PredβTM: A Novel β-Transmembrane Region Prediction Algorithm

Predicting the transmembrane regions is an important aspect of understanding the structures and architecture of different β-barrel membrane proteins. Despite significant efforts, currently available β-transmembrane region predictors are still limited in terms of prediction accuracy, especially in precision. Here, we describe PredβTM, a transmembrane region prediction algorithm for β-barrel proteins. Using amino acid pair frequency information in known β-transmembrane protein sequences, we have trained a support vector machine classifier to predict β-transmembrane segments. Position-specific amino acid preference data is incorporated in the final prediction. The predictor does not incorporate evolutionary profile information explicitly, but is based on sequence patterns generated implicitly by encoding the protein segments using amino acid adjacency matrix. With a benchmark set of 35 β-transmembrane proteins, PredβTM shows a sensitivity and precision of 83.71% and 72.98%, respectively. The segment overlap score is 82.19%. In comparison with other state-of-art methods, PredβTM provides a higher precision and segment overlap without compromising with sensitivity. Further, we applied PredβTM to analyze the β-barrel membrane proteins without defined transmembrane regions and the uncharacterized protein sequences in eight bacterial genomes and predict possible β-transmembrane proteins. PredβTM can be freely accessed on the web at http://transpred.ki.si/.     

Predicting disordered regions in proteins based on decision trees of reduced amino acid composition.

Intrinsically unstructured proteins (IUPs) are proteins lacking a fixed three dimensional structure or containing long disordered regions. IUPs play an important role in biology and disease. Identifying disordered regions in protein sequences can provide useful information on protein structure and function, and can assist high-throughput protein structure determination. In this paper we present a system for predicting disordered regions in proteins based on decision trees and reduced amino acid composition. Concise rules based on biochemical properties of amino acid side chains are generated for prediction. Coarser information extracted from the composition of amino acids can not only improve the prediction accuracy but also increase the learning efficiency. In cross-validation tests, with four groups of reduced amino acid composition, our system can achieve a recall of 80% at a 13% false positive rate for predicting disordered regions, and the overall accuracy can reach 83.4%. This prediction accuracy is comparable to most, and better than some, existing predictors. Advantages of our approach are high prediction accuracy for long disordered regions and efficiency for large-scale sequence analysis. Our software is freely available for academic use upon request.     

IUPred: web server for the prediction of intrinsically unstructured regions of proteins based on estimated energy content

Intrinsically unstructured/disordered proteins and domains (IUPs) lack a well-defined three-dimensional structure under native conditions. The IUPred server presents a novel algorithm for predicting such regions from amino acid sequences by estimating their total pairwise interresidue interaction energy, based on the assumption that IUP sequences do not fold due to their inability to form sufficient stabilizing interresidue interactions. Optional to the prediction are built-in parameter sets optimized for predicting short or long disordered regions and structured domains.     

Voltage-independent sodium-binding events reported by the 4B-4C loop in the human glutamate transporter excitatory amino acid transporter 3

Glutamate transporters (excitatory amino acid transporter (EAATs)) are critical for normal excitatory signaling and maintaining subtoxic glutamate concentrations in mammalian central nervous system. Recently, a crystal structure for a homologous transporter in bacteria was reported. Still, little is understood regarding the mechanism of substrate uptake. In transmembrane domain 4, the mammalian EAATs contain a stretch of over 50 amino acids (4B-4C loop) that are absent in the bacterial protein. These residues have been suggested to be located in the large extracellular vestibule seen in the crystal structure. State-dependent trypsin-cleavage sites have been reported in this region, suggesting that the 4B-4C loop undergoes significant conformational changes. Here we employed substituted cysteine accessibility, voltage clamp fluorometry, and fluorescence resonance energy transfer on oocytes expressing mutant EAAT3 transporters to determine the location and functionality of the 4B-4C loop. We find that this loop extends from near the center of the protein and that the majority of the residues are positioned on the outer perimeter of the protein, rather than inside the vestibule. Our fluorescence resonance energy transfer measurements demonstrated that these residues do not undergo large scale motions during glutamate uptake. However, our voltage clamp fluorometry studies indicate that these residues report on Na(+) and glutamate binding-induced conformational changes, including a previously un-described voltage-independent component of Na(+) binding to the transporter. The finding that residues far from the glutamate-binding site report on several different types of binding events suggests that the series of small conformational changes that accomplish glutamate uptake extend throughout the transporter structure.     

