Shoko Morita

Different vascular permeability between the sensory and secretory circumventricular organs of adult mouse brain

The blood-brain barrier (BBB) prevents free access of circulating molecules to the brain and maintains a specialized brain environment to protect the brain from blood-derived bioactive and toxic molecules; however, the circumventricular organs (CVOs) have fenestrated vasculature. The fenestrated vasculature in the sensory CVOs, including the organum vasculosum of lamina terminalis (OVLT), subfornical organ (SFO) and area postrema (AP), allows neurons and astrocytes to sense a variety of plasma molecules and convey their information into other brain regions and the vasculature in the secretory CVOs, including median eminence (ME) and neurohypophysis (NH), permits neuronal terminals to secrete many peptides into the blood stream. The present study showed that vascular permeability of low-molecular-mass tracers such as fluorescein isothiocyanate (FITC) and Evans Blue was higher in the secretory CVOs and kidney as compared with that in the sensory CVOs. On the other hand, vascular permeability of high-molecular-mass tracers such as FITC-labeled bovine serum albumin and Dextran 70,000 was lower in the CVOs as compared with that in the kidney. Prominent vascular permeability of low- and high-molecular-mass tracers was also observed in the arcuate nucleus. These data demonstrate that vascular permeability for low-molecular-mass molecules is higher in the secretory CVOs as compared with that in the sensory CVOs, possibly for large secretion of peptides to the blood stream. Moreover, vascular permeability for high-molecular-mass tracers in the CVOs is smaller than that of the kidney, indicating that the CVOs are not totally without a BBB.

Astrocytic TRPV1 ion channels detect blood-borne signals in the sensory circumventricular organs of adult mouse brains

The circumventricular organs (CVOs), including the organum vasculosum of the lamina terminalis (OVLT), subfornical organ (SFO), and area postrema (AP) sense a variety of blood‐borne molecules because they lack typical blood‐brain barrier. Though a few signaling pathways are known, it is not known how endogenous ligands for transient receptor potential vanilloid receptor 1 ion channel (TRPV1) are sensed in the CVOs. In this study, we aimed to examine whether or not astrocytic TRPV1 senses directly blood‐borne molecules in the OVLT, SFO, and AP of adult mice. The reverse transcription‐polymerase chain reaction and Western analysis revealed the expression of TRPV1 in the CVOs. Confocal microscopic immunohistochemistry further showed that TRPV1 was localized prominently at thick cellular processes of astrocytes rather than fine cellular processes and cell bodies. TRPV1‐expressing cellular processes of astrocytes surrounded the vasculature to constitute dense networks. The expression of TRPV1 was also found at neuronal dendrites but not somata in the CVOs. The intravenous administration of a TRPV1 agonist resiniferatoxin (RTX) prominently induced Fos expression at astrocytes in the OVLT, SFO, and AP and neurons in adjacent related nuclei of the median preoptic nuclei (MnPO) and nucleus of the solitary tract (Sol) of wild‐type but not TRPV1‐knockout mice. The intracerebroventricular infusion of RTX induced Fos expression at both astrocytes and neurons in the CVOs, MnPO, and Sol. Thus, this study demonstrates that blood‐borne molecules are sensed directly by astrocytic TRPV1 of the CVOs in adult mammalians.

Expression of chondroitin sulfate proteoglycans in barrel field of mouse and rat somatosensory cortex

Chondroitin sulfate proteoglycans (CSPGs) consist of chondroitin sulfate (CS) glycosaminoglycans (GAGs) and core protein and regulate the migration, axonal outgrowth, and synaptogenesis in mammalian brains. In the present study, we investigated the localization of CSPGs, the effects of sensory deprivation on the density of perineuronal nets (PNNs), and the effects of chondroitinase ABC (Chase) on the formation of barrel structures in the posterior medial barrel subfield (PMBSF). In developing mouse and rat brains, the immunoreactivity of chondroitin-6-sulfate containing proteoglycan (CS-6-PG), phosphacan, and neurocan was stronger at barrel septa as compared with barrel hollows and surrounding cortex, while the labeling of Wisteria floribunda agglutinin (WFA) was observed at barrel hollows. In adult brains, CS-6-PG-immunoreactive and WFA-labeled PNNs were observed mainly at barrel hollows of mouse, but they were seen chiefly at barrel septa of rats. Sensory deprivation of facial vibrissae reduced the number of WFA-labeled PNNs at barrel hollows but not at barrel septa. Intracerebral injection of Chase did not affect the formation of barrel structures in the PMBSF. These data indicate species-dependent heterogeneity of CSPG expression and activity-dependent formation of PNNs in the PMBSF, but CS GAGs have no crucial function in constructing the barrel structures during early postnatal development.

A new method for visualization of endothelial cells and extravascular leakage in adult mouse brain using fluorescein isothiocyanate.

