from the striatum (Fig. 4). Specifically, we observed a significant (two-way ANOVA, Treatment F_(1,14700)_ = 579.2, p ≤ 0.001, Distance F_(149,14700)_ = 83.10, p ≤ 0.001, Interaction F_(149,14700)_ = 0.6593, n.s.) reduction in number of neurites at some distances from the nucleus. Post hoc Sidak’s test revealed a significantly reduced number of arborizations at distances of 38 and 41 μm (p ≤ 0.05). Correspondingly, the length of the longest neurite was significantly shorter in neurons isolated from rats prenatally exposed to VPA compared to control samples (Mann-Whitney; U = 639, P < 0,001; Fig. 4B). Next, TH-positive neurons isolated from the striatum were also stained to assess the morphology of specific dopaminergic neurons (Fig. 5). Significant changes in neuronal morphology were observed (two-way ANOVA, factor treatment F_(1, 17250)_ = 133.2, p ≤ 0.001, factor distance F_(149, 17250)_ = 133.4, p ≤ 0.001, factor interaction F_(149,17250)_ = 0.5539, n.s.), however, post hoc analysis revealed no statistically significant differences at specific distances. No significant changes were observed when comparing the length of the longest neurites between groups (Fig. 5B).
Fig. 4. Dendritic arborization in primary neurons isolated from the striatum in neonatal control (CTRL) and prenatally valproate (VPA)-exposed rats. Neurons were incubated in vitro for 9 days and then stained. The arborization of the dendritic tree was assessed using Sholl analysis (A). The neurite outgrowth was measured by the length of the longest neurite (B). Results are presented as mean ± SEM (n = 50 neurons per group). Statistical differences between groups were determined by two-way ANOVA for Sholl analysis (main effects in the graph) or Mann–Whitney test (***p < 0.001) for the longest neurite
Fig. 5. Dendritic arborization in tyrosine hydroxylase (TH) positive primary neurons isolated from the striatum in neonatal control (CTRL) and prenatally valproate (VPA)-exposed rats. Neurons were incubated in vitro for 9 days and then stained. The arborization of the dendritic tree was assessed using Sholl analysis (A). The neurite outgrowth was measured by the length of the longest neurite (B). Results are presented as mean ± SEM (n = 58 resp. 59 neurons per group). Representative images of neurons isolated from CTRL (C) and prenatally VPA-treated (D) rats, immunostained for microtubule-associated protein 2 (MAP2), TH, and DAPI for nuclear visualization. Statistical differences between groups were determined by two-way ANOVA for Sholl analysis (main effects in the graph) or Mann–Whitney test for the longest neurite
Since changes in neurite outgrowth and arborization in prenatally VPA-exposed rats may be linked to dopamine levels in dopaminergic brain regions, as well as to molecular motor protein and dopamine receptor levels during later development, we evaluated dopamine concentrations and gene expression of selected parameters in the striatum and midbrain of 30-day-old male and female rats. We observed a significant increase in total dopamine levels in the midbrain (Fig. 6B) of rats prenatally exposed to VPA, regardless of sex (two-way ANOVA, factor treatment F_(1, 30)_ = 7.192, p ≤ 0.05, factor sex F_(1, 30)_ = 0.424, n.s., interaction of factors F_(1, 30)_ = 0.781, n.s.). Although there was a trend toward increased dopamine levels in the striatum (Fig. 6A) of prenatally VPA-exposed rats, this increase did not reach statistical significance. Similarly, markedly increased TH protein levels (Representative Western blot bands are shown in the supplementary material) were observed in males from prenatally VPA-exposed rats compared to controls in both the striatum (145.89 ± 45.67% of CTRL, n = 4, as 100% ± 27.24, n = 5) and midbrain (181.38 ± 83.97% of CTRL, n = 3, as 100% ± 38.08, n = 4), this difference was not statistically significant and was absent in females (data not shown).
