Astrocytes account for 50% of the total cellular volume [1,2] and 90% of the number of brain cells. These cells were traditionally thought to provide structural support for neuronal growth and differentiation and to remove toxic metabolites from the extracellular environment [1,2,3]. The astrocytes are also known to interact with the endothelial cells which form the blood brain barrier (BBB) and regulate the flow of chemicals in and out of the brain [1,2]. In addition to providing structural support, astrocytes also participate in many other processes crucial for the brain through their abundance and diverse ion channels, receptors, and transporters for neurotransmitters and neuromodulation [4-5].

Neurotoxicity caused by excitatory amino acids (EAAs) such as glutamate contributes to the pathology of the central nervous system (CNS), including neurodegenerative disease, ischemia and trauma. Glutamic acid and /or its structural analogues such as domoic (DOM) or kainic acid (KA) have been proposed to mediate brain toxicity through their interaction with EAA receptors collectively known as the glutamate receptors (GluRs). Sites often damaged are the hypothalamus by glutamate [6], the hippocampal and neocortical pyramidal neurons by domoic acid [6], the cerebellar Purkinje neurons by ibogaine [6-7] and the corpus striatum by 3-Nitropropionic Acid (3-NPA) and amphetamine [6]. The excitotoxic damage presents as neuronal loss and neuronal degeneration with characteristic cytoplasmic vacuolation and swollen dendrites. The regional distribution of the neuronal lesions reflects that of the GluR distribution. Astrocytes with swollen perikarya and karyorrhexis have often been observed in the vicinity of these lesions [8-9]. It has been suggested that astrocytes contribute to a decline of neurologic function by either the accumulation and release of EAAs after ischemia and oxidative stress or by formation of epileptogenic scars in response to CNS injury and metabolism of protoxins to potent toxins. Recent evidence shows that in response to invasion by microorganisms, both astrocytes and microglia mount an immune response by releasing cytokines [10-13]. Although it is known that astrocytes and microglia are involved in the initiation, maintenance, and suppression of immune responses, they have also been shown to propagate CNS damage [10-13]. Therefore, these cells may also be involved in neurodegenerative disorders. Astrocyte activation (reactive gliosis) and neuroinflammation are known to accompany neurotrauma, stroke, neurodegenerative disease or tumors [14]. Since reactive gliosis represents a homotypic response to diverse insults of the CNS, it is assumed that neuroinflammation is induced by a common set of signals. Cytokines, neurotrophins, and neurohormones are potent biological response modifiers that exhibit a spectrum of cellular actions associated with a variety of neurotoxic exposures that induce neuroinflammation [10,15-17]. The signal transduction pathways and gene-activation events that link these chemical messengers to glial activation are not yet fully characterized.

Several mechanisms have been implicated as mediators for the effects of DOM [18,19]. These include the potential role played by astrocytes [20]. Acute toxicity studies, using single parenteral doses of DOM, have shown light microscopy evidence of cell injury of the astrocytes in several brain regions [21]. Here we provide additional morphologic evidence of the effects of DOM on astrocytes as assessed by electron microscopy after subchronic oral administration of DOM in rats. We also further explore the pathogenesis of the glial lesion induced by DOM and regional functionality using in vitro astrocytes isolated from the hippocampus and the brain stem. Currently, limited information is available on the molecular events altered by DOM that may affect the function of astrocytes. Ultimately, the characterization of the early cellular events subsequent to DOM intoxication will yield new insights for the rational development of novel therapeutic interventions for the prevention and treatment of amnesic shellfish poisoning and for other brain pathologies involving excitotoxicity.

 Electron microscopy revealed morphological changes in the CA 3 region of the hippocampus of animals treated for 64 days with 5.0 mg/kg/day of DOM as compared to the control and to animals receiving 0.1 mg/kg/bw. Various degrees of degenerating astrocytes from minimal vacuolation of the cytoplasm to necrosis were observed (Figure 1).