A Multilabel Model Based on Chou’s Pseudo–Amino Acid Composition for Identifying Membrane Proteins with Both Single and Multiple Functional Types

Predicting membrane protein type is a meaningful task because this kind of information is very useful to explain the function of membrane proteins. Due to the explosion of new protein sequences discovered, it is highly desired to develop efficient computation tools for quickly and accurately predicting the membrane type for a given protein sequence. Even though several membrane predictors have been developed, they can only deal with the membrane proteins which belong to the single membrane type. The fact is that there are membrane proteins belonging to two or more than two types. To solve this problem, a system for predicting membrane protein sequences with single or multiple types is proposed. Pseudo–amino acid composition, which has proven to be a very efficient tool in representing protein sequences, and a multilabel KNN algorithm are used to compose this prediction engine. The results of this initial study are encouraging.     

Characterisation and cloning of a Na(+)-dependent broad-specificity neutral amino acid transporter from NBL-1 cells: a novel member of the ASC/B(0) transporter family

Na(+)-dependent neutral amino acid transport into the bovine renal epithelial cell line NBL-1 is catalysed by a broad-specificity transporter originally termed System B(0). This transporter is shown to differ in specificity from the B(0) transporter cloned from JAR cells [J. Biol. Chem. 271 (1996) 18657] in that it interacts much more strongly with phenylalanine. Using probes designed to conserved transmembrane regions of the ASC/B(0) transporter family we have isolated a cDNA encoding the NBL-1 cell System B(0) transporter. When expressed in Xenopus oocytes the clone catalysed Na(+)-dependent alanine uptake which was inhibited by glutamine, leucine and phenylalanine. However, the clone did not catalyse Na(+)-dependent phenylalanine transport, again as in NBL-1 cells. The clone encoded a protein of 539 amino acids; the predicted transmembrane domains were almost identical in sequence to those of the other members of the B(0)/ASC transporter family. Comparison of the sequences of NBL-1 and JAR cell transporters showed some differences near the N-terminus, C-terminus and in the loop between helices 3 and 4. The NBL-1 B(0) transporter is not the same as the renal brush border membrane transporter since it does not transport phenylalanine. Differences in specificity in this protein family arise from relatively small differences in amino acid sequence.     

A combination of compositional index and genetic algorithm for predicting transmembrane helical segments

Transmembrane helix (TMH) topology prediction is becoming a focal problem in bioinformatics because the structure of TM proteins is difficult to determine using experimental methods. Therefore, methods that can computationally predict the topology of helical membrane proteins are highly desirable. In this paper we introduce TMHindex, a method for detecting TMH segments using only the amino acid sequence information. Each amino acid in a protein sequence is represented by a Compositional Index, which is deduced from a combination of the difference in amino acid occurrences in TMH and non-TMH segments in training protein sequences and the amino acid composition information. Furthermore, a genetic algorithm was employed to find the optimal threshold value for the separation of TMH segments from non-TMH segments. The method successfully predicted 376 out of the 378 TMH segments in a dataset consisting of 70 test protein sequences. The sensitivity and specificity for classifying each amino acid in every protein sequence in the dataset was 0.901 and 0.865, respectively. To assess the generality of TMHindex, we also tested the approach on another standard 73-protein 3D helix dataset. TMHindex correctly predicted 91.8% of proteins based on TM segments. The level of the accuracy achieved using TMHindex in comparison to other recent approaches for predicting the topology of TM proteins is a strong argument in favor of our proposed method. Availability: The datasets, software together with supplementary materials are available at: http://faculty.uaeu.ac.ae/nzaki/TMHindex.htm.     

