one or another of these UTRs were found in the rat and all were spliced to the second exon (called exon D) containing the ATG initiator. Thus, transcription of the REST gene is likely to be initiated upstream of each of the three alternative 5'UTRs. This raises the possibility that there may be at least three promoters within REST, the usage of which might be context-dependent. In the present study, we identify and characterize three clusters of transcription initiation sites and three promoters within the 5' end of the mouse REST gene (mREST), investigate the expression in different cell types of exons A, B, and C encoding alternative 5' untranslated regions (UTRs), and delineate certain positive and negative regulatory regions that overlap with the multiple promoter activities of mREST.

Mouse B lymphocytes that were specifi- cally bound to dinitrophenylated bovine serum albumin on nylon fibers exhibited continuous morphological changes, whereas bound T lymphocytes remained more or less spherical. Cinematomicrographic studies showed that the shape changes were associated with local and global move- ments, although the attached cells did not translocate along the fiber. Cap formation induced by anti-immuno- globulin was always found to be opposite to the point of attachment. The movements and the shape changes were prevented by cytochalasin B and colchicine. Treatment with these agents did not prevent cap formation but led to randomization of the position of the caps with respect to the fiber. Exposure to concanavalin A or attachment of cells to concanavalin A fibers prevented both movement and patch and cap formation, suggesting that cellular structures regulating the mobility of various receptors are altered by binding to concanavalin A fibers. These observa- tions also indicate that interactions of local areas of the lymphocyte surface with certain ligands and substrates can strongly affect the movement and morphology of the entire cell. The relationship of cell movement to cell surface events is poorly understood (1). In studying this problem, a number of factors can be distinguished: local movement of cell surface receptors (diffusion and patch formation), global movements

Recent evidence suggests that neurons in primary sensory cortex arrange into competitive groups, representing stimuli by their joint activity rather than as independent feature analysers. A possible explanation for these results is that sensory cortex implements attractor dynamics, although this proposal remains controversial. Here we report that fast attractor dynamics emerge naturally in a computational model of a patch of primary visual cortex endowed with realistic plasticity (at both feedforward and lateral synapses) and mutual inhibition. When exposed to natural images (but not random pixels), the model spontaneously arranges into competitive groups of reciprocally connected, similarly tuned neurons, while developing realistic, orientation-selective receptive fields. Importantly, the same groups are observed in both stimulus-evoked and spontaneous (stimulus-absent) activity. The resulting network is inhibition-stabilized and exhibits fast, non-persistent attractor dynamics. Our results suggest that realistic plasticity, mutual inhibition and natural stimuli are jointly necessary and sufficient to generate attractor dynamics in primary sensory cortex.

Mental imagery occurs "when a representation of the type created during the initial phases of perception is present but the stimulus is not actually being perceived." How does the capability to perform mental imagery arise? Extending the idea that imagery arises from learned associations, we propose that mental rotation, a specific form of imagery, could arise through the mechanism of sequence learning-that is, by learning to regenerate the sequence of mental images perceived while passively observing a rotating object. To demonstrate the feasibility of this proposal, we constructed a simulated nervous system and embedded it within a behaving humanoid robot. By observing a rotating object, the system learns the sequence of neural activity patterns generated by the visual system in response to the object. After learning, it can internally regenerate a similar sequence of neural activations upon briefly viewing the static object. This system learns to perform a mental rotation task in which the subject must determine whether two objects are identical despite differences in orientation. As with human subjects, the time taken to respond is proportional to the angular difference between the two stimuli. Moreover, as reported in humans, the system fills in intermediate angles during the task, and this putative mental rotation activates the same pathways that are activated when the system views physical rotation. This work supports the proposal that mental rotation arises through sequence learning and the idea that mental imagery aids perception through learned associations, and suggests testable predictions for biological experiments.