To further evaluate the effects of DOM on astrocytes, primary astrocytes were cultured from two different regions of the brain focusing on the brain stem and the hippocampus. The MTT assay shows that the astrocytes from the brain stem appeared to be more sensitive to higher (>10 uM) concentrations of DOM than those astrocytes from the hippocampus (Figure 2). Hence, 10 μM of DOM was used for further experiments. After 24 hrs of treatment with DOM, the total RNA was extracted and used for transcriptional analysis. The initial analysis was done on i). the early injury genes (EIGs) including c-fos, c-jun, c-myc, Hsp-72; ii). astrocyte specific markers-GFAP; iii). the glutamate receptor subtypes - GluR 2, NMDAR 1, NMDAR 2A, NMDAR 2B and iv). the proliferating cell nuclear antigen (PCNA).

The c-jun was down-regulated whereas c-fos and Hsp-72 and the astrocyte specific marker (GFAP) were not affected. The glutamate receptor subtype GluR 2 was down-regulated over 2-fold, (Figure 3), expression of NMDAR 1, 2A and 2B were not detected. These observed changes were only in the astrocytes cultured from the hippocampus.

To get a global view of the transcriptional changes, three microarrays RO 1 DNA chips were used for each analysis and the putative positive genes were confirmed using semi-quantitative PCR analysis. From the DNA chip analysis, only the chemokines/cytokines, tyrosine kinases and apoptotic genes showed either up- or down-regulated gene expression. This analysis was supplemented with other genes associated with these pathways absent from the DNA chips. The cytokines were all up-regulated significantly as confirmed by the t-tests. These included IL-α, IL-β, IL-6, small inducible cytokine, interferon protein 10P-10, CXC chemokine LIX and IGF binding proteins (Figure 4). Additional genes which were examined including the transforming growth factor (TGF- α and Β), tumor necrosis factor (TNF-α), nerve growth factor (NGF), MIP-1 and IL-7. Of these, only the TNF-α and TGF- β were affected (Table 1). The expression of the other genes were not affected by the exposure to DOM. Other cytokines such as TGF-α , MIP-1, IL-7 and NGF were not induced after exposure to DOM.

Two specific membrane receptors, Trk A and Trk B were down-regulated 3-fold, whereas, Trk C was unaffected. In addition, the apoptotic genes- Bcl-2 and Bax were down-regulated 2 and 3 fold, respectively, whereas, Bad and caspase-3 were not affected (Figure 3).

The oxidation markers- gamma-glutamylcysteine synthetase (GCS), glutathione-S-transferase (GST), glutathione peroxidase (GPx), cyclooxygenase (COX-2), catalase and superoxide dismutase (Zn and Mn dependent) were not affected.

Although several mechanisms have been implicated on the neurotoxic effect of DOM, the pathogenesis of the glial lesion has received very little attention. This is despite the fact that regardless of the cause of neuronal damage, reactive glial cells always appear at and around the site of degeneration. Earlier studies have shown in addition to the neurons being affected [26], various degrees of injury was observed in hippocampal astrocytes in animals that had received subchronic oral administration of DOM. The present study demonstrated the ultra-structural morphologic changes are similar to those observations reported by light microscopy in acute toxicity studies [21]. In support of the in vivo studies, we also conducted in vitro studies using primary astrocytes from two different brain regions- the hippocampus and the brain stem. These cells were exposed 24 hrs to DOM and a systematic analysis of the different pathways that have been implicated in neurotoxicity were examined. Two different approaches were used- first, we examined the effects on the known genes that have been reported to be affected by DOM exposure, which included the EIGs, astrocytes specific marker (GFAP), and the glutamate receptors (AMPA and the kainate subtypes). Secondly, we used microarray technology to further elucidate the other pathways that might be involved. All the transcriptional changes were observed only in the astrocytes from the hippocampus and not from the brain stem.

