The prediction of organelle-targeting peptides in eukaryotic proteins with Grammatical-Restrained Hidden Conditional Random Fields

Mitochondrial targeting of Candida albicans SPFH proteins and requirement of stomatins for SDS-induced stress tolerance

The SPFH (stomatin, prohibitin, flotillin, and HflK/HflC) protein superfamily is conserved across all domains of life. Fungal SPFH proteins are required for respiration, stress adaptation, and membrane scaffolding. In the yeast Candida albicans, stomatin-like protein 3 (Slp3) forms punctate foci at the plasma membrane, and SLP3 overexpression causes cell death following exposure to the surfactant, SDS, and the oxidative stressor, H2O2. Here, we sought to determine the cellular localization and functionally characterize stomatin-like protein 2 (Slp2), prohibitin-1 (Phb1), prohibitin-2 (Phb2), and prohibitin-12 (Phb12) in C. albicans. Cytological and western blotting results showed that Slp2-Gfp/Rfp and prohibitin-Gfp fusion proteins localize to the mitochondrion in yeast cells. Growth assay results did not identify any respiration defects in a panel of stomatin and prohibitin mutant strains, suggesting that SPFH respiratory function has diverged in C. albicans from other model eukaryotes. However, a slp2Δ/Δ/slp3Δ/Δ double mutant strain grew poorly in the presence of 0.08% SDS, accumulated intracellular reactive oxidative species, and displayed aberrant ergosterol distribution in the plasma membrane. These phenotypes were not observed in slp2Δ/Δ or slp3Δ/Δ single mutants, indicating a possible indirect genetic interaction between SLP2 and SLP3. In addition, slp2Δ/Δ and slp2Δ/Δ/slp3Δ/Δ mutant strains were slightly resistant to the antifungal drug, fluconazole. Collectively, these findings reveal the cellular localization of Slp2, Phb1, Phb2, and Phb12, highlight the significance of stomatins in C. albicans SDS stress tolerance, and, for the first time, associate stomatins with antifungal resistance.

IMPORTANCE

Stomatins and prohibitins coordinate respiration and stress adaptation in fungi. Invasive mycoses caused by Candida albicans are a significant cause of morbidity, and candidemia patients show high mortality rates worldwide. Mitochondria are essential for C. albicans commensalism and virulence, and mitochondrial proteins are targets for antifungal interventions. C. albicans encodes five SPFH proteins: two stomatin-like proteins and three prohibitins. We have previously shown that Slp3 is important for C. albicans adaptation to various types of environmental stress. Moreover, synthetic compounds that bind to mammalian prohibitins inhibit C. albicans filamentation and are fungicidal. However, there is limited information available regarding the remaining SPFH proteins. Our findings show that mitochondrial localization of SPFH proteins is conserved in C. albicans. In addition, we demonstrate the importance of stomatins in plasma membrane and mitochondrial stress tolerance.

Prediction of mitochondrial targeting signals and their cleavage sites

In this chapter we survey prediction tools and computational methods for the prediction of amino acid sequence elements which target proteins to the mitochondria. We will primarily focus on the prediction of N-terminal mitochondrial targeting signals (MTSs) and their N-terminal cleavage sites by mitochondrial peptidases. We first give practical details useful for using and installing some prediction tools. Then we describe procedures for preparing datasets of MTS containing proteins for statistical analysis or development of new prediction methods. Following that we lightly survey some of the computational techniques used by prediction tools. Finally, after discussing some caveats regarding the reliability of such methods to predict the effects of mutations on MTS function; we close with a discussion of possible future directions of computer prediction methods related to mitochondrial proteins.

Computer-Aided Prediction of Protein Mitochondrial Localization

Protein sequences, directly translated from genomic data, need functional and structural annotation. Together with molecular function and biological process, subcellular localization is an important feature necessary for understanding the protein role and the compartment where the mature protein is active. In the case of mitochondrial proteins, their precursor sequences translated by the ribosome machinery include specific patterns from which it is possible not only to recognize their final destination within the organelle but also which of the mitochondrial subcompartments the protein is intended for. Four compartments are routinely discriminated, including the inner and the outer membranes, the intermembrane space, and the matrix. Here we discuss to which extent it is feasible to develop computational methods for detecting mitochondrial targeting peptides in the precursor sequence and to discriminate their final destination in the organelle. We benchmark two of our methods on the general task of recognizing human mitochondrial proteins endowed with an experimentally characterized targeting peptide (TPpred3) and predicting which submitochondrial compartment is the final destination (DeepMito). We describe how to adopt our web servers in order to discriminate which human proteins are endowed with mitochondrial targeting peptides, the position of cleavage sites, and which submitochondrial compartment are intended for. By this, we add some other 1788 human proteins to the 450 ones already manually annotated in UniProt with a mitochondrial targeting peptide, providing for each of them also the characterization of the suborganellar localization.