Organization of the membrane domain of the human liver sodium/bile acid cotransporter

Mammalian sodium/bile acid cotransporters (SBATs) are glycoproteins with an exoplasmic N-terminus, an odd number of transmembrane regions, and a cytoplasmic C-terminus. Various algorithms predict eight or nine membrane-embedded regions derived from nine hydrophobic stretches of the protein (H1-H9). Three methods were used to define which of these were transmembrane or membrane-associated segments in the liver bile acid transporter. The first was in vitro translation/insertion scanning using either single hydrophobic sequences between the N-terminal domain of the alpha-subunit of the gastric H,K-ATPase and the C-terminal domain of the beta-subunit that contains five N-linked glycosylation exoplasmic flags or using constructs beginning with the N-terminus of the transporter of various lengths and again ending in the C-terminus of the H,K-ATPase beta-subunit. Seven of the predicted segments, but not the amphipathic H3 and H8 sequences, insert as both individual signal anchor and stop transfer sequences in the reporter constructs. These sequences, H3 and H8, are contained within two postulated long exoplasmic loops in the classical seven-transmembrane segment model. The H3 segment acts as a partial stop transfer signal when expressed downstream of the endogenous H2. In a similar manner, the other amphipathic segment, H8, inserts as a signal anchor sequence when translated in the context with the upstream transporter sequence in two different glycosylation constructs. Alanine insertion scanning identified regions of the transporter requiring precise alignment of sequence to form competent secondary structures. The transport activity of these mutants was evaluated either in native protein or in a yellow fluorescent protein (YFP) fusion protein construct. All alanine insertions in H3 and H8 abolished taurocholate uptake, suggesting that both these regions have structures with critical intramolecular interactions. Moreover, these insertions also prevented trafficking to the plasma membrane as assessed by confocal microscopy with a polyclonal antibody against either the C-terminus of the transporter or the YFP signal of the YFP-transporter fusion protein. Two glycosylation signals inserted in the first postulated loop region and four of five such signals in the second postulated loop region were not recognized by the oligosaccharide transferase, and the L256N mutation exhibited 10% glycosylation and was inactive. These findings support a topography with nine membrane-spanning or membrane-associated segments.     

Membrane insertion scanning of the human ileal sodium/bile acid co-transporter.

Mammalian sodium-dependent bile acid transporters (SBATs) responsible for bile salt uptake across the liver sinusoidal or ileal/renal brush border membrane have been identified and share approximately 35% amino acid sequence identity. Programs for prediction of topology and localization of transmembrane helices identify eight or nine hydrophobic regions for the SBAT sequences as membrane spanning. Analysis of N-linked glycosylation has provided evidence for an exoplasmic N-terminus and a cytoplasmic C-terminus, indicative of an odd number of transmembrane segments. To determine the membrane topography of the human ileal SBAT (HISBAT), an in vitro translation/translocation protocol was employed using three different fusion protein constructs. Individual HISBAT segments were analyzed for signal anchor or stop translocation (stop transfer) activity by insertion between a cytoplasmic anchor (HK M0) or a signal anchor segment (HK M1) and a glycosylation flag (HK beta). To examine consecutive HISBAT sequences, sequential hydrophobic sequences were inserted into the HK M0 vector or fusion vectors were made that included the glycosylated N-terminus of HISBAT, sequential hydrophobic sequences, and the glycosylation flag. Individual signal anchor (SA) and stop transfer (ST) properties were found for seven out of the nine predicted hydrophobic segments (H1, H2, H4, H5, H6, H7, and H9), supporting a seven transmembrane segment model. However, the H3 region was membrane inserted when translated in the context of the native HISBAT flanking sequences. Furthermore, results from translations of sequential constructs ending after H7 provided support for integration of H8. These data provide support for a SBAT transmembrane domain model with nine integrated segments with an exoplasmic N-terminus and a cytoplasmic C-terminus consistent with a recent predictive analysis of this transporter topology.     

