CoPreTHi: a Web tool which combines transmembrane protein segment prediction methods

CoPreTHi is a Java based web application, which combines the results of methods that predict the location of transmembrane segments in protein sequences into a joint prediction histogram. Clearly, the joint prediction algorithm, produces superior quality results than individual prediction schemes. The program is available at http://o2.db.uoa.gr/CoPreTHi.     

TMSEG: Novel prediction of transmembrane helices

Transmembrane proteins (TMPs) are important drug targets because they are essential for signaling, regulation, and transport. Despite important breakthroughs, experimental structure determination remains challenging for TMPs. Various methods have bridged the gap by predicting transmembrane helices (TMHs), but room for improvement remains. Here, we present TMSEG, a novel method identifying TMPs and accurately predicting their TMHs and their topology. The method combines machine learning with empirical filters. Testing it on a non‐redundant dataset of 41 TMPs and 285 soluble proteins, and applying strict performance measures, TMSEG outperformed the state‐of‐the‐art in our hands. TMSEG correctly distinguished helical TMPs from other proteins with a sensitivity of 98 ± 2% and a false positive rate as low as 3 ± 1%. Individual TMHs were predicted with a precision of 87 ± 3% and recall of 84 ± 3%. Furthermore, in 63 ± 6% of helical TMPs the placement of all TMHs and their inside/outside topology was correctly predicted. There are two main features that distinguish TMSEG from other methods. First, the errors in finding all helical TMPs in an organism are significantly reduced. For example, in human this leads to 200 and 1600 fewer misclassifications compared to the second and third best method available, and 4400 fewer mistakes than by a simple hydrophobicity‐based method. Second, TMSEG provides an add‐on improvement for any existing method to benefit from. Proteins 2016; 84:1706–1716. © 2016 Wiley Periodicals, Inc.     

An unstable transmembrane segment in the cystic fibrosis transmembrane conductance regulator

The cystic fibrosis transmembrane conductance regulator (CFTR), a chloride channel with 12 membrane-spanning sequences, undergoes inefficient maturation in the endoplasmic reticulum (ER). Potentially charged residues in transmembrane segments may contribute to this defect in biogenesis. We demonstrate that transmembrane segment 6 of CFTR, which contains three basic amino acids, is extremely unstable in the lipid bilayer upon membrane insertion in vitro and in vivo. However, two distinct mechanisms counteract this anchoring deficiency: (i) the ribosome and the ER translocon co-operate to prevent transmembrane segment 6 from passing through the membrane co- translationally; and (ii) cytosolic domains of the ion channel post-translationally maintain this segment of CFTR in a membrane-spanning topology. Although these mechanisms are essential for successful completion of CFTR biogenesis, inefficiencies in their function retard the maturation of the protein. It seems possible that some of the disease-causing mutations in CFTR may reduce the efficiency of proper membrane anchoring of the protein.     

Structure of the integrin alphaIIb transmembrane segment

Integrin cell-adhesion receptors transduce signals bidirectionally across the plasma membrane via the single-pass transmembrane segments of each alpha and beta subunit. While the beta3 transmembrane segment consists of a linear 29-residue alpha-helix, the structure of the alphaIIb transmembrane segment reveals a linear 24-residue alpha-helix (Ile-966 -Lys-989) followed by a backbone reversal that packs Phe-992-Phe-993 against the transmembrane helix. The length of the alphaIIb transmembrane helix implies the absence of a significant transmembrane helix tilt in contrast to its partnering beta3 subunit. Sequence alignment shows Gly-991-Phe-993 to be fully conserved among all 18 human integrin alpha subunits, suggesting that their unusual structural motif is prototypical for integrin alpha subunits. The alphaIIb transmembrane structure demonstrates a level of complexity within the membrane that is beyond simple transmembrane helices and forms the structural basis for assessing the extent of structural and topological rearrangements upon alphaIIb-beta3 association, i.e. integrin transmembrane signaling.     