Purified fractions of the neural cell-adhesion molecule N-CAM from embryonic chicken brain contain two similar polypeptides (Mr, 160,000 and 130,000), each contain- ing an amino-terminal external binding region, a carbohy- drate-rich central region, and a carboxyl-terminal region that is associated with the cell. Previous studies indicate that the two polypeptides arise by alternative splicing of mRNAs tran- scribed from a single gene. We report here the 3556-nucleotide sequence of a cDNA clone (pEC208) that encodes 964 amino acids from the carbohydrate and cell-associated domains of the larger N-CAM polypeptide followed by 664 nucleotides of 3' untranslated sequence. The predicted protein sequence con- tains attachment sites for polysialic acid-containing oligosac- charides, four tandem homologous regions of polypeptide resembling those seen in the immunoglobulin superfamily, and a single hydrophobic sequence that appears to be the mem- brane-spanning segment. The cytoplasmic domain carboxyl terminal to this segment includes a block of =250 amino acids present in the larger but not in the smaller N-CAM polypep- tide. We designate these the Id (large domain) polypeptide and the sd (small domain) polypeptide. The intracellular domains of the Id and sd polypeptides are likely to be critical for cell-surface modulation of N-CAM by interacting in a differ- ential fashion with other Intrinsic proteins or with the cytoskeleton.

Eukaryotic transcriptional regulation in different cells invc large numbers and arrangements of cis and trans elements survey the number of cis regulatory elements that are activ different contexts, we have devised a high-throughput selec procedure permitting synthesis of active cis motifs that enh the activity of a minimal promoter. This synthetic promoter struction method (SPCM) was used to identify >100 DNA quences that showed increased promoter activity in the ne blastoma cell line Neuro2A. After determining DNA sequence selected synthetic promoters, database searches for known ments revealed a predominance of eight motifs: AP2, CEBP, I Ebox, ETS, CREB, AP1, and SP1/MAZ. The most active of selected synthetic promoters contain composites of a numbE these motifs. Assays of DNA binding and promoter activity of t exemplary motifs (ETS, CREB, and SP1/MAZ) were used to p the effectiveness of SPCM in uncovering active sequences. U 10% of 133 selected active sequences had no match in curre available databases, raising the possibility that new motifs transcriptional regulatory proteins to which they bind ma! revealed by SPCM. The method may find uses in construe databases of active cis motifs, in diagnostics, and in gene ther

Protein synthesis in neurons is essential for the consolidation of memory and for the stabilization of activity-dependent forms of Ai synaptic plasticity such as long-term potentiation (LTP). Activity- re; dependent translation of dendritically localized mRNAs has been thl proposed to be a critical source of new proteins necessary for ch synaptic change. mRNA for the activity-regulated cytoskeletal inl protein, Arc, is transcribed during LTP and learning, and disruption ra] of its translation gives rise to deficits in both. We have found that inl selective translation of Arc in a synaptoneurosomal preparation is mn induced by the brain-derived neurotrophic factor, a neurotrophin of that is released during high-frequency stimulation patterns used to of elicit LTP. This effect involves signaling through the TrkB receptor ex and is blocked by the N-methyl-D-aspartate-type glutamate recep- mn tor antagonist, MK801. The results suggest there is a synergy bl( between neurotrophic and ionotropic mechanisms that may influ- fec ence the specificity and duration of changes in synaptic efficacy at glutamatergic synapses. tra syi

Animal behavior often involves a temporally ordered sequence of actions learned from experience. Here we describe simulations of interconnected networks of spiking neurons that learn to generate patterns of activity in correct temporal order. The simulation consists of large-scale networks of thousands of excitatory and inhibitory neurons that exhibit short-term synaptic plasticity and spike-timing dependent synaptic plasticity. The neural architecture within each area is arranged to evoke winner-take-all (WTA) patterns of neural activity that persist for tens of milliseconds. In order to generate and switch between consecutive firing patterns in correct temporal order, a reentrant exchange of signals between these areas was necessary. To demonstrate the capacity of this arrangement, we used the simulation to train a brain-based device responding to visual input by autonomously generating temporal sequences of motor actions.

