In response to injury and degeneration, astrocytes hypertrophy, extend processes, and increase production of glial fibrillary acidic protein (GFAP), an intermediate filament protein located within their cytoplasm. The present study tested the hypothesis that GFAP expression alters the vulnerability of neurons to excitotoxic and metabolic insult induced by 3-nitroproprionic acid (3-NP), an irreversible inhibitor of mitochondrial complex II activity or the excitotoxin quinolinic acid (QA). In this respect, adult GFAP knockout mice (KO) and wild-type control mice (WT) received unilateral intrastriatal injections of 3-NP (200 nmol/microl) or QA (100 nmol/microl) and were killed 1, 2, or 4 weeks later. Lesion volume and neuronal counts were quantified using unbiased stereologic principles. For both QA and 3-NP lesions, a significant decrease in lesion volume and an increase in striatal projection neurons were seen in GFAP KO mice compared with WT mice. Enzyme-linked immunoassay analysis revealed increased basal levels of glial cell derived neurotrophic factor (GDNF) relative to WT mice. In contrast, no differences were observed in the expression of ciliary neurotrophic factor or nerve growth factor. These data strongly suggest that the expression of GFAP is implicated with the production of GDNF to a degree that confers neuroprotection after an excitotoxic or metabolic insult.

Although estrogen has been shown to be neuroprotective, studies concerning its effect on some behaviors are contradictory, reporting both ameliorative and detrimental effects. A factor involved in hormone efficacy is the estrogen regimen. We reported an effect of 10 µg estrogen for 14 days on the cyclic AMP response element-binding protein (CREB) pathway, including brain-derived neurotrophic factor, in rat medial amygdala (MeA). To determine the effects of estrogen on neuronal numbers and brain region volume in MeA and central nucleus of the amygdala (CeA), we used stereology to test the effect of various estrogen regimens on the number of neuron-specific protein (NeuN)-labeled neurons and brain region volume of MeA and CeA. Ovariectomized rats were injected with vehicle for 14 days, 2.5 µg estradiol benzoate (E2) for 4 or 14 days, or 10 µg estrogen for 14 days. Because NeuN-labeled neuronal number may be related to neuronal survival and upregulation of CREB signaling, we tested the effect of these regimens on levels of phosphorylated CREB (pCREB) labeling in the MeA and CeA. The 2.5 µg estrogen for 14 days regimen increased the mean number of NeuN-labeled neurons and pCREB-labeled cells in the MeA compared to vehicle or 2.5 µg for 4 days. There was an increase in volume of the MeA with 2.5 µg estrogen for 14 days compared to vehicle or 2.5 µg for 4 days. No differences in these parameters were seen in CeA. These data indicate a neuroanatomical heterogeneity of a time effect of estrogen on cells expressing NeuN and pCREB in the MeA versus CeA.

The neuroprotective properties of cyclosporin A (CsA) are mediated by its ability to prevent mitochondrial permeability transition during exposure to high levels of calcium or oxidative stress. By using the mitochondrial toxin 3-nitropropionic acid (3NP), the present study assessed whether CsA could protect striatal neurons in vitro and in vivo. In vitro, 3NP produced a 20-30% reduction of striatal glutamic acid decarboxylase-immunoreactive (GAD-ir) neurons. A single treatment with CsA protected GAD-ir neurons from 3NP toxicity at lower (0.2 or 1.0 microM), but not at higher (5.0 microM) doses. Similar findings were seen when the cultures were treated twice with cyclosporin. In vivo experiments used the Lewis rat model of Huntington's disease (HD) in which a low 3NP dose was delivered subcutaneously through an osmotic minipump. Rats received unilateral or bilateral intrastriatal saline injections to disrupt the blood-brain barrier (BBB) and facilitate CsA reaching vulnerable neurons. In the first experiment, CsA treated 3NP-lesioned rats displayed significantly more dopamine-and adenosine-3;, 5;-monophosphate-regulated phosphoprotein (DARPP32-ir) neurons ipsilateral to BBB disruption compared to the contralateral intact striatum, indicating that disruption of the BBB maybe necessary for CsA's neuroprotective effects. In the second experiment, stereological counts of DARPP32-ir neurons revealed that CsA protected striatal neurons in a dose-dependent manner following bilateral disruption of the striatal BBB. Rats treated with the higher (15 or 20 mg/kg) but not lower (5 mg/kg) doses of CsA displayed greater numbers of DARRP32-ir striatal neurons relative to vehicle-treated 3NP-lesioned rats. Thus, under conditions in which CsA can gain access to striatal neurons, significant protection from 3NP toxicity is observed. Therefore, CsA or more lipophilic analogues of this compound, may be of potential therapeutic benefit by protecting vulnerable neurons from the primary pathological event observed in HD.