O.P. Ottersen

Aquaporin-4 in the central nervous system: Cellular and subcellular distribution and coexpression with KIR4.1

Aquaporin-4 (AQP4) is the predominant water channel in the neuropil of the central nervous system. It is expressed primarily in astrocytes, but also occurs in ependymocytes and endothelial cells. A striking feature of AQP4 expression is its polarized distribution in brain astrocytes and retinal Muller cells. Thus, immunogold analyses have revealed an enrichment of AQP4 in endfeet membranes in contact with brain microvessels or subarachnoidal space and a low but significant concentration in non-endfeet membranes, including those astrocyte membranes that ensheath glutamate synapses. The subcellular compartmentation of AQP4 mimics that of the potassium channel Kir4.1, which is implicated in spatial buffering of K(+). We propose that AQP4 works in concert with Kir4.1 and the electrogenic bicarbonate transporter NBC and that water flux through AQP4 contributes to the activity dependent volume changes of the extracellular space. Such volume changes are important as they affect the extracellular solute concentrations and electrical fields, and hence neuronal excitability. We conclude that AQP4-mediated water flux represents an integral element of brain volume and ion homeostasis.

The expression of vesicular glutamate transporters VGLUT1 and VGLUT2 in neurochemically defined axonal populations in the rat spinal cord with emphasis on the dorsal horn

Two vesicular glutamate transporters, VGLUT1 and VGLUT2, have recently been identified, and it has been reported that they are expressed by largely nonoverlapping populations of glutamatergic neurons in the brain. We have used immunocytochemistry with antibodies against both transporters, together with markers for various populations of spinal neurons, in an attempt to identify glutamatergic interneurons in the dorsal horn of the mid‐lumbar spinal cord of the rat. The great majority (94–100%) of nonprimary axonal boutons that contained somatostatin, substance P or neurotensin, as well as 85% of those that contained enkephalin, were VGLUT2‐immunoreactive, which suggests that most dorsal horn neurons that synthesize these peptides are glutamatergic. In support of this, we found that most somatostatin‐ and enkephalin‐containing boutons (including somatostatin‐immunoreactive boutons that lacked calcitonin gene‐related peptide and were therefore probably derived from local interneurons) formed synapses at which AMPA receptors were present.

Quantification of immunogold labelling reveals enrichment of glutamate in mossy and parallel fibre terminals in cat cerebellum

The glutamate immunoreactivity of different cell populations was compared quantitatively in the cerebellar cortex of cat, using an antiserum raised against glutamate coupled to bovine serum albumin by glutaraldehyde. Neuronal and glial processes were identified on serial electron microscopic sections which were processed by a postembedding immunogold procedure. The surface density of colloidal gold particles was used for statistical comparison of the relative levels of glutamate in cell populations, or in different parts of the same population. The terminals of mossy and parallel fibres had significantly higher levels of glutamate immunoreactivity than Golgi cell terminals, granule cell dendritic digits, Purkinje cell dendrites or dendritic spines. Golgi cell terminals were identified by their position and GABA immunoreactivity as revealed by immunogold in serial sections. The dendritic digits of the putative glutamatergic granule cells had significantly higher glutamate immunoreactivity than did Purkinje cell dendrites and dendritic spines. Glial cell processes in the molecular layer had lower level of glutamate immunoreactivity than any of the neuronal processes. The results demonstrate that the highest levels of glutamate immunoreactivity occur in mossy and parallel fibre presynaptic terminals that are known to have an excitatory effect. This supports previous suggestions that glutamate may be a transmitter at these synapses. The measurement of the levels of putative amino acid transmitters in identified neuronal populations, or in different parts of the same population, could have wide applications in studies on the chemical neuroanatomy of the nervous system.

Metabolic compartmentation of glutamate and glutamine: Morphological evidence obtained by quantitative immunocytochemistry in rat cerebellum

An electron microscopic, double-labelling immunocytochemical procedure was used to assess the level of fixed glutamate and glutamine in different cell profiles in ultrathin sections of rat cerebellar cortex. The procedure was based on sequential immunolabelling with two rabbit antisera, using gold particles of different sizes as markers and formaldehyde vapour as a means to avoid interference between the two incubations. Model sections containing a series of known concentrations of the respective amino acids (aldehyde--fixed to rat brain protein) were incubated together with the tissue material. These revealed a close to linear relationship between gold particle density and antigen concentration throughout the range of biological relevance. The ratio between the density of the two categories of gold particles was calculated for the individual profile types. This ratio showed a 20-fold variation, with the highest glutamate/glutamine ratios obtained for putative excitatory terminals (terminals of parallel fibres in the outer part of the molecular layer, followed by mossy and climbing fibre boutons) and the lowest for glial cells (Bergmann glia, astrocytes in the granule cell layer, and oligodendrocytes). Granule cell bodies and dendrites, and cell bodies and processes of putative GABAergic cells (Purkinje, basket and Golgi cells) displayed intermediate ratios. The ratios corresponded to millimolar ratios (mM fixed glutamate/mM fixed glutamine) ranging from 4.5 to 0.2, tentatively assessed by adjusting for differences in labelling efficiency of the two antigens. Our results show that the compartmentation of glutamate and glutamine, an issue previously addressed mainly in the test tube, can be studied in morphologically intact preparations at a resolution that matches the complexity of CNS tissue. The data indicate that glutamate is effectively converted to glutamine in all categories of glial cells, and that glutamate synthesis prevails in each of the three types of excitatory terminals in the cerebellar cortex. Terminals of putative GABAergic cells form a distinct low glutamate/low glutamine compartment.

