Diffuse midline glioma (DMG) is a rare and aggressive pediatric brain tumor, with a median survival of less than 12 months. Due to its location in the pons, surgical resection is impossible, leaving radiation therapy as the only palliative treatment option. Unfortunately, radiation therapy yields minimal improvement in survival. Thus, characterization of the vascular component of the DMG microenvironment at the cellular and molecular levels following radiotherapy to improve therapeutic strategies. A human syngeneic blood-brain tumor barrier (BBTB) in vitro model, comprising endothelial cells, pericytes and DMG cells was submitted to a single dose of radiation (2 Gγ to 6 Gγ) and was characterized for its physical and metabolic properties over a period of 7 days post-exposure. The results were then compared to the effects of the same irradiation protocol on a physiologic blood-brain barrier (BBB) model. Following irradiation, the endothelial permeability of the BBB ECs and the BBTB ECs was preserved for up to 7 days but associated with Claudin-5 heterogeneous distribution at the ECs borders and decrease of expression after irradiation. Nevertheless, irradiation was found to potentiate the effect of TNFα on the physical integrity of the BBB, which was less important for the BBTB. The metabolic properties of the BBB and BBTB were modulated by radiation at the transcriptional level. Interestingly, different regulations were observed in endothelial cells and pericytes. Notably, pericytes have demonstrated compensatory effects. Immunoblots confirmed the decrease of BCRP, MRP4 and MFSD2A in BBTB endothelial cells after irradiation. Despite significant reduced efficiency, P-gp/BCRP efflux pump activity remains functional in endothelial cells and pericytes following irradiation. Irradiation sensitizes the BBB, but to a lesser extent the BBTB, to the effects of pro-inflammatory cytokines. The observed decrease in P-gp/BCRP activity, as well as the involvement of MFSD2A, MRP4 and Claudin-5 regulation, warrant further investigations. The online version contains supplementary material available at 10.1186/s12987-026-00778-6.
We aimed to streamline the Tap Test in terms of timing and content in order to reduce the burdensome assessment in tap-test-positive patients and optimize the protocol of CSF-TT in iNPH, and strive for building explicit connection between CSF-TT and outcome of surgery. This retrospective cohort study enrolled a total of 69 inpatients with a positive Tap Test according to strict inclusion and exclusion criteria. Among them, 31 underwent the traditional Tap Test (T-TT, baseline-8–24 h-72 h) and 38 underwent the refined Tap Test (R-TT, baseline-24–48 h). All enrolled patients completed the same battery of assessments and ventriculoperitoneal shunt surgery. The positive predictive value (PPV) at different time points were compared between the two test versions using Chi-Square Test and Fisher’s exact test. The correlation between test results at each time point and surgical effectiveness was expressed using the phi coefficient, kappa value, and p-value. The magnitude of differences in mean scores for each measure between patients with effective versus ineffective surgery was analyzed using t-tests, p-values, and Cohen’s d. The outcome was collected on the outpatients’ records at 3–6 months follow-up. Among 69 probable iNPH patients, 63 sick people were diagnosed definite iNPH. The PPV of the R-TT was comparable to that of the T-TT (PPV: 92.11% vs. 90.32%, p = 1.00). No significant differences in PPV of time points were detected in T-TT (8 h vs. 24 h: p = 0.973; 8 h vs. 72 h: p = 1.00; 24 h vs. 72 h: p = 0.656), which is similar with that in R-TT (24 h vs. 48 h: p = 0.627). The assessment results at 24 h showed a strong correlation and agreement with surgical outcome (phi = 0.7073, kappa = 0.7039, p < 0.0001). The mean improvement in SDMT scores was significantly greater in the surgical responder group compared to the non-responder group at 24 h (Cohen’s d = 0.927, t = 4.25, p = 0.001). Assessment at 24 h demonstrates the strongest concordance with postoperative outcomes among the evaluated time points and appears sufficient for efficient surgical triage in patients with probable iNPH. Within this framework, the SDMT shows the greatest discriminative capacity between surgical responders and non-responders, supporting its potential role as a practical adjunct measure. The assessment protocol proposed in this study offers a feasible and reproducible approach for clinical evaluation and treatment decision-making, with the potential to optimize healthcare resource utilization, and reduce patient burden. The online version contains supplementary material available at 10.1186/s12987-026-00783-9.