PPTPP: a novel therapeutic peptide prediction method using physicochemical property encoding and adaptive feature representation learning

MOTIVATION

Peptide is a promising candidate for therapeutic and diagnostic development due to its great physiological versatility and structural simplicity. Thus, identifying therapeutic peptides and investigating their properties are fundamentally important. As an inexpensive and fast approach, machine learning-based predictors have shown their strength in therapeutic peptide identification due to excellences in massive data processing. To date, no reported therapeutic peptide predictor can perform high-quality generic prediction and informative physicochemical properties (IPPs) identification simultaneously.

RESULTS

In this work, Physicochemical Property-based Therapeutic Peptide Predictor (PPTPP), a Random Forest-based prediction method was presented to address this issue. A novel feature encoding and learning scheme were initiated to produce and rank physicochemical property-related features. Besides being capable of predicting multiple therapeutics peptides with high comparability to established predictors, the presented method is also able to identify peptides' informative IPP. Results presented in this work not only illustrated the soundness of its working capacity but also demonstrated its potential for investigating other therapeutic peptides.

AVAILABILITY AND IMPLEMENTATION

https://github.com/YPZ858/PPTPP.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Differential processing and localization of human Nocturnin controls metabolism of mRNA and nicotinamide adenine dinucleotide cofactors

Nocturnin (NOCT) is a eukaryotic enzyme that belongs to a superfamily of exoribonucleases, endonucleases, and phosphatases. In this study, we analyze the expression, processing, localization, and cellular functions of human NOCT. We find that NOCT protein is differentially expressed and processed in a cell and tissue type-specific manner to control its localization to the cytoplasm or mitochondrial exterior or interior. The N terminus of NOCT is necessary and sufficient to confer import and processing in the mitochondria. We measured the impact of cytoplasmic NOCT on the transcriptome and observed that it affects mRNA levels of hundreds of genes that are significantly enriched in osteoblast, neuronal, and mitochondrial functions. Recent biochemical data indicate that NOCT dephosphorylates NADP(H) metabolites, and thus we measured the effect of NOCT on these cofactors in cells. We find that NOCT increases NAD(H) and decreases NADP(H) levels in a manner dependent on its intracellular localization. Collectively, our data indicate that NOCT can regulate levels of both mRNAs and NADP(H) cofactors in a manner specified by its location in cells.

Mitochondrial localization of Dictyostelium discoideum dUTPase mediated by its N-terminus

OBJECTIVE

The nuclear and mitochondrial genomes of Dictyostelium discoideum, a unicellular eukaryote, have relatively high A+T-contents of 77.5% and 72.65%, respectively. To begin to investigate how the pyrimidine biosynthetic pathway fulfills the demand for dTTP, we determined the catalytic properties and structure of the key enzyme deoxyuridine triphosphate nucleotidohydrolase (dUTPase) that hydrolyzes dUTP to dUMP, the precursor of dTTP.

RESULTS

The annotated genome of D. discoideum identifies a gene encoding a polypeptide containing the five conserved motifs of homotrimeric dUTPases. Recombinant proteins, comprised of either full-length or core polypeptides with all conserved motifs but lacking residues 1-37 of the N-terminus, were active dUTPases. Crystallographic analyses of the core enzyme indicated that the C-termini, normally flexible, were constrained by interactions with the shortened N-termini that arose from the loss of residues 1-37. This allowed greater access of dUTP to active sites, resulting in enhanced catalytic parameters. A tagged protein comprised of the N-terminal forty amino acids of dUTPase fused to green fluorescent protein (GFP) was expressed in D. discoideum cells. Supporting a prediction of mitochondrial targeting information within the N-terminus, localization and subcellular fractionation studies showed GFP to be in mitochondria. N-terminal sequencing of immunoprecipitated GFP revealed the loss of the dUTPase sequence upon import into the organelle.

DeepSig: deep learning improves signal peptide detection in proteins

Motivation

The identification of signal peptides in protein sequences is an important step toward protein localization and function characterization.

Results

Here, we present DeepSig, an improved approach for signal peptide detection and cleavage-site prediction based on deep learning methods. Comparative benchmarks performed on an updated independent dataset of proteins show that DeepSig is the current best performing method, scoring better than other available state-of-the-art approaches on both signal peptide detection and precise cleavage-site identification.

Availability and implementation

DeepSig is available as both standalone program and web server at https://deepsig.biocomp.unibo.it. All datasets used in this study can be obtained from the same website.

Contact

pierluigi.martelli@unibo.it.

Supplementary information

Supplementary data are available at Bioinformatics online.

