2.1 Key resource table
The details are in Table 1.
2.2 Animals and OPC culture
A total of one hundred female mice, subdivided into two distinct strains - fifty SPF C57BL/6J mice and fifty SPF C57B6.Cg-Dock7 +/- Lepr /J mice, both aged between 8–10 weeks and with respective body weights of 20 ± 2 g and 21 ± 2 g, were procured for this study. The former group was sourced from the Animal Experiment Center of Xinqiao Hospital, Army Medical University, while the latter was self-bred, originating from breeder specimens acquired from Jackson Laboratories, USA. All mice were housed under controlled conditions featuring a temperature of 22 ± 2 ℃, a relative humidity of 55 ± 5%, and a 12-hour light-dark cycle, with ad libitum access to water. The entire study was sanctioned by the Army Medical University's Animal Ethics Committee, adhering to guidelines aimed at minimizing both the number of animals utilized and their respective levels of distress.
In vitro assays employed immature oligodendrocytes OLN-93, a readily passagable OPC cell line, cultivated according to the protocol described by Strelau et al.[33]. The cells were maintained in a 5% CO2 incubator at 37 ℃, grown in DMEM fortified with 10% fetal bovine serum and 1% penicillin-streptomycin. These cells were subcultured at a 1:2 ratio every three days, with cells from the 6th to 12th generations being selected for experimental procedures.
2.3 Materials and model induction
The sodium valproate, a soluble short-chain fatty acid, was dissolved in saline to create a 100 mg/ml solution and was stored at -20 ℃, protected from light, in preparation for animal experiments. Clemastine, known for its promyelinating properties, was prepared at a concentration of 35 mg/ml using DMSO as the solvent. For the cellular studies, VPA was initially dissolved in DMSO to obtain a 2.5 M solution. This stock solution was later diluted with DMEM, ensuring the final DMSO concentration did not surpass 1% of the total mixture. Palmitic acid solution (20 mM) was prepared according Xin et al[34]. and utilized for OLN-93. Glucose and gamma-aminobutyric acid (GABA) were dissolved directly in DMEM to obtain a 5 M glucose stock solution and a 1 M GABA stock solution. (+)-Bicuculline (GABAA receptor antagonist), CGP52432 (GABAB receptor antagonist) and U0126 (ERK inhibitor) were all solubilized in DMSO to 10 mM and stored them at -20°C away from light.
In the experiment, fifty female C57BL6 mice in estrus were identified, numbered, and randomly assigned into either the control group (n = 25) or VPA group (n = 25). An additional fifty female C57B6.Cg-Dock7 +/- Lepr /J mice were likewise identified, numbered, and bifurcated randomly into GDM group (n = 25) and GDM + VPA group (n = 25). Mating was facilitated by housing male mice of corresponding genotypes with the females at a ratio of 1:2 for a period of 72 hours. Fertilization success was confirmed through the identification of a vaginal plug, marking the initiation of the embryonic timeline at day 0.5 (E + 0.5 d). Subsequently, on day 12.5 of embryonic development (E + 12.5 d), mice in the VPA and GDM + VPA groups were administered a subcutaneous injection of 0.25 ml of a saline-diluted VPA solution at a dose of 600 mg/kg[35]. Meanwhile, the control group and GDM group were given a 0.25 ml subcutaneous injection of saline. To induce gestational diabetes, the GDM and GDM + VPA groups were subjected to a continuous high-fat diet regimen supplying 45 kcal/day, starting seven days prior to the embryonic timeline (E-7 d) and maintained throughout the pregnancy. Following birth, the offspring were categorized into four groups mirroring their respective maternal groupings: control, VPA, GDM, and GDM + VPA.
To investigate the potential dysmyelination in offspring induced by the promyelinating agent clemastine, pups from different litters were segregated at postnatal day (PND) 28 into four groups: the placebo group (n = 8), the clemastine group (n = 8), the VPA + placebo group (n = 8), and the VPA + clemastine group (n = 8), derived from both control and VPA groups. Saline served as the placebo, while the clemastine and VPA + clemastine groups received an intraperitoneal injection of 10 mg/kg saline-diluted clemastine continuously for 21 days, commencing from the peak of myelination in the corpus callosum[20, 36].