IIIG9 was originally identified as a novel regulatory subunit of protein phosphatase 1 (PP1, PPP1R32) in ependymal cells, where it localizes to cilia and adherens junctions. However, it remains unclear whether IIIG9 forms complexes with PP1α, β, or γ to regulate ciliary function and/or the stability of adherens junctions in polarized cells. MDCK and ependymal cells, both polarized epithelial cell types, provide valuable models for investigating the molecular mechanisms underlying epithelial polarization and neoplastic events. We characterized the expression of PP1 subunits in ependymal cells using RT-qPCR in combination with laser microdissection (LMD) and in situ hybridization. We analyzed the colocalization of catalytic PP1 isoforms and IIIG9 by confocal microscopy, both in situ and in vitro. Similar analyses were performed in monolayers and cysts of MDCK cells. To assess the interaction between IIIG9 and PP1α, we used fluorescence resonance energy transfer (FRET) analysis in HEK cells and proximity ligation assays (PLA) in adult ependymal cells and MDCK cells. PP1α is the predominant PP1 isoform expressed in ependymal cells lining the ventricular wall of the adult brain, similar to IIIG9. Additionally, IIIG9 mRNA was detected in ependymal cilia, hippocampal neurons, and the cerebral cortex. Both proteins were found in the cytoplasm, at adherens junctions (positive for E-cadherin), and in cilia (positive for acetylated α-tubulin) throughout the entire ventricular system. In polarized MDCK cell cultures (monolayers and cysts), IIIG9 and PP1α colocalized with adherens junction proteins, including β-catenin and Par3. FRET analysis revealed a high-efficiency interaction between IIIG9 and PP1α, both in the cytoplasm and in puncta near the plasma membrane in HEK cells. Additionally, PLA detected positive interactions between IIIG9 and PP1α, as well as between IIIG9 and E-cadherin, in adult brain ependymal cells. Our data confirm the interaction between IIIG9 and PP1α in ependymal and MDCK cells. This interaction likely plays a key role in mediating the subcellular functions of PP1 at adherens junctions and cilia in polarized cells, thereby contributing to the regulation of cell polarity. The online version contains supplementary material available at 10.1186/s12987-026-00762-0.
Vascular endothelial growth factors (VEGFs) and their receptors (VEGFRs) are critical regulators of angiogenesis and vascular homeostasis. While VEGF signaling has been extensively studied in endothelial cells, emerging evidence suggests it also plays roles in non-endothelial brain cells. However, its spatial and cell-type-specific function within the hypothalamus, and more specifically at the level of the blood/CSF barrier, remains poorly defined. In particular, little is known about VEGF receptor expression in tanycytes, a specialized glial population that lines the third ventricle and regulates body-brain communication within the median eminence (ME), a key neurovascular interface located at the tuberal region of the hypothalamus. We used a multi-modal approach including single-cell RNA sequencing (scRNA-seq) reanalysis, RNAscope in situ hybridization, immunohistochemistry, FACS-isolated qPCR in male and female mice, and human spatial transcriptomics to map the expression of VEGFR1 (Flt1), VEGFR2 (Kdr), and VEGF ligands in hypothalamic tanycytes across gender, development, and aging. Our data reveal a striking spatial compartmentalization of VEGFR expression in tanycytes within the ME and the arcuate (ARH), ventromedial (VMH) and dorsomedial (DMH) hypothalamus. VEGFR2 is preferentially expressed in ARH-tanycytes, while VEGFR1 is confined to VMH/DMH-tanycytes; and none of these receptors are expressed in ME-tanycytes. This pattern is unique to the ME and not observed in other circumventricular organs. VEGFR1 expression is established neonatally in mice (P0) and remains stable throughout life, whereas VEGFR2 expression becomes progressively refined postnatally, localizing to ARH-tanycytes in adulthood and showing a significant decline with aging. VEGFA is broadly expressed in all hypothalamic tanycytes, including ME-tanycytes, supporting a paracrine model of signaling. Importantly, despite species-specific differences in the spatial organization of VEGF receptors in tanycytes, the presence of VEGF ligands (Vegfa and Vegfb) and receptors in tanycytes of both mice and humans supports partial evolutionary conservation of VEGF signaling at the brain-blood interface. Our findings unveil, for the first time a non-endothelial VEGF signaling system in hypothalamic tanycytes that is spatially compartmentalized, developmentally programmed and age-dependent. These insights reveal new roles for VEGF signaling in neurovascular and neuroendocrine function, raising important considerations for central effects of VEGF-targeted therapies in aging and disease.