A Guide to Computational Methods for Predicting Mitochondrial Localization

Predicting mitochondrial localization of proteins remains challenging for two main reasons: (1) Not only one but several mitochondrial localization signals exist, which primarily dictate the final destination of a protein in this organelle. However, most localization prediction algorithms rely on the presence of a so-called presequence (or N-terminal mitochondrial targeting peptide, mTP), which occurs in only ~70% of mitochondrial proteins. (2) The presequence is highly divergent on sequence level and therefore difficult to identify on the computer.In this chapter, we review a number of protein localization prediction programs and propose a strategy to predict mitochondrial localization. Finally, we give some helpful suggestions for bench scientists when working with mitochondrial protein candidates in silico.

Poldip2 is an oxygen-sensitive protein that controls PDH and αKGDH lipoylation and activation to support metabolic adaptation in hypoxia and cancer

The present work establishes that the addition of the prosthetic group lipoic acid to catabolic enzymes is a dynamically regulated posttranslational modification that increases metabolic plasticity under hypoxia and in cancer cells. We show that that the polymerase-δ interacting protein 2 (Poldip2) is an oxygen-sensitive protein that regulates the lipoylation and activation of the pyruvate and α-ketoglutarate dehydrogenase complexes. Additionally, our work reveals that mitochondrial peptidases participate in an integrated response needed for metabolic adaptation. This study positions Poldip2 as a key regulator of mitochondrial function and cell metabolism.

Although the addition of the prosthetic group lipoate is essential to the activity of critical mitochondrial catabolic enzymes, its regulation is unknown. Here, we show that lipoylation of the pyruvate dehydrogenase and α-ketoglutarate dehydrogenase (αKDH) complexes is a dynamically regulated process that is inhibited under hypoxia and in cancer cells to restrain mitochondrial respiration. Mechanistically, we found that the polymerase-δ interacting protein 2 (Poldip2), a nuclear-encoded mitochondrial protein of unknown function, controls the lipoylation of the pyruvate and α-KDH dihydrolipoamide acetyltransferase subunits by a mechanism that involves regulation of the caseinolytic peptidase (Clp)-protease complex and degradation of the lipoate-activating enzyme Ac-CoA synthetase medium-chain family member 1 (ACSM1). ACSM1 is required for the utilization of lipoic acid derived from a salvage pathway, an unacknowledged lipoylation mechanism. In Poldip2-deficient cells, reduced lipoylation represses mitochondrial function and induces the stabilization of hypoxia-inducible factor 1α (HIF-1α) by loss of substrate inhibition of prolyl-4-hydroxylases (PHDs). HIF-1α–mediated retrograde signaling results in a metabolic reprogramming that resembles hypoxic and cancer cell adaptation. Indeed, we observe that Poldip2 expression is down-regulated by hypoxia in a variety of cell types and basally repressed in triple-negative cancer cells, leading to inhibition of lipoylation of the pyruvate and α-KDH complexes and mitochondrial dysfunction. Increasing mitochondrial lipoylation by forced expression of Poldip2 increases respiration and reduces the growth rate of cancer cells. Our work unveils a regulatory mechanism of catabolic enzymes required for metabolic plasticity and highlights the role of Poldip2 as key during hypoxia and cancer cell metabolic adaptation.

Roles for the Rad27 Flap Endonuclease in Mitochondrial Mutagenesis and Double-Strand Break Repair in Saccharomyces cerevisiae

The structure-specific nuclease, Rad27p/FEN1, plays a crucial role in DNA repair and replication mechanisms in the nucleus. Genetic assays using the rad27-∆ mutant have shown altered rates of DNA recombination, microsatellite instability, and point mutation in mitochondria. In this study, we examined the role of Rad27p in mitochondrial mutagenesis and double-strand break (DSB) repair in Saccharomyces cerevisiae Our findings show that Rad27p is essential for efficient mitochondrial DSB repair by a pathway that generates deletions at a region flanked by direct repeat sequences. Mutant analysis suggests that both exonuclease and endonuclease activities of Rad27p are required for its role in mitochondrial DSB repair. In addition, we found that the nuclease activities of Rad27p are required for the prevention of mitochondrial DNA (mtDNA) point mutations, and in the generation of spontaneous mtDNA rearrangements. Overall, our findings underscore the importance of Rad27p in the maintenance of mtDNA, and demonstrate that it participates in multiple DNA repair pathways in mitochondria, unlinked to nuclear phenotypes.

