Isotypes of alpha-tubulin are differentially regulated during neuronal maturation.

The mRNAs for two isotypes of alpha-tubulin, termed T alpha 1 and T26, are known to be expressed in the rat nervous system. We have compared the expression of these two alpha-tubulin mRNAs during neural development, using RNA blotting and in situ hybridization techniques with probes directed against unique sequences of each mRNA. T alpha 1 mRNA is highly enriched in the embryonic nervous system but is markedly less abundant in the adult brain; T26 mRNA is expressed in many embryonic tissues with little change in abundance during development. Within the nervous system, T alpha 1 mRNA is enriched in regions with neurons actively undergoing neurite extension, such as the cortical plate, whereas T26 mRNA is relatively homogeneous in distribution, with some enrichment in proliferative zones. Expression of T alpha 1 mRNA is also increased in PC12 cells induced to differentiate and extend neurite processes by nerve growth factor. Taken together, the data indicate that T alpha 1-tubulin mRNA is expressed at high levels during the extension of neuronal processes. The abundant expression of T alpha 1-tubulin mRNA may therefore reflect either a means to increase the available pool of alpha-tubulin or a specific requirement for the T alpha 1 isotype for neurite extension.

Intercellular calcium signaling via gap junctions in glioma cells.

Calcium signaling in C6 glioma cells in culture was examined with digital fluorescence video microscopy. C6 cells express low levels of the gap junction protein connexin43 and have correspondingly weak gap junctional communication as evidenced by dye coupling (Naus, C. C. G., J. F. Bechberger, S. Caveney, and J. X. Wilson. 1991. Neurosci. Lett. 126:33-36). Transfection of C6 cells with the cDNA encoding connexin43 resulted in clones with increased expression of connexin43 mRNA and protein and increased dye coupling, as well as markedly reduced rates of proliferation (Zhu, D., S. Caveney, G. M. Kidder, and C. C. Naus. 1991. Proc. Natl. Acad. Sci. USA. 88:1883-1887; Naus, C. C. G., D. Zhu, S. Todd, and G. M. Kidder. 1992. Cell Mol. Neurobiol. 12:163-175). Mechanical stimulation of a single cell in a culture of non-transfected C6 cells induced a wave of increased intracellular calcium concentration ([Ca2+]i) that showed little or no communication to adjacent cells. By contrast, mechanical stimulation of a single cell in cultures of C6 clones expressing transfected connexin43 cDNA induced a Ca2+ wave that was communicated to multiple surrounding cells, and the extent of communication was proportional to the level of expression of the connexin43 cDNA. These results provide direct evidence that intercellular Ca2+ signaling occurs via gap junctions. Ca2+ signaling through gap junctions may provide a means for the coordinated regulation of cellular function, including cell growth and differentiation.

In vivo growth of C6 glioma cells transfected with connexin43 cDNA.

In order to examine the possible role of intercellular communication via gap junctions in the control of tumor growth, we have transfected C6 glioma cells with connexin43 cDNA. We obtained several clones with variable expression of connexin43. The growth rate of these clones in culture was inversely related to the degree of expression of the transfected cDNA. To examine the growth of these transfected cells in vivo, cells were grown in spinner culture flasks to form spheroids 250-300 microns in diameter. Spheroids of nontransfected C6 cells produced large gliomas. Immunohistochemical and in situ hybridization analyses revealed relatively high levels of connexin43 protein and mRNA in the host tissue, while little of this protein was detected in the glioma. In contrast, spheroids of connexin43-transfected cells grew more slowly and exhibited elevated levels of connexin43 protein and mRNA. These findings suggest that the expression of connexin43 may be associated with the control of brain tumor growth in vivo.

Gap junctions promote the bystander effect of herpes simplex virus thymidine kinase in vivo.

