The expression of the glutamate re-uptake transporter excitatory amino acid transporter 1 (EAAT1) in the normal human CNS and in motor neurone disease: an immunohistochemical study.

A monoclonal antibody to excitatory amino acid transporter 1 (EAAT1) has been generated which robustly stains paraffin-embedded, formaldehyde-fixed as well as snap-frozen human post-mortem brain tissue. We have used this antibody to map the distribution of EAAT1 throughout normal human CNS tissue. In addition this antibody has been used to perform a semi-quantitative immunohistochemical analysis of the expression of EAAT1 in motor cortex and cervical cord tissue taken from motor neurone disease cases (n=17) and neurologically normal controls (n=12). By comparing the relative optical density measurements of identical regions of motor cortex and cervical spinal cord an increase in the expression levels of EAAT1 was observed in motor neurone disease tissue compared to the control tissue and in both motor cortex and cervical spinal cord (9-17% and 13-33% increases respectively). EAAT1 was observed to be the most abundant transporter in more "caudal" brain regions such as the diencephalon and brainstem and its expression in other regions was frequently more uniform than that of EAAT2. In the motor cortex, EAAT1 immunoreactivity was present in all grey matter laminae, with some staining of individual astrocytes in the white matter. In spinal cord, EAAT1 immunoreactivity was strongest in the substantia gelatinosa. In the ventral horn, motor neurones were surrounded with a dense rim of perisomatic EAAT1 immunoreactivity, and the neuropil showed diffuse staining. Additional studies using double-labelling immunocytochemistry demonstrated that astrocytic co-localisation of EAAT1 and EAAT2 may occasionally be seen, but was not widespread in the human CNS and that in general astrocytes were positive for either EAAT1 or EAAT2. These results demonstrate that the EAAT1 has a widespread abundance throughout all regions of the human CNS examined and that there exist discrete populations of astrocytes that are positive solely for either EAAT1 or EAAT2. Furthermore, there is evidence to suggest that altered EAAT1 expression in motor neurone disease follows a different pattern to the reported changes of EAAT2 expression in this condition, indicating that the role of glutamate transporters in the pathogenesis of motor neurone disease appears more complex than previously appreciated.

Glutamine synthetase activity and expression are not affected by the development of motor neuronopathy in the G93A SOD-1/ALS mouse.

The expression and activity of the enzyme glutamine synthetase (GS) were examined in the G93A/SOD-1 transgenic mouse model of progressive motor neuronopathy to investigate the mechanisms underlying degeneration of the motor neurones. Clinical signs appeared in G93A/SOD-1 mice at around 90 days, with severe spasticity and loss of self-righting reflex from 120 to 150 days of age. GS expression was examined using western blotting in primary astrocyte cultures derived from newborn (P1-2) G93A/SOD-1 mice and their non-transgenic littermates and in lower spinal cord from animals at 30, 60 and 90 days of age and disease end-stage (120-150 days). There were no differences in the levels of GS expression in the transgenic mice compared to the unaffected littermates at any of the disease stages examined. GS activity was measured spectrophotometrically in spinal cord extracts at these disease stages. There was a decrease in V(max) at 60 days compared to 30 days in both groups of mice (3.48+/-0.58 cf. 6.43+/-1.83 mmol/h/mg protein; non-transgenic littermates), with GS activity highest at end-stage (9.38+/-0.71 mmol/h/mg protein cf. 7.64+/-0.42 mmol/h/mg protein in littermates). Conversely, K(m) was transiently increased at 60 days (2.53+/-0.26 mM cf. 1.32+/-0.20 in littermates), remaining within the range of 30 day measurements from 90 days onwards. There were no differences in V(max) or K(m) values between the G93A/SOD-1 mice and their unaffected non-transgenic littermates at any of the disease stages examined. We conclude that there is no evidence that a change in glutamine synthetase activity or expression contributes to the progressive neurodegeneration observed in the G93A/SOD-1 mice.

Cultured glial cells are resistant to the effects of motor neurone disease-associated SOD1 mutations.

Free radical damage has been implicated in the pathophysiology of motor neurone disease (MND); mutations have been identified in the gene encoding Cu/Zn superoxide dismutase (SOD1). There is evidence that glial cell dysfunction may contribute to motor neurone injury, but the exact role of glial cells in MND has yet to be established. The aim of this study was to determine whether expression of mutant SOD1 affects the response of glia to oxidative stress. Stable C6 glioma cells expressing mutant SOD1 and cortical astrocyte cultures from G93A-SOD1 transgenic mice were exposed to: xanthine/xanthine oxidase; hydrogen peroxide; A23187 and 3-morpholinosydonimine. Cell viability was measured using the 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. Neither C6 glioma cells nor cortical astrocytes expressing mutant SOD1 were more susceptible to any of the free radical generating systems compared to control cells. These results suggest that astrocytes are resistant to the toxic effects of mutant SOD1 widely reported for neuronal cells.

