Analysis of amplitude modulated control features for ECoG neuroprosthetics

Electrocorticogram recordings for neuroprosthetics provide an intermediate level of abstraction between EEG and microwire single neuron recordings. For adaptive filtering methodologies used in neuroprosthetics, extraction of spatio-control parameters remains a difficulty. Since amplitude modulation in extracellular recordings plays a key role in both neuronal activation and rate coding, seeking spatial pattern classification and temporally intermittent population synchronization in terms of increased voltage may provide viable control signals. This study seeks to explore preprocessing modalities that emphasize amplitude modulation in the ECoG above the level of noise and background fluctuations in order to derive the commands for complex control tasks. The decoding performance of the amplitude modulation across the recording spectra was found to be spatially specific in the cortex.

Wireless transmission of neuronal recordings using a portable real-time discrimination/compression algorithm

A design challenge of portable wireless neural recording systems is the tradeoff between bandwidth and power consumption. This paper investigates the compression of neuronal recordings in real-time using a novel discriminating Linde-Buzo-Gray algorithm (DLBG) that preserves spike shapes while filtering background noise. The technique is implemented in a low power digital signal processor (DSP) which is capable of wirelessly transmitting raw neuronal recordings. Depending on the signal to noise ratio of the recording, the compression ratio can be tailored to the data to maximally preserve power and bandwidth. The approach was tested in real and synthetic data and achieved compression ratios between 184:1 and 10:1.

Electrocorticographic interictal spike removal via denoising source separation for improved neuroprosthesis control

Electrocorticographic (ECoG) neuroprosthesis is a promising area of research that could provide channels of communication and control for patients who have lost their motor functions due to damage to the nervous system. However, implantation of subdural electrodes are clinically restricted to diagnostics of pre-surgical epileptic patients. Hence, interictal activity is present in the recordings across various areas of the sensorimotor cortex and suppresses the amplitude modulated features extracted to model hand trajectories. Denoising source separation is a recently introduced framework which extracts hidden structures of interest within the data through denoising the source estimates with filters designed around prior knowledge on the observations. Herein, we exploit the high amplitude quasiperiodic nature of the observed interictal spikes and show that removal of the interictal activity improves linear prediction of hand trajectories.

Inferring the stability of LIFE through Brain Machine Interfaces

We examine neural signals from Longitudinally implanted Intra-Fascicular Electrodes (LIFE) in a chronic, rabbit model. Translation-invariant wavelet de-noising methods are used to improve S%R. Then traditional template-based spike sorting is applied to discriminate single units. We investigate the effect of discriminating between identified units on Brain Machine Interface (BMI) decoding performance. We infer the stability of LIFE based on decoding performance with and without current BMI methods to counter-act electrode neural signal degradation.

BMI cyberworkstation: enabling dynamic data-driven brain-machine interface research through cyberinfrastructure

Dynamic data-driven brain-machine interfaces (DDDBMI) have great potential to advance the understanding of neural systems and improve the design of brain-inspired rehabilitative systems. This paper presents a novel cyberinfrastructure that couples in vivo neurophysiology experimentation with massive computational resources to provide seamless and efficient support of DDDBMI research. Closed-loop experiments can be conducted with in vivo data acquisition, reliable network transfer, parallel model computation, and real-time robot control. Behavioral experiments with live animals are supported with real-time guarantees. Offline studies can be performed with various configurations for extensive analysis and training. A Web-based portal is also provided to allow users to conveniently interact with the cyberinfrastructure, conducting both experimentation and analysis. New motor control models are developed based on this approach, which include recursive least square based (RLS) and reinforcement learning based (RLBMI) algorithms. The results from an online RLBMI experiment shows that the cyberinfrastructure can successfully support DDDBMI experiments and meet the desired real-time requirements.

Detection of synchronized firings in multivariate neural spike trains during motor tasks

This paper describes and compares two classical methods for the detection of neuron groups which exhibit synchronized firings in multivariate spike trains. These methods were compared on experimental and randomized data corresponding to the firing activity of 104 neurons located in motor, premotor, and parietal cortices in a monkey during movement tasks. Both methods exhibited high false positive rates in randomized data, but results showed that this rate can be advantageously reduced with a simple postprocessing. Otherwise, one method permitted to detect a significant number of synchronized groups of neurons related to the behavioral task.

Neural signal sampling via the low power wireless pico system

This paper presents a powerful new low power wireless system for sampling multiple channels of neural activity based on Texas Instruments MSP430 microprocessors and Nordic Semiconductor's ultra low power high bandwidth RF transmitters and receivers. The system's development process, component selection, features and test methodology are presented.

Gravity transform for input conditioning in brain machine interfaces

Gravity transform measures cooperative neural activity being utilized for the analysis of neural assemblies. In this paper we verify the applicability of the gravity transform to specify components of neural assemblies, which could be combined, leading ultimately to a reduction of the input dimensionality in brain-machine interface models. Our analysis was performed on data collected from rats performing a lever pressing task. We compare the results from the gravity transform analysis with the assignment obtained through a sensitivity analysis applied on a linear optimal filter.

Simultaneus prediction of four kinematic variables for a brain-machine interface using a single recurrent neural network

Implementation of brain-machine interface neural-to-motor mapping algorithms in low-power, portable digital signal processors (DSPs) requires efficient use of model resources especially when predicting signals that show interdependencies. We show here that a single recurrent neural network can simultaneously predict hand position and velocity from the same ensemble of cells using a minimalist topology. Analysis of the trained topology showed that the model learns to concurrently represent multiple kinematic parameters in a single state variable. We further assess the expressive power of the state variables for both large and small topologies.

