Content-specific fronto-parietal synchronization during visual working memory.

Lateral prefrontal and posterior parietal cortical areas exhibit task-dependent activation during working memory tasks in humans and monkeys. Neurons in these regions become synchronized during attention-demanding tasks, but the contribution of these interactions to working memory is largely unknown. Using simultaneous recordings of neural activity from multiple areas in both regions, we find widespread, task-dependent, and content-specific synchronization of activity across the fronto-parietal network during visual working memory. The patterns of synchronization are prevalent among stimulus-selective neurons and are governed by influences arising in parietal cortex. These results indicate that short-term memories are represented by large-scale patterns of synchronized activity across the fronto-parietal network.

Response preparation and inhibition: the role of the cortical sensorimotor beta rhythm.

Paradigms requiring either a GO or a NO-GO response are often used to study the neural mechanisms of response inhibition. Here this issue is examined from the perspective of event-related beta (14-30 Hz) oscillatory activity. Two macaque monkeys performed a task that began with a self-initiated lever depression and maintenance (sustained motor output) and required a visual pattern discrimination followed by either a lever release (GO) or continued lever-holding (NO-GO) response. Analyzing simultaneous local field potentials (LFPs) from primary somatosensory, frontal motor, and posterior parietal cortices, we report two results. First, beta oscillation desynchronized shortly after stimulus presentation, the onset of which was approximately the same for both the GO and NO-GO conditions ( approximately 110 ms). Since it is well known that beta desynchronization is a reliable indicator of movement preparation, this result suggests that early motor preparation took place in both conditions. Second, following the GO/NO-GO decision ( approximately 190 ms), beta activity rebounded significantly ( approximately 300 ms) only in the NO-GO condition. Coherence and Granger causality measures revealed that the dynamical organization of the rebounded beta network was similar to that existing during the sustained motor output prior to stimulus onset. This finding suggests that response inhibition led to the restoration of the sensorimotor network to its prestimulus state.

Evaluating causal relations in neural systems: granger causality, directed transfer function and statistical assessment of significance.

We consider the question of evaluating causal relations among neurobiological signals. In particular, we study the relation between the directed transfer function (DTF) and the well-accepted Granger causality, and show that DTF can be interpreted within the framework of Granger causality. In addition, we propose a method to assess the significance of causality measures. Finally, we demonstrate the applications of these measures to simulated data and actual neurobiological recordings.

Cortical coordination dynamics and cognition.

New imaging techniques in cognitive neuroscience have produced a deluge of information correlating cognitive and neural phenomena. Yet our understanding of the inter-relationship between brain and mind remains hampered by the lack of a theoretical language for expressing cognitive functions in neural terms. We propose an approach to understanding operational laws in cognition based on principles of coordination dynamics that are derived from a simple and experimentally verified theoretical model. When applied to the dynamical properties of cortical areas and their coordination, these principles support a mechanism of adaptive inter-area pattern constraint that we postulate underlies cognitive operations generally.

Causal influences in primate cerebral cortex during visual pattern discrimination.

Anatomical studies of the visual cortex demonstrate the existence of feedforward, feedback and lateral pathways among multiple cortical areas. Yet relatively little evidence has previously been available to show the causal influences of these areas on one another during visual information processing. We simultaneously recorded event-related local field potentials (LFPs) from surface-to-depth bipolar electrodes at six sites in the ventral region of the right hemisphere visual cortex in a highly trained macaque monkey during performance of a visual pattern discrimination task. Applying a new statistical measure, the short-time directed transfer function (STDTF), to the LFP data set, we charted the changing strength and direction of causal influence between these cortical sites on a fraction-of-a-second time scale. We present results showing, for the first time, the dynamics of distinct feedforward, feedback and lateral influences in the ventral portion of the primate visual cortex during visual pattern processing.

Short-window spectral analysis of cortical event-related potentials by adaptive multivariate autoregressive modeling: data preprocessing, model validation, and variability assessment.

In this article we consider the application of parametric spectral analysis to multichannel event-related potentials (ERPs) during cognitive experiments. We show that with proper data preprocessing, Adaptive MultiVariate AutoRegressive (AMVAR) modeling is an effective technique for dealing with nonstationary ERP time series. We propose a bootstrap procedure to assess the variability in the estimated spectral quantities. Finally, we apply AMVAR spectral analysis to a visuomotor integration task, revealing rapidly changing cortical dynamics during different stages of task processing.