Fig. 6. Dopamine levels in the striatum (A) and midbrain (B) of 30-day-old control (CTRL) rats and those prenatally exposed to valproate (VPA). Dopamine concentrations were measured by ELISA in protein extracts from each brain region, with levels normalized to total protein content. Statistical differences between groups were analyzed using two-way ANOVA, with main effects displayed in the graph. Results are presented as mean ± SEM (n = 6–10 per group)
Analysis of gene expression for dopamine receptors, molecular motors and neurite outgrowth-related proteins revealed a significant increase (Student’s t-test, t = 2.48; df = 12; p ≤ 0.05) in the D4 receptor subtype in the ventral tegmental area of the midbrain in prenatally VPA-exposed male rats (Table 2) compared to CTRLs at P30. No other gene expression changes in males or females reached statistical significance (Table 2).Table 2. Gene expression of dopamine receptors, molecular motors, and neurite outgrowth-related proteins in dopaminergic brain regions of control (CTRL) and prenatally valproate (VPA)-exposed male (A) and female (B) rats at postnatal day 30GeneStriatumTegmentumCTRLVPACTRLVPAA. Male* Dr11.26 ± 0.342.80 ± 0.791.18 ± 0.281.88 ± 0.28 Dr22.42 ± 0.961.51 ± 0.431.68 ± 0.581.49 ± 0.21 Dr31.45 ± 0.501.80 ± 0.561.12 ± 0.201.16 ± 0.17 Dr41.09 ± 0.251.21 ± 0.49**1.18*** ± 0.306.98**** ± 1.99 Dr51.05 ± 0.121.07 ± 0.061.14 ± 0.221.18 ± 0.19 kif1a1.11 ± 0.191.71 ± 0.431.14 ± 0.201.05 ± 0.16 kif1B1.14 ± 0.200.89 ± 0.061.21 ± 0.241.46 ± 0.21 wnt5a1.04 ± 0.111.00 ± 0.101.07 ± 0.141.21 ± 0.13 dynein1.05 ± 0.121.33 ± 0.221.13 ± 0.192.51 ± 0.81B. Female Dr11.54 ± 0.581.37 ± 0.361.28 ± 0.397.72 ± 3.44 Dr21.25 ± 0.311.50 ± 0.474.35 ± 2.221.67 ± 0.74 Dr31.02 ± 0.090.90 ± 0.081.25 ± 0.370.88 ± 0.10 Dr41.25 ± 0.382.24 ± 0.481.25 ± 0.330.64 ± 0.64 Dr51.13 ± 0.231.05 ± 0.111.07 ± 0.140.83 ± 0.09 kif1a1.15 ± 0.220.74 ± 0.111.15 ± 0.281.03 ± 0.14 kif5b1.11 ± 0.200.80 ± 0.101.52 ± 0.710.60 ± 0.08 wnt5a1.11 ± 0.200.79 ± 0.101.09 ± 0.171.03 ± 0.21 dynein*1.21 ± 0.281.84 ± 0.491.29 ± 0.342.56 ± 1.40Results represent relative changes compared to CTRL group calculated by the 2^−∆∆CT^ method. Glyceraldehyde 3-phosphate dehydrogenase was selected as the reference gene. Results are expressed as mean ± SEM (n = 7–8 per group). Statistical significance was determined by Student’s t-test (*p < 0.01 ). Dr1-Dr5, dopamine receptors (1–5); Kif1A-5B, Kinesin Family Member 1A-5B; Wnt5A, Wingless-Type Family Member 5A
Discussion
In this study, we observed altered neuronal arborization in primary neurons isolated from the midbrain and striatum following prenatal VPA exposure, a known animal model exhibiting autistic-like behaviors. These morphological changes were accompanied by increased dopamine levels and elevated Dr4 gene expression, specifically in the VTA of the midbrain at P30. Although an overall difference in dendritic arborization was observed in TH-positive primary neurons from the striatum of prenatally VPA-exposed rats, no significant changes were detected at specific distances from the nucleus or in the length of the longest neurites, nor were there any alterations in the gene expression of selected molecular motors or neurite outgrowth-related proteins in the striatum or VTA of the midbrain.
Our finding that neurons isolated from the midbrain and striatum of prenatally VPA-exposed rats have a reduced dendritic arborisation at short distance from the nucleus is novel. It suggests potential connectivity alterations in dopaminergic brain regions during early stages of development in this model, as well as in other models exhibiting autistic-like features (Cezar et al. 2025; Finszter et al. 2023). The overall neurite length of TH-positive tegmental neurons did not change following prenatal VPA treatment, the arborization, particularly of their short dendrites was reduced. This result may imply impaired growth of dopaminergic neurons during development in autistic conditions.
