Protein and Peptide Letters - Volume 20, Issue 10, 2013

Knowledge of structural classes is applied in numerous important predictive tasks that address structural and functional features of proteins, although the prediction accuracy of the protein structural classes is not high. In this study, 45 different features were rationally designed to model the differences between protein structural classes, among which, 30 of them reflect the combined protein sequence information. In terms of correlation function, the protein sequence can be converted to a digital signal sequence, from which we can generate 20 discrete Fourier spectrum numbers. According to the segments of amino with different characteristics occurring in protein sequences, the frequencies of the 10 kinds of segments of amino acid (motifs) in protein are calculated. Other features include the secondary structural information :10 features were proposed to model the strong adjacent correlations in the secondary structural elements and capture the long-range spatial interactions between secondary structures, other 5 features were designed to differentiate α/β from α+ β classes , which is a major problem of the existing algorithm. The methods were proposed based on a large set of lowidentity sequences for which secondary structure is predicted from their sequence (based on PSI-PRED). By means of this method, the overall prediction accuracy of four benchmark datasets were all improved. Especially for the dataset FC699, 25PDB and D1189 which are 1.26%, 1% and 0.85% higher than the best previous method respectively.

Protein insertion sequences and their biological role in many organisms have been largely unknown. Here we study proteomes of 12 organisms of diverse genomes for insertion length and amino acid preferences. A total of 871 common proteins were catalogued amongst the 12 organisms for structure based sequence alignment. This underscores the key observations: (i) AT-richness seems to have no implication on the average protein length in an organism as only Dictyostelium discoideum and Plasmodium falciparum encode proteins of high average length (ii) all studied organisms possess insertion in their proteins, however >40 residue length insertions and unique insertions were abundant in pathogen proteomes of Plasmodium falciparum followed by Toxoplasma gondii and Leishmania major, suggesting accessory structural and functional features that may favour evolutionary fitness. (iii) Glu and Asp residues are over-represented in most proteomes irrespective of AT/GC compositions or pathogenecity with an exception of Plasmodium falciparum where Asn dominates (iv) Abundance of Asn residues in Plasmodium falciparum is exceptional given that this feature is not common to other AT-rich genomes. In conclusion, this bioinformatics based study provides comprehensive knowledge of insertions and residue’s preference among pathogen proteins, which can be exploited for further inhibitor studies.

Kallikrein-related peptidases (KLKs) are trypsin-like and chymotrypsin-like serine proteases which are expressed in several tissues. Their activity is tightly controlled by inhibitors including members of the serine protease Kazal-type (SPINK) family. These enzymes are promising targets for the treatment of skin desquamation, inflammation and cancer. Spink3 or caltrin I is expressed in mouse pancreas and males accessory glands and the resulting mature protein has been associated with different activities such as an inhibitor of trypsin and acrosin activity, calcium transport inhibitor in sperm and inhibitor of cell proliferation during embryogenesis. In this study, we produced a soluble recombinant Spink3 from mouse seminal vesicle (rmSpink3) that inhibited the activity of human KLKs. Using FRET substrates, rmSpink3 exhibited a potent inhibitory activity against human KLK2, KLK3, KLK5 (Ki ranging from 260 to 1500 nM), and to a lesser extent against KLK6, KLK1 and KLK7 (Ki around 3000 nM). As shown by mass spectrometry analysis of rmSpink3 incubated with trypsin, the inhibitor was not truncated by the target enzyme. Based on the in silico analysis of the expression of Spink3/SPINK1 and KLKs it is speculated that some KLKs may be natural targets of Spink3/SPINK1, however experimental confirmation using both proteins from mouse or human origin is needed. This work shows that rmSpink3 is a potent inhibitor of various human KLK members suggesting the potential of this molecule in the diagnosis/prevention of several human diseases.

Classifying sequences is one of the central problems in computational biosciences. Several tools have been released to map an unknown molecular entity to one of the known classes using solely its sequence data. However, all of the existing tools are problem-specific and restricted to an alphabet constrained by relevant biological structure. Here, we introduce TRAINER, a new online tool designed to serve as a generic sequence classification platform to enable users provide their own training data with any alphabet therein defined. TRAINER allows users to select among several feature representation schemes and supervised machine learning methods with relevant parameters. Trained models can be saved for future use without retraining by other users. Two case studies are reported for effective use of the system for DNA and protein sequences; candidate effector prediction and nucleolar localization signal prediction. Biological relevance of the results is discussed.