Extraction and localization of mesoscopic motor control signals for human ECoG neuroprosthetics

Electrocorticogram (ECoG) recordings for neuroprosthetics provide a mesoscopic level of abstraction of brain function between microwire single neuron recordings and the electroencephalogram (EEG). Single-trial ECoG neural interfaces require appropriate feature extraction and signal processing methods to identify and model in real-time signatures of motor events in spontaneous brain activity. Here, we develop the clinical experimental paradigm and analysis tools to record broadband (1Hz to 6kHz) ECoG from patients participating in a reaching and pointing task. Motivated by the significant role of amplitude modulated rate coding in extracellular spike based brain-machine interfaces (BMIs), we develop methods to quantify spatio-temporal intermittent increased ECoG voltages to determine if they provide viable control inputs for ECoG neural interfaces. This study seeks to explore preprocessing modalities that emphasize amplitude modulation across frequencies and channels in the ECoG above the level of noisy background fluctuations in order to derive the commands for complex, continuous control tasks. Preliminary experiments show that it is possible to derive online predictive models and spatially localize the generation of commands in the cortex for motor tasks using amplitude modulated ECoG.

Self-organizing maps with dynamic learning for signal reconstruction

Wireless Brain Machine Interface (BMI) communication protocols are faced with the challenge of transmitting the activity of hundreds of neurons which requires large bandwidth. Previously a data compression scheme for neural activity was introduced based on Self Organizing Maps (SOM). In this paper we propose a dynamic learning rule for improved training of the SOM on signals with sparse events which allows for more representative prototype vectors to be found, and consequently better signal reconstruction. This work was developed with BMI applications in mind and therefore our examples are geared towards this type of signals. The simulation results show that the proposed strategy outperforms conventional vector quantization methods for spike reconstruction.

Deep Brain Stimulation of the Internal Segment of the Globus Pallidus in Delayed Runaway Dyskinesia

BACKGROUND

Dyskinesias that occur during a period without medication after embryonic cell transplantation have been commonly reported in double-blind trials; however, to date, they have not been reported in the few patients who participated in open-label pilot studies.

DESIGN

Single case observation with preoperative and postoperative data, and intraoperative single-cell physiology.

PATIENT

A patient who underwent embryonic cell transplantation in 1993 as part of the University of South Florida open-label study was referred for evaluation of intractable dyskinesia of the right arm. The dyskinesia was present during evaluation of the patient after a 12-hour period without medication and was clinically disabling. It was manifested as a severe groping movement of the hand. Intraoperative physiologic evaluation revealed decreased firing rates in the internal segment of the globus pallidus.

RESULTS

Deep brain stimulation of the internal segment of the globus pallidus resulted in resolution of the dyskinesia.

CONCLUSION

This case highlights the delayed development of runaway dyskinesia after a period without medication as an important potential long-term adverse effect of embryonic cell transplantation in patients with Parkinson disease.

Structural modifications in chronic microwire electrodes for cortical neuroprosthetics: a case study

Long-term viability of chronic invasive neural probes is a necessary condition for extracting robust control signals directly from neural tissue. Although immune/tissue response is a leading factor in the degradation of single neuron recording, we investigate a second component of signal degradation connected to the structural changes associated with microwire electrodes chronically exposed to extracelluar environments in vivo. Scanning electron microscopy is used to assess the surface modifications to the electrodes after an implantation duration of four weeks in rats. The electrode developed a smooth fracture surface, a reduction of the metal diameter, and pitting in the insulation of the electrode structure. Over the duration of implantation, recording properties of the electrode were marked by a reduction in the peak-to-peak amplitude in neuronal firing.

A comparison of optimal MIMO linear and nonlinear models for brain-machine interfaces

The field of brain-machine interfaces requires the estimation of a mapping from spike trains collected in motor cortex areas to the hand kinematics of the behaving animal. This paper presents a systematic investigation of several linear (Wiener filter, LMS adaptive filters, gamma filter, subspace Wiener filters) and nonlinear models (time-delay neural network and local linear switching models) applied to datasets from two experiments in monkeys performing motor tasks (reaching for food and target hitting). Ensembles of 100-200 cortical neurons were simultaneously recorded in these experiments, and even larger neuronal samples are anticipated in the future. Due to the large size of the models (thousands of parameters), the major issue studied was the generalization performance. Every parameter of the models (not only the weights) was selected optimally using signal processing and machine learning techniques. The models were also compared statistically with respect to the Wiener filter as the baseline. Each of the optimization procedures produced improvements over that baseline for either one of the two datasets or both.

Evolving into epilepsy: Multiscale electrophysiological analysis and imaging in an animal model

Epilepsy research for the design of seizure detection/prediction neuroprosthetics has been faced with the search for electrophysiologic control parameters that can be used to infer the epileptic state of the animal and be leveraged at a later time to deliver neurotherapeutic feedback. The analysis presented here uses multi-microelectrode array technology to provide an electrophysiologic quantification of a hippocampal neural ensemble during the latent period of epileptogenesis. Through the use of signal processing system identification methodologies, we were able to assess the spatial and temporal interrelations of ensembles of hippocampal neurons and relate them to the evolution of the epileptic condition. High-field magnetic resonance (MR) imaging was used to determine the location of electrode placement and to evaluate hippocampal pyramidal cell structural damage. Long-term single unit activity analysis suggests that hippocampal neurons in both CA1-2 and dentate regions increase the number of occurrences and duration of their bursting activity after injury to the contra-lateral hippocampus. The trends inferred from both single neuron and ensemble analysis suggests that the evolution into epilepsy is not abrupt but modulates gradually from the time of injury.