The RNA of the glutamate transporter EAAT2 is variably spliced in amyotrophic lateral sclerosis and normal individuals.

Impaired re-uptake of synaptic glutamate, and a reduced expression of the glutamate transporter EAAT2 have been found in the motor cortex of patients with amyotrophic lateral sclerosis (ALS). Two splice forms of the EAAT2 RNA resulting from retention of intronic sequences (EAAT2/Int) and deletion of one protein coding exon (EAAT2/C1) have been reported to account for the EAAT2 protein loss in ALS. In this study we investigated the presence of two known (EAAT2/C1; EAAT2/Int) and three novel (EAAT2/C2-4) EAAT2 RNA in motor cortex of 17 ALS cases and 11 controls. Reverse transcription and PCR were carried out to amplify the complementary DNA of the complete and variably spliced EAAT2 transcripts. Nested PCR was followed to generate amplicons specific for EAAT2/C1-4 and EAAT2/Int. EAAT2/Int was detected in 59% of ALS specimens as compared to 36% of controls showing a trend but no statistical significance of a more frequent expression in ALS (Type I error 24.6%). EAAT2/C1-4 were found to be equally expressed in ALS patients and controls. Our results indicate that the involvement of EAAT2 transcripts in ALS is unlikely to be primary, and more complex than previously recognized. Alterations of quantitative expression of distinct EAAT2 splice forms in ALS cannot be excluded from this study and remain to be investigated.

The role of astrocytes and noradrenaline in neuronal glucose metabolism.

In the classical model the energy requirements during neuronal activation are provided by the delivery of additional glucose directly into the extracellular compartment that results from the increase in local cerebral blood flow (rCBF). The present review proposes that astrocytes play a key role in the response to neuronal activation. Arginine for the synthesis of NO, which has a major role in the increase in rCBF, is released from astrocytes in response to stimulation of astrocytic glutamate receptors. The increased delivery of glucose by the blood stream enters astrocytes, where some of it is converted to glycogen. During neuronal activation there is a decrease in extracellular glucose owing to increased utilization followed by a delayed increase; this results from stimulation of astrocytic beta-adrenergic receptors, which leads to a breakdown of glycogen and the export of glucose.

The expression of the glial glutamate transporter protein EAAT2 in motor neuron disease: an immunohistochemical study.

Emerging evidence suggests that a disturbance of the glutamate neurotransmitter system may be a contributory factor to motor neuron injury in motor neuron disease. Previous autoradiographic and immunoblotting studies have suggested that there may be reduced expression of glutamate transporter proteins in pathologically affected areas of the CNS in motor neuron disease. This study further explores the possible alteration in expression of the excitatory amino acid transporter protein EAAT2 in MND, by examining the protein expression in situ, in frozen sections, using immunohistochemistry. The aim of the study was to compare the distribution and density of EAAT2 in the motor cortex and spinal cord of MND cases (n = 16) compared with neurologically normal controls (n = 12), matched for relevant parameters. A novel, previously characterized, monoclonal antibody to EAAT2 was employed. EAAT2 immunoreactivity in motor neuron disease and control cases was compared using relative optical density measurements generated by computerized image analysis. In the motor cortex, EAAT2 immunoreactivity was laminated comprising a superficial intense band (corresponding to layers 1 and 2); a paler middle band (layer 3 and part of 5) and a more intense deep layer (layers 5 and 6). In the spinal cord, the ventral horn showed strong immunoreactivity with dense perisomatic staining around motor neuron cell bodies, the substantia gelatinosa showed moderate diffuse staining and the intermediate spinal laminae showed weak staining. This general pattern of immunoreactivity was preserved in the motor neuron disease cases. However, in the motor neuron disease cases compared with controls, the optical density values for EAAT2 immunoreactivity were significantly reduced in all grey matter regions of the lumbar spinal cord (P < 0.001) and were increased in the middle laminae of the motor cortex (P < 0.05). This study indicates that glutamate transporter pathology in motor neuron disease may be a more complex phenomenon than previously recognized.

Extracellular glucose turnover in the striatum of unanaesthetized rats measured by quantitative microdialysis.