Alternatively spliced isoforms of FE65 serve as neuron-specific and non-neuronal markers.

FE65 is predominantly expressed in brain and is especially rich in the regions with the highest densities of neurons. The FE65 protein binds to an intracellular domain of the beta-amyloid precursor protein (betaPP) and may modulate the production of beta-amyloid peptide (AP). One of FE65 exons, a mini-exon (exon 9, 6 bp), is alternatively spliced, giving rise to two isoforms varying only in 6 base pairs. We quantitated the two isoforms by a sensitive reverse transcription-competitive polymerase chain reaction technique, and characterized their expressions in various tissues and cell cultures, and the kinetics of expression of the two isoforms in P19 embryonal carcinoma cell lines during neuronal differentiation. Our results show that the exon 9-inclusive (E9) form, the more abundant form in brain, was exclusively expressed in neurons, while the exon 9-exclusive (DeltaE9) form was widely expressed in all non-neuronal cells, but was not expressed in differentiated neurons. When P19 cells were differentiated to neurons, expression of FE65 was significantly up regulated ( approximately 30-fold) and the splicing pattern of the FE65 pre-mRNA was switched from the DeltaE9 pattern to the E9 form. Based upon their distinctive expression patterns, these two isoforms may serve as neuronal and non-neuronal markers, and determination of their ratios may have applications in neuropathological diagnosis.

Spatiotemporal reorganization of electrical activity in the human brain associated with a timing transition in rhythmic auditory-motor coordination.

We used a 61-channel electrode array to investigate the spatiotemporal dynamics of electroencephalographic (EEG) activity related to behavioral transitions in rhythmic sensorimotor coordination. Subjects were instructed to maintain a 1:1 relationship between repeated right index finger flexion and a series of periodically delivered tones (metronome) in a syncopated (anti-phase) fashion. Systematic increases in stimulus presentation rate are known to induce a spontaneous switch in behavior from syncopation to synchronization (in-phase coordination). We show that this transition is accompanied by a large-scale reorganization of cortical activity manifested in the spatial distributions of EEG power at the coordination frequency. Significant decreases in power were observed at electrode locations over left central and anterior parietal areas, most likely reflecting reduced activation of left primary sensorimotor cortex. A second condition in which subjects were instructed to synchronize with the metronome controlled for the effects of movement frequency, since synchronization is known to remain stable across a wide range of frequencies. Different, smaller spatial differences were observed between topographic patterns associated with synchronization at low versus high stimulus rates. Our results demonstrate qualitative changes in the spatial dynamics of human brain electrical activity associated with a transition in the timing of sensorimotor coordination and suggest that maintenance of a more difficult anti-phase timing relation is associated with greater activation of primary sensorimotor areas.

The human FE65 gene: genomic structure and an intronic biallelic polymorphism associated with sporadic dementia of the Alzheimer type.

The FE65 protein binds to the intracellular domain of the beta-amyloid precursor protein (betaPP) and may modulate the internalization of betaPP. This gene is highly expressed in regions of the brain specifically affected in dementia of the Alzheimer type (DAT). As a prelude to further investigations of the role of FE65 in the metabolism of betaPP and in the pathogenesis of DAT, we have determined the entire genomic structure and sequence of human FE65 and have discovered several polymorphisms in this gene. Human FE65 contains 14 exons ranging in size from 6 to 735 bp. All splice sites conform to consensus sequences except for the donor site of intron 10. The 5' end of FE65 mRNA was identified by rapid amplification of the cDNA 5' end and is 31 bp longer than the previously published cDNA sequence. The 5'-flanking region of this gene is TATA-less and is very GC-rich with at least five putative Sp1 binding sites. In comparison to the genomic rat FE65 sequence, the human FE65 5'-untranslated region is 134 bp longer and has an extra exon (exon 1, 86 bp). To identify mutations/polymorphisms of the coding regions of this gene, we performed blinded analysis of 457 Caucasian case-control samples from a large epidemiological study of sporadic DAT. Screening was conducted by single-strand conformation polymorphism. Four minor variants were found within the coding region, with frequencies between 0.002 and 0.015; two of the four result in amino acid substitutions. The more informative biallelic polymorphism (a trinucleotide deletion and a single base substitution) was found within intron 13 (84 bp), which interrupts two exons encoding the betaPP binding site. The frequency of the minor allele in this intron was 0.097 in DAT cases and 0.161 in controls (chi2=7.78, P=0.0054). Having at least one copy of the minor allele was associated with a decreased risk for DAT (chi2=9.20, P<0.005, odds ratio=0.49, 95% CI 0.31-0.77). Multivariate analysis showed that this association was independent of the APOE genotype. These results suggest that either FE65 itself or a closely linked gene influences the pathogenesis of sporadic DAT. The interaction of FE65 with betaPP and the association of a FE65 polymorphism with DAT lend credence to the hypothesis that the metabolism of betaPP is central to the pathogenesis of common sporadic forms of DAT.