Design and fabrication of a flexible substrate microelectrode array for brain machine interfaces

We report a neural microelectrode array design that leverages the recording properties of conventional microwire electrode arrays with the additional features of precise control of the electrode geometries. Using microfabrication techniques, a neural probe array is fabricated that possesses a flexible polyimide-based cable. The performance of the design was tested with electrochemical impedance spectroscopy and in vivo studies. The gold-plated electrode site has an impedance value of 0.9 M Omega at 1 kHz. Acute neural recording provided high neuronal yields, peak-to-peak amplitudes (as high as 100 microV), and signal-to-noise ratios (27 dB).

Analysis of the correlation between local field potentials and neuronal firing rate in the motor cortex

Neuronal firing rate has been the signal of choice for invasive motor brain machine interfaces (BMI). The use of local field potentials (LFP) in BMI experiments may provide additional dendritic information about movement intent and may improve performance. Here we study the time-varying amplitude modulated relationship between local field potentials (LFP) and single unit activity (SUA) in the motor cortex. We record LFP and SUA in the primary motor cortex of rats trained to perform a lever pressing task, and evaluate the correlation between pairs of peri-event time histograms (PETH) and movement evoked local field potentials (mEP) at the same electrode. Three different correlation coefficients were calculated and compared between the neuronal PETH and the magnitude and power of the mEP. Correlation as high as 0.7 for some neurons occurred between the PETH and the mEP magnitude. As expected, the correlations between the single trial LFP and SUV are much lower due to the inherent variability of both signals.

The effect of a bellows leak in an Ohmeda 7810 ventilator on room contamination, inspired oxygen, airway pressure, and tidal volume

We investigated the effect of a small bellows leak (bellows full at end-expiration) on inspired oxygen fraction (Fio(2)), exhaled tidal volume (Vt), airway pressure, and room contamination in an oxygen-driven anesthesia ventilator (Ohmeda 7810, Madison, WI). CO(2) concentration at the ventilator exhalation valve, Fio(2), Vt, and airway pressure were measured (n = 3) while ventilating a CO(2)-producing test lung at 8 breaths/min and an inspiratory/expiratory ratio of 1:2, with and without a bellows leak (4-mm-long tear). Set Vt was 400, 600, 800, and 1000 mL. Fresh gas flow (FGF) was 0.3 L/min O(2) and (a) 5.0 L/min air, (b) 2.0 L/min air, and (c) 0.2 L/min nitrogen. There was no clinical difference in Fio(2), Vt, PIP (peak inspiratory pressure) and PEEP (positive end-expiratory pressure), with and without a 4-mm bellows tear, at all FGFs and Vt settings. CO(2) at the ventilator exhalation valve was always nonzero with a bellows leak, indicating that CO(2)-laden circuit gas was contaminating the drive gas via the bellows leak. A 4-mm bellows tear in an Ohmeda 7810 ventilator allows anesthetic gases to contaminate ambient air but does not cause clinically significant changes in Fio(2), exhaled Vt, PIP, or PEEP.

Interpreting spatial and temporal neural activity through a recurrent neural network brain-machine interface

We propose the use of optimized brain-machine interface (BMI) models for interpreting the spatial and temporal neural activity generated in motor tasks. In this study, a nonlinear dynamical neural network is trained to predict the hand position of primates from neural recordings in a reaching task paradigm. We first develop a method to reveal the role attributed by the model to the sampled motor, premotor, and parietal cortices in generating hand movements. Next, using the trained model weights, we derive a temporal sensitivity measure to asses how the model utilized the sampled cortices and neurons in real-time during BMI testing.

Staged Deep Brain Stimulation for Refractory Craniofacial Dystonia with Blepharospasm: Case Report and Physiology

OBJECTIVE AND IMPORTANCE

We report the intraoperative results, subsequent course, and 1-year follow-up evaluation of a patient with medication-refractory craniofacial dystonia for whom we planned bilateral globus pallidus internus (GPi) deep brain stimulation (DBS) implantation but delayed the left GPi DBS implantation because of robust intraoperative effects of right GPi DBS.

CLINICAL PRESENTATION

A 47-year-old patient had a 5-year history of progressively severe, bilateral craniofacial dystonia with blepharospasm (Meige's syndrome) that was refractory to medications and to botulinum toxin (A and B) injections. Blepharospasm interfered with his ability to perform his duties as a Special Forces soldier and ended his military career.

INTERVENTION

Under stereotactic guidance (magnetic resonance imaging and computed tomographic image fusion, Cosman-Roberts-Wells frame, and University of Florida surgical navigation software) and with detailed microelectrode mapping (four microelectrode passes), a DBS electrode was implanted in the right posteroventral GPi. Microelectrode recordings were taken to document electrophysiological activity of neurons in the region, and intraoperative macrostimulation was performed. The patient was followed up for 6 months with right unilateral GPi DBS, and later a left GPi DBS electrode was placed.

CONCLUSION

Although DBS for primary generalized dystonia is commonly performed by simultaneously implanting bilateral GPi electrodes, it may be reasonable in cases of refractory blepharospasm and/or craniofacial dystonia to use a staged procedure for implantation in selected patients. Additionally, the physiology, especially that encountered in the striatum, may help to elucidate the pathophysiological basis for refractory blepharospasm and Meige's syndrome. More cases will be needed to determine the significance of the results reported in this article.