1. Steady-state and time-resolved quantitative microdialysis was used to measure dialysate concentration, extracellular concentration and the in vivo recovery of glucose in rat striatum. 2. The extracellular concentration of glucose, determined by the zero net flux method of Lönnroth, was 350 +/- 20 microM and the in vivo recovery was 39 +/- 2%. 3. Veratridine caused a steep decrease in dialysate glucose after an initial delay of 7.5 min. When steady-state glucose levels had been reached in the presence of veratridine the extracellular concentration was reduced to zero, but there was no significant change in in vivo recovery. 4. Measurement of the dynamic changes during the administration of veratridine showed an immediate decrease in extracellular glucose concentration and a steep rise in in vivo recovery, which accounted for the delay in the delay in the decrease in dialysate glucose. When extracellular concentration reached zero, in vivo recovery returned to control levels.

Expression of the glial glutamate transporter EAAT2 in the human CNS: an immunohistochemical study.

Glutamate transporters play an essential role in terminating the excitatory glutamatergic signal at post-synaptic receptors and in protecting neurones from excitotoxic effects, as well as replenishing the neurotransmitter supply at glutamatergic synapses. The distribution and density of glutamate transporters may be important determinants of vulnerability to glutamate-mediated injury. There is emerging evidence that glutamate transporter dysfunction may be present in motor neurone disease (MND). In this study, a monoclonal antibody, suitable for immunohistochemistry (IHC) in human post-mortem tissue, was produced to the human astrocytic glutamate transporter EAAT2 (excitatory amino acid transporter 2). Western blotting of homogenates of human cortical tissue with the EAAT2 antibody produced a discrete band at 66 kDa. Detailed IHC analysis of the expression of the EAAT2 protein in the human CNS was undertaken. EAAT2 was exclusively localised to astrocytes, with preferential expression in the caudate nucleus, nucleus basalis of Meynert, spinal ventral horn, cerebral cortex and hippocampus, but with lower levels of expression throughout many other CNS regions. Motor neurone groups vulnerable to neurodegeneration in MND appeared distinctive in being surrounded by extensive, coarse, strongly immunoreactive perisomatic glial profiles. Motor neurone groups which tend to be spared in MND, such as those present in the oculomotor nucleus, showed a lower expression of EAAT2, with fewer perisomatic profiles. The EAAT2 antibody will provide a useful tool for increasing our understanding of the role of EAAT2 in excitatory neurotransmission in health and disease states.

The mechanisms controlling physiologically stimulated changes in rat brain glucose and lactate: a microdialysis study.

1. This study is concerned with the supply of metabolic substrates for neuronal metabolism. Experiments were carried out to investigate whether mechanisms demonstrated in cultured astrocytes also occurred in vivo; these were cAMP-mediated breakdown of glycogen and glutamate uptake-stimulated release of lactate. 2. In vivo microdialysis was used in freely moving rats. Lactate and glucose in the dialysate were assayed using enzyme-based on-line assays. Drugs were given locally through the dialysis probe. Regional cerebral blood flow was measured using the hydrogen clearance method. 3. There was an increase in dialysate glucose in response to the beta-adrenoceptor agonist isoprenaline and to 8-bromo-cAMP, an analogue of cAMP, the second messenger of beta-adrenoceptor stimulation. The effect of isoprenaline was blocked by the antagonist propranolol. Isoprenaline had no effect on dialysate lactate, which was increased by the glutamate uptake blocker beta-D,L-threohydroxyaspartate (THA). 4. Physiological stimulation of neuronal activity produced an increase in both lactate and glucose. The increase in lactate was depressed in the presence of THA but was unaffected by propranolol. The increase in glucose was blocked by propranolol. Regional cerebral blood flow was increased by physiological stimulation but was unaffected by propranolol. 5. These results demonstrate that physiologically stimulated increases in glucose and lactate in the brain are mediated by different mechanisms.

A role for astrocytes in glucose delivery to neurons?