Thy-1: a hydrophobic transmembrane segment at the carboxyl terminus

Thy-1 is a membrane glycoprotein distributed throughout mammalian species and expressed on brain tissue and various cells of the hematopoietic lineage. The nucleotide sequences of rat and mouse Thy-1 genes revealed a similar gene organization and the existence of an extra 31 amino acid segment at the carboxyl terminus not described previously. These extra amino acids, highly conserved in both species, include a 20 amino acid hydrophobic segment at the carboxyl terminus, which may be responsible for integration of Thy-1 within the plasma membrane.     

Wavelet change-point prediction of transmembrane proteins

MOTIVATION: A non-parametric method, based on a wavelet data-dependent threshold technique for change-point analysis, is applied to predict location and topology of helices in transmembrane proteins. A new propensity scale generated from a transmembrane helix database is proposed. RESULTS: We show that wavelet change-point performs well for smoothing hydropathy and transmembrane profiles generated using different scales. We investigate which wavelet bases and threshold functions are overall most appropriate to detect transmembrane segments. Prediction accuracy is based on the analysis of two data sets used as standard benchmarks for transmembrane prediction algorithms. The analysis of a test set of 83 proteins results in accuracy per segment equal to 98.2%; the analysis of a 48 proteins blind-test set, i.e. containing proteins not used to generate the propensity scales, results in accuracy per segment equal to 97.4%. We believe that this method can also be applied to the detection of boundaries of other patterns such as G + Cisochores and dot-plots. AVAILABILITY: The transmembrane database, TMALN and source code are available upon request from the authors.     

The HMMTOP transmembrane topology prediction server

The HMMTOP transmembrane topology prediction server predicts both the localization of helical transmembrane segments and the topology of transmembrane proteins. Recently, several improvements have been introduced to the original method. Now, the user is allowed to submit additional information about segment localization to enhance the prediction power. This option improves the prediction accuracy as well as helps the interpretation of experimental results, i.e. in epitope insertion experiments. Availability: HMMTOP 2.0 is freely available to non-commercial users at http://www.enzim.hu/hmmtop. Source code is also available upon request to academic users.     

Towards genome-scale structure prediction for transmembrane proteins

In this paper we briefly review some of the recent progress made by ourselves and others in developing methods for predicting the structures of transmembrane proteins from amino acid sequence. Transmembrane proteins are an important class of proteins involved in many diverse biological functions, many of which have great impact in terms of disease mechanism and drug discovery. Despite their biological importance, it has proven very difficult to solve the structures of these proteins by experimental techniques, and so there is a great deal of pressure to develop effective methods for predicting their structure. The methods we discuss range from methods for transmembrane topology prediction to new methods for low resolution folding simulations in a knowledge-based force field. This potential is designed to reproduce the properties of the lipid bilayer. Our eventual aim is to apply these methods in tandem so that useful three-dimensional models can be built for a large fraction of the transmembrane protein domains in whole proteomes.     

Mutational analysis of tetracycline resistance protein transmembrane segment insertion

The tetracycline resistance proteins (TetA) of gram-negative bacteria are secondary active transport proteins that contain buried charged amino acids that are important for tetracycline transport. Earlier studies have shown that insertion of TetA proteins into the cytoplasmic membrane is mediated by helical hairpin pairs of transmembrane (TM) segments. However, whether helical hairpins direct spontaneous insertion of TetA or are required instead for its interaction with the cellular secretion (Sec) machinery is unknown. To gain insight into how TetA proteins are inserted into the membrane, we have investigated how tolerant the class C TetA protein encoded by plasmid pBR322 is to placement of charged residues in TM segments. The results show that the great majority of charge substitutions do not interfere with insertion even when placed at locations that cannot be shielded internally within helical hairpins. The only mutations that frequently block insertion are proline substitutions, which may interfere with helical hairpin folding. The ability of TetA to broadly tolerate charge substitutions indicates that the Sec machinery assists in its insertion into the membrane. The results also demonstrate that it is feasible to engineer charged residues into the interior of TetA proteins for the purpose of structure-function analysis.     