Ascertaining the Importance of Neurons to Develop Better Brain-Machine Interfaces

In the design of brain-machine interface (BMI) algorithms, the activity of hundreds of chronically recorded neurons is used to reconstruct a variety of kinematic variables. A significant problem introduced with the use of neural ensemble inputs for model building is the explosion in the number of free parameters. Large models not only affect model generalization but also put a computational burden on computing an optimal solution especially when the goal is to implement the BMI in low-power, portable hardware. In this paper, three methods are presented to quantitatively rate the importance of neurons in neural to motor mapping, using single neuron correlation analysis, sensitivity analysis through a vector linear model, and a model-independent cellular directional tuning analysis for comparisons purpose. Although, the rankings are not identical, up to sixty percent of the top 10 ranking cells were in common. This set can then be used to determine a reduced-order model whose performance is similar to that of the ensemble. It is further shown that by pruning the initial ensemble neural input with the ranked importance of cells, a reduced sets of cells (between 40 and 80, depending upon the methods) can be found that exceed the BMI performance levels of the full ensemble.

Divide-and-conquer approach for brain machine interfaces: nonlinear mixture of competitive linear models

This paper proposes a divide-and-conquer strategy for designing brain machine interfaces. A nonlinear combination of competitively trained local linear models (experts) is used to identify the mapping from neuronal activity in cortical areas associated with arm movement to the hand position of a primate. The proposed architecture and the training algorithm are described in detail and numerical performance comparisons with alternative linear and nonlinear modeling approaches, including time-delay neural networks and recursive multilayer perceptrons, are presented. This new strategy allows training the local linear models using normalized LMS and using a relatively smaller nonlinear network to efficiently combine the predictions of the linear experts. This leads to savings in computational requirements, while the performance is still similar to a large fully nonlinear network.

Contribution of proteomics to the molecular analysis of renal cell carcinoma with an emphasis on manganese superoxide dismutase

Renal cell carcinoma (RCC) originates in the renal cortex. It accounts for 2-3 percent of all cancers occurring in adults and it is characterised by lack of early clinical manifestations, unpredictable outcome, and absence of effective treatment modalities except early surgery. RCC comprises a heterogeneous group of tumours with various molecular and cytogenetic abnormalities and different histological features as cell types and tumour architecture. Molecular genetic and proteomic tools led to the discovery of potential diagnostic prognostic and therapeutic biomarkers of RCC. In this review we discuss recent developments in understanding genotype-phenotype relationships, with attention to manganese superoxide dismutase, a mitochondrial enzyme related to the redox cycle which affects various regulatory functions of cells. The expression of this protein has been evaluated in numerous human tumour types including RCC, and post-translational modifications are being investigated.

Cellular distribution of translationally controlled tumor protein in rat and human testes

In a recent proteomic study we identified 53 spermatogonial proteins among which was the translationally controlled tumor protein (TCTP). This is a protein previously reported as being implicated in proliferation events in normal and tumoral tissues that had never previously been seen in the testis. The present study was aimed at establishing the complete cellular distribution of TCTP and its transcript and the ontogenetic expression of this gene within the testis. Using an immunohistochemistry technique, an intense TCTP signal was detected in gonocytes (the prespermatogonia) in the fetal rat testis and in spermatogonia within adult human and neonatal and adult rat testes. Meiotic spermatocytes and postmeiotic haploid spermatids were also strongly immunostained in a stage-dependent manner in human and rat testes. In addition, different levels of TCTP expression were also observed in the testicular somatic cells, with strong expression in Leydig cells and peritubular cells, and weak expression in Sertoli cells. Western and Northern blot analyses confirmed the presence of TCTP at all ages studied, with higher levels of RNA expression at 9 and 20 d postpartum, when spermatogonia and primary spermatocytes represent the highest proportion of germ cells: it was also confirmed that TCTP is present in all populations of isolated testicular cells. A transcript of 0.85 kb corresponding to TCTP, was expressed at all ages studied. This transcript was found to be expressed strongly in spermatogonia, somewhat less in isolated Leydig, resident macrophage, peritubular and Sertoli cells, weakly in the primary spermatocytes but not at all in spermatids. Interestingly, in the latter, a different transcript of 1.1 kb was present. The same 1.1 kb transcript appeared in testis extracts from 35 days postpartum onwards, corresponding to an age when spermatids accumulate within the tubules. Of note is that resident macrophages were found to express both the 0.85 and the 1.1 kb transcripts. We conclude that the strong expression of TCTP in spermatogonia makes it highly likely that the protein plays a significant role in spermatogenesis.

Proteomic dissection of dome formation in a mammary cell line: role of tropomyosin-5b and maspin

In this work we extended the study of genes controlling the formation of specific differentiation structures called "domes" formed by the rat mammary adenocarcinoma cell line LA7 under the influence of DMSO. We have reported previously that an interferon-inducible gene, rat-8, and the beta-subunit of the epithelial sodium channel (ENaC) play a fundamental role in this process. Now, we used a proteomic approach to identify proteins differentially expressed either in DMSO-induced LA7 or in 106A10 cells. Two differentially expressed proteins were investigated. The first, tropomyosin-5b, strongly expressed in DMSO-induced LA7 cells, is needed for dome formation because its synthesis inhibition by the antisense RNA technology abolished domes. The second protein, maspin, strongly expressed in the uninduced 106A10 cell line, inhibits dome formation because 106A10 cells, transfected with rat8 cDNA (the function of which is required for the organization of these structures), acquired the ability to develop domes when cultured in presence of an antimaspin antibody. Dome formation in these cultures are accompanied by ENaC beta-subunit expression in the absence of DMSO. Therefore, dome formation requires the expression of tropomyosin-5b, in addition to the ENaC beta-subunit and the rat8 proteins, and is under the negative control of maspin.