The present paper examines the possible role of astrocytes in the delivery of glycogen-derived glucose for neuronal metabolism. Such a process would require astrocytic expression of glucose-6-phosphatase. The degree and significance of brain expression of glucose-6-phosphatase (EC 3.1.3.9) has been a subject of controversy. Published immunohistochemical data are consistent with expression of glucose-6-phosphatase by astrocytes, both in vivo and in vitro. In this paper additional confirmation of the expression of glucose-6-phosphatase mRNA in rat brain is presented. Although cultured astrocytes demonstrate glucose-6-phosphatase activity in vitro under assay conditions, there is very limited in vitro evidence that this activity confers a glucose-export capacity on astrocytes. Under most conditions in vitro, lactate export predominates, however this may relate to aspects of the in vitro phenotype. Data relating to astrocytic glucose and lactate export are considered in the context of hypotheses of trafficking by astrocytes of substrates for neuronal metabolism, hypotheses that imply and require compartmentation of these substances, in contrast with current formulations of glucose transport into and within brain that imply no glucose compartmentation. Microdialysis studies of the properties of the brain extracellular fluid (ECF) glucose pool in the freely moving rat were performed seeking evidence of glucose compartmentation. Results of these studies do imply compartmentalisation of brain glucose, and are consistent with a model envisaging the majority of glucose reaching the neuron via the astrocytic intracellular space and the ECF. In addition, such studies provide evidence that rises in ECF glucose concentration are not the direct result of local recruitment of cerebral blood flow, but suggest the influence of intermediate, astrocyte-based mechanisms. Astrocytic glucose-6-phosphatase may permit astrocytes to modulate the trans-astrocytic flux of glucose to adjacent neurons in response to signals reflecting increased neuronal demand.

In vivo and in vitro anti-tumour activity of dimaprit.

Dimaprit, a histamine H2-receptor agonist, injected daily i.p. to fibrosarcoma-bearing mice, induced a decrease in tumour growth and an increase in survival. Dimaprit, added to tumour cell cultures (10(-4) M), inhibited the incorporation of 3H-thymidine while embryonic cell cultures were unaffected. This particular anti-tumour activity is probably H2-independent as histamine and impromidine have no effect on tumour cell cultures.

Inverse correlation between tumor incidence and tissue histamine levels in W/WV, WV/+, and +/+ mice.

The influence of mast cells on tumor incidence and growth rate was studied in 2 grafted tumor models (fibrosarcoma MC-B6-1 and the Lewis lung carcinoma 3LL). Three kinds of WBB6F1 mice (a cross between WB/ReJ-W/+ and C57BL/6J-WV/+ mice) were used: W/WV (deeply mast cell depleted), WV/+ (partially mast cell depleted), and +/+ (normal mast cell number). The presumed resistance of F1 hybrids to tumor cells of parental origin was observed in 12 of 13 +/+ mice, but only in 11 of 22 WV/+ mice and in none of 39 W/WV mice. Tumor incidence and metastasis incidence were inversely correlated with tissue histamine levels and mast cell number. Growth rates of tumors were similar in W/WV and WV/+ mice, but the tumor growth rate was much slower in the only +/+ mouse in which the tumor grew. These results confirm the protective role of mast cells against tumors.

Tissue histamine levels and mast cell numbers in tumour-bearing mice.

In C57BL/6 mice bearing the 3LL carcinoma and in C3H mice bearing the McC3 -1 fibrosarcoma (18th passage), the increase in skin histamine levels was correlated with the increase in mast cell number. The number of cells able to incorporate tritiated thymidine was proportional to the mast cell number. These results strengthen the notion that, in tumour-bearing rodents, the increase in tissue histamine is an active phenomenon.

Selective increased tissue histamine levels in tumour-bearing rodents.

Histamine levels increased in the fundus of mice bearing a primary 3-methylcholanthrene-induced fibrosarcoma, and in the ventral skin, skeletal muscle and rumen of rats bearing a D.M.B.A. induced mammary adenocarcinoma; they did not increase in the tissues of mice bearing a McC3-1 fibrosarcoma (38th passage) or a Lewis lung carcinoma before the appearance of metastasis, but an increase in histamine levels was observed in dorsal skin, ventral skin and fundus, after the appearance of metastasis.

Enhancement of Lewis lung carcinoma growth by sera or spleen cells from tumour-bearing mice.

Lewis lung carcinoma(3LL) cells grafted in syngeneic adult C57BL/6 mice produced local tumours associated with lung metastasis in 78% of recipients. Adult thymectomized, lethally irradiated, bone marrow cell-reconstituted animals (B mice) were more resistant since this tumour grew in only 46% of recipients and especially the weight of lung metastasis was almost 8 times less than in normal animals. Sera from tumour-bearing mice transferred into B mice enhanced tumour incidence and weight of metastasis to the level observed in normal mice. In vitro cultured 3LL cells displayed an intense mitogenic activity as measured by 3H-thymidine incorporation. Addition into these cultures of serum from tumour-bearing animals did not alter this activity. Addition of normal spleen cells in a ratio 40/1 reduced this mitogenic activity to a half or a third. Spleen cells from tumour-bearing mice, either normal or serum-enhanced B mice, mixed in vitro with 3LL cells, produced a 4-fold increase of mitogenic activity of the latter. These results indicate that the lymphoid system may contribute to the growth and spreading of 3LL tumours.