Cerebrospinal fluid tap test (CSF TT) is a key predictive method commonly used to identify candidates for shunt surgery in idiopathic Normal Pressure Hydrocephalus (iNPH), however, the sensitivity of this procedure is limited. The aim was to compare sensitivity of a broad set of physiotherapy-based clinical tests for assessing changes following CSF TT in a group of selected iNPH patients and to explore potential sex-based differences in the responses. Ninety-five selected iNPH patients (mean age 77, SD 6, 57% male), underwent CSF TT prior to shunt surgery. Clinical tests comprised different static balance tests in varying conditions including feet and heels apart and together; eyes open and closed; on floor and foam cushion, followed by gait and functional performance measures including 10-meter walk test (10MWT), 3-meter backward walking (3MBW), Timed-up and Go (TUG) and 6-minute walking test (6MWT). Duration in seconds (sec) and distance were noted. Assessments were conducted at set time points before and after the drainage procedure. Changes in performance were analyzed, and responses compared between sexes. Significant increase was observed across most static balance tests except tandem eyes closed (EC) and one-legged stance tests. The largest enhancements were seen in the Romberg test (4.9 s), followed by foam cushion-eyes open (EO) tests (feet together: 3.9 s and heels together: 3.8 s), heels together-EC (3.8 s) and tandem-EO (3.1 s). Gait and functional tests significantly improved (p < 0.001) and 3MBW showed the largest change, with a 32% reduction in sec and a 23% reduction in steps, followed by decrease in TUG sec (27%) and 10-MWT sec (21%), and 6MWT distance increase (25%). Males demonstrated significant increase in heels together-EO and feet together-EO duration which were not observed in females; after adjusting for baseline performances, no significant sex-related differences in responsiveness to CSF TT were found. Challenging comprehensive physiotherapy tests, including gait assessments in direction and capacity, such as 3MBW and 6MWT, together with postural control evaluations using foam cushion and tandem stance increases the sensitivity to change and are suggested to be used to detect functional changes after CSF TT in patients with iNPH. The project is registered in "FoU inSweden” (Research and Development in Sweden) ID: 285356. The online version contains supplementary material available at 10.1186/s12987-026-00776-8.
Investigation of the failure of shunts used to treat hydrocephalus has been a topic of research for several decades, though it is difficult to understand the role of heterogeneous patient ventriculomegaly shifts over time. In this work, we present the design, verification, and validation of a novel benchtop model of the human lateral and third ventricles. 3D models were rendered from MRI of pre-revision hydrocephalic patients (n = 6), printed in hollow High Impact Polystyrene (HIPS) molds, and injected with silicone rubber allowed to rotationally cure for three hours. Wall thickness of the ventricle models was measured via random point sampling (n = 300), assessing distribution of silicone rubber within the mold. One sample z-test was used to compare mean wall thickness to the target thickness of 1 mm. To visualize the inner volume post-hoc, expanding polyurethane foam was injected into ventricle models, creating negatives of the hollow molds. The negatives were 3D scanned and measured for frontal horn diameter, occipital horn diameter, Evan’s Index, Frontal-Occipital-Horn Ratio, and frontal horn volume. A single MRI was chosen for verification of repeatability (n = 11). Ventricular size was validated across models from patients (n = 6). Accuracy of ventricular expansion (indirect compliance assessment) was tested. For all tests, a confidence interval was set at 0.95 (α = 0.05). The verification methods indicated that low variance of experimental clinical values were observed between ventricle model replicates. Nominal-actual comparison results consistently showed similar datapoint displacement densities between models. Validation showed that most ventricle geometries fell within 5% of MRI measurements. A strong positive correlation was observed between internal pressure and volume (R2 = 0.9932). Experimental Frontal Horn and Occipital Horn Diameter measurements compared to clinical observations showed anatomical accuracy. This study validates and verifies a model of the human ventricular system with arbitrarily chosen heterogeneous patient ventricles of varying morphologies and volumes. Paired with a pump, this model can be used to recapitulate cerebrospinal fluid flow. We show the recreation of patient-specific clinical lateral ventricle characteristics of size, shape, and static compliance control necessary to study the influence of these parameters on shunt function. The manufacturing process has the capacity to create accurate benchtop models of the lateral and third ventricles with geometric detail that should be refined over time with additive systems accounting for cranial viscoelastic compliance and varying material properties with elastic and shear moduli more similar to brain. The online version contains supplementary material available at 10.1186/s12987-025-00742-w.
