What are artificial neural networks?

Artificial neural networks have been applied to problems ranging from speech recognition to prediction of protein secondary structure, classification of cancers and gene prediction. How do they work and what might they be good for?     

A new 'spin' on neural stem cells?

The existence of neural stem cells in the adult brain was essentially denied until the last decade. Within the past ten years, considerable progress has been made in examining the fundamental properties of neural stem cells. Most recently there has been much interest in the identification and precise location of the adult neural stem cells in vivo. Studies examining the localization of neural stem cells are controversial and suggest two distinct locations within the adult brain: the ependymal layer lining the ventricles, and the subependymal layer immediately adjacent to the ependyma.     

Converted neural cells: induced to a cure?

Many neurodegenerative disorders such as Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS) and others often occur as a result of progressive loss of structure or function of neurons. Recently, many groups were able to generate neural cells, either differentiated from induced pluripotent stem cells (iPSCs) or converted from somatic cells. Advances in converted neural cells have opened a new era to ease applications for modeling diseases and screening drugs. In addition, the converted neural cells also hold the promise for cell replacement therapy (Kikuchi et al., 2011; Krencik et al., 2011; Kriks et al., 2011; Nori et al., 2011; Rhee et al., 2011; Schwartz et al., 2012). Here we will mainly discuss most recent progress on using converted functional neural cells to treat neurological diseases and highlight potential clinical challenges and future perspectives.     

Gene therapy: can neural stem cells deliver?

Neural stem cells are a self-renewing population that generates the neurons and glia of the developing brain. They can be isolated, proliferated, genetically manipulated and differentiated in vitro and reintroduced into a developing, adult or pathologically altered CNS. Neural stem cells have been considered for use in cell replacement therapies in various neurodegenerative diseases, and an unexpected and potentially valuable characteristic of these cells has recently been revealed--they are highly migratory and seem to be attracted to areas of brain pathology such as ischaemic and neoplastic lesions. Here, we speculate on the ways in which neural stem cells might be exploited as delivery vehicles for gene therapy in the CNS.     

TGFβ superfamily signaling in the neural crest lineage

The neural crest cell (NCC) lineage is often referred to as the fourth germ layer in embryos, as its wide range of migration and early colonization of multiple tissues and organ systems throughout the developing body is astounding. Many human birth defects are thought to have their origins within the NCC lineage. Exciting recent conditional mouse targeting and transgenic combinatorial suppression approaches have revealed that the TGFβ superfamily is a key signaling pathway within the cardiac and cranial NCC subpopulations. Given the complexity of TGFβ superfamily signaling and that multiple ligand and receptor combinations have already been shown to be expressed within the NCC subpopulations, and the difficulty in transgenically targeting entire signaling cascades, we review several up-to-date transgenic approaches that are revealing unexpected consequences.Key words: TGFβ, neural crest, heart, cranial crest, mouse transgenics     

The Notch—Hes pathway in mammalian neural development

W wide variety of neurons and glial cells differentiate from common precursor cells in the developing nervous system.During this process,Notch-mediated cell-cell interaction is essential for maintenance of dividing cells and subsequent generation of cell type diversity.Activation of Notch inhibits cellular differentiation,and abnormality of the Notch pathway leads to premature neuronal differentiation,the lack of some cell types,and severe defects of tissue morphogenesis.Recent data demonstrate that Notch fails to inhibit cellular differentiation in the absence of the bHLH genes Hes1 and Hes5,which functionally antagonize the neuronal bHLH genes such as Mash1.These results indicate that the two Hes genes are essential effectors for the Notch pathway and that neuronal differentiation is controlled by the pathway “Notch→Hes1/Hes5-Mash”.     

Spontaneous symmetry breaking in self–organizing neural fields

We extend the theory of self-organizing neural fields in order to analyze the joint emergence of topography and feature selectivity in primary visual cortex through spontaneous symmetry breaking. We first show how a binocular one-dimensional topographic map can undergo a pattern forming instability that breaks the underlying symmetry between left and right eyes. This leads to the spatial segregation of eye specific activity bumps consistent with the emergence of ocular dominance columns. We then show how a 2-dimensional isotropic topographic map can undergo a pattern forming instability that breaks the underlying rotation symmetry. This leads to the formation of elongated activity bumps consistent with the emergence of orientation preference columns. A particularly interesting property of the latter symmetry breaking mechanism is that the linear equations describing the growth of the orientation columns exhibits a rotational shift-twist symmetry, in which there is a coupling between orientation and topography. Such coupling has been found in experimentally generated orientation preference maps     

LewisX: A neural stem cell specific glycan?