[Anti-tumour cytostatic and cytotoxic properties of peritoneal exudate cells of conventional and germ-free mice, stimulated or not by "Corynebacterium parvum" (author's transl)]. / Propriétés antitumorales cytostatiques et cytotoxiques des cellules péritonéales de souris conventionnelles et axéniques stimulées ou non par Corynebacterium parvum.

The treatment by Corynebacterium parvum induced an increase of peritoneal cell number in conventional mice but no modifications in germ-free mice. Against YC8 tumoral target cells, cytostatic properties of peritoneal cells were of the same intensity in conventional and germ-free after C. parvum treatment. Against K.BALB cells, C. parvum treatment induced an increase of cytostatic properties from 9 to 93% in conventional mice and from 51 to 84% in germ-free mice. Cytotoxic properties were increased by C. parvum in conventional mice but were unchanged in germ-free mice. The bacterial flora could play a role in the cytotoxic and cytostatic properties of peritoneal cells in conventional mice.

[Influence of the irritant used to provoke the peritoneal exudate used in the macrophage migration inhibition test (author's transl)]. / Influence de l'agent irritant utilisé dans la préparation de l'exsudat péritonéal dans le test d'inhibition de migration des macrophages.

The effect of the irritant chosen to provoke the peritoneal exudate in donors of cells for the in vitro migration inhibition test was studied. Guinea-pigs with delayed hypersensitivity to tubercle bacilli were used. Irritants studied included starch, Bayol and thioglycollate medium. The exudates obtained with thioglycollate were found to show greater inhibition than those obtained with the other irritants in the in vitro response to PPD, Inhibition of migration was seen when 1 microgram/ml of PPD was added to the culture medium; with this dose the inhibition of migration was 53%. The state of activation of macrophages caused by the irritant thus appears to have some importance for their responsiveness to factors liberated by specifically sensitized lymphocytes.

[Equal sensitivity of normal or tumoral fibroblasts to cytotoxic and cytostatic effects of normal or "Corynebacterium parvum" activated peritoneal exudate cells (author's transl)]. / Sensibilité identique des fibroblastes cultivés normaux et tumoraux aux effets cytotoxiques et cytostatiques des cellules d'exsudat péritonéal normales ou activées par Corynebacterium parvum.

Cytotoxic and cytostatic properties of peritoneal exudate cells from BALB/c mice either normal or pretreated with Corynebacterium parvum normal were studied using as target cells two lines of BALB/c fibroblasts one normal (BALB) and one transformed by Kirsten virus (K.BALB).. C. parvum activated cells displayed higher cytotoxic as measured by 51Cr or 3H-thymidine release and cytostatic, as measured by 3H-thymidine incorporation, activities against these lines of syngeneic cells. Cytotoxic effect of normal peritoneal cells was evident when 51Cr or 3H-thymidine release was used whereas no such effect was found when cytolytic plaque formation test was used. Sensitivity of both BALB and K.BALB target cell lines was found to be equal to cytotoxic and cytostatic activities of C. parvum activated or normal syngeneic peritoneal cells.

In vitro anti-tumour properties of peritoneal exudate cells of conventional germ-free and stimulated mice.

Adherent cells of peritoneal exudates were obtained from conventional, or from germ-free mice or from mice having received in intraperitoneal infection of a variety of phlogogenic substances such as Corynebacterium parvum (C. parvum) (Mérieux) 500 microgram, thioglycolate (Difco) 3 ml, Bayol (Esso) 0.50 ml + 0.50 ml culture medium, glycogene 1.2 ml. The cytotoxic properties adherent cells were studied in vitro by the chromium release technique (CRT) and their cytostatic properties by the inhibition of the incorporation of tritiated thymidine by YC8 lymphoma cells. C. parvum was found to be the most active agent in enhancing the cytotoxic properties of adherent cells, followed by BCG and Bayol. Glycogen peptone and other substrates were without effect. The unstimulated peritoneal macrophages of conventional mice were found to inhibit thymidine incorporation by tumour cells, whereas those of germ-free mice could not do so. C. parvum markedly increased the cytostatic property adhered cells from both germ-free and conventional mice.