Claudin-5 plays a crucial role in maintaining blood-brain barrier (BBB) integrity through endothelial tight junction formation. Alternative splicing of claudin-5 within the microvascular endothelium may modulate BBB structural and functional dynamics, thereby affecting neuronal damage and recovery after ischemic stroke. The aim of this study was to investigate temporal and hemisphere-specific changes in claudin-5 isoform expression after ischemic stroke and to evaluate their correlation with BBB dysfunction and neurological outcomes. Male Wistar rats underwent thromboembolic stroke. Claudin-5 isoform expression was assessed at 3, 6, and 24 h post-stroke onset, with additional groups receiving recombinant tissue plasminogen activator (rt-PA) at 4 h post-stroke. Brain edema, infarct volume, hemorrhage, and cerebral blood flow were evaluated using 9.4T MRI. Ipsilateral and contralateral cerebrovascular claudin-5 expression was quantified via western blotting while neurological function was assessed by 28-point neuroscore. RNA sequencing analysis was performed to identify novel splice variants. In addition, molecular dynamics simulations of AlphaFold-predicted claudin-5 isoform 1 and 2 tetramers embedded in lipid bilayers were performed to quantify steric–electrostatic barrier properties and ion permeability. A time-dependent increase in claudin-5 isoform 1 (35 kDa) expression levels in the ipsilateral cerebrovasculature at 6 h was observed. Isoform 2 (25 kDa) and fragment (10 kDa) isoforms of claudin-5 remained unchanged. Treatment with rt-PA maintained the elevated levels of isoform 1 claudin-5 protein expression within the ipsilateral hemisphere. Increased isoform 1 expression correlated with edema, hemorrhage, and worsened neurological function at 24 h post-stroke onset. RNA sequencing revealed novel CLDN5 splice isoforms in post-stroke rat brain tissue resembling known human CLDN5 isoforms. Computational modeling suggested that isoform 1 forms a wide but electrostatically exclusionary pore with strong anion selectivity, whereas isoform 2 permits markedly greater cation permeability. These isoform-specific barrier properties are consistent with the in vivo association of isoform 1 upregulation with BBB dysfunction after stroke. These findings demonstrate that ischemic stroke induces temporal, hemisphere-specific alterations in claudin-5 isoform expression that correlate with BBB dysfunction and poor neurological outcomes. The combination of RNA sequencing and molecular dynamics simulations indicates that alternative splice variants of claudin-5 confer distinct structural and permeability profiles, representing a previously unrecognized mechanism of endothelial tight junction dysfunction in stroke. These results highlight claudin-5 isoform expression as both a novel biologically relevant and a potential therapeutic target for preserving BBB integrity following cerebral ischemia. The online version contains supplementary material available at 10.1186/s12987-026-00798-2. [Image: see text]
Dynamic contrast-enhanced MRI has become an invaluable tool for mapping cerebrospinal fluid (CSF) flow and delineating solute transport pathways in the brain. However, the influence of experimental parameters, including infusion location, infusion rate, and anesthetic regimen, on tracer transport kinetics and distribution patterns remains incompletely characterized, potentially confounding interpretation of results. We investigated the impact of infusion parameters on tracer transport in mice using Gd-DTPA as a CSF tracer. The effect of infusion rate was assessed by administering Gd-DTPA into the cisterna magna (ICM) at two different rates under isoflurane anesthesia. The influence of anesthesia was evaluated by comparing tracer transport patterns under isoflurane versus ketamine/xylazine and low-dose isoflurane at the slower infusion rate. To examine the impact of infusion site, Gd-DTPA was also delivered into the lateral ventricle (ICV) for comparison with ICM delivery. Region-specific signal time courses were analyzed using cross-correlation and hierarchical clustering to characterize tracer transport pathways. Tracer transport within the brain was significantly influenced by infusion location, rate, and anesthetic regimen. ICV infusion produced rapid, extensive transport into deep brain structures, while ICM infusion promoted transport toward ventral regions. Cross-correlation analysis revealed that ICM infusion primarily facilitated tracer transport along periarterial spaces, whereas ICV infusion favored transport across the ventricular-parenchymal interface. Hierarchical clustering of region-specific signal time courses further demonstrated distinct transport patterns associated with different infusion sites. Experimental conditions substantially impact tracer transport kinetics and its spatial distribution in CSF dynamics studies. While infusion site determines primarily transport pathways, i.e., via periarterial spaces versus across ventricular-parenchymal interface, infusion rate and anesthesia modulate transport efficiency. These findings underscore the importance of methodological considerations in the design and interpretation of CSF tracer studies and emphasize the need for standardized protocols to facilitate meaningful comparisons across research groups. The online version contains supplementary material available at 10.1186/s12987-026-00758-w.
Preclinical neurological research relies predominantly on male animals, despite well-documented sex differences in neurological diseases, which ultimately may result in sex-dependent treatment efficiency. A key player in neurological disease treatment is the blood-brain barrier (BBB), the barrier property of brain capillaries, which tightly regulates molecular exchange between the blood and the brain. The BBB represents a major obstacle to brain drug delivery due to its tightness and presence of drug efflux pumps, with some studies suggesting that the BBB properties may differ between sexes. However, in vivo evidence is limited, and whether primary in vitro BBB models, commonly used to evaluate the permeability of novel drug candidates, display sex-dependent differences, lacks attention. With this study, we therefore aimed to investigate if a mouse in vitro model of the BBB displayed sex-dependent differences in BBB morphology and phenotype, and therefore whether sex should be considered a critical variable in its use. Primary mouse brain endothelial cells (PMBEC) were isolated from cortices of sexually mature C57Bl/6 mice. Transendothelial electrical resistance (TEER) measurements and transport of the paracellular marker [14C]-mannitol were used to evaluate monolayer tightness. Gene and protein expression of tight junction proteins, selected transporters and receptors as well as efflux transporters were assessed. P-glycoprotein (P-gp) function was evaluated in bidirectional [3H]-digoxin transport studies. Female- and male-derived PMBECs grew in monolayers, expressed the endothelial marker von Willebrand factor and showed elongated spindle-shaped morphology typical of endothelial cells. Female- and male derived PMBEC monolayers exhibited comparable barrier properties as reflected by TEER measurements and mannitol permeability. Tight junction mRNA and protein expression did not differ between sexes and displayed similar expression levels of key transporters. Lastly, P-gp and breast cancer resistance protein were detected in PMBECs of both female and male origin and P-gp function was similar in the two sexes. Our present study shows that PMBECs do not differ substantially between sexes. However, as this is the first study of its kind, it warrants further investigations into sex differences in PMBECs and whether these fully translates to the in vivo BBB. [Image: see text] The online version contains supplementary material available at 10.1186/s12987-026-00781-x.
