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.     

Sequence and conformational preferences at termini of α‐helices in membrane proteins: Role of the helix environment

α‐helices are amongst the most common secondary structural elements seen in membrane proteins and are packed in the form of helix bundles. These α‐helices encounter varying external environments (hydrophobic, hydrophilic) that may influence the sequence preferences at their N and C‐termini. The role of the external environment in stabilization of the helix termini in membrane proteins is still unknown. Here we analyze α‐helices in a high‐resolution dataset of integral α‐helical membrane proteins and establish that their sequence and conformational preferences differ from those in globular proteins. We specifically examine these preferences at the N and C‐termini in helices initiating/terminating inside the membrane core as well as in linkers connecting these transmembrane helices. We find that the sequence preferences and structural motifs at capping (Ncap and Ccap) and near‐helical (N' and C') positions are influenced by a combination of features including the membrane environment and the innate helix initiation and termination property of residues forming structural motifs. We also find that a large number of helix termini which do not form any particular capping motif are stabilized by formation of hydrogen bonds and hydrophobic interactions contributed from the neighboring helices in the membrane protein. We further validate the sequence preferences obtained from our analysis with data from an ultradeep sequencing study that identifies evolutionarily conserved amino acids in the rat neurotensin receptor. The results from our analysis provide insights for the secondary structure prediction, modeling and design of membrane proteins. Proteins 2014; 82:3420–3436. © 2014 Wiley Periodicals, Inc.     

Prediction of buried helices in multispan alpha helical membrane proteins

Analysis of a database of structures of membrane proteins shows that membrane proteins composed of 10 or more transmembrane (TM) helices often contain buried helices that are inaccessible to phospholipids. We introduce a method for identifying TM helices that are least phospholipid accessible and for prediction of fully buried TM helices in membrane proteins from sequence information alone. Our method is based on the calculation of residue lipophilicity and evolutionary conservation. Given that the number of buried helices in a membrane protein is known, our method achieves an accuracy of 78% and a Matthew's correlation coefficient of 0.68. A server for this tool (RANTS) is available online at http://gila.bioengr.uic.edu/lab/.     

Helix kinks are equally prevalent in soluble and membrane proteins

Helix kinks are a common feature of α‐helical membrane proteins, but are thought to be rare in soluble proteins. In this study we find that kinks are a feature of long α‐helices in both soluble and membrane proteins, rather than just transmembrane α‐helices. The apparent rarity of kinks in soluble proteins is due to the relative infrequency of long helices (≥20 residues) in these proteins. We compare length‐matched sets of soluble and membrane helices, and find that the frequency of kinks, the role of Proline, the patterns of other amino acid around kinks (allowing for the expected differences in amino acid distributions between the two types of protein), and the effects of hydrogen bonds are the same for the two types of helices. In both types of protein, helices that contain Proline in the second and subsequent turns are very frequently kinked. However, there are a sizeable proportion of kinked helices that do not contain a Proline in either their sequence or sequence homolog. Moreover, we observe that in soluble proteins, kinked helices have a structural preference in that they typically point into the solvent. Proteins 2014; 82:1960–1970. © 2014 The Authors. Proteins: Structure, Function, and Bioinformatics Published by Wiley Periodicals, Inc.     

Transmembrane helix predictions revisited

Methods that predict membrane helices have become increasingly useful in the context of analyzing entire proteomes, as well as in everyday sequence analysis. Here, we analyzed 27 advanced and simple methods in detail. To resolve contradictions in previous works and to reevaluate transmembrane helix prediction algorithms, we introduced an analysis that distinguished between performance on redundancy-reduced high- and low-resolution data sets, established thresholds for significant differences in performance, and implemented both per-segment and per-residue analysis of membrane helix predictions. Although some of the advanced methods performed better than others, we showed in a thorough bootstrapping experiment based on various measures of accuracy that no method performed consistently best. In contrast, most simple hydrophobicity scale-based methods were significantly less accurate than any advanced method as they overpredicted membrane helices and confused membrane helices with hydrophobic regions outside of membranes. In contrast, the advanced methods usually distinguished correctly between membrane-helical and other proteins. Nonetheless, few methods reliably distinguished between signal peptides and membrane helices. We could not verify a significant difference in performance between eukaryotic and prokaryotic proteins. Surprisingly, we found that proteins with more than five helices were predicted at a significantly lower accuracy than proteins with five or fewer. The important implication is that structurally unsolved multispanning membrane proteins, which are often important drug targets, will remain problematic for transmembrane helix prediction algorithms. Overall, by establishing a standardized methodology for transmembrane helix prediction evaluation, we have resolved differences among previous works and presented novel trends that may impact the analysis of entire proteomes.     