In Alzheimer disease, elevated levels of the BACE1 enzyme are correlated with increased production of amyloid peptides and disease pathology. The increase in BACE1 levels is post-transcriptional and may involve altered translation efficiency. Earlier studies have indicated that translation of BACE1 mRNA is cap-dependent. As ribosomal subunits move from the cap-structure to the initiation codon, they fail to recognize several AUG codons in the 5' leader. In this study, we looked for physical evidence of the mechanism underlying ribosomal scanning or shunting along the BACE1 5' leader by investigating structural stability in the 5' leaders of endogenous mRNAs in vivo. To perform this analysis, we probed RNAs using lead(II) acetate, a cell-permeable chemical that induces cleavage of unpaired nucleotides having conformational flexibility. The data revealed that the ≈440-nt 5' leader was generally resistant to cleavage except for a region upstream of the initiation codon. Cleavage continued into the coding region, consistent with destabilization of secondary structures by translating ribosomes. Evidence that a large segment of the BACE1 5' leader was not cleaved indicates that this region is structurally stable and suggests that it is not scanned. The data support a mechanism of translation initiation in which ribosomal subunits bypass (shunt) part of the BACE1 5' leader to reach the initiation codon. We suggest that a nucleotide bias in the 5' leader may predispose the initiation codon to be more accessible than other AUG codons in the 5' leader, leading to an increase in its relative utilization.

We describe simulations of large-scale networks of excitatory and inhibitory spiking neurons that can generate dynamically stable winner-take-all (WTA) behavior. The network connectivity is a variant of center-surround architecture that we call center-annular-surround (CAS). In this architecture each neuron is excited by nearby neighbors and inhibited by more distant neighbors in an annular-surround region. The neural units of these networks simulate conductance-based spiking neurons that interact via mechanisms susceptible to both short-term synaptic plasticity and STDP. We show that such CAS networks display robust WTA behavior unlike the center-surround networks and other control architectures that we have studied. We find that a large-scale network of spiking neurons with separate populations of excitatory and inhibitory neurons can give rise to smooth maps of sensory input. In addition, we show that a humanoid brain-based-device (BBD) under the control of a spiking WTA neural network can learn to reach to target positions in its visual field, thus demonstrating the acquisition of sensorimotor coordination.

Reentry in nervous systems is the ongoing bidirectional exchange of signals along reciprocal axonal fibers linking two or more brain areas. The hypothesis that reentrant signaling serves as a general mechanism to couple the functioning of multiple areas of the cerebral cortex and thalamus was first proposed in 1977 and 1978 (Edelman, 1978). A review of the amount and diversity of supporting experimental evidence accumulated since then suggests that reentry is among the most important integrative mechanisms in vertebrate brains (Edelman, 1993). Moreover, these data prompt testable hypotheses regarding mechanisms that favor the development and evolution of reentrant neural architectures.

Short-term correlations among neurons have been observed in different animals and brain regions. In a series of anatomically and physiologically based simulations of neuronal networks we have explored the emergence and the temporal dynamics of such correlations, and we have suggested potential functional roles for correlated neural activity in visual perception. A main focus of this paper is the role of short-term correlations in the control of behavior. Several examples of behavior that depend on the presence of a pattern of correlations among modeled neurons are discussed.

The Dynamic Core and Global Workspace hypotheses were independently put forward to provide mechanistic and biologically plausible accounts of how brains generate conscious mental content. The Dynamic Core proposes that reentrant neural activity in the thalamocortical system gives rise to conscious experience. Global Workspace reconciles the limited capacity of momentary conscious content with the vast repertoire of long-term memory. In this paper we show the close relationship between the two hypotheses. This relationship allows for a strictly biological account of phenomenal experience and subjectivity that is consistent with mounting experimental evidence. We examine the constraints on causal analyses of consciousness and suggest that there is now sufficient evidence to consider the design and construction of a conscious artifact.