Transfer of the herpes simplex thymidine kinase gene (HSVtk) into tumor cells followed by the administration of ganciclovir (GCV) provides a potential strategy for the treatment of some malignancies. During GCV treatment, not only the cells that express the HSVtk gene are killed but also frequently neighboring tumor cells that are not genetically altered. This has been called the "bystander effect." Although the mechanism of the bystander effect in vivo remains elusive, our results suggest that gap junction formation between neighboring cells is an important contributing factor. The C6 rat glioma cell line, which exhibits a low level of intercellular communication by gap junctions and connexin43 (Cx43)-transfected clones of this cell line forming gap junctions from a moderate level (Cx43-12 and Cx43-14) to a high level (Cx43-13), were transduced with HSVtk. Transduced and nontransduced cells were mixed in various concentrations and then cultured in vitro or injected s.c. into C.B-17/SCID-beige mice followed by i.p. injections of GCV. Cx43-transfected clones showed a significant increase of the bystander effect compared with the less coupled C6 parental cell line. In 11 of 12 mice injected with cells of Cx43-transfected clones, no tumors were seen at the inoculation site when a mixture of 50% HSVtk-negative and HSVtk-positive cells was used. Moreover, in mice injected with cells of clone Cx43-13, which exhibits the highest intercellular communication, tumors were frequently undetectable at the inoculation site when using mixtures of 75% HSVtk-negative and 25% HSVtk-positive cells, and even mixtures containing 5% HSVtk-positive cells of Cx43-transfected clones showed tumor size reduction. All animals in control groups (n = 26) developed large tumors at every injection site. These results demonstrate that gap junctions are an important component in mediating the bystander effect in vivo.

Connexin43 suppresses MFG-E8 while inducing contact growth inhibition of glioma cells.

Gap junction expression has been reported to control the growth of a variety of transformed cells. We undertook parallel analysis of connexins Cx32 and Cx43 in glioma cells, which revealed potential mechanisms underlying this phenomenon and led to several novel findings. Cx43, but not Cx32, suppressed C6 glioma cell growth. Paradoxically, Cx32 transfection resulted in severalfold more dye transfer than Cx43. However, Cx43 transfectants shared endogenous metabolites more efficiently than Cx32 transfectants. Interestingly, a significant portion of Cx43 permeants were incorporated into macromolecules more readily than those that transferred via Cx32. Cx43 induced contact inhibition of cell growth but in contrast to other reports, did not affect log phase growth rates. Cell death, senescence, or suppression of growth factor signaling was not involved because no significant alterations were seen in cell viability, telomerase, or mitogen-activated protein kinase activity. However, suppression of cell growth by Cx43 entailed the secretion of growth-regulatory factors. Most notably, a major component of conditioned medium that was affected by Cx43 was found to be MFG-E8 (milk fat globule epidermal growth factor 8), which is involved in cell anchorage and integrin signaling. These results indicate that Cx43 regulates cell growth by the modulation of extracellular growth factors including MFG-E8. Furthermore, the ability of a Cx to regulate cell growth may rely on its ability to mediate the intercellular transfer of endogenous metabolites but not artificial dyes.

Astrocytes in neuroprotection and neurodegeneration: The role of connexin43 and pannexin1.

The World Health Organization has predicted that by 2040 neurodegenerative diseases will overtake cancer to become the world's second leading cause of death after cardiovascular disease. This has sparked the development of several European and American brain research initiatives focusing on elucidating the underlying cellular and molecular mechanisms of neurodegenerative diseases. Connexin (Cx) and pannexin (Panx) membrane channel proteins are conduits through which neuronal, glial, and vascular tissues interact. In the brain, this interaction is highly critical for homeostasis and brain repair after injury. Understanding the molecular mechanisms by which these membrane channels function, in health and disease, might be particularly influential in establishing conceptual frameworks to develop new therapeutics against Cx and Panx channels. This review focuses on current insights and emerging concepts, particularly the impact of connexin43 and pannexin1, under neuroprotective and neurodegenerative conditions within the context of astrocytes.

Astrocytes promote glioma invasion via the gap junction protein connexin43.

