Peptide Design Aided by Neural Networks:  Biological Activity of Artificial Signal Peptidase I Cleavage Sites

Wrede, Paul; Landt, Olfert; Klages, S.; Fatemi, Afshin; Hahn, Ulrich; Schneider, Gisbert

doi:10.1021/bi9726032

Cited by 27 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remarkably, these all contained Trp, especially at positions − 2 and − 5, and they had h-regions dominated by Phe. The highest-scoring cleavage site region was subsequently tested in vivo for their ability to promote secretion in an E. coli expression system [ 55 ]. Indeed, the Phe- and Trp-rich construct (FFFFGWYGWA↓RE) was fully cleavable, but so were the wild type (LAGFATVAQA↓AC) and a “consensus” pattern derived from a simpler, weight matrix-like approach (VVIMSASAMA↓AC).…”

Section: Signal Peptide Predictionmentioning

confidence: 99%

A Brief History of Protein Sorting Prediction

et al. 2019

View full text Add to dashboard Cite

Ever since the signal hypothesis was proposed in 1971, the exact nature of signal peptides has been a focus point of research. The prediction of signal peptides and protein subcellular location from amino acid sequences has been an important problem in bioinformatics since the dawn of this research field, involving many statistical and machine learning technologies. In this review, we provide a historical account of how position-weight matrices, artificial neural networks, hidden Markov models, support vector machines and, lately, deep learning techniques have been used in the attempts to predict where proteins go. Because the secretory pathway was the first one to be studied both experimentally and through bioinformatics, our main focus is on the historical development of prediction methods for signal peptides that target proteins for secretion; prediction methods to identify targeting signals for other cellular compartments are treated in less detail.

show abstract

Section: Signal Peptide Predictionmentioning

confidence: 99%

A Brief History of Protein Sorting Prediction

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Knowing which proteins are cytosolic and which are targeted to an organelle will help assembling metabolic pathways that putatively occur in the organelle [1]. The prediction of the subcellular localization of proteins has a tradition in bioinformatics, and many such computational tools have been developed over the past two decades, facilitating the identification and even the design of targeting signal features [3][4][5][6]. The first prediction methods yielded 70-80% accuracy for secretory proteins [7,8], current techniques reach up to 95% accuracy with a reduced risk of false-positive predictions [9,10].…”

Section: Introductionmentioning

confidence: 99%

Advances in the prediction of protein targeting signals

Schneider

Fechner

2004

Proteomics

Self Cite

View full text Add to dashboard Cite

Enlarged sets of reference data and special machine learning approaches have improved the accuracy of the prediction of protein subcellular localization. Recent approaches report over 95% correct predictions with low fractions of false-positives for secretory proteins. A clear trend is to develop specifically tailored organism- and organelle-specific prediction tools rather than using one general method. Focus of the review is on machine learning systems, highlighting four concepts: the artificial neural feed-forward network, the self-organizing map (SOM), the Hidden-Markov-Model (HMM), and the support vector machine (SVM).

show abstract

“…Various approaches including computational designs have been attempted with limited success in search of more highly functional peptides to serve as substrates for the E. coli enzyme (17)(18)(19)(20). For instance, peptide libraries were created by incorporating randomized sequences into the signal peptide of TEM-1 ␤-lactamase, varying six amino acid residues between Ϫ4 and ϩ2 positions around the signal peptidase cleavage site (19).…”

mentioning

confidence: 99%

Discovery of Substrate for Type I Signal Peptidase SpsB fromStaphylococcus aureus

Sharkov¹,

Cai

2002

Journal of Biological Chemistry

View full text Add to dashboard Cite

Based on the kinetic model of substrate phage proteolysis, we have formulated a strategy for best manipulating the conditions in screening phage display libraries for protease substrates (Sharkov, N. A., Davis, R. M., Reidhaar-Olson, J. F., Navre, M., and Cai, D. (2001) J. Biol. Chem. 276, 10788 -10793). This strategy is exploited in the present study with signal peptidase SpsB from Staphylococcus aureus. We demonstrate that highly active substrate phage clones can be isolated from a phage display library by systematically tuning the selection stringency in screening. Several of the selected clones exhibit superior reactivity over a control, the best clone, SIIIRIII-8, showing >100-fold improvement. Because no conserved sequence features were readily revealed that could allow delineation of the active and unreactive clones, the sequences identified in five of the active clones were tested as synthetic dodecamers, Ac-AGX 8 GA-NH 2 . Using electrospray ionization mass spectrometry, we show that four of these peptides can be cleaved by SpsB and that Ala is the P1 residue exclusively and Ala or Leu the P3 residue, in keeping with the (؊3, ؊1) rule for substrate recognition by signal peptidase. Our successful screening with SpsB demonstrated the general applicability of the screening strategy and allowed us to isolate the first peptide substrates for the enzyme.Bacterial type I signal peptidase is responsible for cleaving the signal peptide from precursor proteins, and its activity is an integral part of the export and maturation of secreted proteins in vivo. The essential function of the enzyme to bacterial cell viability has been demonstrated using genetic approaches with both Gram-positive and Gram-negative organisms (1-3), supporting the notion that the signal peptidase is potentially an antibacterial target (4). Drug discovery efforts with the enzyme, however, may be hampered by the lack of an effective in vitro assay employing a nonprotein substrate such as a peptide (4).Our current understanding is that signal peptides are highly variable in sequence (5). Based on the studies carried out over the past 2 decades, it has been established that the recognition sites for signal peptidases lie between Ϫ6 and ϩ1 in sequences encompassing the site of cleavage (6 -12). Sequence conservation analyses of a large panel of naturally occurring signal peptides in bacteria and eukaryotes reveal that the predominant residue at the P1 site is Ala and that the predominant residues at the P3 site are large aliphatic residues (Leu, Ile, Val) as well as Ala and Ser, a consensus dubbed the (Ϫ3, Ϫ1) rule (9 -11). The (Ϫ3, Ϫ1) rule also holds for the cleavage of engineered preproteins in vivo as well as in vitro (6 -8, 12). The reaction of signal peptidases with synthetic peptides, on the other hand, is not as well explored as with protein substrates. For the signal peptidase LepB from Escherichia coli, the best characterized signal peptidase, Ala was found as the only residue permitted at the P1 site through single amino acid replace...

show abstract

Peptide Design Aided by Neural Networks: Biological Activity of Artificial Signal Peptidase I Cleavage Sites

Cited by 27 publications

References 39 publications

A Brief History of Protein Sorting Prediction

A Brief History of Protein Sorting Prediction

Advances in the prediction of protein targeting signals

Discovery of Substrate for Type I Signal Peptidase SpsB fromStaphylococcus aureus

Contact Info

Product

Resources

About