Assessing cerebrospinal fluid rhinorrhea: a two-dimensional electrophoresis approach

Assessment of nasal cerebrospinal fluid (CSF) fistula commonly relies on the determination of CSF markers in an aqueous rhinorrhea, such as the beta2-transferrin immunofixation assay. While generally reliable, false positive and false negative results have been reported for most of the laboratory tests yet available. Based on the hypothesis that the simultaneous assessment of several CSF markers may yield an increased sensitivity and specificity, we used a proteomics, two-dimensional electrophoresis 2-DE based approach to study samples of nasal secretions obtained from 18 patients suspected of CSF rhinorrhea. Since CSF, nasal mucus and plasma may coexist in the nasal cavities, we first defined five specific markers for each of these biological fluids (transferrin, prostaglandin-D synthase, transthyretin, and two unknown trains of spots for CSF, immunoglobulin A (IgA) S-chain, lipocortin-1, lipocalin-1, prolactine-inducible protein and palatal lung nasal epithelium clone protein for mucus, haptoglobin alpha1/2- and beta-chains, fibrinogen alpha-, beta- and gamma-chains for plasma). Gels from the rhinorrhea patients were then compared to these 2-DE reference maps to determine the presence or absence of the defined markers, and clinical data were independently compared to the results of the 2-DE study. In all cases, the biological fluid(s) anticipated to be present in the nasal secretions based on clinical data were correctly identified by 2-DE. Moreover, an excellent correlation was found in nine patients who underwent extensive workup for suspected CSF rhinorrhea, since CSF was found by the 2-DE method in four patients in whom a CSF fistula was confirmed, whereas the test was negative in five patients in whom a CSF fistula was excluded. In the remaining patients, mucus, sometimes contamined with blood, was found to be the major component of the nasal secretions, confirming that clear mucus may mimick CSF rhinorrhea. These preliminary results suggest that a 2-DE-based multimarker approach is a valid, sensitive, and specific method to assess the presence of CSF in occult rhinorrhea.

Identification of ribosome-associated viral and cellular basic proteins during the course of infection with herpes simplex virus type 1

Herpes simplex virus type 1 (HSV-1) infection induces severe alterations of the translational apparatus, including the phosphorylation of a few ribosomal proteins, and the progressive association of several nonribosomal proteins to ribosomes. Therefore, we hypothesized that ribosomes themselves could contribute to the HSV-1-induced translational control of host and viral gene expression. As a prerequisite to test this hypothesis, we undertook the identification of the nonribosomal proteins associated to the ribosomes during the course of HSV-1 infection. After separation by two-dimensional polyacrylamide gel electrophoresis of basic proteins extracted from the ribosomal fraction, the identification of unknown protein spots was carried out by N-terminal sequencing and peptide mass determination by mass spectrometry. This allowed us to identify HSV-1 VP19C and VP26 that associated to ribosomes with different kinetics. Another nonribosomal protein turned out to be the poly(A)-binding protein 1 (PAB1P). Newly synthesized PAB1P continued to associate to ribosomes all along infection.

Proteomics meets cell biology: the establishment of subcellular proteomes

Proteome research aims to unravel the biological complexity encoded by the genome. Due to the complexity of higher eukaryotic cells, single-step characterization of a proteome is likely to be difficult to achieve. However, advantage can be taken of the macromolecular architecture of a cell, e.g., subcellular compartments, organelles, macromolecular structures and multiprotein complexes, to establish subcellular proteomes. This review highlights recent developments in this area of proteomics, namely the establishment of two-dimensional electrophoresis (2-DE) reference maps of subcellular compartments and organelles as well as the characterization of macromolecular structures and multiprotein complexes using a proteomics approach.

New perspectives in the Escherichia coli proteome investigation

Escherichia coli is a model organism for biochemical and biological studies as it is one of the best characterised prokaryote. Two-dimensional polyacrylamide gel electrophoresis, computer image analysis and different protein identification techniques gave rise, in 1995, to the Escherichia coli SWISS-2D PAGE database (http://www.expasy.ch/ch2d/). In the E. coli 3.5-10 SWISS-2D PAGE map, 40% of the E. coli proteome was displayed. The present study demonstrated that the use of narrow range pH gradients is able to potentially display up to a few copies of protein per E. coli cell. Moreover, the six new E. coli SWISS-2D PAGE maps (pH 4-5, 4.5-5.5, 5-6, 5.5-6.7, 6-9 and 6-11) presented here displayed altogether more than 70% of the entire E. coli proteome.