Reactive astrocytes are integral to the glioma microenvironment. Connexin43 (Cx43) is a major gap junction protein in astrocytes and its expression is enhanced significantly in glioma-associated astrocytes, especially at the peri-tumoral region. Although downregulation of Cx43-mediated intercellular communication is associated with increased malignancy in tumor cells, the role of Cx43 in stromal cells in glioma progression is not defined. Using a mouse model consisting of syngeneic intracranial implantation of GL261 glioma cells into Nestin-Cre:Cx43(fl/fl) mice where Cx43 was eliminated in astrocytes, we demonstrate a role of astrocytic Cx43 in the dissemination of glioma cells from the tumor core. To determine whether heterocellular communication between astrocytes and glioma cells is essential for reduced invasion in the absence of astrocytic Cx43, we abolished channel formation between glioma cells and astrocytes by either knocking down Cx43 in glioma cells with short hairpin RNA (shRNA) or overexpressing a dominant-negative channel-defective Cx43-T154A mutant in these cells. Although Cx43shRNA in glioma cells reduced invasion, expression of Cx43-T154A had no effect on glioma invasion, suggesting tumoral Cx43 may influence motility independently from its channel function. Alteration in astrocytic Cx43 function, such as by replacing the wild-type allele with a C-terminal truncated Cx43 mutant exhibiting reduced intercellular coupling, is sufficient to reduce glioma spreading into the brain parenchyma. Our results reveal a novel role of astrocytic Cx43 in the formation of an invasive niche and raise the possibility to control glioma progression by manipulating the microenvironment.

ATP-mediated glia signaling.

Glia calcium signaling has recently been identified as a potent modulator of synaptic transmission. We show here that the spatial expansion of calcium waves is mediated by ATP and subsequent activation of purinergic receptors. Ectopic expression of gap junction proteins, connexins (Cxs), leads to an increase in both ATP release and the radius of calcium wave propagation. Cx expression was also associated with a phenotypic transformation, and cortical neurons extended longer neurites when co-cultured with Cx-expressing than with Cx-deficient cells. Purinergic receptor activation mediated both these effects, because treatment with receptor antagonists restored the glia phenotype and slowed neurite outgrowth. These results identify a key role of ATP in both short-term calcium signaling events and in long-term differentiation regulated by glia.

Colocalization of prosomatostatin-derived peptides in the caudate-putamen of the rat.

In the striatum of rat, somatostatin 14, somatostatin 28, and somatostatin 28(1-12) have previously been localized within a small population of medium aspiny local circuit neurons. Because all three peptide fragments are generated through the cleavage of prosomatostatin by different converting enzymes, the possibility for differential expression of these peptides exists. In order to investigate this possibility, frozen sections were collected from the brains of adult female Wistar rats fixed with 4% paraformaldehyde and double labelled using immunocytochemistry and in situ hybridization. Sections were first processed for somatostatin 14, somatostatin 28, or somatostatin 28(1-12) by using the avidin-biotin complex immunocytochemical technique followed by in situ hybridization using 35S-labelled antisense riboprobes to somatostatin mRNA. The results of such analysis revealed that somatostatin 28 and somatostatin mRNA are 100% colocalized. Somatostatin 14 and somatostatin 28(1-12), in contrast, are only present within 66% of the neurons that express somatostatin mRNA. Examination of the anatomical distribution of neurons that express both somatostatin mRNA and somatostatin 14 or somatostatin 28(1-12) protein reveals that these neurons are present throughout the caudate-putamen of rat but are more prevalent in the ventromedial regions. Neurons that express somatostatin mRNA but not somatostatin 14 or somatostatin 28(1-12) are also present throughout the caudate-putamen but are most numerous within a dorsolateral strip just beneath the corpus callosum. These results suggest that the somatostatin neuron population within the rat caudate-putamen is actually composed of two smaller subpopulations based on neuropeptide content. The first subpopulation contains somatostatin 28 and constitutes one-third of the total somatostatin population, whereas the other contains somatostatin 28, somatostatin 14, and somatostatin 28(1-12) and represents the remaining two-thirds of the cells that express somatostatin mRNA.

Colocalization of somatostatin, neuropeptide Y, and NADPH-diaphorase in the caudate-putamen of the rat.