Although we did not directly examine dendritic spines in striatal neurons, other studies reported increased dendritic spine density in the nucleus accumbens and ventral hippocampus following VPA administration, accompanied by retracted neuronal arborization in the hippocampus (Bringas et al. 2013). These findings correspond to our observations in neurons from dopaminergic brain areas. On the other hand, changes in dendritic arborization of TH-positive neurons isolated from the striatum of prenatally VPA-exposed animals were less pronounced in our results, suggesting that other neuronal populations within the striatum might be more affected. Striatal projections to the midbrain, along with reciprocal connections to and from the cortex, are well established, involving both gamma-aminobutyric acid (GABA)-ergic and glutamatergic neurons, whose morphology and function was demonstrated to be altered in autism-like conditions in the striatum (Peixoto et al. 2019; Ibáñez-Sandoval et al. 2024). Nevertheless, it is a fact that striatal TH-positive neurons, although sparsely distributed within the striatum (Garma et al. 2024; Keber et al. 2015), may have their own significance in relation to motor functions. However, there is ongoing debate regarding which neurotransmitter they secrete (Xenias et al. 2015). Overall, neuronal morphology is highly variable, and it is possible that only certain TH-positive neurons or neurons sharing neurotransmitters are affected by VPA exposure or associated with the autistic phenotype. Therefore, functional studies correlating neuronal changes with behavioral alterations are needed for further understanding.
Our hypothesis that altered neurite outgrowth in vitro would be reflected by reduced expression of molecular motors and neurite outgrowth-related proteins in striatal or VTA tissue was not confirmed. It is possible that subtle differences in the expression of proteins associated with neurite growth were undetected, or that these changes had already normalized by P30, emphasizing the necessity for investigations during earlier developmental time points. This aspect also represents a limitation of our study, as other developmental stages were not examined.
Another important finding is that morphological changes in primary neurons isolated from dopaminergic brain regions were accompanied by increased dopamine levels in the VTA of the midbrain in animals prenatally exposed to VPA in both sexes. In this context, it is interesting that increased monoamine levels in the striatum and ventral midbrain were observed in an animal model of Angelman syndrome, which also exhibits autistic-like symptoms (Farook et al. 2012). Surprisingly, in our study, no significant changes in TH protein levels, which are crucial for synthesizing dopamine, were detected in the midbrain. A slight increase in TH levels was observed in the striatum of males prenatally exposed to VPA compared to controls, while no differences were found in females. In contrast, a study with a similar design found decreased dopamine levels in the dorsal striatum in females prenatally exposed to VPA (Maisterrena et al. 2022) at P45. These authors also reported no alterations in the number of mesencephalic dopamine neurons or in TH protein levels in the ventral striatum. Other study (Ádám et al. 2020) reported a decrease in VTA neuron numbers and reduced dopamine levels in the ventral striatum of prenatally VPA-exposed mice, with no such dopamine reduction in the non-limbic caudate-putamen. In contrast, the number of TH-positive cells in the substantia nigra of the midbrain increased at P7. Conversely, other authors report that male mice prenatally exposed to VPA have fewer TH-positive cells in the substantia nigra compared to controls (Zarate-Lopez et al. 2024) in P31. These findings, together with our data, indicate that dopaminergic dysfunction occurs in ASD models, with its manifestations depending on the developmental stage and the specific brain regions involved.
Relatively minor changes were found in dopamine receptor expression. Specifically, increased gene expression of the Dr4 was observed specifically in the midbrain VTA of male rats prenatally exposed to VPA at P30. Similar upregulation has been reported in peripheral lymphocytes of autistic patients (Emanuele et al. 2010). Since the DR4 inhibits adenylate cyclase and lowers cAMP when activated by dopamine, its increased or altered expression in the midbrain’s tegmental area may serve as a compensatory response to reduced neurite outgrowth from other brain regions. It is important to note that DR4 expression normally decreases drastically postnatally, so its elevation in autistic conditions may reflect abnormal development (Nair and Mishra 1995; Ijomone et al. 2025). This interpretation should be considered cautiously, as little is currently known about DR4 changes during development in autism-like models or in ASD patients.
Overall, our findings extend current knowledge on the effects of prenatal VPA exposure, particularly regarding neuronal morphology and neurite development in dopaminergic brain regions during early developmental stages. A limitation of the work is the absence of functional electrophysiological or behavioral correlates. Nevertheless, these results demonstrate that prenatal VPA exposure induces region-specific neuronal and dopaminergic changes in the midbrain and striatum, enhancing our understanding of the neurodevelopmental mechanisms potentially underlying ASD.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1