The gut microbiota plays a vital role in shaping brain development through complex bidirectional communication within the microbiota-gut-brain axis. Emerging evidence highlights neural, immune, endocrine, metabolic, and epigenetic pathways by which gut microbes influence neurodevelopmental processes. This review synthesizes current knowledge on the temporal dynamics of gut colonization and brain maturation. Drawing on mechanistic insights from animal models, we emphasize the central role of the maternal microbiota and particularly, microbially derived metabolites that cross the feto-placental barrier and shape fetal brain development. We also discuss molecular and cellular targets of microbial influence, implications for neurodevelopmental disorders, and potential therapeutic strategies. Understanding these interactions opens new avenues for early-life interventions aimed at optimizing neurodevelopment and preventing neuropsychiatric conditions.

The developing brain shows remarkable capacity for adaptation following early adversity, but the behaviors that influence neural compensation mechanisms remain unclear. Prenatal stress exposure provides a natural model for studying these mechanisms, as it alters neurodevelopment while allowing examination of potential protective factors. However, whether early adaptive behaviors-the skills needed to meet everyday demands such as self-care and communication-can buffer against the neural consequences of prenatal stress has not been established. Natural disasters provide a unique opportunity to examine these mechanisms, as they serve as measurable prenatal stressors with well-defined exposure timing. In this pilot study, using a quasi-experimental design, we examined children with (n=11) and without (n=23) prenatal exposure to Superstorm Sandy (SS) to investigate how early adaptive behaviors (ages 2-6) moderate the association between prenatal stress (i.e., exposure to a natural disaster) and later brain activity during emotional processing (age 8). We first examined main effects of SS on both adaptive behaviors over time (ages 2-6 years) and functional brain activation at age 8 in brain regions responsible for facial emotional processing. Moderation models subsequently explored whether early-life adaptive behaviors influenced the association between SS and later brain activation. The Behavior Assessment System for Children, Second Edition (BASC-2) measured child adaptive behaviors. Functional Magnetic Resonance Imaging (fMRI) measured regional brain activation using an emotional face processing task. Prenatal stress exposure was associated with non-significant trends toward reduced adaptive behaviors over time and reduced brain activation in the right ventral anterior insula. Critically, early adaptive behaviors moderated the association between prenatal stress and later brain activation in the left amygdala and both hemispheres of the hippocampus, ventral anterior insula, and rostral anterior cingulate cortex. Simple slopes analyses revealed that prenatal stress was associated with significantly reduced brain activation at low adaptive skills. However, this association was attenuated among children that exhibited higher adaptive skills such that activation patterns were comparable to their unexposed peers. Our preliminary moderation (i.e., interaction) findings provide initial evidence that adaptive behaviors may serve as a neural buffer against prenatal stress. This protective pattern indicates that early adaptive skills may help maintain neural responsiveness following prenatal stress exposure. If validated in larger, adequately powered samples, interventions targeting adaptive behaviors in early childhood could potentially reduce the neural burden of prenatal stress and support more resilient brain development in at-risk populations. These findings highlight adaptive behaviors as potential targets for early intervention to promote neural resilience following prenatal adversity.

Neonatal hypoxic-ischemic (HI) brain injury is a major cause of mortality and long-term neurological disability, yet effective neuroprotective strategies remain limited. Microglia are central mediators of injury and repair, with arginase 1 (ARG1) marking anti-inflammatory, reparative states. However, the functional roles of ARG1+ microglia in tissue remodeling after HI are poorly understood. Neonatal mice (P10) underwent HI using the Vannucci procedure. ARG1 activity was inhibited pharmacologically using N-omega-hydroxy-nor-L-arginine (nor-NOHA). ARG1 expression, microglial morphology, efferocytosis, tissue scar, and injury volume were assessed via immunohistochemistry, Western blotting, and arginase activity assays at 1 and 5 days post-injury. ARG1+ microglia rapidly engaged apoptotic neurons, exhibiting phagocytic activity confirmed by CD68 expression. Nor-NOHA treatment reduced ARG1 enzymatic activity, impaired microglial process extension, attenuated efferocytosis, and increased injury volume. ARG1+ microglia persisted in the glial scar and colocalized with collagen I alpha-1 (Col1a1), suggesting a role in extracellular matrix (ECM) deposition. Inhibition of ARG1 decreased Col1a1 expression, highlighting its contribution to tissue remodeling. ARG1+ microglia are pivotal in neonatal HI, mediating early efferocytosis and later ECM remodeling, thereby limiting injury and shaping scar architecture. Pharmacological blockade of ARG1 exacerbates injury, underscoring its reparative function. These findings establish ARG1 as a critical regulator of microglial-mediated neuroprotection and tissue repair, providing a potential therapeutic target for neonatal HI brain injury.