In the in vitro experiments, GABA, VPA, (+)-Bicuculline, and CGP52432 were all diluted in cell growth medium prior to used. The effects of various concentrations of these drugs on OLN-93 cells were assessed using the Cell Counting Kit-8 (CCK-8) cell activity assay to evaluate cell activity. Following a 12-hour period to allow for complete adhesion, the OLN-93 cells were exposed to 100 mM GABA for 24 hours to activate cellular GABA receptors. Subsequently, 100 µM (+)-Bicuculline and 25 µM CGP52432 were administered as positive controls to demonstrate GABA receptor antagonism, with the results compared to those obtained following a 24-hour treatment with 0.5 mM VPA.
In order to mimic the conditions of GDM-induced injury, an in vitro high-fat, high-glucose (HFHG) culture environment was established using 200 µM palmitic acid and 50 mM glucose, according to the protocol outlined by Liu et al[37]. Under these conditions, OLN-93 cells were subjected to treatments with GABA, VPA, and the established positive controls once more. Furthermore, to analyze the repercussions of inhibiting ERK signaling in oligodendrocyte precursor (OPC) cells, the fully adhesive OLN-93 cells were initially treated with 10 µM U0126 for 30 minutes. The earlier established GABA and VPA treatment protocols were then replicated under both normal and HFHG conditions to observe any consequent alterations.
2.4 Maternal blood glucose and development of offspring
On pre-embryonic day 7 and embryonic day 0.5, 7.5, 15.5, and 21.5 (E-7 d, E + 0.5 d, E + 7.5 d, E + 15.5 d, E + 21.5 d), the pregnant mice were weighed. The second drop of blood was collected from the tip of the tail to measure fasting blood glucose levels using a glucometer (Sinocare, GA-3, China), following 8 hours of food and water deprivation. On E + 12.5 d, the oral glucose tolerance test (OGTT) was conducted in accordance with standard protocols after an 8-hour period of food and water deprivation. The pups' blood glucose levels and body weight were monitored at PND 28, 42, and 56. The development of the pups between PND 8 and PND 12 was assessed using eye-opening scores and orientation tests.
2.5 Social behavior
As per the methodology outlined by Cohen et al.[38], repetitive/stereotypical behaviors, juvenile reciprocal social interactions, and three-chamber sociability were evaluated using the Noldus EthoVision-XT animal motion tracking system (Noldus, Netherlands). Following acclimatization, PND 28 pups were placed in an open-field experimental box to monitor repetitive/stereotypical behaviors. Subsequently, pairs of pups were introduced into the same boxes to observe social interactions. At PND 56, the pups were subjected to the three-chamber sociability test (Fig. 1G).
2.6 Sample collection
At PND 14, 28, 42, and 56, brain tissues were collected from the control and VPA groups (n = 4) to investigate temporal differences. To study the effects of remyelination, brain tissues were harvested from the placebo, clemastine, VPA + placebo, and VPA + clemastine groups (n = 6) at PND 56. Additionally, PND 28 brain tissues, utilized to explore the dysmyelination induced by GDM and prenatal VPA exposure, were obtained from pups in the control, VPA, GDM, and GDM + VPA groups (n = 6). The mice were administered a 50 mg/kg dose of 1% sodium pentobarbital as anesthesia and were then perfused with a 4% paraformaldehyde solution. The brains were extracted, weighed, and left to dehydrate in gradient ethanol overnight before being embedded in paraffin. Based on the anatomical location of the corpus callosum[39], these paraffin-embedded brains were sectioned into 3.5 µm slices.
Brain samples from identical mouse groups were harvested concurrently, employing a consistent methodology (n = 6). The samples underwent dehydration in a graduated sucrose solution overnight before being encapsulated in an optimal cutting temperature compound (SAKURA, YZ4583). Subsequently, the specimens were flash-frozen at -80 ℃ for 5 minutes, sectioned into 15.0 µm slices using a freezing microtome (Leica, CM1950, USA), and preserved at -20°C.
2.7 Chemical staining
Selected cerebral paraffin sections from the control, VPA, GDM, and GDM + VPA groups were utilized for further analysis. These sections were first incubated at 60 ℃ for 2 hours, followed by dewaxing with xylene for 30 minutes and a successive rehydration through a gradient alcohol procedure. Luxol fast blue (LFB) staining was then implemented, adhering to the methodology outlined by Xie et al[40]. Imagery was captured at 400× magnification using the fluorescence microscope (Olympus, BX63, Japan).