Arginines in the first transmembrane segment promote maturation of a P-glycoprotein processing mutant by hydrogen bond interactions with tyrosines in transmembrane segment 11

A key goal is to correct defective folding of mutant ATP binding cassette (ABC) transporters, as they cause diseases such as cystic fibrosis. P-glycoprotein (ABCB1) is a useful model system because introduction of an arginine at position 65 of the first transmembrane (TM) segment could repair folding defects. To determine the mechanism of arginine rescue, we first tested the effects of introducing arginines at other positions in TM1 (residues 52-72) of a P-glycoprotein processing mutant (G251V) that is defective in folding and trafficking to the cell surface (20% maturation efficiency). We found that arginines introduced into one face of the TM1 helix (positions 52, 55, 56, 59, 60, 62, 63, 66, and 67) inhibited maturation, whereas arginines on the opposite face of the helix promoted (positions 64, 65, 68, and 71) or had little effect (positions 61, and 69) on maturation. Arginines at positions 61, 64, 65, and 68 appeared to lie close to the drug binding sites as they reduced the apparent affinity for drug substrates such as vinblastine and verapamil. Therefore, arginines that promoted maturation may face an aqueous drug translocation pathway, whereas those that inhibited maturation may face the lipid bilayer. The highest maturation efficiencies (60-85%) were observed with the Arg-65 and Arg-68 mutants. Mutations that removed hydrogen bond acceptors (Y950F/Y950A or Y953F/Y953A) in TM11 predicted to lie close to Arg-65 or Arg-68 inhibited maturation but did not affect maturation of the G251V parent. Therefore, arginine may rescue defective folding by promoting packing of the TM segments through hydrogen bond interactions.     

Val133 and Cys137 in transmembrane segment 2 are close to Arg935 and Gly939 in transmembrane segment 11 of human P-glycoprotein

P-glycoprotein (P-gp; ABCB1) transports a wide variety of structurally diverse compounds out of the cell. The protein has two homologous halves joined by a linker region. Each half consists of a transmembrane (TM) domain with six TM segments and a nucleotide-binding domain. The drug substrate-binding pocket is at the interface between the TM segments in each half of the protein. Preliminary studies suggested that the arrangement of the two halves of P-gp shows rotational symmetry (i.e. "head-to-tail" arrangement). Here, we tested this model by determining whether the cytoplasmic ends of TM2 and TM3 in the N-terminal half are in close contact with TM11 in the C-terminal half. Mutants containing a pair of cysteines in TM2/TM11 or TM3/TM11 were subjected to oxidative cross-linking with copper phenanthroline. Two of the 110 TM2/TM11 mutants, V133C(TM2)/G939C(TM11) and C137C(TM2)/A935C (TM11), were cross-linked at 4 degrees C, when thermal motion is reduced. Cross-linking was specific since no cross-linked product was detected in the 100 double Cys TM3/TM11 mutants. Vanadate trapping of nucleotide or the presence of some drug substrates inhibited cross-linking of mutants V133C(TM2)/G939C(TM11) and C137C(TM2)/A935C(TM11). Cross-linking of TM2 and TM11 also blocked drug-stimulated ATPase activity. The close proximity of TM2/TM11 and TM5/TM8 (Loo, T. W., Bartlett, M. C., and Clarke, D. M. (2004) J. Biol. Chem. 279, 7692-7697) indicates that these regions between the two halves must enclose the drug-binding pocket at the cytoplasmic side of P-gp. They may form the "hinges" required for conformational changes during the transport cycle.     

The MEMPACK alpha-helical transmembrane protein structure prediction server

MOTIVATION: The experimental difficulties of alpha-helical transmembrane protein structure determination make this class of protein an important target for sequence-based structure prediction tools. The MEMPACK prediction server allows users to submit a transmembrane protein sequence and returns transmembrane topology, lipid exposure, residue contacts, helix-helix interactions and helical packing arrangement predictions in both plain text and graphical formats using a number of novel machine learning-based algorithms. AVAILABILITY: The server can be accessed as a new component of the PSIPRED portal by at http://bioinf.cs.ucl.ac.uk/psipred/.     

Solution structure of LC4 transmembrane segment of CCR5

CC-chemokine receptor 5 (CCR5) is a specific co-receptor allowing the entry of human immunodeficiency virus type 1 (HIV-1). The LC4 region in CCR5 is required for HIV-1 entry into the cells. In this study, the solution structure of LC4 in SDS micelles was elucidated by using standard 1H two-dimensional NMR spectroscopy, circular dichroism, and fluorescence quenching. The LC4 structure adopts two helical structures, whereas the C-terminal part remains unstructured. The positions in which LC4 binds to the HIV-1 inhibitory peptide LC5 were determined by docking calculations in addition to NMR data. The poses showed the importance of the hydrophobic interface of the assembled structures. The solution structure of LC4 elucidated in the present work provides a structural basis for further studies on the HIV-1 inhibitory function of the LC4 region.     