Peroxisomal bifunctional enzyme binds and activates the activation function-1 region of the peroxisome proliferator-activated receptor alpha

The transcriptional activity of peroxisome proliferator-activated receptors (PPARs), and of nuclear hormone receptors in general, is subject to modulation by cofactors. However, most currently known co-activating proteins interact in a ligand-dependent manner with the C-terminal ligand-regulated activation function (AF)-2 domain of nuclear receptors. Since PPARalpha exhibits a strong constitutive transactivating function contained within an N-terminal AF-1 region, it can be speculated that a different set of cofactors might interact with this region of PPARs. An affinity purification approach was used to identify the peroxisomal enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase (bifunctional enzyme, BFE) as a protein which strongly and specifically interacted with the N-terminal 92 amino acids of PPARalpha. Protein-protein interaction assays with the cloned BFE confirmed this interaction, which could be mapped to amino acids 307-514 of the BFE and the N-terminal 70 amino acids of PPARalpha. Moreover, transient transfection experiments in hepatoma cells revealed a 2.2-fold increase in the basal and ligand-stimulated transcriptional activity of PPARalpha in the presence of BFE. This stimulatory effect is preferentially observed for the PPARalpha isoform and it is significantly stronger (4.8-fold) in non-hepatic cells, which presumably express lower levels of endogenous BFE. Hence, the BFE represents the first known cofactor capable of activating the AF-1 domain of PPAR without requiring additional regions of this receptor. These data are compatible with a model whereby the PPAR-regulated BFE is able to modulate its own expression through an enhancement of the activity of PPARalpha, representing a novel peroxisomal-nuclear feed-forward regulatory loop.

The mouse SWISS-2D PAGE database: a tool for proteomics study of diabetes and obesity

A number of two-dimensional electrophoresis (2-DE) reference maps from mouse samples have been established and could be accessed through the internet. An up-to-date list can be found in WORLD-2D PAGE (http://www.expasy.ch/ch2d/2d- index.html), an index of 2-DE databases and services. None of them were established from mouse white and brown adipose tissues, pancreatic islets, liver nuclei and skeletal muscle. This publication describes the mouse SWISS-2D PAGE database. Proteins present in samples of mouse (C57BI/6J) liver, liver nuclei, muscle, white and brown adipose tissue and pancreatic islets are assembled and described in an accessible uniform format. SWISS-2D PAGE can be accessed through the World Wide Web (WWW) network on the ExPASy molecular biology server (http://www.expasy.ch/ ch2d/).

Proteomics: new perspectives, new biomedical opportunities

Proteomics-based approaches, which examine the expressed proteins of a tissue or cell type, complement the genome initiatives and are increasingly being used to address biomedical questions. Proteins are the main functional output, and the genetic code cannot always indicate which proteins are expressed, in what quantity, and in what form. For example, post-translational modifications of proteins, such as phosphorylation or glycosylation, are very important in determining protein function. Similarly, the effects of environmental factors or multigenic processes such as ageing or disease cannot be assessed simply by examination of the genome alone. This review describes the underlying technology and illustrates several areas of biomedical research, ranging from pathogenesis of neurological disorders to drug and vaccine design, in which potential clinical applications are being explored.

Proteomics meets cell biology: the establishment of subcellular proteomes

Proteome research aims to unravel the biological complexity encoded by the genome. Due to the complexity of higher eukaryotic cells, single-step characterization of a proteome is likely to be difficult to achieve. However, advantage can be taken of the macromolecular architecture of a cell, e.g., subcellular compartments, organelles, macromolecular structures and multiprotein complexes, to establish subcellular proteomes. This review highlights recent developments in this area of proteomics, namely the establishment of two-dimensional electrophoresis (2-DE) reference maps of subcellular compartments and organelles as well as the characterization of macromolecular structures and multiprotein complexes using a proteomics approach.

Proteomics meets cell biology: the establishment of subcellular proteomes

Proteome research aims to unravel the biological complexity encoded by the genome. Due to the complexity of higher eukaryotic cells, single-step characterization of a proteome is likely to be difficult to achieve. However, advantage can be taken of the macromolecular architecture of a cell, e.g., subcellular compartments, organelles, macromolecular structures and multiprotein complexes, to establish subcellular proteomes. This review highlights recent developments in this area of proteomics, namely the establishment of two-dimensional electrophoresis (2-DE) reference maps of subcellular compartments and organelles as well as the characterization of macromolecular structures and multiprotein complexes using a proteomics approach.

The establishment of a human liver nuclei two-dimensional electrophoresis reference map

This short communication describes the establishment of a two-dimensional electrophoresis (2-DE) reference map of nuclear proteins isolated from human liver. The human liver nuclei 2-DE reference map contains 1497 spots. In an initial identification study using peptide mass fingerprinting as a means of protein identification we were able to identify 26 spots corresponding to 15 different proteins. The human liver nuclei 2-DE reference map is now included in the SWISS-2DPAGE database, which can be accessed through the ExPASy server (http://www.expasy.ch/ch2d/).

Changes induced by oxygen in rat liver proteins identified by high-resolution two-dimensional gel electrophoresis

Molecular oxygen (O2) regulates the expression of a variety of genes. Several of the proteins that respond to changes in oxygen concentration have been identified in a variety of cell lines. We extend these previous studies by analyzing the effect of oxygen on the entire protein expression profile of an intact organ using high-resolution two-dimensional gel electrophoresis. To this end, we used an isolated, in vitro perfused organ preparation to produce two groups of rat livers perfused with high (95% O2, 5% CO2) or low (95% N2, 5% CO2) oxygen concentrations. Using two-dimensional gel electrophoresis we compared the protein expression profiles of both groups of livers. Computer analysis of the files obtained after laser densitometry of the two-dimensional gels revealed two spots that were strongly up-regulated in high PO2 perfused livers compared with low PO2 perfused livers. These spots were analyzed by peptide mass fingerprinting analysis. These spots were identified as arginase 1 (liver-type arginase; EC 3.5.3.1) and mitochondrial enoyl-CoA hydratase 1 (EC 4.2.1.17). The possible role of these proteins in its new context of oxygen availability is discussed.