OLN-93 cells were seeded in 96-well plates at 4×103/well. After 12 h of growth, they were divided into eight groups with varying concentrations of GABA (0 mM, 2.5 mM, 5 mM, 10 mM, 25 mM, 50 mM, 100 mM, 250 mM), eight groups with varying concentrations of VPA (0 mM, 0.25 mM, 0.5 mM, 1.0 mM, 2.5 mM, 5.0 mM, 5.0 mM 10.0 mM, 25.0 mM), four groups with varying concentrations of Bicuculline (0 µM, 25 µM, 100 µM, 400 µM), and four groups with varying concentrations of CGP52432 (0 µM, 5 µM, 25 µM, 125 µM). The drugs were administered to each group for 24 h. The Cell Counting Kit-8 (CCK-8) assay was conducted according to standard procedures. Absorbance values indicative of cell activity were measured using the Multiskan™ FC Microplate Photometer (Thermo Fisher Scientific, 1410101, USA).
OLN-93 cells were seeded in 6-well plates at 2×105/well and subsequently divided into twelve groups. Six of these groups (blank, GABA, GABA + Bicuculline, GABA + CGP52432, GABA + 0.25 mM VPA, and GABA + 0.5 mM VPA) were cultured under normal conditions, while the other six groups (HFHG, HFHG + GABA, HFHG + GABA + Bicuculline, HFHG + GABA + CGP52432, HFHG + GABA + 0.25 mM VPA, and HFHG + GABA + 0.5 mM VPA) were cultivated in a HFHG medium. These groups underwent the same treatment protocol as delineated in the model induction section. Following fixation and crystal violet staining, images were captured from each well at 200× magnification using the fluorescence microscope (Olympus, IXplore, Japan). Additionally, in the 96-well plates, OLN-93 cells were seeded at 4×103/well (6 wells per group), and separated into the same 12 groups as described for the crystal violet staining procedure. The previous treatment regimen and CCK-8 assay were utilized to evaluate the cell activity of OLN-93.
2.8 Transmission electron microscopy
On PND 28, pups from the control, VPA, GDM, and GDM + VPA groups (n = 4) were administered 50 mg/kg of sodium pentobarbital for anesthesia. Subsequently, pre-cooled saline was infused, and a 1×1×1 mm section of the corpus callosum was extracted from each brain and immersed in a 2.5% glutaraldehyde solution for 24 hours. These sections were then fixed and sectioned into 70 nm slices. Following lead citrate staining, images were acquired at 1200×, 10,000×, and 100,000× magnification using a transmission electron microscope (TEM, Hitachi, H-7500, Japan).
2.9 Immunostaining
Paraffin-embedded cerebral sections from selected pups in the placebo, clemastine, VPA + placebo, and VPA + clemastine groups, as well as in the control, VPA, GDM, and GDM + VPA groups were randomly selected. Following the dewaxing and rehydration protocols outlined in the LFB staining methodology, antigen retrieval was performed using enhanced sodium citrate. Immunohistochemical staining was conducted using the universal mouse/rabbit polymer method detection system (ZSGB Biotech, PV-6000)[41] and the DAB chromogenic kit (ZSGB Biotech, ZLI-9019). Primary antibodies, CNPase (1:400, ab277621) and myelin basic protein (1:800, 78896s), were diluted with a 3% BSA solution. Following hematoxylin staining and dehydration processes, images were captured at a 400× magnification using a fluorescence microscope (Olympus, BX63, Japan).