Towards understanding the evolution and functional diversification of DNA-containing plant organelles

Plastids and mitochondria derive from prokaryotic symbionts that lost most of their genes after the establishment of endosymbiosis. In consequence, relatively few of the thousands of different proteins in these organelles are actually encoded there. Most are now specified by nuclear genes. The most direct way to reconstruct the evolutionary history of plastids and mitochondria is to sequence and analyze their relatively small genomes. However, understanding the functional diversification of these organelles requires the identification of their complete protein repertoires – which is the ultimate goal of organellar proteomics. In the meantime, judicious combination of proteomics-based data with analyses of nuclear genes that include interspecies comparisons and/or predictions of subcellular location is the method of choice. Such genome-wide approaches can now make use of the entire sequences of plant nuclear genomes that have emerged since 2000. Here I review the results of these attempts to reconstruct the evolution and functions of plant DNA-containing organelles, focusing in particular on data from nuclear genomes. In addition, I discuss proteomic approaches to the direct identification of organellar proteins and briefly refer to ongoing research on non-coding nuclear DNAs of organellar origin (specifically, nuclear mitochondrial DNA and nuclear plastid DNA).

TPpred3 detects and discriminates mitochondrial and chloroplastic targeting peptides in eukaryotic proteins

MOTIVATION

Molecular recognition of N-terminal targeting peptides is the most common mechanism controlling the import of nuclear-encoded proteins into mitochondria and chloroplasts. When experimental information is lacking, computational methods can annotate targeting peptides, and determine their cleavage sites for characterizing protein localization, function, and mature protein sequences. The problem of discriminating mitochondrial from chloroplastic propeptides is particularly relevant when annotating proteomes of photosynthetic Eukaryotes, endowed with both types of sequences.

RESULTS

Here, we introduce TPpred3, a computational method that given any Eukaryotic protein sequence performs three different tasks: (i) the detection of targeting peptides; (ii) their classification as mitochondrial or chloroplastic and (iii) the precise localization of the cleavage sites in an organelle-specific framework. Our implementation is based on our TPpred previously introduced. Here, we integrate a new N-to-1 Extreme Learning Machine specifically designed for the classification task (ii). For the last task, we introduce an organelle-specific Support Vector Machine that exploits sequence motifs retrieved with an extensive motif-discovery analysis of a large set of mitochondrial and chloroplastic proteins. We show that TPpred3 outperforms the state-of-the-art methods in all the three tasks.

AVAILABILITY AND IMPLEMENTATION

The method server and datasets are available at http://tppred3.biocomp.unibo.it.

CONTACT

gigi@biocomp.unibo.it

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Proteome-scale identification and characterization of mitochondria targeting proteins of Mycobacterium avium subspecies paratuberculosis: Potential virulence factors modulating host mitochondrial function

Mycobacterium avium subsp. paratuberculosis is the etiological agent of Johne's Disease among ruminants. During the course of infection, it expresses a number of proteins for its successful persistence inside the host that cause variety of physiological abnormalities in the host. Mitochondrion is one of the attractive targets for pathogenic bacteria. Employing a proteome-wide sequence and structural signature based approach we have identified 46 M. avium subsp. paratuberculosis proteins as potential targets for the host mitochondrial targeting. These may act as virulence factors modulating mitochondrial physiology for bacterial survival and immune evasion inside the host cells.

Computer-based prediction of mitochondria-targeting peptides

Computational methods are invaluable when protein sequences, directly derived from genomic data, need functional and structural annotation. Subcellular localization is a feature necessary for understanding the protein role and the compartment where the mature protein is active and very difficult to characterize experimentally. Mitochondrial proteins encoded on the cytosolic ribosomes carry specific patterns in the precursor sequence from where it is possible to recognize a peptide targeting the protein to its final destination. Here we discuss to which extent it is feasible to develop computational methods for detecting mitochondrial targeting peptides in the precursor sequences and benchmark our and other methods on the human mitochondrial proteins endowed with experimentally characterized targeting peptides. Furthermore, we illustrate our newly implemented web server and its usage on the whole human proteome in order to infer mitochondrial targeting peptides, their cleavage sites, and whether the targeting peptide regions contain or not arginine-rich recurrent motifs. By this, we add some other 2,800 human proteins to the 124 ones already experimentally annotated with a mitochondrial targeting peptide.

TPpred2: improving the prediction of mitochondrial targeting peptide cleavage sites by exploiting sequence motifs

SUMMARY

Targeting peptides are N-terminal sorting signals in proteins that promote their translocation to mitochondria through the interaction with different protein machineries. We recently developed TPpred, a machine learning-based method scoring among the best ones available to predict the presence of a targeting peptide into a protein sequence and its cleavage site. Here we introduce TPpred2 that improves TPpred performances in the task of identifying the cleavage site of the targeting peptides. TPpred2 is now available as a web interface and as a stand-alone version for users who can freely download and adopt it for processing large volumes of sequences. Availability and implementaion: TPpred2 is available both as web server and stand-alone version at http://tppred2.biocomp.unibo.it.

CONTACT

gigi@biocomp.unibo.it

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.