Identification and characterization of SV31, a novel synaptic vesicle membrane protein and potential transporter

Synaptic vesicle proteins govern all relevant functions of the synaptic vesicle life cycle, including vesicle biogenesis, vesicle transport, uptake and storage of neurotransmitters, and regulated endocytosis and exocytosis. In spite of impressive progress made in the past years, not all known vesicular functions can be assigned to defined protein components, suggesting that the repertoire of synaptic vesicle proteins is still incomplete. We have identified and characterized a novel synaptic vesicle membrane protein of 31 kDa with six putative transmembrane helices that, according to its membrane topology and phylogenetic relation, may function as a vesicular transporter. The vesicular allocation is demonstrated by subcellular fractionation, heterologous expression, immunocytochemical analysis of brain sections and immunoelectron microscopy. The protein is expressed in select brain regions and contained in subpopulations of nerve terminals that immunostain for the vesicular glutamate transporter 1 and the vesicular GABA transporter VGaT (vesicular amino acid transporter) and may attribute specific and as yet undiscovered functions to subsets of glutamatergic and GABAergic synapses.     

Isolation and characterization of Golgi apparatus-specific GODZ with the DHHC zinc finger domain

We identified a novel Golgi apparatus-specific protein with the DHHC zinc finger domain and four putative transmembrane regions, designated as GODZ. The amino acid sequences were highly conserved among mouse and human GODZs and homologous proteins in human, mouse, rat, Drosophila melanogaster, and Caenorhabditis elegans, implying a functional significance of the GODZ protein family. Overexpression of mouse GODZ in COS7 cells suppressed the sorting of the glutamate receptor GluRalpha1 from the Golgi apparatus. These results suggest that GODZ plays a role in the membrane protein trafficking.     

Estimation of amino acid residue substitution rates at local spatial regions and application in protein function inference: a Bayesian Monte Carlo approach

The amino acid sequences of proteins provide rich information for inferring distant phylogenetic relationships and for predicting protein functions. Estimating the rate matrix of residue substitutions from amino acid sequences is also important because the rate matrix can be used to develop scoring matrices for sequence alignment. Here we use a continuous time Markov process to model the substitution rates of residues and develop a Bayesian Markov chain Monte Carlo method for rate estimation. We validate our method using simulated artificial protein sequences. Because different local regions such as binding surfaces and the protein interior core experience different selection pressures due to functional or stability constraints, we use our method to estimate the substitution rates of local regions. Our results show that the substitution rates are very different for residues in the buried core and residues on the solvent-exposed surfaces. In addition, the rest of the proteins on the binding surfaces also have very different substitution rates from residues. Based on these findings, we further develop a method for protein function prediction by surface matching using scoring matrices derived from estimated substitution rates for residues located on the binding surfaces. We show with examples that our method is effective in identifying functionally related proteins that have overall low sequence identity, a task known to be very challenging.     

Cross-validation of protein structural class prediction using statistical clustering and neural networks.

We present an approach to predicting protein structural class that uses amino acid composition and hydrophobic pattern frequency information as input to two types of neural networks: (1) a three-layer back-propagation network and (2) a learning vector quantization network. The results of these methods are compared to those obtained from a modified Euclidean statistical clustering algorithm. The protein sequence data used to drive these algorithms consist of the normalized frequency of up to 20 amino acid types and six hydrophobic amino acid patterns. From these frequency values the structural class predictions for each protein (all-alpha, all-beta, or alpha-beta classes) are derived. Examples consisting of 64 previously classified proteins were randomly divided into multiple training (56 proteins) and test (8 proteins) sets. The best performing algorithm on the test sets was the learning vector quantization network using 17 inputs, obtaining a prediction accuracy of 80.2%. The Matthews correlation coefficients are statistically significant for all algorithms and all structural classes. The differences between algorithms are in general not statistically significant. These results show that information exists in protein primary sequences that is easily obtainable and useful for the prediction of protein structural class by neural networks as well as by standard statistical clustering algorithms.     