Postembedding light- and electron microscopic immunocytochemistry of amino acids: description of a new model system allowing identical conditions for specificity testing and tissue processing

SummarySpecificity testing should be performed under conditions identical to or closely similar to those of the immunocytochemical procedure. This paper describes a new model system that meets this requirement for postembedding immunocytochemistry of amino acids at the light- and electron microscopic levels. Test conjugates, obtained by reacting different amino acids with brain macromolecules in the presence of glutaraldehyde, were freeze-dried and embedded in an epoxy resin (Durcupan) exactly as for brain tissue. One section from each of the embedded amino acid conjugates and from a brain protein-glutaraldehyde conjugate (without amino acid) were piled on top of each other and embedded anew. Transverse semithin (0.5 μm) and ultrathin sections were cut through the stack. These test sections, in which all the different conjugates were represented, were then processed in the same drops of sera as the tissue sections to permit identical conditions for testing and immunocytochemistry. After immunogold labelling for electron microscopy, a quantitative expression of crossreactivity was obtained by computer-assisted calculation of gold particle densities over the different conjugates. The antisera used in the present study (glutamate anti-serum 13, taurine antiserum 20, and GABA antiserum 26) showed highly selective labelling of the respective amino acid conjugates and produced distinct labelling patterns in simultaneously processed cerebellar sections.

An atlas of glycine- and GABA-like immunoreactivity and colocalization in the cochlear nuclear complex of the guinea pig

SummaryThe distribution and colocalization of γ-aminobutyric acid (GABA)- and glycine-like immunoreactivity in the cochlear nuclear complex of the guinea pig have been studied to produce a light microscopic atlas. The method used was based on post-embedding immunocytochemistry in pairs of 0.5-μm-thick plastic sections treated with polyclonal antibodies against conjugated GABA and glycine respectively. Immunoreactive cells, presumably short axon neurones, predominated in the dorsal cochlear nucleus, with mostly single-GABA-labelled cells in the superficial layer, double-labelled in the middle, and single-glycine-labelled in the deep layers. A few large single-glycine-labelled cells, interpreted as commissural neurons, occurred in the ventral nucleus. Scattered double-labelled cells, probably Golgi cells, were seen in the granule cell domain. Immunolabelled puncta of all three staining categories occurred in large numbers throughout the complex, apposed to somata and in the neuropil, showing a differential distribution onto different types of neuron. Three immunolabelled tracts were noted: the tuberculoventral tract, the commissural acoustic stria, and the trapezoidal descending fibres. Most of the fibres in these tracts were single-labelled for glycine, although in the last mentioned tract single-GABA- and double-labelled fibres were also found. Some of the immunolabelled cell types described here are proposed as the origins of the similarly labelled puncta and fibres on the basis of known intrinsic connections.

Anchoring of aquaporin-4 in brain: Molecular mechanisms and implications for the physiology and pathophysiology of water transport

Astrocytes show an enrichment of aquaporin-4 (AQP4) in those parts of the plasma membrane that are apposed to pial or perivascular basal laminae. This observation begged the following questions: 1, What are the molecular mechanisms that are responsible for the site specific anchoring of AQP4? 2, What are the physiological and pathophysiological roles of the AQP4 pools at these specialized membrane domains? Recent studies suggest that the site specific anchoring depends on the dystrophin complex. Further, alpha-syntrophin (a member of the dystrophin complex) is required to maintain a polarized expression of AQP4 in the perivascular membranes. Hence transgenic mice deficient in alpha-syntrophin provided a model where the perivascular pool of AQP4 could be removed for assessment of its functional roles. Data suggest that the perivascular pool of AQP4 plays a role in edema formation and that this pool (through its serial coupling with the AQP4 pools in other astrocyte membranes) is involved in K(+) siphoning. In the cerebral cortex, the astrocyte membrane domain contacting the pial basal lamina differs from the perivascular membrane domain in regard to the mechanisms for AQP anchoring. Thus deletion of alpha-syntrophin causes only a 50% loss of AQP4 from the former membrane (compared with a 90% loss in the latter), pointing to the existence of additional anchoring proteins. We will also discuss the subcellular distribution and anchoring of AQP4 in the other cell types that express this protein: endothelial cells, ependymal cells, and the specialized astrocytes of the osmosensitive organs.