Ischemic stroke is a leading cause of death and disability worldwide. Post-stroke neuroinflammation is critically shaped by the dynamic activation and infiltration of lymphocytes, yet their precise temporal patterns, trafficking mechanisms, and ultimate impact on recovery remain poorly defined. The present study aims to define the spatiotemporal dynamics of T cells, B cells, and NK cells in peripheral and central compartments after stroke, specifically examining their migration across the cerebral vascular endothelium and their role in reshaping the brain’s immune landscape. The transient middle cerebral artery occlusion (tMCAO) model was employed. Cell adoptive transfer or antibody blockade was administrated. Integrated with bioinformatics analysis of single-cell RNA sequencing data, the neurobehavioral assessment, fluorescent-activated cell sorting (FACS), flow cytometry, RT-PCR, immunofluorescence staining and laser confocal microscopy analysis, multicolor immunohistochemistry (mIHC), and tissue loss volume measurement were performed. T, B, and NK cells exhibited divergent temporal dynamics in peripheral blood and brain infiltration following cerebral ischemia. Specifically, T cells, particularly NKT and regulatory T cells (Tregs), expanded during the subacute phase, orchestrating inflammation resolution and aiding stroke recovery. B cells mounted a rapid response, marked by a significant infiltration in the acute phase, with a distinct subset identified to facilitate white matter repair. NK cells were activated in the late phase, contributing to the reduction of brain tissue loss via immunomodulatory mechanisms. Remodeling of the cerebral vasculature post-stroke was also identified, highlighting reactive endothelial venule (REV) as a key portal for the trafficking of T, B, and NK cells into the brain parenchyma. The dynamic equilibrium of lymphocytes is a critical, stage-specific regulator of the post-stroke microenvironment, targeting REV to modulate their infiltration may constitute a key therapeutic avenue for advancing immunotherapy in ischemic stroke. The online version contains supplementary material available at 10.1186/s12987-026-00794-6.
With the progression of late-onset Alzheimer disease (LOAD), there is a dysregulation and then a breakdown of the blood-brain barrier (BBB). An important pathological feature in the brains of patients is the accumulation of amyloid beta (Aβ) peptides. Their aggregation leads to the formation of particularly harmful Aβ oligomers (Aβ-O). Unfortunately, our understanding of changes in the blood-brain barrier, particularly with regard to the effects of Aβ-O, is still very limited. This study investigated a LOAD-specific and induced pluripotent stem cell (hiPSC)-based in vitro model of the BBB for disease mechanisms and validated the findings in two independent laboratories. This study also investigated Aβ transport across the BBB. Furthermore, obtained in vitro findings were confirmed in the cerebrospinal fluid proteome of a LOAD patient cohort. Control and LOAD hiPSCs exhibited comparable efficiency in forming brain capillary endothelial-like cells (BCECs). Although transendothelial electrical resistance (TEER) assessments indicated no significant differences in barrier tightness between LOAD and control BCECs, high-throughput multiplex qPCR analysis revealed subtle alterations in barrier integrity. This included changes in various barrier markers, such as mucins (MUC1, MUC20), aquaporins (AQP5, AQP10), junctional transcripts (CLDNs, TJP1, OCLN), and receptors (LRP1, INSR, LSR), which were confirmed in LOAD patients. High-content imaging and flow cytometry indicated reduced cadherin 5 (CDH5) levels in LOAD BCECs. Importantly, the results also highlighted a difference in the transport of Aβ-O across the BBB. This model demonstrates a LOAD-relevant phenotype with decreased Aβ transport and alterations in key transcripts and could thus serve for future translational studies to rescue pathogenic phenotypes. [Image: see text] The online version contains supplementary material available at 10.1186/s12987-025-00753-7.