cDNA cloning and chromosome mapping of the human Fe65 gene: interaction of the conserved cytoplasmic domains of the human beta-amyloid precursor protein and its homologues with the mouse Fe65 protein.

Using the yeast two hybrid system, a mouse embryo cDNA library was screened for proteins that interact with the C-terminus of the human beta-amyloid precursor protein (beta PP). A fusion protein was identified that interacts specifically with the cytoplasmic domain of beta PP and does not interact with the beta-amyloid region. The protein encoded by this partial mouse cDNA is identical to the C-terminus of the rat Fe65 protein. This mouse protein also interacts with the homologous C-terminal domains of the mouse amyloid precursor-like proteins, APLP1 and APLP2. These conserved cytoplasmic regions contain a common amino acid motif, Asn-Pro-Thr-Tyr, which has previously been shown to influence both the secretion and internalization of beta PP. Fe65 has been implicated in regulatory and cell signaling mechanisms because it contains two different motifs involved in protein binding, a WW domain (a variant of Src homology 3 domains) and a phosphotyrosine interaction domain (PID). Interestingly, the PID domain binds to the same motif present in the conserved cytoplasmic domains of the beta PP and beta PP-like proteins. RNA analyses reveal that Fe65 is predominantly expressed in brain and in the regions most affected by Alzheimer's disease (AD)-associated neuropathology. The human Fe65 mRNA was cloned from a fetal brain cDNA library. The message encodes a protein of 735 amino acids that is 95% identical to the rat Fe65 protein. The human Fe65 gene was mapped on human metaphase chromosomes to band 11p15 using fluorescence in situ hybridization.

Interareal synchronization in the visual cortex.

The primary visual cortex (V1) is part of a highly interconnected network of cortical areas, hierarchically organized but operating concurrently across hierarchical levels. The high degree of reciprocal interconnection among visual cortical areas provides a framework for their interaction during the performance of visual scene analysis. The functional interdependency of visual cortical areas which develops during scene analysis can be investigated by techniques which measure interareal correlated activity. Evidence from monkeys performing a visual pattern discrimination suggests that synchronization of aperiodic activity from neuronal ensembles in cortical areas at different hierarchical levels is a relevant aspect of visual function. The near-periodic nature of the synchronized response to moving light bars in earlier studies may have been a result of the type of stimulus used. Various models of visual cortex are discussed in which interareal synchronization plays a functional role.

Spatio-temporal correlations in human gamma band electrocorticograms.

Animal electrocorticogram (ECoG) studies have shown that spatial patterns in the gamma band (>20 Hz) reflect perceptual categorization. Spatio-temporal correlations were investigated in the 20-50 Hz range in search for similar phenomena in human ECoG. ECoGs were recorded in a somatosensory discrimination task from 64-electrode subdural grid arrays, with inter-electrode spacing of 1 cm, overlying somatosensory, motor and superior temporal cortices in 2 patients with intractable epilepsy. Bootstrap techniques were devised to analyze the spatial and temporal characteristics of the correlations. Despite an extensive search, no evidence was found for globally correlated activity related to behavior either in narrow (1.e., 35-45 Hz) or broad (i.e., 20-50 Hz) bands. Spatial patterns, extracted using principal component analysis, could not be classified with respect to stimulus type in any time interval. Instead, spatially and temporally intermittent synchronization was observed between pairs of electrodes in 1 cm X 1 cm regions with high variability within and across trials. The distribution of correlation coefficients differed substantially from background levels at inter-electrode distances of 1 cm and 1.4 cm but not 2 cm or more. The minimum duration of correlation, the decorrelation time, of the ECoG was about 50 msec; the average correlation duration at 1 cm inter-electrode distance was about 150 msec; and the recurrence rate of significant correlation peaks was about 1.3/sec. The findings suggest that the surface diameters of domains of spatially correlated activity underlying perceptual categorization in human gamma band ECoG are limited to less than 2 cm and that the intermittent synchronization observed across separations of 1 cm and 1.4 cm is not solely due to volume conduction. Thus, if such gamma band spatial patterns exist in the human brain, no existing technology would be capable of measuring them at the scalp, and subdural electrode arrays for cortical surface recording would have to have spacings under 5 mm.