LewisX (LeX) detecting antibodies are routinely used for cell sorting of neural stem- and progenitor cells (NSPCs). Applications include the enrichment of NSPCs after neural differentiation of human induced pluripotent- or embryonic stem cells, as well as their direct isolation from mouse neural tissue. Nevertheless, only little is known about the role of LeX in the central nervous system. Here we review the current knowledge on LeX-containing glycans expressed by neural stem cells and their progeny. New LeX-carrier proteins and ligands have recently been identified which reveal further insights into the potential function(s) of LeX-glycans. Moreover, evidence accumulates that individual LeX detecting antibody clones vary in their suitability as neural stem cell specific biomarker. Each antibody clone detects a unique LeX-containing glycan epitope. This allows a versatile utilization of anti-LeX antibodies that goes beyond neural stem cell sorting applications.     

Downregulation of Rho-GDI γ promotes differentiation of neural stem cells

Rho-GDIγ belongs to the Rho-GDI protein family, which was observed to have high level expression in the entire brain. Although it exists in neuronal population, its physiological function is poorly understood. This study shows that Rho-GDIγ is a key factor in the G13 signaling pathway based on an analysis of global gene expression. By using RNAi technology to downregulate expression of Rho-GDIγ we found distinct morphological changes in neural stem cell line C17.2. More important, RT-PCR confirmed that RNAi-mediated downregulation of Rho-GDIγ decreased expression of Rho-GDIγ-regulated genes RhoA, Cdc42, Limk2, and N-WASP and slightly increased expression of Rac1. Further, immunochemical staining indicated that downregulation of Rho-GDIγ increased the tendency of C17.2 cells to differentiate. These data strongly suggest that Rho-GDIγ plays a key role in the differentiation of neural stem cells.     

What are the molecular mechanisms of neural tube defects?

Neural tube defects (NTDs) are some of the most common human malformations. The vast majority of NTDs can be prevented by the administration of folic acid; however, to date there has been no effective treatment of folic acid-resistant NTDs. A recent paper¹ has confirmed an earlier report² that the administration of inositol to the curly tail mutant mouse, which is a model of folate-resistant NTDs, can cure such defects. The molecular pathway by which this is achieved is thought to occur by the up-regulation of the retinoic acid receptor β in the underlying hindgut endoderm, correcting a proliferation defect. However, alternative explanations also may account for NTDs. BioEssays 20:6–8, 1998. © 1998 John Wiley & Sons, Inc.     

How robust are neural network models of stimulus generalization?

Artificial feed-forward neural networks are commonly used as a tool for modelling stimulus selection and animal signalling. A key finding of stimulus selection research has been generalization: if a given behaviour has been established to one stimulus, perceptually similar novel stimuli are likely to induce a similar response. Stimulus generalization, in feed-forward neural networks, automatically arises as a property of the network. This network property raises understandable concern regarding the sensitivity of the network to variation in its internal parameter values used in relation to its structure and to its training process. Researchers must have confidence that the predictions of their model follow from the underlying biology that they deliberately incorporated in the model, and not from often arbitrary choices about model implementation. We study how network training and parameter perturbations influence the qualitative and quantitative behaviour of a simple but general network. Specifically, for models of stimulus control we study the effect that parameter variation has on the shape of the generalization curves produced by the network. We show that certain network and training conditions produce undesirable artifacts that need to be avoided (or at least understood) when modelling stimulus selection.     

Direct or indirect? Graphical models for neural oscillators.

Univariate and bivariate time series analysis techniques have enabled new insights into neural processes. However, these techniques are not feasible to distinguish direct and indirect interrelations in multivariate systems. To this aim multivariate times series techniques are presented and investigated by means of simulated as well as physiological time series. Pitfalls and limitations of these techniques are discussed.