N. Joan Abbott

The organization of the cerebral ganglion in the shore crab, Carcinus maenas

SummaryThe organization of the cerebral ganglion of the shore crab Carcinus maenas, is investigated by conventional histological and electronmicroscopic techniques. This study forms part of a comprehensive survey of the blood-brain interface, particularly interesting in this group, as decapod Crustacea are unusual among invertebrates in possessing an intracerebral blood supply. Apart from the intracerebral blood vessels, tissue organization is closely similar to that observed in insect central neural ganglia. The ganglion is surrounded by the neural lamella, an acellular connective tissue sheath, probably containing mucopolysaccharide and collagen. A layer of specialised glia, the perineurium, immediately underlies the neural lamella, and appears to contribute to its formation. Large glia occupying a conspicuous cortical zone below the perineurium may be involved in glycogen metabolism and storage. Further morphologically distinct glial types are observed associated with neurones and blood vessels, but all neuroglia within the ganglion are probably of common origin. Neurone cell bodies are generally situated peripherally in groups, and send axons into neuropil (synaptic) areas in the ganglion core. Large lacunae in the cortical region and narrower 20 nm clefts deeper in the ganglion, constitute the interstitial space, and contain deposits of fibrillar material. Possible physiological implications are discussed.

Access of ferritin to the interstitial space of Carcinus brain from intracerebral blood vessels.

Abstract The penetration of ferritin across the blood-brain interface in the shore crab, Carcinus, is investigated in an EM study. The bulk of cerebral vascular spaces have no mesodermal endothelial lining, but are surrounded by neuroglia. It is shown that neither glial basement membrane, junctions between perivascular glia, nor material in the extracellular space, form an absolute barrier to ferritin movement. Ferritin is observed in expanded lacunae and narrow (200 A) clefts adjacent to axons and glia several μ from the vascular lumen.

Fine-structural investigation of rat brain microvascular endothelial cells: Tight junctions and vesicular structures in freshly isolated and cultured preparations

SummaryA comparison was made between endothelial cells in freshly-isolated rat brain microvessels, and following culture of the cells for 1–10 days during growth to confluence. Attention focused on tight junctions and vesicular structures, as seen in thin sections and freeze-fracture replicas. Freshly-isolated vessels had an abnormal appearance, with a profusion of luminal microvillar processes, and extensive cytoplasmic vacuolation. There were numerous vesicular profiles, reaching a density of ∼60 μm−2, and with a large proportion open to the surface, as shown by labelling with cationized ferritin at 4 °C for 5 min. Junctional zones were relatively loosely organized, with evidence for some cell: cell separation, as well as some residual tight junctional sites withinzonula adhaerens junctions. In freeze-fracture replicas, Junctional strands showed segments of tightly packed intramembrane particles, generally on the P face. After 1 day in culture, the cells appeared more normal, with no vacuolation or luminal processes. Vesicles were still numerous, some associated with junctional zones, while tight junctions were relatively sparse; freeze-fracture showed some incomplete tight junctional strands, with some of the intramembrane particles fracturing onto the E face. The double offset fibrillar nature of the strands could occasionally be seen. Cells cultured for 4 and 10 days showed a progressive increase in the completeness of the junctional zone, with more tight junctional contacts within the length of theadhaerens junction, and an aggregation of microfilaments in the underlying cytoplasm. The number of vesicular profiles declined, and they were progressively excluded from the junctional zone. These observations have relevance for studies on the physiology of the brain endotheliumin vitro, and for comparisons with thein vivo condition.

Mechanisms for the Passive Regulation of Extracellular K+ in the Central Nervous System: The Implications of Invertebrate Studies

Homeostasis of the brain extracellular K concentration is a necessary prerequisite for neuronal functioning and synaptic integration3,7. There is considerable evidence for such regulation in vertebrate brain, and when the interstitial K concentration is maintained constant in the face of chronically altered K levels in blood, an active, energy-dependent mechanism is suggested7. However, it is not known to what extent the brain can maintain a degree of internal homeostasis by purely passive ‘K buffering’ systems. In this paper we shall examine the evidence for active and passive regulation of K in the brain microenvironment and discuss some results from a crustacean preparation which indicate that passive K buffering may occur. Possible involvement of glia, and alternative mechanisms, will be discussed. As astrocytes have been suggested as a major site for K buffering in the vertebrate brain, evidence for a homeostatic function of astrocytes will be considered.

The Na‐K ATPase of the Blood‐Brain Barrier: A Microelectrode Studya

[K+] in the interstitial fluid of the brain of higher vertebrates is kept remarkably constant, despite fluctuations in plasma concentration and the activity of neurons. This is due in part to an efflux system that transports K+ from brain to blood,’ located in the endothelial blood-brain barrier (BBB)? Although ion regulation at the BBB has been studied in whole brain and isolated microvessels, little has been done on single vessels in situ. Crone and Olesen’ have shown that the microvessels at the frog brain surface have a high resistance (2,000 ncm’), and therefore tight endothelium. Crone4 has described diffusion potentials due to passive ion fluxes across these vessel walls. We have extended the method to study Na+-K+ transport at the BBB, in individual microvessels in situ, with circulation intact. The transendothelial potential difference, +, was studied with 3 M KC1 glass microelectrodes in brain surface microvessels (15-50 pm diameter) of the frogs Rana temporaria and R. pipiens anesthetized in 0.2% MS-222 (Sigma). Junction potentials were minimized with salineand KC1-agar bridges? Mean control potential +c in frog Ringer (2 rnM [K+]) was 4.5 * 0.24 mV, lumen negative (145 vessels in 33 frogs). Changing the [K+] of the bathing solution (CSF side), in the range 0100 mM [K+] caused changes in I) that could be separated into a component Jld due to passive dSusion,4 and an active component +*, blocked by Mouabain, which we attribute to an electrogenic Na+-K+ pump on the abluminal membrane of the endothelial BBB.6 +a was reduced or absent in unhealthy frogs. Reduction of I)a by ouabain was maximal at lo-’ M; lo-’ M had no effect, and lop4 M ouabain caused a slight increase in

The organization of the cerebral ganglion in the shore crab, Carcinus maenas. II. The relation of intracerebral blood vessels to other brain elements.

a, possibly by stimulating the pump; or by altering barrier tightness or K+ permeability. A five-minute exposure to lop3 M ouabain was used to assess Na+-K+-ATPase kinetics, since longer exposures caused animal deterioration and increased barrier permeability. +a/+%m (a measure of Na+-K+ pump rate) showed a sigmoidal dependence on [K+], saturating around 40 m M [K’], and with a half-saturation concentration of 6 mM [K+]. A plot of l/<pump rate X [K+] versus [K+] gave a straight line, regression coefficient 0.999, typical of a Na-K ATPase requiring binding of two K+ ions for pump activation. The K, for the K+ sites was 2.6 mM. The frog BBB Na+-K+ ATPase thus saturates at a higher [K+] than does the frog choroid plexus enzyme