Large-scale cortical networks and cognition.

The well-known parcellation of the mammalian cerebral cortex into a large number of functionally distinct cytoarchitectonic areas presents a problem for understanding the complex cortical integrative functions that underlie cognition. How do cortical areas having unique individual functional properties cooperate to accomplish these complex operations? Do neurons distributed throughout the cerebral cortex act together in large-scale functional assemblages? This review examines the substantial body of evidence supporting the view that complex integrative functions are carried out by large-scale networks of cortical areas. Pathway tracing studies in non-human primates have revealed widely distributed networks of interconnected cortical areas, providing an anatomical substrate for large-scale parallel processing of information in the cerebral cortex. Functional coactivation of multiple cortical areas has been demonstrated by neurophysiological studies in non-human primates and several different cognitive functions have been shown to depend on multiple distributed areas by human neuropsychological studies. Electrophysiological studies on interareal synchronization have provided evidence that active neurons in different cortical areas may become not only coactive, but also functionally interdependent. The computational advantages of synchronization between cortical areas in large-scale networks have been elucidated by studies using artificial neural network models. Recent observations of time-varying multi-areal cortical synchronization suggest that the functional topology of a large-scale cortical network is dynamically reorganized during visuomotor behavior.

Episodic multiregional cortical coherence at multiple frequencies during visual task performance.

The way in which the brain integrates fragmentary neural events at multiple locations to produce unified perceptual experience and behaviour is called the binding problem. Binding has been proposed to involve correlated activity at different cortical sites during perceptuomotor behaviour, particularly by synchronization of narrow-band oscillations in the gamma-frequency range (30-80 Hz). In the rabbit olfactory system, inhalation induces increased gamma-correlation between sites in olfactory bulb and cortex. In the cat visual system, coherent visual stimuli increase gamma-correlation between sites in both the same and different visual cortical areas. In monkeys, some groups have found that gamma-oscillations transiently synchronize within striate cortex, superior temporal sulcus and somatosensorimotor cortex. Others have reported that visual stimuli produce increased broad-band power, but not gamma-oscillations, in several visual cortical areas. But the absence of narrow-band oscillations in itself does not disprove interregional synchronization, which may be a broad-band phenomenon. We now describe episodes of increased broad-band coherence among local field potentials from sensory, motor and higher-order cortical sites of macaque monkeys performing a visual discrimination task. Widely distributed sites become coherent without involving other intervening sites. Spatially selective multiregional cortical binding, in the form of broad-band synchronization, may thus play a role in primate perceptuomotor behaviour.

Mapping of the distal boundary of the X-inactivation center in a rearranged X chromosome from a female expressing XIST.

A female patient with primary amenorrhea, immature secondary sexual characteristics, and tall stature was found to have a normal X chromosome and a rearranged X [rea(X)] chromosome that resembled an 'isochromosome' Xp, but retained the proximal portion of Xq. The rea(X) was interpreted as rec(X)dup p,inv(X)(p11.4q13). Replication studies demonstrated that the rea(X) was always the late-replicating and, therefore, presumably inactive X chromosome, which must contain the X-inactivation center. Consistent with this interpretation, fluorescence in situ hybridization demonstrated that the rea(X) retained the XIST gene, and reverse transcription polymerase chain reaction analysis showed that XIST was expressed in the patient's cells. By fluorescence in situ hybridization with previously mapped probes, the breakpoint of the rea(X) was located within an approximately 500-kb region located approximately 200 to 700 kb distal to the XIST locus. This is the closest breakpoint distal to XIST in an inactivated X chromosome and, therefore, defines a new distal boundary for the X-inactivation center in humans.

