Human rotation-mediated fetal mixed brain cell aggregate culture: characterization and N-methyl-D-aspartate toxicity

Experimental in vivo and in vitro models of multiple sclerosis: EAE and beyond

Although the primary cause of multiple sclerosis (MS) is unknown, the widely accepted view is that aberrant (auto)immune responses possibly arising following infection(s) are responsible for the destructive inflammatory demyelination and neurodegeneration in the central nervous system (CNS). This notion, and the limited access of human brain tissue early in the course of MS, has led to the development of autoimmune, viral and toxin-induced demyelination animal models as well as the development of human CNS cell and organotypic brain slice cultures in an attempt to understand events in MS. The autoimmune models, collectively known as experimental autoimmune encephalomyelitis (EAE), and viral models have shaped ideas of how environmental factors may trigger inflammation, demyelination and neurodegeneration in the CNS. Understandably, these models have also heavily influenced the development of therapies targeting the inflammatory aspect of MS. Demyelination and remyelination in the absence of overt inflammation are better studied in toxin-induced demyelination models using cuprizone and lysolecithin. The paradigm shift of MS as an autoimmune disease of myelin to a neurodegenerative disease has required more appropriate models reflecting the axonal and neuronal damage. Thus, secondary progressive EAE and spastic models have been crucial to develop neuroprotective approaches. In this review the current in vivo and in vitro experimental models to examine pathological mechanisms involved in inflammation, demyelination and neuronal degeneration, as well as remyelination and repair in MS are discussed. Since this knowledge is the basis for the development of new therapeutic approaches for MS, we particularly address whether the currently available models truly reflect the human disease, and discuss perspectives to further optimise and develop more suitable experimental models to study MS.

Characterization of long-term mouse brain aggregating cultures: Evidence for maintenance of neural precursor cells

An extensive characterization of fetal mouse brain cell aggregates has been performed using immunohistochemical and stereological methods. Single cell suspensions from mechanically dissociated cortex and hippocampus were cultured in serum-free, B27-supplemented medium under constant gyratory agitation for up to 56 days. Three-dimensional aggregates started to form immediately after seeding and reached a final average size of 500 microm in diameter. Among the cell types identified, neurons were the most abundant cells in the aggregates, followed by astrocytes, microglia, and oligodendrocytes. Western blotting for synaptophysin and immunostaining for neurotransmitter-related molecules indicated the presence of well-defined phenotypic characteristics of the neurons in this culture system, suggesting functionality. Proliferating cells, many with neural precursor cell properties, were seen throughout the culture period and could be isolated from the aggregates even after 2 months in culture. Neural precursor cells were isolated from the aggregates after more than 1 month in culture; these cells were successfully differentiated into neurons, astrocytes, and oligodendrocytes. The aggregate culture system may provide a versatile tool for molecular dissection of processes identified in mouse models, including transgenic animals and manipulation of neural precursor cells.

Sequestration of nNOS in neurofilamentous aggregate bearing neurons in vitro leads to enhanced NMDA-mediated calcium influx

The significance of copper/zinc superoxide dismutase (SOD1) and neuronal nitric oxide synthase (nNOS) co-localization to neurofilamentous (NF) aggregates in amyotrophic lateral sclerosis (ALS) is unknown. In this study, we have used dissociated motor neurons from either C57BL/6 or mice that over-express the human low molecular weight neurofilament protein (hNFL+/+) to examine the relationship between NF aggregate formation, SOD1 and nNOS co-localization, and the regulation of NMDA-mediated calcium influx in vitro. The intracellular distribution of NF aggregates, SOD1 and nNOS was examined by confocal microscopy and NMDA-induced alterations in intracellular calcium levels using either Oregon green fluorescence or FURA-2 photometric imaging. Cell death was assessed using an antibody to activated caspase-3. C57 Bl/6 motor neurons expressed nNOS in a punctate manner, whereas SOD1 was distributed homogeneously throughout the cytosol. In contrast, hNFL+/+ motor neurons demonstrated co-localization of SOD1 and nNOS by day 9 post-plating, preceding the formation of NF aggregates. Both proteins co-localized to NF aggregates once formed. With NMDA stimulation, aggregate-bearing hNFL+/+ motor neurons demonstrated significant increases in intracellular calcium, whereas only a minimal alteration in intracellular calcium was observed in C57 Bl/6 neurons. Following stimulation with 100 microM NMDA, 75.5+/-5.5% of hNFL+/+ neurons became apoptotic, whereas only 16.3+/-5.3% of C57 Bl/6 were. These observations suggest that the presence of NF aggregates results in a failure of regulation of NMDA-mediated calcium influx, and that this occurs due to the sequestration of nNOS to the NF aggregate, preventing its down-regulation of the NMDA receptor.