Bihelix: Towards de novo structure prediction of an ensemble of G‐protein coupled receptor conformations

G‐Protein Coupled Receptors (GPCRs) play a critical role in cellular signal transduction pathways and are prominent therapeutic targets. Recently there has been major progress in obtaining experimental structures for a few GPCRs. Each GPCR, however, exhibits multiple conformations that play a role in their function and we have been developing methods aimed at predicting structures for all these conformations. Analysis of available structures shows that these conformations differ in relative helix tilts and rotations. The essential issue is, determining how to orient each of the seven helices about its axis since this determines how it interacts with the other six helices. Considering all possible helix rotations to ensure that no important packings are overlooked, and using rotation angle increments of 30° about the helical axis would still lead to 12⁷ or 35 million possible conformations each with optimal residue positions. We show in this paper how to accomplish this. The fundamental idea is to optimize the interactions between each pair of contacting helices while ignoring the other 5 and then to estimate the energies of all 35 million combinations using these pair‐wise interactions. This BiHelix approach dramatically reduces the effort to examine the complete set of conformations and correctly identifies the crystal packing for the experimental structures plus other near‐native packings we believe may play an important role in activation. This approach also enables a detailed structural analysis of functionally distinct conformations using helix‐helix interaction energy landscapes and should be useful for other helical transmembrane proteins as well. Proteins 2012. © 2011 Wiley Periodicals, Inc.     

Comprehensive analysis of the numbers,lengths and amino acid compositions of transmembrane helices in prokaryotic,eukaryotic and viral integral membrane proteins of high-resolution structure

We report a comprehensive analysis of the numbers, lengths and amino acid compositions of transmembrane helices in 235 high-resolution structures of integral membrane proteins. The properties of 1551 transmembrane helices in the structures were compared with those obtained by analysis of the same amino acid sequences using topology prediction tools. Explanations for the 81 (5.2%) missing or additional transmembrane helices in the prediction results were identified. Main reasons for missing transmembrane helices were mis-identification of N-terminal signal peptides, breaks in α-helix conformation or charged residues in the middle of transmembrane helices and transmembrane helices with unusual amino acid composition. The main reason for additional transmembrane helices was mis-identification of amphipathic helices, extramembrane helices or hairpin re-entrant loops. Transmembrane helix length had an overall median of 24 residues and an average of 24.9 ± 7.0 residues and the most common length was 23 residues. The overall content of residues in transmembrane helices as a percentage of the full proteins had a median of 56.8% and an average of 55.7 ± 16.0%. Amino acid composition was analysed for the full proteins, transmembrane helices and extramembrane regions. Individual proteins or types of proteins with transmembrane helices containing extremes in contents of individual amino acids or combinations of amino acids with similar physicochemical properties were identified and linked to structure and/or function. In addition to overall median and average values, all results were analysed for proteins originating from different types of organism (prokaryotic, eukaryotic, viral) and for subgroups of receptors, channels, transporters and others.     

The Positive Inside Rule Is Stronger When Followed by a Transmembrane Helix

The translocon recognizes transmembrane helices with sufficient level of hydrophobicity and inserts them into the membrane. However, sometimes less hydrophobic helices are also recognized. Positive inside rule, orientational preferences of and specific interactions with neighboring helices have been shown to aid in the recognition of these helices, at least in artificial systems. To better understand how the translocon inserts marginally hydrophobic helices, we studied three naturally occurring marginally hydrophobic helices, which were previously shown to require the subsequent helix for efficient translocon recognition. We find no evidence for specific interactions when we scan all residues in the subsequent helices. Instead, we identify arginines located at the N-terminal part of the subsequent helices that are crucial for the recognition of the marginally hydrophobic transmembrane helices, indicating that the positive inside rule is important. However, in two of the constructs, these arginines do not aid in the recognition without the rest of the subsequent helix; that is, the positive inside rule alone is not sufficient. Instead, the improved recognition of marginally hydrophobic helices can here be explained as follows: the positive inside rule provides an orientational preference of the subsequent helix, which in turn allows the marginally hydrophobic helix to be inserted; that is, the effect of the positive inside rule is stronger if positively charged residues are followed by a transmembrane helix. Such a mechanism obviously cannot aid C-terminal helices, and consequently, we find that the terminal helices in multi-spanning membrane proteins are more hydrophobic than internal helices.     