Introduction: Precise regulation of neurite initiation, elongation, and branching is critical for neuronal network formation. Rac1, a key regulator of cytoskeletal remodeling, influences neurite morphogenesis through protein phosphorylation-mediated signaling, but the global phosphorylation landscape that governs Rac1-mediated morphogenesis remains unknown. Method: To address this knowledge gap, we performed phosphoproteomics profiling of primary rat cortical neurons treated with 3, 10, or 30 µm of a Rac1 inhibitor for 48 h to evaluate phosphoprotein dynamics. Phosphorylation levels of 167 signaling proteins were quantified using a targeted phospho-antibody array and correlated with neuron count, neurite count and length, and branch point count. Results: Correlation analysis identified morphology-specific phosphoproteins, such as Tau (Thr181), GluR1 (Ser863), TrkB (Tyr515), and Merlin (Ser10), whose phosphorylation levels were significantly altered across three Rac1 inhibitor concentrations, and multiple phosphorylation sites showed dose-specific correlations with neurite morphology features. Conclusion: These results define a correlation-based framework linking phosphoprotein signaling to neurite morphology and offer novel insights into neurodevelopmental processes, neuronal disorders, and developmental neurotoxicity.

The long-term neurological consequences of SARS-CoV-2, the virus responsible for the COVID-19 pandemic, are an area of growing concern, particularly for prenatally exposed individuals. Prior research has shown that APOE4, the leading genetic risk factor for late-onset Alzheimer's disease, is associated with increased COVID-19 severity and enhanced SARS-CoV-2 neurotropism. However, whether the interaction between APOE4 and SARS-CoV-2 infection leads to adverse neurodevelopmental outcomes remains unclear. Using human induced pluripotent stem cell derived cortical and ganglionic eminence organoids (COs and GEOs) to model neurodevelopment, we have previously reported that SARS-CoV-2 preferentially infects glial cells, and that APOE4 promotes gliogenesis in COs and accelerates GABAergic neuron differentiation in GEOs. Here, we build upon our previous work by using COs and GEOs to examine how APOE4 modifies cellular responses to SARS-CoV-2 during late gestational development. Using low viral titers to better mimic natural infection, COs and GEOs were infected at 220-270 DIV, aligning with the third trimester, and were analyzed 7 days post infection. We observed region-specific, APOE4-dependent changes. In infected COs, APOE4 elevated immature astrocyte marker, suggesting a genotype-dependent glial response. Additionally, infected GEOs exhibited reduced marker expression for mature neurons within both genotypes. Notably, APOE4 and infection interacted to modulate immature neuron expression in a region-specific manner. Taken together, this study suggests that APOE4 modulates region-specific responses to low-grade SARS-CoV-2 infection, underscoring the importance of exploring how genetic risk factors alter neurodevelopmental vulnerability to prenatal viral infection.

Introduction: We recently identified variants in 10 genes that are members of either the p53 pathway or Fanconi Anemia Complex (FAC), regulators of DNA repair (DNA damage response [DDR]) in 17 cases with pediatric acute-onset neuropsychiatric syndrome (PANS) or regression in autism spectrum disorder and other neurodevelopmental disorders (NDD). We aimed to identify additional cases with genetic vulnerabilities in DDR and related pathways. Methods: Whole-exome sequencing (WES) and whole-genome sequencing (WGS) data from 32 individuals were filtered and analyzed to identify ultrarare pathogenic or likely pathogenic variants. Results: Variants affecting DDR were found in 14 cases diagnosed with PANS or regression (CUX1, USP45, PARP14, UVSSA, EP300, TREX1, SAMHD1, STK19, MYTl1, TEP1, PIDD1, ADNP, FANCD2, and RAD54L). The CUX1 variant is de novo, as are two cases that had mutations in genes that affect mitochondrial functions that are connected directly or indirectly to mitophagy (PRKN and POLG), which can trigger the same innate immune pathways when disrupted as abnormal DDR. We also found pathogenic or likely pathogenic secondary mutations in several genes that are primarily expressed in the gut that have been implicated in gut microbiome homeostasis (e.g., LGALS4, DUOX2, CCR9). Conclusion: These findings align with previous genetic findings and strengthen the hypothesis that abnormal DDR and mitochondrial dysfunction underlie pathogenic processes in at least some cases of neuropsychiatric decompensation. The potential involvement of genetic variants in gut microbiome homeostasis is a novel aspect of our study. Functional characterization of the downstream impact of DDR deficits may point to novel treatment strategies.