Glutamate‐like Immunoreactivity in Retinal Terminals of the Mouse Suprachiasmatic Nucleus

With a view to identifying the neurotransmitter content of retinal terminals within the mouse suprachiasmatic nucleus, a highly specific antiserum to glutaraldehyde‐coupled glutamate was used in a postembedding immunogold procedure at the ultrastructural level. Retinal terminals were identified by cholera toxin–horseradish peroxidase transported anterogradely from the retina and reacted with tetramethyl benzidine/tungstate/H2O2, or by their characteristically pale and distended mitochondria with irregular cristae. Controls included model ultrathin sections containing high concentrations of various amino acids. Alternate serial sections were labelled with anti‐glutamate and anti‐γ‐aminobutyric acid (GABA). Data were analysed by computer‐assisted image analysis. Density of glutamate labelling (gold particles per μm2) on whole retinal terminals was > 3 times higher than that on postsynaptic dendrites, and > 5 times higher than that on miscellaneous non‐retinal non‐glutamatergic terminals in the suprachiasmatic nucleus. The overall density of gold particles over retinal terminals was ∼ 3 times higher than that over GABAergic terminals, in which glutamate‐like immunoreactivity was mainly mitochondrial. Labelling of vesicles in retinal terminals was almost 5 times greater than the apparent labelling of vesicles in GABAergic terminals, underscoring the location of transmitter glutamate within synaptic vesicles in retinal terminals. In the retino‐recipient region of the suprachiasmatic nucleus there was also a small population of non‐retinal glutamatergic terminals. Their overall immunoreactivity was similar to or exceeded that of retinal terminals, but morphological features clearly distinguished between these two types of glutamate‐containing terminals. The present results indicate that the vast majority of retinal terminals may use glutamate as a transmitter, in keeping with electrophysiological and neuropharmacological data from other sources. The possibility of cotransmitters within retinal terminals, suggested by the presence of dense‐core vesicles among the glutamate‐containing synaptic vesicles, has still to be addressed.

GABA, glycine, aspartate, glutamate and taurine in the vestibular nuclei: an immunocytochemical investigation in the cat.

SummaryThe distributions of five amino acids with well-established neuroexcitatory or neuroinhibitory properties were investigated in the feline vestibular complex. Consecutive semithin sections of plastic-embedded tissue were incubated with antisera raised against protein-glutaraldehyde conjugates of GABA, glycine, aspartate, glutamate and taurine. This approach allowed us to study the relative densities of the different immunoreactivities at the level of individual cell profiles. The results indicate that in the vestibular nuclei, neuronal colocalization of two or more neuroactive amino acids is the rule rather than an exception. Colocalization was found of immunoreactivities for GABA and glycine; glycine, aspartate and glutamate; glycine and aspartate, and glutamate and aspartate. GABA immunoreactive neurons were generally small and were found scattered throughout the vestibular complex. Glycine immunoreactive neurons were similarly distributed, except in the superior nucleus where the latter type of neuron could not be detected. Neuronal profiles colocalizing immunoreactivities for GABA and glycine occurred in all nuclei, but were most numerous in the lateral nucleus. The vast majority of the neurons showed noteworthy staining for glutamate and aspartate, although the level of immunoreactivities varied (e.g., the large neurons in the lateral and descending nuclei were more intensely aspartate immunoreactive than the smaller ones). Taurine-like immunoreactivity did not occur in neuronal cell bodies but appeared in Purkinje cell axons and in glial cell profiles. The functional significance of the complex pattern of amino acid colocalization remains to be clarified. In particular it needs to be distinguished between metabolic and transmitter pools of the different amino acids. The present results call for caution when attempts are made to conclude about transmitter identity on the basis of amino acid contents alone.

Highly differential expression of the monocarboxylate transporters MCT2 and MCT4 in the developing rat brain

Monocarboxylate transporters (MCTs) play an important role in the metabolism of all cells. They mediate the transport of lactate and pyruvate but also some other substrates such as ketone bodies. It has been proposed that glial cells release monocarboxylates to fuel neighbouring neurons. A key element in this hypothesis is the existence of neuronal MCTs. Amongst the three MCTs known to be expressed in the brain (MCT1, 2 and 4) only MCT2 has been found in neurons. Here we have studied the expression pattern of MCT2 during postnatal development. By use of immunoperoxidase and double immunofluorescence microscopy we report that neuronal MCT2 occurs in most brain areas, including the hippocampus and cerebellum, from birth to adult. MCT2 is also expressed in specific subpopulations of astrocytes. Neuronal MCT2 is most abundant in the first 3 postnatal weeks and thereafter decreases toward adulthood. In contrast to MCT2, MCT4 is exclusively present in astroglia during all stages of development. Furthermore, MCT4 expression is very low at birth and reaches adult level by P14. Our results are consistent with previous data suggesting that in the immature brain much of the energy demand is met by monocarboxylates and ketone bodies.