Somatostatin, neuropeptide Y, and nicotinamide adenine dinucleotide phosphate-diaphorase are colocalized within a small population of medium aspiny neurons in the caudate-putamen of the rat. The extent of colocalization, however, appears to be in dispute. In order to examine the question of colocalization between these three neuroactive substances, a series of double-labelling experiments was performed. This was accomplished by combining immunocytochemistry for somatostatin or neuropeptide Y or enzyme histochemistry for nicotinamide adenine dinucleotide phosphate-diaphorase with in situ hybridization for somatostatin and/or neuropeptide Y mRNA. The results of such analysis indicate that nicotinamide adenine dinucleotide phosphate-diaphorase and somatostatin mRNA are 100% colocalized throughout the caudate-putamen, except for the area bordering the globus pallidus. All neurons that contain neuropeptide Y contain somatostatin message. Only 84% of the neurons that contain somatostatin mRNA, however, also contain neuropeptide Y. Neurons that contain somatostatin 28 but not neuropeptide Y are found throughout the caudate-putamen. These results indicate that the somatostatin neuron population in the rat caudate-putamen is not homogeneous. Instead, the medium aspiny neuron population is actually composed of several subpopulations based on the content of neuroactive substances.

Somatostatin and the patch/matrix compartments of the rat caudate-putamen.

In the caudate-putamen of the rat, two subpopulations of medium aspiny neurons exist that contain somatostatin. The first subpopulation contains somatostatin 14, somatostatin 28, and somatostatin 28(1-12). The other subpopulation contains only somatostatin 28. To examine the relationship between somatostatin-containing neurons and the patch/matrix compartments, a series of double-labelling experiments using antisera directed against different somatostatin peptides and calbindin were used. Sections stained in this manner were examined with the aid of a confocal microscope. The results of these experiments indicate that somatostatin 28(1-12)-containing neurons may play a role in matrix integration, with some input directed from the patch compartment. In addition, somatostatin 28-containing neurons are more numerous in the patch compartment than somatostatin 28(1-12)-containing neurons, suggesting a possible role for these neurons in patch integration.

Neurochemical compartmentalization of the globus pallidus in the rat: an immunocytochemical study of calcium-binding proteins.

The globus pallidus external segment forms a major target center of the mammalian striatum which is characterized by neurochemically distinct compartments. The present study was undertaken to determine if a corresponding compartmentalization exists within the globus pallidus external segment in the rat. Immunocytochemical examination of the calcium-binding proteins parvalbumin and calbindin D28kDa, which are present in neurons of the striatal matrix compartment, was employed. The results indicate three neurochemically distinct compartments within the globus pallidus external segment: 1) an area in the medial aspect of the entire length of the globus pallidus that contains dense immunoreactivity for calbindin D28kDa; 2) a narrow rim at the striatopallidal junction in the rostral two-thirds of the globus palidus that contains calbindin D28kDa immunoreactivity designated as the "border zone" of the globus pallidus; and 3) an area between these two zones showing very poor immunoreactivity for calbindin D28kDa but containing parvalbumin immunoreactive neurons. The calbindin D28kDa immunoreactive border zone corresponds to the area of the globus pallidus where striatal inputs converge extensively, whereas the rest of the nucleus is involved in segregated, topographically organized pathways. Parvalbumin-containing neurons are involved in the propagation of striatal output related to striosomal and sensorimotor aspects of basal ganglia function. The present results also indicate that calbindin D28kDa immunoreactivity is completely absent from striosomal neurons and is therefore a useful marker for striatal compartments.

Immunohistochemical and in situ hybridization analysis of the development of the rat somatostatin-containing neocortical neuronal system.