A knowledge-based scale for amino acid membrane propensity

In this article, a membrane-propensity scale for amino acids is derived using only two ingredients: (i) a set of transmembrane helices segments from membrane protein crystal structures and (ii) the request that each component of the set has a free energy lower than that of a typical soluble protein sequence of the same length. Although the most widely used hydropathy scales satisfy this request, we use an optimization procedure that allows for extraction of an optimal scale, which correlates equally well with those scales. We show that, if the choice of the sequence database is accurate, significant knowledge-based scales, which are robust with respect to changes in the learning set, can be easily derived. The obtained scales can be used for transmembrane helices prediction. The predictive power of one of these scales is tested on membrane proteins, soluble proteins, and signal peptides databases, finding that its performances is comparable with those of the hydropathy scales.     

Large Tilts in Transmembrane Helices Can Be Induced during Tertiary Structure Formation

While early structural models of helix-bundle integral membrane proteins posited that the transmembrane α-helices [transmembrane helices (TMHs)] were orientated more or less perpendicular to the membrane plane, there is now ample evidence from high-resolution structures that many TMHs have significant tilt angles relative to the membrane. Here, we address the question whether the tilt is an intrinsic property of the TMH in question or if it is imparted on the TMH during folding of the protein. Using a glycosylation mapping technique, we show that four highly tilted helices found in multi-spanning membrane proteins all have much shorter membrane-embedded segments when inserted by themselves into the membrane than seen in the high-resolution structures. This suggests that tilting can be induced by tertiary packing interactions within the protein, subsequent to the initial membrane-insertion step.     

Marginally hydrophobic transmembrane α‐helices shaping membrane protein folding

Cells have developed an incredible machinery to facilitate the insertion of membrane proteins into the membrane. While we have a fairly good understanding of the mechanism and determinants of membrane integration, more data is needed to understand the insertion of membrane proteins with more complex insertion and folding pathways. This review will focus on marginally hydrophobic transmembrane helices and their influence on membrane protein folding. These weakly hydrophobic transmembrane segments are by themselves not recognized by the translocon and therefore rely on local sequence context for membrane integration. How can such segments reside within the membrane? We will discuss this in the light of features found in the protein itself as well as the environment it resides in. Several characteristics in proteins have been described to influence the insertion of marginally hydrophobic helices. Additionally, the influence of biological membranes is significant. To begin with, the actual cost for having polar groups within the membrane may not be as high as expected; the presence of proteins in the membrane as well as characteristics of some amino acids may enable a transmembrane helix to harbor a charged residue. The lipid environment has also been shown to directly influence the topology as well as membrane boundaries of transmembrane helices—implying a dynamic relationship between membrane proteins and their environment.     

In silico predictions of LH2 ring sizes from the crystal structure of a single subunit using molecular dynamics simulations

Most of the currently known light‐harvesting complexes 2 (LH2) rings are formed by 8 or 9 subunits. As of now, questions like “what factors govern the LH2 ring size?” and “are there other ring sizes possible?” remain largely unanswered. Here, we investigate by means of molecular dynamics (MD) simulations and stochastic modeling the possibility of predicting the size of an LH2 ring from the sole knowledge of the high resolution crystal structure of a single subunit. Starting with single subunits of two LH2 rings with known size, that is, an 8‐ring from Rs. moliscianum (MOLI) and a 9‐ring from Rps. acidophila (ACI), and one with unknown size (referred to as X), we build atomic models of subunit dimers corresponding to assumed 8‐, 9‐, and 10‐ring geometries. After inserting each of the dimers into a lipid‐water environment, we determine the preferred angle between the corresponding subunits by three methods: (1) energy minimization, (2) free MD simulations, and (3) potential of mean force calculations. We find that the results from all three methods are consistent with each other, and when taken together, it allows one to predict with reasonable level of confidence the sizes of the corresponding ring structures. One finds that X and ACI very likely form a 9‐ring, while MOLI is more likely to form an 8‐ring than a 9‐ring. Finally, we discuss both the merits and limitations of all three prediction methods. Proteins 2011; © 2011 Wiley‐Liss, Inc.     

How strongly do sequence conservation patterns and empirical scales correlate with exposure patterns of transmembrane helices of membrane proteins?