Characterisation and cloning of a Na-dependent broad-specificity neutral amino acid transporter from NBL-1 cells: a novel member of the ASC/B transporter family

Na⁺-dependent neutral amino acid transport into the bovine renal epithelial cell line NBL-1 is catalysed by a broad-specificity transporter originally termed System B⁰. This transporter is shown to differ in specificity from the B⁰ transporter cloned from JAR cells [J. Biol. Chem. 271 (1996) 18657] in that it interacts much more strongly with phenylalanine. Using probes designed to conserved transmembrane regions of the ASC/B⁰ transporter family we have isolated a cDNA encoding the NBL-1 cell System B⁰ transporter. When expressed in Xenopus oocytes the clone catalysed Na⁺-dependent alanine uptake which was inhibited by glutamine, leucine and phenylalanine. However, the clone did not catalyse Na⁺-dependent phenylalanine transport, again as in NBL-1 cells. The clone encoded a protein of 539 amino acids; the predicted transmembrane domains were almost identical in sequence to those of the other members of the B⁰/ASC transporter family. Comparison of the sequences of NBL-1 and JAR cell transporters showed some differences near the N-terminus, C-terminus and in the loop between helices 3 and 4. The NBL-1 B⁰ transporter is not the same as the renal brush border membrane transporter since it does not transport phenylalanine. Differences in specificity in this protein family arise from relatively small differences in amino acid sequence.     

Identification of residues in the first cytoplasmic loop of P-glycoprotein involved in the function of chimeric human MDR1-MDR2 transporters.

The human MDR1 gene encodes the multidrug transporter (P-glycoprotein), a multidrug efflux pump. The highly homologous MDR2 gene product does not appear to be a functional multidrug pump. We have constructed a chimeric protein in which the first intracytoplasmic loop and the third and fourth transmembrane domains of the MDR1 protein were replaced by the analogous region of MDR2. Substitution of the MDR2 sequences encompassing amino acid residues 140 to 229 resulted in 17 amino acid changes, 10 in the intracytoplasmic loop (amino acids 141-188) and 7 in the transmembrane regions. This chimeric protein was expressed on the surface of NIH 3T3 cells where it bound [3H]azidopine but did not confer drug resistance. When only 4 residues, 165, 166, 168, and 169, were changed back to MDR1 amino acids, a functional drug transporter was recovered. When residues 165, 166, 168, and 169 from MDR2 were substituted into a functional MDR1 cDNA, the resulting construction was not able to confer drug resistance. These results indicate that the major functional differences between MDR1 and MDR2 in this region of P-glycoprotein reside in a small segment of the first intracytoplasmic loop. We also independently analyzed the effect of replacing Asn183 of MDR1 with Ser which occurs in MDR2. Substitution of Ser at position 183 in combination with Val at position 185 in P-glycoprotein resulted in a relative increase in resistance to actinomycin D, vinblastine, and doxorubicin in transfected NIH 3T3 cells. These results emphasize the importance of the first intracytoplasmic loop in P-glycoprotein in determining function and relative drug specificity of the transporter.     

Analysis of the nitrous oxide reduction genes, nosZDFYL, of Achromobacter cycloclastes.

The structural gene, nosZ, for the monomeric N2O reductase has been cloned and sequenced from the denitrifying bacterium Achromobacter cycloclastes. The nosZ gene encodes a protein of 642 amino acid residues and the deduced amino acid sequence showed homology to the previously derived sequences for the dimeric N2O reductases. The relevant DNA region of about 3.6 kbp was also sequenced and found to consist of four genes, nosDFYL based on the similarity with the N2O reduction genes of Pseudomonas stutzeri. The gene product of A. cycloclastes nosF (299 amino acid residues) has a consensus ATP-binding sequence, and the nos Y gene encodes a hydrophobic protein (273 residues) with five transmembrane segments, suggesting the similarity with an ATP-binding cassette (ABC) transporter which has two distinct domains of a highly hydrophobic region and ATP-binding sites. The nosL gene encodes a protein of 193 amino acid residues and the derived sequence showed a consensus sequence of lipoprotein modification/processing site. The expression of nosZ gene in Escherichia coli cells and the comparison of the translated sequences of the nosDFYL genes with those of bacterial transport genes for inorganic ions are discussed.     