Infection of stationary human brain aggregates with HIV-1 SF162 and IIIB results in transient neuronal damage and neurotoxicity

The cellular basis of HIV associated dementia has been correlated with microglial activation and neuronal dysfunction in symptomatic HIV-1 disease. As a cellular model of HIV-1 infection of brain tissue in vitro, we established a stationary human brain aggregate (SHBA) system to compare infection of HIV-1 SF162 (R5 virus) to that of IIIB (X4 virus). Aggregates were analysed by immunohistochemistry, morphometry, flow cytometry and p24 ELISA. SHBAs had a 1 mm(3) size with a mixed cellular composition of 36% neurones, 27% astrocytes, 2% macrophages/microglia and 14% oligodendrocytes. Infection of SHBA's with the R5 HIV-1 SF162 virus led to the expression of HIV-1 p24 antigen in 6% of cells. Infection with this R5 using virus culminated in transient neuronal damage and a decrease in mitotically active progenitor cells within aggregates. Infection with X4 using HIV-1 IIIB was associated with astrocytosis and neurotoxicity. We propose that: (1) the pattern of cellular damage elicited by HIV-1 infection of brain tissue in vitro depends on virus subtype as determined by its preferential use of R5 or X4 chemokine receptors for entry into cells; (2) SHBAs are a reliable and readily established model of the cellular complexity of human brain tissue in vitro.

Production of beta-amyloid by primary human foetal mixed brain cell cultures and its modulation by exogenous soluble beta-amyloid

Previous studies on beta-amyloid production have been carried out using transfected cells and cell lines. We measured the 40 and 42 amino acid forms of beta-amyloid released into the culture medium by primary human foetal mixed brain cell aggregate culture over 3 months. In this model, neurones and supporting cells are maintained in serum-free defined medium. The secretion of significant amounts of beta-amyloid 40 and 42 was observed throughout culture for three separate cultures. Levels of beta-amyloid 40 and 42 closely followed the neuronal content of the cultures as estimated by cellular neurone-specific enolase. Addition of synthetic beta-amyloid 1-40 to the cultures for 1 week at 35 days in vitro resulted in a dose-related reduction in cellular neurone-specific enolase levels. Primary human aggregate brain cell cultures produced multimeric beta-amyloid, as determined by immunoassay. beta-Amyloid-treated cultures released diminishing amounts of multimeric beta-amyloid and contained increasing amounts of intracellular multimeric beta-amyloid with increasing exogenous beta-amyloid. These results suggest that release of multimeric beta-amyloid into the extracellular environment by human primary neurones can be affected by the presence of extracellular beta-amyloid. This has implications for Alzheimer's disease in that beta-amyloid released into the extracellular environment by dead/dying neurones could modulate beta-amyloid release by surrounding neurones, potentially causing amplification of toxicity. Moreover, intracellular beta-amyloid oligomer-dependent neurotoxicity may be a component of neurodegeneration in Alzheimer's disease, and other conditions with increased beta-amyloid synthesis, suggesting anti-amyloid therapies for Alzheimer's disease may have to target intracellular beta-amyloid.

A reliable primary human CNS culture protocol for morphological studies of dendritic and synaptic elements

Primary dissociated human fetal forebrain cultures were grown in defined serum-free conditions. At 4 weeks in vitro the cultures contained abundant morphologically well differentiated neurons with complex dendritic arbors. Astrocytic proliferation was negligible without the use of antimitotic agents. Confocal scanning laser microscopy (CSLM) and electron microscopy confirmed the presence of a dense neuropil, numerous cell-cell contacts and synapses. Neurons expressed a variety of proteins including growth associated protein-43 (GAP43), microtubule associated protein-2ab (MAP), class-III beta tubulin (C3BT), neurofilaments (NF), synaptophysin (SYN), parvalbumin (PA) and calbindin (CB). The cultures have proven to be reliable and simple to initiate and maintain for many weeks without passaging. They are useful in investigations of dendritic growth and injury of primary human CNS neurons.