The dynamic range of protein expression: a challenge for proteomic research

Proteomic research, for its part, is benefiting enormously from the last decade of genomic research as we now have archived, annotated and audited sequence databases to correlate and query experimental data. While the two-dimensional electrophoresis (2-DE) gels are still a central part of proteomics, we reflect on the possibilities and realities of the current 2-DE technology with regard to displaying and analysing proteomes. Limitations of analysing whole cell/tissue lysates by 2-DE alone are discussed, and we investigate whether extremely narrow p/ranges (1 pH unit/25 cm) provide a solution to display comprehensive protein expression profiles. We are confronted with a challenging task: the dynamic range of protein expression. We believe that most of the existing technology is capable of displaying many more proteins than is currently achievable by integrating existing and new techniques to prefractionate samples prior to 2-DE display or analysis. The availability of a "proteomics toolbox", consisting of defined reagents, methods, and equipment, would assist a comprehensive analysis of defined biological systems.

The dynamic range of protein expression: a challenge for proteomic research

Proteomic research, for its part, is benefiting enormously from the last decade of genomic research as we now have archived, annotated and audited sequence databases to correlate and query experimental data. While the two-dimensional electrophoresis (2-DE) gels are still a central part of proteomics, we reflect on the possibilities and realities of the current 2-DE technology with regard to displaying and analysing proteomes. Limitations of analysing whole cell/tissue lysates by 2-DE alone are discussed, and we investigate whether extremely narrow p/ranges (1 pH unit/25 cm) provide a solution to display comprehensive protein expression profiles. We are confronted with a challenging task: the dynamic range of protein expression. We believe that most of the existing technology is capable of displaying many more proteins than is currently achievable by integrating existing and new techniques to prefractionate samples prior to 2-DE display or analysis. The availability of a "proteomics toolbox", consisting of defined reagents, methods, and equipment, would assist a comprehensive analysis of defined biological systems.

Human gingival crevicular fluid contains MRP8 (S100A8) and MRP14 (S100A9), two calcium-binding proteins of the S100 family

Human gingival crevicular fluid contains unidentified proteins which might play a role as markers in periodontal diseases. Therefore, low-molecular-weight proteins found in human gingival crevicular fluid (GCF), but absent from serum, were identified in the present study by means of two-dimensional electrophoresis (2-D PAGE) analysis. GCF, serum, and whole saliva were collected from periodontitis and healthy subjects, as well as from edentulous and newborn subjects. Protein samples were separated by two-dimensional polyacrylamide gel electrophoresis, stained with silver, and compared with reference protein maps in the SWISS-2D PAGE database. In GCF and saliva from periodontitis patients and healthy subjects, four dominant low-molecular-mass (from 8 to 14 kDa) acidic spots were observed. They were not found in serum and were less visible in saliva from edentulous and newborn subjects. From N-terminal amino acid sequencing, the two 2-D protein spots of 8 kDa and isoelectric points between 6.5 and 7.0 were both identified as protein MRP8 (SI00A8), a member of the S100 family of calcium-binding proteins. Using peptide mass fingerprinting and matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF-MS), we identified the other two protein spots, with mass of 14 kDa and isoelectric points between 5.5 and 6.0, as protein MRP14 (S100A9), also belonging to the S100 family. The presence of MRP8 and MRP14 in GCF was confirmed by Western blot, with monoclonal antibodies. The two polypeptides, MRP8 and MRP14, identified in GCF represent the major difference between the 2-D PAGE patterns of serum and GCF, and we hypothesize that they may play an important role in the gingival sulcus and could represent possible markers for periodontal diseases.

Modified expression of plasma glutathione peroxidase and manganese superoxide dismutase in human renal cell carcinoma

Two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) is a powerful tool to separate thousands of polypeptides and to highlight the modification of protein expression in malignant diseases. By applying 2-D PAGE to ten normal human kidney and ten homologous renal cell carcinoma (RCC) tissues, we found two peptides in all ten normal tissues but not in RCCs and, conversely, two peptides were detected in all RCCs but not in normal tissues. Using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) and internal sequence analysis, the two first peptides were identified as two isoforms of plasma glutathione peroxidase (GPxP). The two other peptides isolated in all RCCs but not in normal tissues were identified by N-terminal sequence analysis as multimeric forms of manganese superoxide dismutase (Mn-SOD). No multimeric Mn-SODs and only two monomeric forms were detected in normal tissues. GPxP and Mn-SOD are metallo-enzymes encoded on chromosome 5q32 and on chromosome 6p25, respectively. Their regions are within the locus 5q21-->qter and 6q21-6q27 on which deletions and translocations are described in some cytogenetic studies of RCC transformation. Therefore, our results might suggest a correlation between the modified expression of GPxP and Mn-SOD in tumor tissues and chromosomal modifications, and that the two proteins may be putative markers for diagnosis of RCC.