Given the difficulty in determining high-resolution structures of helical membrane proteins, sequence-based prediction methods can be useful in elucidating diverse physiological processes mediated by this important class of proteins. Predicting the angular orientations of transmembrane (TM) helices about the helix axes, based on the helix parameters from electron microscopy data, is a classical problem in this regard. This problem has triggered the development of a number of different empirical scales. Recently, sequence conservation patterns were also made use of for improved predictions. Empirical scales and sequence conservation patterns (collectively termed as "prediction scales") have also found frequent applications in other research areas of membrane proteins: for example, in structure modeling and in prediction of buried TM helices. This trend is expected to grow in the near future unless there are revolutionary developments in the experimental characterization of membrane proteins. Thus, it is timely and imperative to carry out a comprehensive benchmark test over the prediction scales proposed so far to determine their pros and cons. In the current analysis, we use exposure patterns of TM helices as a golden standard, because if one develops a prediction scale that correlates perfectly with exposure patterns of TM helices, it will enable one to predict buried residues (or buried faces) of TM helices with an accuracy of 100%. Our analysis reveals several important points. (1) It demonstrates that sequence conservation patterns are much more strongly correlated with exposure patterns of TM helices than empirical scales. (2) Scales that were specifically parameterized using structure data (structure-based scales) display stronger correlation than hydrophobicity-based scales, as expected. (3) A nonnegligible difference is observed among the structure-based scales in their correlational property, suggesting that not every learning algorithm is equally effective. (4) A straightforward framework of optimally combining sequence conservation patterns and empirical scales is proposed, which reveals that improvements gained from combining the two sources of information are not dramatic in almost all cases. In turn, this calls for the development of fundamentally different scales that capture the essentials of membrane protein folding for substantial improvements.     

The effect of nucleotide bias upon the composition and prediction of transmembrane helices

Transmembrane helices are the most readily predictable secondary structure components of proteins. They can be predicted to a high degree of accuracy in a variety of ways. Many of these methods compare new sequence data with the sequence characteristics of known transmembrane domains. However, the known transmembrane sequences are not necessarily representative of a particular organism. We attempt to demonstrate that parameters optimized for the known transmembrane domains are far from optimal when predicting transmembrane regions in a given genome. In particular, we have tested the effect of nucleotide bias upon the composition and hence the prediction characteristics of transmembrane helices. Our analysis shows that nucleotide bias of a genome has a strong and predictable influence upon the occurrences of several of the most important hydrophobic amino acids found within transmembrane helices. Thus, we show that nucleotide bias should be taken into account when determining putative transmembrane domains from sequence data.     

Analysis of protein transmembrane helical regions by a neural network.

Neural networks were used to generalize common themes found in transmembrane-spanning protein helices. Various-sized databases were used containing nonoverlapping sequences, each 25 amino acids long. Training consisted of sorting these sequences into 1 of 2 groups: transmembrane helical peptides or nontransmembrane peptides. Learning was measured using a test set 10% the size of the training set. As training set size increased from 214 sequences to 1,751 sequences, learning increased in a nonlinear manner from 75% to a high of 98%, then declined to a low of 87%. The final training database consisted of roughly equal numbers of transmembrane (928) and nontransmembrane (1,018) sequences. All transmembrane sequences were entered into the database with respect to their lipid membrane orientation: from inside the membrane to outside. Generalized transmembrane helix and nontransmembrane peptides were constructed from the maximally weighted connecting strengths of fully trained networks. Four generalized transmembrane helices were found to contain 9 consensus residues: a K-R-F triplet was found at the inside lipid interface, 2 isoleucine and 2 other phenylalanine residues were present in the helical body, and 2 tryptophan residues were found near the outside lipid interface. As a test of the training method, bacteriorhodopsin was examined to determine the position of its 7 transmembrane helices.     

Computational differentiation of N-terminal signal peptides and transmembrane helices

Signal peptides and transmembrane helices both contain a stretch of hydrophobic amino acids. This common feature makes it difficult for signal peptide and transmembrane helix predictors to correctly assign identity to stretches of hydrophobic residues near the N-terminal methionine of a protein sequence. The inability to reliably distinguish between N-terminal transmembrane helix and signal peptide is an error with serious consequences for the prediction of protein secretory status or transmembrane topology. In this study, we report a new method for differentiating protein N-terminal signal peptides and transmembrane helices. Based on the sequence features extracted from hydrophobic regions (amino acid frequency, hydrophobicity, and the start position), we set up discriminant functions and examined them on non-redundant datasets with jackknife tests. This method can incorporate other signal peptide prediction methods and achieve higher prediction accuracy. For Gram-negative bacterial proteins, 95.7% of N-terminal signal peptides and transmembrane helices can be correctly predicted (coefficient 0.90). Given a sensitivity of 90%, transmembrane helices can be identified from signal peptides with a precision of 99% (coefficient 0.92). For eukaryotic proteins, 94.2% of N-terminal signal peptides and transmembrane helices can be correctly predicted with coefficient 0.83. Given a sensitivity of 90%, transmembrane helices can be identified from signal peptides with a precision of 87% (coefficient 0.85). The method can be used to complement current transmembrane protein prediction and signal peptide prediction methods to improve their prediction accuracies.     