Ventricular shunt catheters are the standard of care for managing hydrocephalus, yet 40% of implanted catheters fail within two years of implantation, mainly due to proximal catheter obstruction. Prior literature suggests that cerebrospinal fluid (CSF) preferentially enters the ventricular catheter’s holes farthest from the catheter tip, where obstruction has been thought to frequently occur. However, previous studies have relied on simplified geometries and idealized catheter positioning, limiting the ability to capture patient-specific anatomical variability. Segmental flow distribution in ventricular catheters was analyzed using ventricular models derived from pediatric patients with hydrocephalus. Two pediatric shunted patients and one pediatric externalized patient with varied sizes, ranging from enlarged (FOHR: 0.45), moderate (0.30), and small (0.29), were segmented from MRI scans and reconstructed into closed-volume 3D models. We simulated the insertion of a 4-row, 4-hole ventricular catheter using frontal, parietal, and occipital surgical approaches. A choroid plexus mimic modeled from confocal microscopy was attached to the ventricular system for inlet flow, and a constant flow rate of 0.35 mL/min was applied to simulate CSF flow inside the ventricular domain. Steady-state, laminar 3D computational fluid dynamics simulations were performed to quantify mass flow into each catheter drainage segment. Occipital placement consistently concentrated flow (> 81%) in the fourth (most distal) segment across all ventricular sizes. In contrast, frontal and parietal approaches showed variable flow distributions dependent on ventricular morphology. In the moderate ventricle, parenchymal contact obstructed distal holes, redirecting flow proximally (Segments 1–2: >99%). Segmental flow distribution was influenced by drainage hole patency, ventricular morphology, and catheter orientation. Altogether, these data suggest that anatomical or approach dependent shifts result in specific tissue contact changes and therefore change in flow distribution. A larger sample scale study is needed to understand variability resulting from heterogeneous catheter placement in varying ventricular morphologies. The online version contains supplementary material available at 10.1186/s12987-026-00786-6.
Mechanisms guiding the induction of blood-brain barrier (BBB) properties in central nervous system (CNS) endothelial cells during human development are incompletely understood. For example, there is a limited understanding of signaling pathways that influence the unique property of low vesicular endocytosis and transcytosis in brain microvascular endothelial cells (BMECs) relative to peripheral endothelial cells. Mouse studies suggest the importance of BBB-relevant developmental pathways, including Wnt and Notch signaling, for the induction of this BBB feature in developing BMECs. To explore induction of reduced vesicular endocytosis and transcytosis in human in vitro model of the BBB, we used human pluripotent stem cell (hPSC)-derived endothelial progenitor cells (EPCs) in which Wnt/β-catenin signaling was activated to generate hPSC-derived CNS-like ECs (hPSC-CECs). We assessed the effects of Notch signaling through overexpression of the Notch1 receptor intracellular domain (N1ICD). N1ICD overexpression in hPSC-CECs resulted in upregulation of GLUT-1, a BBB-enriched glucose transporter, and decreased expression of both PLVAP and caveolin-1, two vesicular endocytosis-associated proteins. The combination of Wnt/β-catenin activation and N1ICD overexpression resulted in fewer vesicles and reduced albumin uptake. These findings indicate that Notch signaling reduces vesicular endocytosis and transcytosis in a human model of the developing BBB and contribute to our understanding of how Notch signaling induces these specific BBB properties in this model of human CNS EC development. The online version contains supplementary material available at 10.1186/s12987-025-00754-6.
Cerebrospinal fluid (CSF) over-drainage is an unintended sequela of using a shunt valve to treat hydrocephalus, and a second valve may be added in series to better control drainage. In this study, we tested whether the opening pressures of two valves in series-Codman Hakim ball valve (HBV) and the Miethke M.blueⓇ adjustable gravitational valve (AGV)-can be treated as being additive using a novel in vitro benchtop model. Six dual-valve circuits were tested in triplicates for different combinations of HBV and AGV settings adding up to a summed opening pressure (OP) of either 20 or 40 cmH2O. This "theoretical" sum was based on the addition of the nominal OP settings as stated by the manufacturer. Kruskal-Wallis tests and linear mixed-effect models were employed to analyze the flow rates for HBV-AGV settings with the same theoretical OP. Kruskal-Wallis analysis demonstrated that HBV-AGV settings for each circuit, at a theoretical OP of 20 and 40 cmH2O, showed statistically significant differences in flow rate. In linear mixed-effects models, as HBV OP was increased and composed a larger fraction of the theoretical OP, the flow rate decreased (effect size: -1.567 × 10- 3 mL/min/cmH2O, standard error: 1.663 × 10- 4 mL/min/cmH2O, p < 2 × 10- 16 at 20 cmH2O, and effect size: -3.735 × 10- 3 mL/min/cmH2O, standard error: 1.475 × 10- 4 mL/min/cmH2O, p < 2 × 10- 16 at 40 cmH2O, respectively). An analysis of the flow data demonstrated statistically significant differences between combinations of HBV-AGV settings adding up to the same theoretical opening pressure. These findings suggest that in practice, the nominal OPs of Hakim and M.blueⓇ valves may not be additive. Clinicians should be aware that drainage patterns can change when adjusting OPs of HakimⓇ and M.blueⓇ valves under the assumption that HBV-AGV settings with the same summed theoretical OP behave alike.