We described a new method for the visualization of vasculature and endothelial cells and the assessment of extravascular leakage in adult mouse brain by using fluorescein isothiocyanate (FITC), or a reactive fluorescent dye. FITC is the fluorescein derivative that reacts covalently with amine groups at alkaline pH. In this method, strong fluorescence of FITC was seen at vasculature throughout the brain and spinal cord, when mice received intracardiac perfusion with FITC-containing saline at pH 7.0 followed by paraformaldehyde (PFA) fixative at pH 8.0. The fluorescence of FITC was faint when animals were fixed with PFA fixative at pH 7.0 after the perfusion of FITC-containing saline at pH 7.0. The fluorescence of FITC was not detected when mice was fixed with PFA fixative before the perfusion of FITC-containing saline. Double labeling immunohistochemistry using an endothelial cell marker CD31 or a pericyte marker desmin revealed that FITC was accumulated at nuclei of endothelial cells but not at those of pericytes. Extravascular leakage of FITC was prominent in the area postrema or a brain region of the circumventricular organs that lacks the blood-brain barrier. Moreover, strong extravascular leakage of FITC was detected at damaged sites of the cerebral cortex with cryoinjury. Thus, FITC method is useful technique for examining the architecture of brain vasculature and endothelial cells and the assessment of extravascular leakage in adult rodents. Moreover, FITC binds covalently to cellular components, so that makes it possible to perform double labeling immunohistochemistry and long-term storage of the preparation.

Tissue plasminogen activator and plasminogen are critical for osmotic homeostasis by regulating vasopressin secretion.

Systemic osmotic homeostasis is regulated mainly by neuroendocrine system of arginine‐vasopressin (AVP) in mammalians. In the present study, we demonstrated that the immunoreactivity of tissue plasminogen activator (tPA) was observed specifically at neurosecretory granules of AVP‐positive magnocellular terminals and that of plasminogen was seen at astrocytes in the neurohypophysis (NH). Both tPA and plasminogen knockout (KO) mice revealed higher plasma osmolarity upon water deprivation, a chronic osmotic stimulation, as compared with their wild‐type (WT) animals, indicating abnormal osmotic control in these KO mice. tPA KO mice but not plasminogen ones revealed lower ability in secreting AVP into the blood circulation upon an acute osmotic stimulation. Both tPA and plasminogen KO animals showed lower ability in secreting AVP into the blood circulation upon a chronic osmotic stimulation. The recombinant tPA was able to promote the release of AVP from isolated NH. Chronic osmotic stimulation decreased the laminin expression level of neurohypophysial microvessel in WT mice but not in plasminogen KO ones. We suggest that AVP secretion is critically regulated by tPA‐dependent facilitation of AVP release from terminals and plasminogen‐dependent increase of AVP permeability across microvessels possibly via laminin degradation.

Astrocytic TLR4 expression and LPS-induced nuclear translocation of STAT3 in the sensory circumventricular organs of adult mouse brain.

The sensory circumventricular organs (CVOs) comprise the organum vasculosum of the lamina terminalis (OVLT), subfornical organ (SFO), and area postrema (AP) and lack the blood-brain barrier. The expression of Toll-like receptor 4 (TLR4) was seen at astrocytes throughout the sensory CVOs and at microglia in the AP and solitary nucleus around the central canal. The peripheral and central administration of lipopolysaccharide induced a similar pattern of nuclear translocation of STAT3. A microglia inhibitor minocycline largely suppressed lipopolysaccharide-induced astrocytic nuclear translocation of STAT3 in the OVLT and AP, but its effect was less in the SFO.

ACTIVITY-DEPENDENT REMODELING OF CHONDROITIN SULFATE PROTEOGLYCANS EXTRACELLULAR MATRIX IN THE HYPOTHALAMO- NEUROHYPOPHYSIAL SYSTEM

The hypothalamo-neurohypophysial system (HNS) consisting of arginine vasopressin (AVP) and oxytocin (OXT) magnocellular neurons shows the structural plasticity including the rearrangement of synapses, dendrites, and neurovascular contacts during chronic physiological stimulation. In this study, we examined the remodeling of chondroitin sulfate proteoglycans (CSPGs), main extracellular matrix (ECM), in the HNS after salt loading known as a chronic stimulation to cause the structural plasticity. In the supraoptic nucleus (SON), confocal microscopic observation revealed that the immunoreactivity of 6B4 proteoglycans (PG) was observed mainly at AVP-positive magnocellular neurons but that of neurocan was seen chiefly at OXT-positive magnocellular neurons. The immunoreactivity of phosphacan and aggrecan was seen at both AVP- and OXT-positive magnocellular neurons. Electron microscopic observation further showed that the immunoreactivity of phosphacan and neurocan was observed at astrocytic processes to surround somata, dendrites, and terminals, but not synaptic junctions. In the neurohypophysis (NH), the immunoreactivity of phosphacan, 6B4 PGs, and neurocan was observed at AVP-positive magnocellular terminals, but the reactivity of Wisteria floribunda agglutinin lectin was seen at OXT-positive ones. The immunoreactivity of versican was found at microvessel and that of aggrecan was not detected in the NH. Quantitative morphometrical analysis showed that the chronic physiological stimulation by 7-day salt loading decreased the level of 6B4 PGs in the SON and the level of phosphacan, 6B4 PGs, and neurocan in the NH. These results suggest that the extracellular microenvironment of CSPGs is different between AVP and OXT magnocellular neurons and activity-dependent remodeling of CSPGs could be involved in the structural plasticity of the HNS.