Interaction and conformational dynamics of membrane‐spanning protein helices

Within 1 or 2 decades, the reputation of membrane‐spanning α‐helices has changed dramatically. Once mostly regarded as dull membrane anchors, transmembrane domains are now recognized as major instigators of protein–protein interaction. These interactions may be of exquisite specificity in mediating assembly of stable membrane protein complexes from cognate subunits. Further, they can be reversible and regulatable by external factors to allow for dynamic changes of protein conformation in biological function. Finally, these helices are increasingly regarded as dynamic domains. These domains can move relative to each other in different functional protein conformations. In addition, small‐scale backbone fluctuations may affect their function and their impact on surrounding lipid shells. Elucidating the ways by which these intricate structural features are encoded by the amino acid sequences will be a fascinating subject of research for years to come.     

Simultaneous prediction of protein secondary structure and transmembrane spans

Prediction of transmembrane spans and secondary structure from the protein sequence is generally the first step in the structural characterization of (membrane) proteins. Preference of a stretch of amino acids in a protein to form secondary structure and being placed in the membrane are correlated. Nevertheless, current methods predict either secondary structure or individual transmembrane states. We introduce a method that simultaneously predicts the secondary structure and transmembrane spans from the protein sequence. This approach not only eliminates the necessity to create a consensus prediction from possibly contradicting outputs of several predictors but bears the potential to predict conformational switches, i.e., sequence regions that have a high probability to change for example from a coil conformation in solution to an α‐helical transmembrane state. An artificial neural network was trained on databases of 177 membrane proteins and 6048 soluble proteins. The output is a 3 × 3 dimensional probability matrix for each residue in the sequence that combines three secondary structure types (helix, strand, coil) and three environment types (membrane core, interface, solution). The prediction accuracies are 70.3% for nine possible states, 73.2% for three‐state secondary structure prediction, and 94.8% for three‐state transmembrane span prediction. These accuracies are comparable to state‐of‐the‐art predictors of secondary structure (e.g., Psipred) or transmembrane placement (e.g., OCTOPUS). The method is available as web server and for download at www.meilerlab.org . Proteins 2013; 81:1127–1140. © 2013 Wiley Periodicals, Inc.     

Identification of novel sphingolipid-binding motifs in mammalian membrane proteins

Specific interactions between transmembrane proteins and sphingolipids is a poorly understood phenomenon, and only a couple of instances have been identified. The best characterized example is the sphingolipid-binding motif VXXTLXXIY found in the transmembrane helix of the vesicular transport protein p24. Here, we have used a simple motif-probability algorithm (MOPRO) to identify proteins that contain putative sphingolipid-binding motifs in a dataset comprising proteomes from mammalian organisms. From these motif-containing candidate proteins, four with different numbers of transmembrane helices were selected for experimental study: i) major histocompatibility complex II Q alpha chain subtype (DQA1), ii) GPI-attachment protein 1 (GAA1), iii) tetraspanin-7 TSN7, and iv), metabotropic glutamate receptor 2 (GRM2). These candidates were subjected to photo-affinity labeling using radiolabeled sphingolipids, confirming all four candidate proteins as sphingolipid-binding proteins. The sphingolipid-binding motifs are enriched in the 7TM family of G-protein coupled receptors, predominantly in transmembrane helix 6. The ability of the motif-containing candidate proteins to bind sphingolipids with high specificity opens new perspectives on their respective regulation and function.     

Inter-helical hydrogen bond formation during membrane protein integration into the ER membrane

Recent work has shown that efficient di- or trimerization of hydrophobic transmembrane helices in detergent micelles or lipid bilayers can be driven by inter-helix hydrogen bonding involving polar residues such as Asn or Asp. Using in vitro translation in the presence of rough microsomes of a model integral membrane protein, we now show that the formation of so-called helical hairpins, two tightly spaced transmembrane helices connected by a short loop, can likewise be promoted by the introduction of Asn-Asn or Asp-Asp pairs in a long transmembrane hydrophobic segment. These observations suggest that inter-helix hydrogen bonds can form within the context of the Sec61 translocon in the endoplasmic reticulum, implying that hydrophobic segments in a nascent polypeptide chain in transit through the Sec61 channel have immediate access to a non-aqueous subcompartment within the translocon.