The chemical differentiation of somatostatin (SS) neurons in rat neocortex was characterized by molecular biochemical and morphological methods. Northern (RNA) blotting indicates that regional distribution of SS mRNA correlates with the known distribution patterns of SS-containing neurons in the adult, while similar analysis of poly (A)+ RNA isolated from telencephalon at various times postnatally shows an increase between P9 and P15, with a slight decrease in the adult. In situ hybridization with a probe specific to SS mRNA, and immunohistochemistry using antisera specific for the N-terminally extended form of SS, SS28, and SS28(1-12), were used to detect neocortical neurons containing this mRNA or its translation product. The appearance of SS mRNA is coincident with detectable immunoreactivity for SS peptides. The expression of the SS gene by cortical neurons occurs in two waves. From P1 to P11, hybridizing neurons are predominant below the cortical plate in the developing infragranular layers. Immunohistochemical analysis of immunoreactivity to SS28 reveals a significant development of this neocortical system by late gestation (E20). At this point SS28(1-12), the predominant SS form detected, is mainly in neurons of the subplate, with less detectable immunoreactivity in the intermediate zone and cortical plate. By P2, neurons in the subplate exhibit detectable SS28 and SS28(1-12). Although immunoreactive perikarya are no longer detectable at P2 in the cortical plate or marginal zone, a very dense plexus of SS28(1-12) fibers is seen in the subplate, marginal zone, and intermediate zone; relatively few immunoreactive fibers are found in the cortical plate. By P12, a dramatic shift occurs; a large supragranular population of these SS neurons is observed by both mRNA and antibody methods, as is a subsequent decrease in number in the adult. The shift in immunoreactivity occurs with supragranular SS28-containing neurons now prominent, and SS28(1-12)-containing neurons and fibers greatly diminished. The number of neurons containing SS mRNA or SS28 immunoreactivity decreases from P12 to adult, when these neurons exhibit a bilaminar distribution. Neurons immunoreactive for SS28(1-12) are now sparsely distributed throughout the cortex, while SS28(1-12) fibers densely innervate layers I and V/VI.

Connexin43 null mutation increases infarct size after stroke.

Glial-neuronal interactions have been implicated in both normal information processing and neuroprotection. One pathway of cellular interactions involves gap junctional intercellular communication (GJIC). In astrocytes, gap junctions are composed primarily of the channel protein connexin43 (Cx43) and provide a substrate for formation of a functional syncytium implicated in the spatial buffering capacity of astrocytes. To study the function of gap junctions in the brain, we used heterozygous Cx43 null mice, which exhibit reduced Cx43 expression. Western blot analysis showed a reduction in the level of Cx43 protein and GJIC in astrocytes cultured from heterozygote mice. The level of Cx43 is reduced in the adult heterozygote cerebrum to 40% of that present in the wild-type. To assess the effect of reduced Cx43 and GJIC on neuroprotection, we examined brain infarct volume in wild-type and heterozygote mice after focal ischemia. In our model of focal stroke, the middle cerebral artery was occluded at two points, above and below the rhinal fissure. Four days after surgery, mice were killed, the brains were sectioned and analyzed. Cx43 heterozygous null mice exhibited a significantly larger infarct volume compared with wild-type (14.4 +/- 1.4 mm(3) vs. 7.7 +/- 0.82 mm(3), P < 0.002). These results suggest that augmentation of GJIC in astrocytes may contribute to neuroprotection after ischemic injury.

Characterization of CArG-binding protein A initially identified by differential display.

While investigating differences in the pattern of gene expression in functionally distinct areas of the rat caudate-putamen employing differential display, we identified a gene that is highly enriched in tissue adjacent to the lateral ventricle. To characterize the gene, a complementary DNA containing the complete coding sequence was obtained and sequenced. In addition, radiolabelled DNA and riboprobes were generated to examine the expression levels and anatomical distribution of the identified gene in the brain. The sequencing data suggests that the identified gene is a member of the heterogeneous nuclear ribonucleoprotein family and likely represents the rat homolog of CArG-binding protein A initially isolated from mouse C2 myogenic cells. CArG-binding protein A is widely distributed and moderately expressed in the rat brain and present within both neurons and astrocytes. Since the CArG box motif forms the core of the serum response element and the serum response element is involved in immediate early gene regulation, the expression level of CArG-binding protein A was examined following treatment of PC12 cells with nerve growth factor and correlated with changes in c-fos and zif268 expression. The results show that CArG-binding protein A is up-regulated following nerve growth factor treatment and that the up-regulation of CArG-binding protein A can be correlated with the down-regulation of c-fos and zif268. The results of the current study leads us to suggest that CArG-binding protein A may be involved in brain development and the regulation of the serum response element.