The GFAP biomarker of gliosis [27] was not changed in either of the in vivo [22,26] or the in vitro studies conducted here. Of the EIG genes, only c-jun was decreased 2-fold, whereas the c-myc, c-fos and the Hsp72 were not altered. Previous studies with neuron cultures have shown that c-fos reaches a peak after 6 hrs of exposure to the stimulus and then declines, whereas c-jun is activated after 24 hrs of exposure and is followed by the induction of Hsp-72 [28,29]. The lack of altered expression in GFAP and some of the EIGs, is simply due to the fact that the expression of these is transient and measurements were done outside of the relevant time window [28,29].

 DOM is structurally similar to the excitatory neurotransmitter glutamate and glutamate is known to exert its affect by binding to and activating the ionotropic glutamate receptors (iGluRs) including the subtypes NMDAR 1, GluR1, GluR 2, GluR 5, and GluR 6 in the neurons. Metabotropic receptors are not involved in DOM toxicity [30]. In our study with astrocytes, only GluR 2 was increased 2-fold, whereas no expression was detected for any of the NMDAR subtypes.

 From the microarray analysis and confirmed by semi-quantitative analysis, the chemokines / cytokines, tyrosine kinases (Trk), and apoptotic genes showed either up- or down-regulated gene expression. Several pro- and anti-inflammatory cytokines including TNF-α, IL-1α, IL-1B, IL-6, CXC chemokine LIX, interferon protein 10P-10, small inducible cytokine and IGF binding protein were up-regulated 2x fold or higher. However, not all cytokines were altered since cytokines are regulated in cascades. The IL-1Β and TNF-α are known to initiate the early events which are pleiotrophic and trigger general responses that are not specific for the initiating stimulus or antigen [15] and these were both up-regulated. The cytokines act by binding to specific membrane receptors, which signal the cell via second messengers, such as the tyrosine kinases (Trk), to alter its gene expression. The Trk A and B were down-regulated 3-fold, whereas, Trk C was unaffected. The apoptotic genes, Bcl-2 and Bax were altered, whereas, BAD and caspase-3 were not affected. The R01 chip also had other neural markers including the different ion channels such as potassium and sodium, γ-aminobutyric acid, cholinergic and the serotonin receptors. None of these were changes.

The oxidation stress markers are known to occur at the site of inflammation, these can also contribute to damage. The activation of astrocytes can lead to the expression of inducible nitric acid oxide synthase (inos), superoxides and nitric oxide which are involved in the removal of free radicals. In our study, the GSTα, GCS, catalase, superoxide dismutase (Zn and Mn dependent), inducible nitric oxide (inos) and COX-2 were not altered after 24 hrs exposure to DOM.

In our study the injury to the astrocytes was characterized by ultra-structural changes, inflammatory,immuno, growth factor, cytokines, and changes in the cell death genes. It has been shown that the activation of the astrocytes is an important defence mechanism. Localized production of cytokines and chemokines has been shown in many debilitating neurologic disorders such as multiple sclerosis, Alzheimer's disease and AIDS [12,16,31]. As the brain ages, inflammatory events become less well-regulated and hence environmental exposure can exacerbate the endogenous heightened CNS inflammation [32]. Many chemokines are capable of directly regulating signal-transduction pathways that are involved in a variety of cellular functions, which range from synaptic transmission to growth. Clearly, the potential use of chemokines and their receptors as targets for therapeutic intervention in CNS disease might now have to be considered in the context of the broader physiological functions of these molecules. Therefore, an understanding of the different interactions between inflammatory mediators leading to restricted local reaction is important in the development of novel therapeutics for inflammatory CNS diseases [33,34,35]. This is facilitated further by the fact that astrocytes from different regions show functional heterogeneity. Recently, Tomita et al. [36] showed that the spice curcumin strongly inhibited the transcription of the cytokine MIP-2 which is mainly produced by the astrocytes. Ananth et al. [37] demonstrated that DOM-induced astrogliosis and neurodenegeration were reduced after 5 days of melatonin administration. Hence, it might be possible to use anti-inflammatory intervention to slow down the progression of many different forms of neurodegenerative diseases.