Conductance of connexin hemichannels segregates with the first transmembrane segment

Gap junction channels are intercellular channels that mediate the gated transfer of molecules between adjacent cells. To identify the domain determining channel conductance, the first transmembrane segment (M1) was reciprocally exchanged between Cx46 and Cx32E(1)43. The resulting chimeras exhibited conductances similar to that of the respective M1 donor. Furthermore, a chimera with the carboxy-terminal half of M1 in Cx46 replaced by that of Cx32 exhibited a conductance similar to that of Cx32E(1)43, whereas the chimera with only the amino-terminal half of M1 replaced retained the unitary conductance of wild-type Cx46. Extending the M1 domain swapping to other connexins by replacing the carboxy-terminal half of M1 in Cx46 with that of Cx37 yielded a chimera channel with increased unitary conductance close to that of Cx37. Furthermore, a point mutant of Cx46, with leucine substituted by glycine in position 35, displayed a conductance much larger than that of the wild type. Thus, the M1 segment, especially the second half, contains important determinants of conductance of the connexin channel.     

Role of transmembrane segment 5 of the plant vacuolar H-pyrophosphatase

Vacuolar H⁺-translocating inorganic pyrophosphatase (V-PPase; EC 3.6.1.1) is a homodimeric proton translocase consisting of a single type of polypeptide with a molecular mass of approximately 81 kDa. Topological analysis tentatively predicts that mung bean V-PPase contains 14 transmembrane domains. Alignment analysis of V-PPase demonstrated that the transmembrane domain 5 (TM5) of the enzyme is highly conserved in plants and located at the N-terminal side of the putative substrate-binding loop. The hydropathic analysis of V-PPase showed a relatively lower degree of hydrophobicity in the TM5 region as compared to other domains. Accordingly, it appears that TM5 is probably involved in the proton translocation of V-PPase. In this study, we used site-directed mutagenesis to examine the functional role of amino acid residues in TM5 of V-PPase. A series of mutants singly replaced by alanine residues along TM5 were constructed and over-expressed in Saccharomyces cerevisiae; they were then used to determine their enzymatic activities and proton translocations. Our results indicate that several mutants displayed minor variations in enzymatic properties, while others including those mutated at E225, a GYG motif (residues from 229 to 231), A238, and R242, showed a serious decline in enzymatic activity, proton translocation, and coupling efficiency of V-PPase. Moreover, the mutation at Y230 relieved several cation effects on the V-PPase. The GYG motif presumably plays a significant role in maintaining structure and function of V-PPase.     

Secondary structure formation of a transmembrane segment in Kv channels

Transmembrane segments in the intact voltage-gated potassium (Kv) channel are helical. To ascertain whether this helicity could first be manifested inside the ribosomal tunnel, we generated biogenic peptide intermediates of Kv1.3 and mass-tagged the cysteine-scanned S6 transmembrane segment using pegylation (PEG-MAL) and calmodulation (CaM-MAL). For reference, we created an extended peptide that was used as a "molecular tape measure" of the ribosomal tunnel and determined that the functional length of the tunnel is 99-112 A. We demonstrate that the S6 segment forms a compact structure inside the ribosomal tunnel and that the N-terminal half of S6 compacts more than the C-terminal half of S6. These results bear on the earliest folding events during biogenesis of ion channels.     

The prediction of transmembrane protein sequences and their conformation: an evaluation

The prediction of transmembrane protein sequences can be performed with reasonable accuracy. However, the prediction of their conformation leaves room for significant improvement.     