Upregulation of gap junction connexin 32 with epileptiform activity in the isolated mouse hippocampus.

Gap junctions, which serve as intercellular channels providing direct cytoplasmic continuity and ionic current flow between adjacent cells, are constituted by connexin proteins. Using an in vitro model of bicuculline-induced epileptiform activity, we asked whether increased connexin levels occur during epileptiform activity in the intact whole hippocampus, freshly isolated from young (15-day-old) mouse brain. Exposure to bicuculline (10 microM), for 2-10 h, induced persistent changes in electrical activities that included enhanced spontaneous field activity (4 h), an epileptiform response to single electrical stimulation (6 h), and spontaneous epileptiform activity (6 h). These electrophysiological changes were not reversed by up to 60 min perfusion with normal artificial cerebrospinal fluid, but were greatly depressed by the gap junction uncoupler, carbenoxolone (120 microM, 10 min). Data from RNase protection assay and immunoblotting showed that among several detected gap junctions, only connexin 32 was affected. After 2-6 h exposure to bicuculline, the connexin 32 mRNA expression was upregulated to 2-3-fold control (P < 0.01), and its protein level was significantly elevated the following 6 h (P < 0.01), at which time electrophysiologically measured evidence of clearly epileptiform activity was apparent. In addition, the transcription factor, c-fos protein, but not the cAMP response element-binding protein, was also found to be increased at the early stage of bicuculline exposure (2 h) compared to control (P < 0.05).Thus, we have found that exposing the acutely isolated hippocampus to bicuculline, induced increased c-fos protein, followed by increased connexin 32 transcript and protein, and concurrently, persistent epileptiform activity that was depressed by carbenoxolone.

Changes in CArG-binding protein A expression levels following injection(s) of the D1-dopamine agonist SKF-82958 in the intact and 6-hydroxydopamine-lesioned rat.

We recently characterized the rat brain homolog of mouse muscle CArG-binding protein A initially identified in C2 myogenic cells and showed an inverse temporal correlation between increased expression levels of this messenger RNA, c-fos and zif268 messenger RNA levels following the addition of nerve growth factor to PC12 cells. In addition, we found an inverse correlation between c-Fos protein and CArG-binding protein A messenger RNA levels in the lateral caudate-putamen of rats treated acutely and chronically with the D2 receptor antagonist fluphenazine (phenothiozine typical psychotic). To determine whether D1 receptor stimulation is also capable of inducing CArG-binding protein A up-regulation, drug naive or dopamine-depleted (i.e. 6-hydroxydopamine-lesioned) D1 hypersensitized rats (i.e. rats given repeated daily injections of SKF-82958 for 14days) were acutely injected with the D1 agonist SKF-82958 and examined using a combination of in situ hybridization for CArG binding protein A and immunocytochemistry for c-Fos. Both acutely treated animals and dopamine-depleted hypersensitized animals showed increases in CArG-binding protein A. Moderate increases were found in the medial caudate-putamen and nucleus accumbens core and shell regions following acute treatment whereas large increases in CArG-binding protein A expression levels were found in the medial and lateral caudate-putamen and the shell and core of the nucleus accumbens following hypersensitization. No change in CArG-binding protein A expression level was found in the dopamine-depleted, drug naive animals relative to controls. Regions of the basal ganglia where increases in CArG-binding protein A were detected following each treatment correlated perfectly with c-Fos protein induction. The results demonstrate that CArG-binding protein A responds to SKF-82958 and that the changes in CArG-binding protein A match perfectly with the pattern of c-Fos induction induced by the D1 agonist.

Effects of graft-derived dopaminergic innervation on the target neurons of patch and matrix compartments of the striatum.