Plain Language SummaryRecent studies show that stress experienced early in life can make a person more sensitive to stress later on. This can lead to various emotional disorders including anxiety and depression. Many children may experience stress due to inappropriate parental care related to poor life situation. In the present study, we reproduced such conditions in a model of limited bedding and nesting material in rats. We looked at what happens when newborn rats live under these conditions from day 2 to day 9 of their life and studied expression of several groups of genes in their brain at the age of 1 month. We also studied how these alterations may affect their response to other stressful conditions, such as restraint. Our findings show that having poor life conditions as a baby can cause subtle changes in how certain genes work in the brain. The work of genes was mostly altered in the ventral hippocampus, a brain region which is responsible for emotions and memory. However, even these subtle changes can affect how the brain responds to other stresses later in life.

Introduction: Neonatal rats, but not juvenile rats, show spontaneous hindlimb locomotor recovery after complete thoracic spinal cord transection (SCT). Significant increases in parvalbumin-positive proprioceptive nerve terminals are observed on motoneurons in both neonatal and juvenile rats with SCT compared with intact rats. Methods: In the present study, we focused on Chx10-positive V2a interneurons, which partially comprise the central pattern generator, and examined parvalbumin-positive nerve terminals on Chx10 neurons and the perineuronal net formation around these neurons using Wisteria floribunda agglutinin (WFA) as a marker 2 weeks after SCT on postnatal day 5 (neonatal) or day 20 (juvenile). Results: Rats with CST during the neonatal period had a significantly greater number of parvalbumin-positive terminals on Chx10 neurons compared to age-matched intact rats, whereas no significant difference was detected between rats with SCT during the juvenile period and age-matched intact rats. Chx10 neurons for which ≥50% of the circumference was surrounded by WFA were identified as WFA-positive. The proportion of WFA-positive neurons among Chx10-positive neurons did not differ significantly between neonatal SCT and age-matched intact rats, but was significantly higher in juvenile SCT and age-matched intact rats. Conclusion: These findings suggest that SCT promotes the formation of proprioceptive afferent terminals on Chx10-positive neurons. The significant increase in terminals following SCT in neonatal rats might facilitate spontaneous motor recovery, whereas enhanced perineuronal net formation around Chx10 neurons following juvenile SCT might restrict synaptic formation and impair motor recovery.