Maintenance of X inactivation of the Rps4, Zfx, and Ube1 genes in a mouse in vitro system.

Several genes, including RPS4X (ribosomal protein subunit 4), ZFX (zinc finger on the X chromosome), and UBE1 (ubiquitin-activating enzyme), have been shown to be expressed from the inactive X chromosome of cultured human cells. By contrast, these genes are subject to X-chromosome inactivation in tissues from adult mice. We have now examined the inactivation status of these genes in cultured mouse cells to determine whether the differences in X-chromosome inactivation between species is due to an intrinsic difference between human and mouse X-chromosome genes or whether it is a function of gene reactivation in cell culture per se. The expression of three mouse X-chromosome genes, Rps4, Zfx, and Ube1 was examined by reverse transcriptase polymerase chain reaction (RT-PCR) in heterozygous cultured cells from a cross of a laboratory mouse by Mus spretus, which were selected to uniformly express the X chromosome from the laboratory mouse parent. No expression of the M. spretus alleles of these genes was observed in the cell line (Hobmski), which is consistent with the patterns of expression previously observed in mouse in vivo and indicates that these genes remain stably inactivated in an immortalized mouse cell line. By cytogenetic and RT-PCR analyses the Hobmski cell line was shown to retain a late-replicating X chromosome from M. spretus, which expressed the M. spretus allele of the X (inactive) specific transcript (Xist). The Hobmski cell line will be a useful resource for studying the features that maintain X-chromosome genes inactive.

Mapping and expression of the ubiquitin-activating enzyme E1 (Ube1) gene in the mouse.

The nucleotide sequence of the human cDNA encoding ubiquitin-activating enzyme E1 is more than 99% identical with the human A1S9T cDNA, a gene that has been shown to complement the temperature-sensitive mutant mouse cell line, tsA1S9. The amino acid sequences of the proteins encoded by these two cDNA sequences are identical, and both cDNAs were previously shown to be located in the same region of the human X chromosome; thus, ubiquitin-activating enzyme E1 and A1S9T appear to be the same gene, designated UBE1. By in situ hybridization to metaphase chromosomes from male mice and by Southern blot analysis of male and female mouse DNA, we show that, in the mouse, a human UBE1 cDNA probe identified both X- and Y-linked loci. Ube1 is located at band A2 of the mouse X Chromosome (Chr) and Ube2 on the short arm of the Y Chr. This is in contrast to the situation in the human, where there is no evidence for Y-linked sequences related to UBE1. Mapping of the Ube1 gene in interspecific backcrosses between Mus spretus and C57BL/6 shows that the Ube1 locus maps close to Timp, in a conserved region of the mouse and human X Chrs that include Otc, Cybb, Syn1, Timp, and Araf. Expression of Ube1 on the inactive X Chr was examined to determine whether this gene is subject to X-Chr inactivation in the mouse, as there is previous evidence that the human UBE1 gene escapes, at least partially, X inactivation. Sequencing of reverse transcriptase polymerase chain reaction (RT-PCR) products from M. spretus, C57BL/6J, and T(X;16)16H x M. spretus F1 female mice indicates that the mouse Ube1 gene is subject to X-Chr inactivation in vivo. This represents a new example of differences between the sex chromosomes of mouse and human.

Inactivation of the Rps4 gene on the mouse X chromosome.

The human RPS4X and RPS4Y genes, located on the X and Y chromosomes, appear to encode isoforms of ribosomal protein S4. Haploinsufficiency of these genes may contribute to the human phenotype known as Turner syndrome. Although RPS4X maps near the X-inactivation center, the gene is expressed on inactive human X chromosomes. We cloned Rps4, the mouse homolog of RPS4X. Exploiting allelic variation in Rps4, we examined transcription of the gene from active and inactive mouse X chromosomes in vivo, in female mice carrying an X-autosome translocation. We report that mouse Rps4, unlike human RPS4X, is subject to X inactivation. This finding may explain, at least in part, why the phenotypic consequences of X monosomy are less severe in mice than in humans.