The 1999 SWISS-2DPAGE database update

SWISS-2DPAGE (http://www.expasy.ch/ch2d/ ) is an annotated two-dimensional polyacrylamide gel electro-phoresis (2-DE) database established in 1993. The current release contains 24 reference maps from human and mouse biological samples, as well as from Saccharomyces cerevisiae, Escherichia coli and Dictyostelium discoideum origin. These reference maps have now 2824 identified spots, corresponding to 614 separate protein entries in the database, in addition to virtual entries for each SWISS-PROT sequence or any user-entered amino acids sequence. Last year improvements in the SWISS-2DPAGE database are as follows: three new maps have been created and several others have been updated; cross-references to newly built federated 2-DE databases have been added; new functions to access the data have been provided through the ExPASy proteomics server.

Improving protein identification from peptide mass fingerprinting through a parameterized multi-level scoring algorithm and an optimized peak detection

We have developed a new algorithm to identify proteins by means of peptide mass fingerprinting. Starting from the matrix-assisted laser desorption/ionization-time-of-flight (MALDI-TOF) spectra and environmental data such as species, isoelectric point and molecular weight, as well as chemical modifications or number of missed cleavages of a protein, the program performs a fully automated identification of the protein. The first step is a peak detection algorithm, which allows precise and fast determination of peptide masses, even if the peaks are of low intensity or they overlap. In the second step the masses and environmental data are used by the identification algorithm to search in protein sequence databases (SWISS-PROT and/or TrEMBL) for protein entries that match the input data. Consequently, a list of candidate proteins is selected from the database, and a score calculation provides a ranking according to the quality of the match. To define the most discriminating scoring calculation we analyzed the respective role of each parameter in two directions. The first one is based on filtering and exploratory effects, while the second direction focuses on the levels where the parameters intervene in the identification process. Thus, according to our analysis, all input parameters contribute to the score, however with different weights. Since it is difficult to estimate the weights in advance, they have been computed with a generic algorithm, using a training set of 91 protein spectra with their environmental data. We tested the resulting scoring calculation on a test set of ten proteins and compared the identification results with those of other peptide mass fingerprinting programs.

Two-dimensional electrophoresis resources available from ExPASy

This paper describes the set of two-dimensional electrophoresis (2-DE) resources currently available from the ExPASy proteomics Web server. These resources include the SWISS-2DPAGE database, 2-DE software packages, 2-DE technical and educational services, as well as indexes and search engines for 2-DE related sites over the Internet.

Information transfer between large and small two-dimensional polyacrylamide gel electrophoresis

To determine the feasibility of data transfer, an interlaboratory comparison was conducted on colon carcinoma cell line (DLD-1) proteins resolved by two-dimensional polyacrylamide gel electrophoresis either on small (6 x 7 cm) or large (16x18 cm) gels. The gels were silver-stained and scanned by laser densitometry, and the image obtained was analyzed using Melanie software. The number of spots detected was 1337+/-161 vs. 2382+/-176 for small vs. large format gels, respectively. After gel calibration using landmarks determined using pl and Mr markers, large- and small-format gels were matched and 712+/-36 proteins were found on both types of gels. Having performed accurate gel matching it was possible to acquire additional information after accessing a 2-D PAGE reference database (http://www.expasy.ch/ cgibin/map2/def?DLD1_HUMAN). Thus, the difference in gel size is not an obstacle for data transfer. This will facilitate exchanges between laboratories or consultation concerning existing databases.

A molecular scanner to automate proteomic research and to display proteome images

Identification and characterization of all proteins expressed by a genome in biological samples represent major challenges in proteomics. Today's commonly used high-throughput approaches combine two-dimensional electrophoresis (2-DE) with peptide mass fingerprinting (PMF) analysis. Although automation is often possible, a number of limitations still adversely affect the rate of protein identification and annotation in 2-DE databases: the sequential excision process of pieces of gel containing protein; the enzymatic digestion step; the interpretation of mass spectra (reliability of identifications); and the manual updating of 2-DE databases. We present a highly automated method that generates a fully annoated 2-DE map. Using a parallel process, all proteins of a 2-DE are first simultaneously digested proteolytically and electro-transferred onto a poly(vinylidene difluoride) membrane. The membrane is then directly scanned by MALDI-TOF MS. After automated protein identification from the obtained peptide mass fingerprints using PeptIdent software (http://www.expasy.ch/tools/peptident.html + ++), a fully annotated 2-D map is created on-line. It is a multidimensional representation of a proteome that contains interpreted PMF data in addition to protein identification results. This "MS-imaging" method represents a major step toward the development of a clinical molecular scanner.

Toward a clinical molecular scanner for proteome research: parallel protein chemical processing before and during western blot

To increase the throughput of protein identification and characterization in proteome studies, we investigated three methods of performing protein digestion in parallel. The first, which we term "one-step digestion-transfer" (OSDT), is based on protein digestion during the transblotting process. It involves the use of membranes containing immobilized trypsin which are intercalated between the gel and a PVDF collecting membrane. During electrotransfer, some digestion of the transferred proteins occurs, although poorly for basic and/or high molecular weight proteins. The second method is based on "in-gel" digestion of all proteins in parallel and termed "parallel in-gel digestion" (PIGD) to denote this fact. The PIGD led to more efficient digestion of basic and high molecular weight proteins (> 40,000) but suffered from a major drawback: loss of resolution for low molecular weight polypeptides (< 60,000) through diffusion during the digestion process. The third method examined was the combination of PIGD and OSDT procedures. This combination, called "double parallel digestion" (DPD), led to greatly improved digestion of high molecular weight and basic proteins without losses of low molecular weight polypeptides. Peptides liberated during transblotting of proteins through the immobilized trypsin membrane were trapped on a PVDF membrane and identified by mass spectrometry in scanning mode.