In the article "Caffeine as a Treatment for Perinatal Hypoxic-Ischemic Brain Injury: The Potential Risks and Benefits" [Dev Neurosci. 2025; Online ahead of print. https://doi.org/10.1159/000545126] by Zhou et al., the authors noted that there were several errors within their article. The corrections are listed below.In the section "Adenosine and Adenosine Receptors" currently reads, "By contrast, A2 adenosine receptor knockout mice that were subjected to common carotid ligation and hypoxia at P7 had more severe brain injury with worse performance in motor behavioural tests, compared with wild-type mice [30]." Should correctly read, "By contrast, A2A adenosine receptor knockout mice that were subjected to common carotid ligation and hypoxia at P7 had more severe brain injury with worse performance in motor behavioural tests, compared with wild-type mice [30]."Reference 47 was erroneously included and is not an intended citation, this reference should be deleted. Under the subheading, "Caffeine: Perspectives and Future Directions", the following sentences should correctly read:It is unclear why studies in rodents suggest benefit with prophylactic caffeine, before HI, whereas limited benefit with prophylactic caffeine in lambs [39], and deleterious effects of adenosine A1 receptor blockade [20]. Speculatively, this difference may reflect that these rodent studies used inhalational hypoxia and so caffeine may help avoid apnea [48] and so reduce the risk of deep hypoxemia. Regardless of the precise mechanism, these findings strongly suggest that considerable caution is needed before considering human studies.The following two errors table 1 should be corrected. For the study Yang et al., 2022, under key findings was missing before "Microglia M2 polarisation". For the study, Sabir et al., 2023 and in row 8 the study year was corrected to 2023.The corrected table is shown below:Table 1.Summary of preclinical studies on the effects of caffeine for perinatal hypoxic-ischemic brain injuryStudySpecies and ageN per groupInsultDose and timingKey findingsDi Martino et al. [34] (2020)P10 micen = 8-10Common carotid artery ligation and hypoxia (1 h)Caffeine 5 mg/kg, i.p. immediately after HI↓ Grey and white matter lesion size↓ Amoeboid microglia and apoptotic cellsn = 6-8Caffeine started at 6, 12 or 24 h after HINo neuroprotective effectWinerdal et al. [35] (2017)P10 micen = 13-29Common carotid artery ligation and hypoxia (1 h)Caffeine 5 mg/kg i.p. immediately after HI↓ Brain atrophy↑ Time on the rotorod behavioral testPotter et al. [37] (2018)P6 ratsn = 6Common carotid artery ligation and hypoxia (2 h)Caffeine citrate 20 mg/kg i.p. administration immediately after HI↑ Performance in rotarod and water maze behavioral tests↑ Silent gap detection (speech detection)Bernis et al. [38] (2025)P7 ratsn = 5-50Common carotid artery ligation and hypoxia (90 min)Caffeine citrate 15, 20 or 40 mg/kg i.p. administration immediately before HI or 40 mg/kg immediately after HI repeated at 24 and 48 h↓ Brain area loss (greatest effect with 40 mg/kg before HI)↓ Microgliosis (40 mg/kg before HI)n = 14Caffeine citrate 120 mg/kg before HI and at 24 h↑ MortalityYang et al. [33] (2022)P3 ratsn = 6Common carotid artery ligation and hypoxia (2.5 h)Caffeine citrate 20 mg/kg/day, i.p. from day 2-6↓ Ventricle dilation↑ MBP expression↓ NLRP3 inflammasome activation↑ Microglia M2 polarisation↓ Microglial activation and microglia M1 polarizationKilicdag et al. [32] (2014)P7 ratsn = 8Common carotid artery ligation and hypoxia (2 h)Caffeine citrate 20 mg/kg/day, i.p. immediately before HI and at 0, 24, 48 and 72 h after HI↓ Cell loss in hippocampus and cortexAlexander et al. [36] (2013)P7 rats (males only)n = 8-15Common carotid artery ligation and hypoxia (2 h)Caffeine 10 mg/kg i.p. immediately after HIPartially ↑ cortical volume↑ Performance in morris water mazeSabir et al. [39] (2023)P7 ratsn = 12Common carotid artery ligation and hypoxia (1.5 h)Caffeine 40 mg/kg i.p. 1 h before hypoxia and at 24 and 48 h↓ Brain area lossMike et al. [40] (2024)Gestational day 141-143 fetal sheepn = 4-41Umbilical cord occlusion, until the onset of asysole1 g i.v. to ewe before delivery, 20 mg/kg caffeine citrate, 2 doses of 10 mg/kg i.v. at 24 and 48 h (lambs)↔ Hippocampal neuronal survival↓ Neuronal apoptosis hippocampus (CA3 only)↑ Feeding and activity↓ Microgliosisn = 8Caffeine citrate 60 mg/kg i.v. and 30 mg/kg at 24 and 48 h (lambs only)↑ Mortality.

Introduction: Lysophosphatidic acid (LPA) is a bioactive phospholipid that mediates a variety of biological actions through binding to G protein-coupled receptors known as LPA receptors (LPARs). In mammals, six LPAR subtypes (LPAR1-6) have been identified. This study aimed to determine the expression of LPAR4 in the developing mouse brain. Methods: Brains samples were prepared from mice in various stages of development and biochemical and immunohistochemical analyses were conducted using anti-LPAR4. Results: Western blot analysis detected two LPAR4-immunoreactive species at ∼50 kDa and ∼42 kDa from embryonic day 16.5 (E16.5). The ∼50 kDa molecule increased during development, reaching a peak at postnatal day 3 (P3), and then gradually decreased through P22. In contrast, the ∼42 kDa molecule continued to increase up to P22. Immunohistochemical analyses demonstrated strong LPAR4 expression in neural cells in the intermediate zone and cortical plate of the E15.5 cerebral cortex, whereas neural progenitors in the ventricular and subventricular zones exhibited weaker expression. At P15, fiber-like staining resembling the apical dendrites of cortical neurons and hippocampal pyramidal cells was also observed. Conclusion: This study demonstrated dynamic, spatiotemporal changes of LPAR4 expression in the brain from embryonic to postnatal stages. These findings support a potential role for LPAR4 in neural development.