Dysregulation of cerebrospinal fluid (CSF) volume results in hydrocephalus, a disease that leads to over 30,000 surgical shunt-valve implantations annually in the US. These shunt-valves require a trial-and-error process to determine optimal settings for each individual, sometimes resulting in implantation of multiple valves in series. This work sought to evaluate two mathematical models of the relationship between valve opening pressure settings in series and resultant drainage using a benchtop system to aid clinicians in determination of optimal shunt-valve settings. A gravity-driven in-vitro flow system at 37 °C with a simulated ICP of 22 cmH2O + 60 cmH2O from valve to simulated peritoneal cavity was built. Differential pressure and gravitational valves were tested in isolation and series at various settings. The relationship between flow rate and the pressure drop across a valve is expressed with a valve coefficient. Results of isolated valve trials were used to calculated valve coefficients for each valve, which were then used to calculate combined valve coefficients to predict flowrate of valves in series. Flowrate predictions were compared to experimental results to evaluate each mathematical model presented here. In isolation, differential pressure and gravitational valves had low intra- and inter-valve variability (p > 0.05). Valves in series had highly variable flowrates across trials and sets of valves in both supine and upright positions (p < 0.05). Using calculated combined valve coefficients to predict flowrates of valves in series, the average percent error was 15 ± 7% and 23 ± 18% in the supine and upright positions, respectively. In all, neither of the two models outperformed the other and both were insufficient to properly characterize the relationship between drainage and opening pressures of valves in series. These results indicate low flowrate variability of isolated valves but high variability of valves placed in series. Without a consistent model from which opening pressure setting of valves in series can be determined, physicians must rely on a trial-and-error method in optimal opening pressure determination which directly impacts patient outcomes. These findings underscore the difficulties faced by physicians in determination of optimal valve settings for shunted patients.
The brain extracellular space (ECS) is a convoluted compartment of nano- to microscale interconnected ducts filled with interstitial fluid and lined by neural cell membranes. A key step in signaling between neural cells is diffusion through the ECS of transmitter molecules released from point sources distributed throughout the parenchyma. Yet, signaling is generally considered solely from the stance of cellular properties, while disregarding ECS sojourn time and putative signal modulation at this phase. Where ECS diffusion is considered, it is commonly done based on volume-averaging techniques blind to individual signaling events or actual ECS structural geometry. This has precluded knowledge on how specific ECS geometries may impact diffusion and modulate signaling arising from individual transmitter release events. We hypothesized that ECS geometry can shape local diffusion gradients resulting from individual point source release events and thereby tune signaling, and we further propose that this modulation can impart non-random functionality. To access the scale of individual transmitter release events and true ECS geometries, we used super-resolution STED microscopy to image the ECS in the hippocampal CA1 stratum radiatum of live mouse brain slices. We then developed a computational diffusion model, DifFlux, based on super-resolved images of hippocampal ECS and applied this along single molecule Monte Carlo diffusion simulations. Our approach allows us to simulate diffusion of molecules of our choosing in actual live ECS geometries. We observed local anomalous and anisotropic diffusion imposed by ECS geometry, whereby diffusion along larger structures was more directional than in denser neuropil of finer cellular structures. Further, we identified that the perisynaptic ECS geometries around respective glutamatergic and GABAergic synapses imposed distinct functional advantages, shedding light on the longstanding conundrum of why glutamatergic and GABAergic synapses are so conspicuously morphologically different. Our modelling broadly identifies ECS structure as a direct modulator of extrasynaptic signaling that can operate in parallel to conventional regulation mechanisms. This ultimately provides a metabolic and computational advantage to the brain. The online version contains supplementary material available at 10.1186/s12987-026-00797-3.
Altered glymphatic function is observed for many neurological diseases. Glioma, one of the most common brain cancers, is known to have altered fluid dynamics in terms of edema and blood-brain barrier breakdown, both features potentially impacting the glymphatic function. To study glioma and its fluid dynamics, we propose a flexible mathematical model, including the tumor, the peri-tumoral edema and the healthy tissue. From a mechanical point of view, we consider the brain as a multicompartment porous medium and model both the fluid movement and the clearance of solutes within the brain. Our results indicate that the impairment of the glymphatic system due to glioma growth is two-fold. First, edema resulting from the leakage of fluid at the blood-brain barrier and/or the occlusion of the interstitial fluid exit routes (notably the perivascular spaces) due to migratory tumor cells result in a slight localized increase of pressure, consequently impairing negatively glymphatic clearance. Second, local changes of porosity (i.e. the volume fraction of certain compartments such as perivascular or extracellular spaces), result in a disruption of the transport of solutes in the brain. Our results indicate that an effect similar to the enhanced permeability and retention is obtained using biologically relevant changes of parameter values of our model. Our mathematical model is the first step towards a digital twin for drug or contrast product delivery within the cerebro-spinal fluid directly (e.g. from intrathecal injection) for patients suffering from gliomas.