Frozen brain slices were randomly selected from the pups in the placebo, clemastine, VPA + placebo, and VPA + clemastine group. Following antigen retrieval and blocking, mixed primary antibodies were applied, including Olig2 (1:200, 65915S, anti-rabbit) and PDGFRα (1:100, 12140181, anti-rat), Olig2 (1:200, anti-rabbit) and CC1 (1:50, OP80, anti-mouse), and PDGFRα (1:100, ab203491, anti-rabbit) and PCNA (1:100, ab29, anti-mouse) were used for 24 h at 4 ℃. Subsequently, corresponding secondary antibodies, labeled with Alexa Fluor 647-labeled Goat Anti-Rabbit IgG (1:500, A0468) and Alexa Fluor™ 488 Goat anti-Rabbit IgG (1:500, A11008), Alexa Fluor 647-labeled Goat Anti-Rabbit IgG (1:500) and Alexa Fluor™ 488 Goat anti-Rat IgG (1:500, A11006), and Alexa Fluor Cy3-labeled Goat Anti-Mouse IgG (1:500, A0521) and Alexa Fluor 488® Goat Anti-Rabbit IgG H&L (1:1000, ab150077), were applied in accordance with the primary antibodies used. The sections were covered with antifade mounting medium with DAPI on the slides. In the control, VPA, GDM, and GDM + VPA group, two other sets of mixed antibodies, including p44/42 MAPK (1: 100, 4695T, anti-rabbit) and NG2 (1:200, mab5384-I, anti-mouse) and phospho-p44/42 MAPK (1:100, 4370T, anti-rabbit) and NG2 (1:200, anti-mouse), were added. Alexa Fluor Cy3-labeled Goat Anti-Mouse IgG (1:500) and Alexa Fluor 488® Goat Anti-Rabbit IgG H&L (1:1000) were used as the mixed secondary antibodies. All of the pictures were taken at 400× magnification using the laser scanning confocal microscope (Zeiss, LSM880, Germany). Fluorescence was excited at 380 nm (DAPI), 488 nm (488), 550 nm (Cy3), or 647 nm (647) wavelengths.
2.10 Immunoblotting
The total proteins from OLN-93 cells were extracted from the groups described in the crystal violet staining using RIPA lysis buffer, phosphatase inhibitor cocktail A, protease inhibitor cocktail, and 0.1M EDTA (pH 8.0). Following the method of Liu et al.[42], the nuclear protein extraction kit (Solarbio, R0050) was utilized to extract nuclear proteins from OLN-93 cells. The protein concentration of the samples was determined using a bicinchoninic acid protein assay kit (Beyotime, P0012). After adjusting the concentration, the protein was combined with SDS-PAGE sample loading buffer. Lysates of 10 µl were separated on BeyoGel™ plus precast PAGE gel (Beyotime, P0452) and transferred to 0.2 µm PVDF membranes following standard procedures. The primary antibodies utilized in the western blotting included p44/42 MAPK (1:1000, ab200708), phospho-p44/42 MAPK (1:1000, anti-rabbit), MEK1/2 (1:1000, 8727T, anti-rabbit), mTOR (1:1000, 2983T, anti-rabbit), phospho-mTOR (1:1000, 2976S, anti-rabbit), Akt (1:1000, 4685S, anti-rabbit), phospho-Akt (1:1000, 4060T, anti-rabbit), DUSP-5 (1:1000, ab200708, anti-rabbit), HDAC-1 (1:1000, 34589T, anti-rabbit), HDAC-2 (1:1000, AF1555, anti-rabbit), HDAC-3 (1:1000, 85057S, anti-rabbit), HDAC-8 (1:1000, AF2737, anti-rabbit), β-actin (1:3000, MA1-140, anti-rabbit), β-tubulin (1:3000, MA5-16308, anti-mouse), and LaminB1 (1:1000, GTX103292, anti-rabbit). Following a secondary antibody incubation, images were captured with the assistance of BeyoECL star kit (Beyotime, P0018AS) and ImageQuant (GE, LAS4000, America).
2.11 Statistical analysis
The results of the behavioural tests were analyzed using Noldus EthoVision-XT 15. ImageJ 2.35 facilitated the analysis of chemical and immunohistochemical staining results, evaluating both the mean and integral optical density values. Additionally, the proportion of double-labeled cell number, area, and intensity derived from the immunofluorescence staining, the cell density of crystal violet staining, and band intensity in western blotting were analyzed using ImageJ. Myelinated axon density, total axon density, myelinated fiber ratio, and g-ratio were evaluated as results of transmission electron microscopy using the software developed by Zaimi et al[43]. Quantitative data were expressed as mean ± SD and analyzed using ANOVA through SPSS 21.0. Raw data were compiled using Microsoft Excel. Count data were presented as median ± 95% confidence interval and examined with Pearson's chi-squared test. A significant difference was denoted by P < 0.05.