VEGF-dependent continuous angiogenesis in the median eminence of adult mice.

Brain vasculature forms the blood–brain barrier (BBB) that restricts the movement of molecules between the brain and blood, but the capillary of the median eminence (ME) lacks the BBB for secretion of adenohypophysial hormone‐releasing peptides. In the present study, we aimed to elucidate whether continuous angiogenesis occurs in the ME of adult mice. By using a mitotic marker, bromodeoxyuridine (BrdU), we demonstrated that new endothelial cells were born continuously in the ME of adults. Prominent expression of NG2, platelet‐derived growth factor receptor B (PDGFRB), and delta‐like ligand 4 was observed at pericytes of adults, although the expression of these angiogenesis‐associated proteins has been shown to be at low or trace levels in adult mature capillary. In addition, vascular endothelial growth factor (VEGF), a key regulator of angiogenesis, was expressed highly in the nervous parenchyma of the ME. Expression of VEGF receptor 2 (VEGFR2) was observed at endothelial cells in the external zone and at somatodendrites in the internal zone. Finally, a VEGFR‐ and PDGFR‐associated tyrosine kinase inhibitor, SU11248, significantly decreased the number of BrdU‐positive proliferating endothelial cells and parenchyma cells. In conclusion, the present study demonstrates VEGF‐dependent continuous angiogenesis in the ME of adult mouse brains under normal conditions, which provides new insight into our understanding of neurosecretion in the ME.

Accessibility of low-molecular-mass molecules to the median eminence and arcuate hypothalamic nucleus of adult mouse

Blood‐derived molecules are able to access to the median eminence (ME) and arcuate hypothalamic nucleus (Arc) due to the lack of the blood–brain barrier. In the present study, we examined the accessibility of low‐molecular‐mass (LMM) molecules into parenchyma in the ME and Arc of adult mice by administration of Dextran 3000 (Dex3k), Dex10k, Evans blue (EB) and fluorescein isothiocyanate (FITC). In the external zone of the ME, the fluorescence of Dex3k, EB and FITC tracers generated an intensity gradient from fenestrated capillary, but that of Dex10k was detected only between the inner and outer basement membrane of pericapillary space. The fluorescence of FITC in the external zone of the ME was closely associated with axonal terminals and surrounded by cellular processes of tanycytes‐like cells and astrocytes. In the ependymal/internal zone of the ME and Arc, the fluorescence of all LMM tracers was seen at tanycytes‐like cells and neurons. The fluorescence of EB and FITC in these regions was not detected when brains were fixed during or before the administration of tracers. The inhomogeneity of accessibility for fluorescent tracers depended on routes for tracer administration. Thus, the present study indicates that the accessibility of LMM blood‐derived molecules to parenchyma depends on fenestration of the capillary in the external zone of the ME and active transport of ependymal cells in the ependymal/internal zone of the ME and Arc. Copyright

Localization of plasminogen in mouse hippocampus, cerebral cortex, and hypothalamus

Although the tissue plasminogen activator/plasminogen system contributes to numerous brain functions, such as learning, memory, and anxiety behavior, little attention has as yet been given to the localization of plasminogen in the brain. We have investigated the localization of plasminogen in the adult mouse brain by using immunohistochemistry. In the hippocampus, plasminogen immunoreactivity was seen in the pyramidal cell layer as numerous punctate structures in neuronal somata. An electron-microscopic study further demonstrated that the plasminogen-immunoreactive punctate structures represented secretory vesicles and/or vesicle clusters. In the cerebral cortex, plasminogen immunoreactivity was evident in the somata of the layer II/III and V neurons. A quantitative analysis revealed that parvalbumin (PV)-positive neurons had more plasminogen-immunoreactive puncta compared with those of PV-negative neurons in the hippocampus and cerebral cortex. Plasminogen immunoreactivity was present throughout the hypothalamus, being particularly prominent in the neuronal somata of the organum vasculosum laminae terminalis, ventromedial preoptic nucleus, supraoptic nucleus, subfornical organ, medial part of the paraventricular nucleus (PVN), posterior part of the PVN, and arcuate hypothalamic nucleus. Thus, plasminogen is highly expressed in specific populations of hippocampal, cortical, and hypothalamic neurons, and plasminogen-containing vesicles are mainly observed at neuronal somata.