Individuals with both Alzheimer’s Disease (AD) and cerebrospinal fluid (CSF) dynamics disorders, such as idiopathic normal-pressure hydrocephalus (iNPH), exhibit reduced CSF Aβ42 levels, complicating the interpretation of AD biomarkers. However, the influence of CSF dynamics on blood-based AD biomarkers remains unclear. This study investigated whether immunoprecipitation mass spectrometry (IP-MS)-based AD plasma biomarkers were associated with disproportionately enlarged subarachnoid space hydrocephalus (DESH) on MRI. This retrospective, cross-sectional study included 509 participants from the Mayo Clinic Study of Aging who underwent MRI, [¹¹C] Pittsburgh compound B (Aβ) PET imaging, and plasma assessments of phosphorylated tau 217 (p-Tau 217), Aβ40, and Aβ42. Age-adjusted logistic regression analyses were conducted to evaluate whether abnormal levels of p-Tau 217, %p-Tau 217 (ratio of phosphorylated to non-phosphorylated tau × 100%), Aβ42, or the Aβ42/40 ratio were associated with DESH, as identified by a previously validated automated algorithm. In addition, combinations of binary amyloid PET and plasma biomarkers to create 4 categories were tested for an association with DESH as a sensitivity analysis. Average age was 72.2 (8.8), with 54% male participants, and 30% were APOE4 carriers. Thirty-two participants (6.3%) were classified as having DESH. Abnormal (low) plasma Aβ42 (OR: 2.56 [95% CI: 1.19, 5.74]) and abnormal (high) p-tau217 (OR: 2.74 [95% CI: 1.09, 7.85]) were associated with DESH, but not the respective ratios of Aβ42/40 (OR: 1.95 [95% CI: 0.94, 4.17]) or %p-tau (OR: 1.82 [95% CI: 0.74, 4.20]). Analysis of the combined variables (amyloid PET and plasma biomarkers) demonstrated that an association with DESH was observed only in participants who were abnormal on both measures, compared with participants who were negative on both plasma biomarkers and PET. This association was attenuated when biomarker ratios were used. The observed association between abnormal plasma biomarker levels and DESH aligns with previous CSF studies and can be mitigated by utilizing the biomarker ratios. However, further validation is needed in populations with more severe CSF flow abnormalities to better understand the role of plasma biomarkers in this population. The online version contains supplementary material available at 10.1186/s12987-026-00789-3.
The meninges form a highly specialized barrier and immune interface that protects the central nervous system (CNS), regulates cerebrospinal fluid (CSF) dynamics, and coordinates communication between the CNS and the periphery. Each layer—dura mater, arachnoid mater and pia mater—possesses distinct structural, vascular and immunological features that collectively shape CNS homeostasis. A broad range of anatomical and molecular studies has revealed that meningeal compartments are far more heterogeneous and functionally complex than traditionally recognized, particularly with respect to their barrier architecture and immune interactions. In this review, we summarize current knowledge of meningeal structure and function, with a focus on barrier properties and immune-cell trafficking. We further discuss how meningeal dysfunction contributes to pathology in bacterial meningitis, multiple sclerosis and Alzheimer’s disease. Emerging evidence highlights the meninges as an active neuroimmune organ rather than a passive covering, critically influencing inflammation, solute clearance and disease progression. Understanding these mechanisms may open new therapeutic avenues targeting meningeal pathways across neurological disorders.
Cerebrospinal fluid (CSF) is vital to maintain brain homeostasis by facilitating waste removal, nutrient transport, and regulation of intracranial pressure. The choroid plexus is generally assigned as the primary source of CSF secretion, but it remains unresolved to what extent water produced during cellular metabolism in the central nervous system contributes to the CSF pool. Brain and spinal cord parenchyma volumes in rats (n = 6) and humans (n = 6) were assessed with volumetric magnetic resonance imaging. Segmentation techniques were applied to obtain the mass of metabolically active tissue. The volume of H2O produced as a biproduct of glucose oxidation was estimated based on the established oxygen consumption of these tissues. We estimate a metabolic H2O production within the brain and spinal cord tissue of 0.078 µL/min in rats and 28.5 µL/min in humans, which amounts to 1–8% of the volume of CSF secreted into the ventricular compartment. Our results demonstrate a minor contribution of the metabolically produced water to the collective CSF pool in rats and humans, notwithstanding that part of the estimated H2O production is likely absorbed by other H2O-utilizing enzymatic processes. Our findings thus support the longstanding view that the choroid plexus is a key contributor to CSF secretion, with potential implications for future understanding and management of pressure-related brain disorders. The online version contains supplementary material available at 10.1186/s12987-026-00769-7.