Regulation of glial glutamate transporters by C-terminal domains

Excitatory amino acid transporter 2 (EAAT2) is a high affinity glutamate transporter predominantly expressed in astroglia. Human EAAT2 encompasses eight transmembrane domains and a 74-amino acid C-terminal domain that resides in the cytoplasm. We examined the role of this region by studying various C-terminal truncations and mutations using heterologous expression in mammalian cells, whole-cell patch clamp recording and confocal imaging. Removal of the complete C terminus (K498X EAAT2) results in loss of function because of intracellular retention of truncated proteins in the cytoplasm. However, a short stretch of amino acids (E500X EAAT2) within the C terminus results in correctly processed transporters. E500X reduced glutamate transport currents by 90%. Moreover, the voltage and substrate dependence of E500X EAAT2 anion currents was significantly altered. WT and mutant EAAT2 anion channels are modified by external Na(+) in the presence as well as in the absence of L-glutamate. Whereas Na(+) stimulates EAAT2 anion currents in the presence of L-glutamate, increased [Na(+)] reduces such currents without glutamate. In cells internally dialyzed with Na(+), WT, and truncated EAAT2 display comparable Na(+) dependence. With K(+) as main internal cation, E500X drastically increased the apparent dissociation constant for external Na(+). The effects of E500X can be represented by a kinetic model that allows translocation of the empty transporter from the outward- to the inward-facing conformation and stabilization of the inward-facing conformation by internal K(+). Our results demonstrate that the C terminus modifies the glutamate uptake cycle, possibly affecting the movements of the translocation domain of EAAT2 glutamate transporter.     

Molecular cloning of cDNA encoding human and rabbit forms of the Ca2+ release channel (ryanodine receptor) of skeletal muscle sarcoplasmic reticulum

We have cloned cDNAs encoding the rabbit and human forms of the Ca2+ release channel of sarcoplasmic reticulum. The human cDNA encodes a protein of 5032 amino acids, with a molecular weight of 563,584, which is made without an NH2-terminal signal sequence. Amino acid substitutions between rabbit and human sequences were noted in 163 positions and deletions or insertions in eight regions accounted for additional sequence differences between the two proteins. Analysis of the sequence indicates that 10 potential transmembrane sequences in the COOH-terminal fifth of the molecule and two additional, potential transmembrane sequences nearer to the center of the molecule could contribute to the formation of the Ca2+ conducting pore. The remainder of the molecule is hydrophilic and presumably constitutes the cytoplasmic domain of the protein. A 114-120 amino acid motif is repeated four times in the protein, in residues 841-954, 955-1068, 2725-2844, and 2845-2958 and a 16 amino acid part of the motif is repeated twice more in residues 1344-1359 and 1371-1386. Although the channel is modulated by Ca2+, ATP, and calmodulin, no clear high affinity Ca2(+)-binding domain of the EF hand type and no clear high affinity ATP-binding domain were detected in the primary sequence. An acidic sequence in residues 1872-1923 contains 79% glutamate or aspartate residues and this sequence is a potential low affinity Ca2(+)-binding site. Several potential calmodulin-binding sites were observed in the sequence, in the region 2800 to 3050.     

Identify Beta-Hairpin Motifs with Quadratic Discriminant Algorithm Based on the Chemical Shifts

Successful prediction of the beta-hairpin motif will be helpful for understanding the of the fold recognition. Some algorithms have been proposed for the prediction of beta-hairpin motifs. However, the parameters used by these methods were primarily based on the amino acid sequences. Here, we proposed a novel model for predicting beta-hairpin structure based on the chemical shift. Firstly, we analyzed the statistical distribution of chemical shifts of six nuclei in not beta-hairpin and beta-hairpin motifs. Secondly, we used these chemical shifts as features combined with three algorithms to predict beta-hairpin structure. Finally, we achieved the best prediction, namely sensitivity of 92%, the specificity of 94% with 0.85 of Mathew’s correlation coefficient using quadratic discriminant analysis algorithm, which is clearly superior to the same method for the prediction of beta-hairpin structure from 20 amino acid compositions in the three-fold cross-validation. Our finding showed that the chemical shift is an effective parameter for beta-hairpin prediction, suggesting the quadratic discriminant analysis is a powerful algorithm for the prediction of beta-hairpin.