An ion-responsive motif in the second transmembrane segment of rhodopsin-like receptors

A L(M)xxxD(N, E) motif (x=a non-ionic amino acid residue, most frequently A, S, L or F; small capitals indicating a minor representation) is found in the second transmembrane (tm2) segment of most G-protein coupling metazoan receptors of the rhodopsin family (Rh-GPCRs). Changes in signal transduction, agonist binding and receptor cycling are known for numerous receptors bearing evolved or experimentally introduced mutations in this tm2 motif, especially of its aspartate residue. The [Na⁺] sensitivity of the receptor-agonist interaction relates to this aspartate in a number of Rh-GPCRs. Native non-conservative mutations in the tm2 motif only rarely coincide with significant changes in two other ubiquitous features of the rhodopsin family, the seventh transmembrane N(D)PxxY(F) motif and the D(E)RY(W,F) or analogous sequence at the border of the third transmembrane helix and the second intracellular loop. Native tm2 mutations with Rh-GPCRs frequently result in constitutive signaling, and with visual opsins also in shifts to short-wavelength sensitivity. Substitution of a strongly basic residue for the tm2 aspartate in Taste-2 receptors could be connected to a lack of sodium sensing by these receptors. These properties could be consistent with ionic interactions, and even of ion transfer, that involve the tm2 motif. A decrease in cation sensing by this motif is usually connected to an enhanced constitutive interaction of the mutated receptors with cognate G- proteins, and also relates to both the constitutive and the overall activity of the short-wavelength opsins.     

The first transmembrane segment of the dopamine D2 receptor: accessibility in the binding-site crevice and position in the transmembrane bundle

The binding site of the dopamine D2 receptor, like that of homologous G-protein-coupled receptors (GPCRs), is contained within a water-accessible crevice formed among its seven transmembrane segments (TMs). Using the substituted-cysteine-accessibility method (SCAM), we are mapping the residues that contribute to the surface of this binding-site crevice. We have now mutated to cysteine, one at a time, 21 consecutive residues in TM1. Six of these mutants reacted with charged sulfhydryl reagents, whereas bound antagonist only protected N52(1.50)C from reaction. Except for A38(1.36)C, none of the substituted cysteine mutants in the extracellular half of TM1 appeared to be accessible. Pro(1.48) is highly conserved in opsins, but absent in catecholamine receptors, and the high-resolution rhodopsin structure showed that Pro(1.48) bends the extracellular portion of TM1 inward toward TM2 and TM7. Analysis of the conversation of residues in the extracellular portion of TM1 of opsins showed a pattern consistent with alpha-helical structure with a conserved face. In contrast, this region in catecholamine receptors is poorly conserved, suggesting a lack of critical contacts. Thus, in catecholamine receptors in the absence of Pro(1.48), TM1 may be straighter and therefore further from the helix bundle, consistent with the apparent lack of conserved contact residues. When examined in the context of a model of the D2 receptor, the accessible residues in the cytoplasmic half of TM1 are at the interface with TM7 and with helix 8 (H8). We propose the existence of critical contacts of TM1, TM7, and H8 that may stabilize the inactive state of the receptor.     

A combined transmembrane topology and signal peptide prediction method

An inherent problem in transmembrane protein topology prediction and signal peptide prediction is the high similarity between the hydrophobic regions of a transmembrane helix and that of a signal peptide, leading to cross-reaction between the two types of predictions. To improve predictions further, it is therefore important to make a predictor that aims to discriminate between the two classes. In addition, topology information can be gained when successfully predicting a signal peptide leading a transmembrane protein since it dictates that the N terminus of the mature protein must be on the non-cytoplasmic side of the membrane. Here, we present Phobius, a combined transmembrane protein topology and signal peptide predictor. The predictor is based on a hidden Markov model (HMM) that models the different sequence regions of a signal peptide and the different regions of a transmembrane protein in a series of interconnected states. Training was done on a newly assembled and curated dataset. Compared to TMHMM and SignalP, errors coming from cross-prediction between transmembrane segments and signal peptides were reduced substantially by Phobius. False classifications of signal peptides were reduced from 26.1% to 3.9% and false classifications of transmembrane helices were reduced from 19.0% to 7.7%. Phobius was applied to the proteomes of Homo sapiens and Escherichia coli. Here we also noted a drastic reduction of false classifications compared to TMHMM/SignalP, suggesting that Phobius is well suited for whole-genome annotation of signal peptides and transmembrane regions. The method is available at as well as at