Fetal dopaminergic neurons grafted into the dopamine-depleted striatum have previously been shown to normalize neurochemical and behavioural abnormalities. However, the extent of graft-induced recovery of striatal compartments, which differ in their ontogeny, neurochemical properties and function, is still not clear. The striosome and matrix compartments of the striatum provide a segregated projection to somatostatin-containing GABAergic neurons of the rostral part of the entopeduncular nucleus and somatostatin-negative GABAergic neurons of the caudal part of the entopeduncular nucleus, respectively. In the present study, preprosomatostatin and glutamate decarboxylase messenger RNA levels in the rostral and caudal parts of the entopeduncular nucleus were determined six and 18 months postgrafting in rats with complete recovery of rotational behaviour following apomorphine challenge, and in rats with unilateral 6-hydroxydopamine lesions or sham lesions and no grafts. Sections were processed for in situ hybridization using 35S-labelled cRNA probes for glutamate decarboxylase (67,000 mol. wt isoform; GAD67) and preprosomatostatin. Autoradiographs showed a marked increase in preprosomatostatin messenger RNA within the ipsilateral entopeduncular nucleus in 6-hydroxydopamine-lesioned rats, and a substantially lower increase six months postgrafting. At 18 months postgrafting, the preprosomatostatin messenger RNA levels were symmetrical within the entopeduncular nucleus. Unilateral depletion of striatal dopamine resulted in a moderate increase in GAD67 messenger RNA levels within the ipsilateral entopeduncular nucleus, along with a substantial decrease in GAD67 levels within the contralateral nucleus. By six months postgrafting, the GAD67 levels had decreased considerably within the ipsilateral entopeduncular nucleus, while the messenger RNA levels had returned to normal within the contralateral nucleus. Interestingly, at 18 months postgrafting, the GAD67 levels remained decreased within the ipsilateral entopeduncular nucleus and were significantly lower than the normal value. The results indicate that fetal nigral grafts placed within the dopamine-depleted striatum can restore the neurochemical alterations seen in striatal target areas such as the entopeduncular nucleus. This may form the neurochemical basis of graft-induced behavioural recovery, as the normalization of neurotransmitter messenger RNA levels in the entopeduncular nucleus reflects the restoration of overall activity in both direct and indirect striatal output pathways. The results also indicate that the graft-derived dopaminergic innervation restores the output of both striosome and matrix compartments of the striatum. The present results also showed a progressive recovery leading to over-compensation of neurotransmitter messenger RNA levels following grafting, perhaps indicating the importance of feedback regulation of grafted dopaminergic neurons by the host.

A pre-loading method of evaluating gap junctional communication by fluorescent dye transfer.

We describe a simple method for evaluating gap junctional communication (GJC) between cells in culture. The procedure involves pre-loading cells with two fluorescent dyes: calcein and DiI. Calcein is able to pass through gap junctions, while DiI is not. These pre-loaded cells are then plated with unlabeled cells. The number of cells receiving calcein from each pre-loaded cell can then be quantified after the cells settle on the plate. Potent and reversible inhibitors of GJC can be used in this system to evaluate dye transfer within a given period of time.

A neuroprotective role for gap junctions.

Glial-neuronal interactions have been implicated in both normal information processing and neuroprotection. One pathway of cellular interactions involves gap junctional intercellular communication (GJIC). In astrocytes, gap junctions are composed primarily of the channel protein, connexin43 (Cx43), and provide a substrate for formation of a functional syncytium implicated in the process of spatial buffering in the CNS. Thus gap junctional communication may be neuroprotective following a CNS insult that entails glutamate cytotoxicity (i.e. ischemia). We have shown that blocking gap junctions during a glutamate insult to co-cultures of astrocytes and neurons results in increased neuronal injury. To assess the effect of reduced Cx43 and GJIC on neuroprotection, we examined brain infarct volume in wild type and Cx43 heterozygote null mice following focal ischemia. Cx43 heterozygous null mice exhibited a significantly larger infarct volume compared to wild type. At the cellular level, a significant increase in TUNEL positive cells was observed in the penumbral region of the Cx43 heterozygote mice. These results suggest that augmentation of GJIC in astrocytes may contribute to neuroprotection following ischemic injury. These findings support the hypothesis that gap junctions play a neuroprotective role against glutamate cytotoxicity.