In this study, the problem of predicting interspecies transmission of avian influenza viruses (AIVs) was investigated with machine learning methods. We identified 87 signature positions in AIV protein sequences with information entropy method and encoded these positions with five amino acid factor scores (AAFactors) concentrated from 491 physicochemical and biochemical properties of amino acids. We constructed four prediction models by integrating these five features with commonly used machine learning technologies including Decision Tree, Naive Bayes, Random Forest and Support Vector Machine. The cross validation experiment results demonstrated the power of AAFactors in predicting avian-to-human transmission of AIVs. Comparative analysis revealed the strengths and weaknesses of different machine learning methods, and the importance of different AAFactors to the prediction.

We have developed a microwave (MW)-assisted peptide synthesis using Fmoc-amino acid nanoparticles in water previously. It is an organic solvent-free, environmentally friendly method for peptide synthesis. In this study, we have investigated the racemization of cysteine during an aqueous based coupling reaction with MW irradiation. Under our MW-assisted protocol using WSCI and DMTMM, the coupling reaction can be performed with low levels of racemization of cysteine. We also demonstrated the synthesis of the nonapeptide oxytocin analogue, Cys(Acm)-Tyr-Ile-Gln-Asn- Cys(Acm)-Pro-Leu-Gly-NH2 using our water based MW-assisted protocol with Fmoc-amino acid nanoparticles.

Thrombin, a highly specific protease of blood coagulation, has two exosites that modulate its specificity. We designed two sets of synthetic substrate FRET peptides with 25- or 11- amino acids (aa) each, based on the PAR 1 sequence, to characterize the effect of exosite 1 engagement on substrate catalysis and preference. The 25-aa set encompassed a sequence binding to exosite 1, and structural modeling showed that binding to thrombin did not differ significantly from that of PAR 1 peptide. Modification at the P3´position of the 25 or 11-aa peptides resulted in small effect on kinetic parameters. Ionic strength higher than physiologic depressed thrombin action on the 25-aa peptides. Addition of ligands of the exosite 1 negatively modulated the catalysis of 25-aa substrates. In conclusion, we succeeded to mimic and study in real time, using these synthetic peptides, the influence of ligand binding to exosite 1 on thrombin activity.

Epstein-Barr virus (EBV) is a human oncogenic herpesvirus associating with several malignant diseases. Latent membrane protein 2 (LMP2) of EBV is considered to be an ideal candidate for immunotherapy or prophylactic EBV vaccine. We designed a LMP2 multiepitope containing T and B-cell epitope-rich peptides and constructed a recombinant plasmid containing mammalian codonoptimization EBV LMP2 multiepitope (pcDNA3.1(+)/EBV-LMP2 multiepitope). After pcDNA3.1(+)/EBV-LMP2 multiepitope was transfected into COS-7 cells, significant expression of the multiepitope in COS-7 cells was confirmed by RT-PCR and immunofluorescence assay. Western blot analysis indicated that serum from immunized mice could be discerned by the EBV-LMP2 protein and the EBV-LMP2 multiepitope specifically. The plasmid DNA of EBV-LMP2 multiepitope induced high levels anti-EBV membrane protein and anti-EBV LMP2 multiepitope IgG in mice. T lymphocytes from spleen of immunized mice showed a strong CTL activity. The present study suggested that plasmid DNA encoding EBV LMP2 multiepitope capable of stimulating enough cellular and humoral immunity could have potential for preventing or controlling EBV infection and EBV associated disease in mice.

Ozone stress induces leaf injuries and yield losses in soybean. To identify the proteins associated with the stress response, ozone tolerant and sensitive cultivars were analyzed by 2D-PAGE. After 11 days of ozone stress unifoliate leaves of tolerant soybean cv. Enrei developed 15% and sensitive cv. Nakasennari developed 52% leaf injury symptoms. Analysis of proteins in the unifoliate leaves identified six proteins in tolerant cv. Enrei and three proteins in sensitive cv. Nakasennari, responded significantly to ozone stress. The significantly responded proteins identified in this study were ATP synthase α-subunit, ATP synthase β-subunit, phosphoglycerate kinase, aldo/ketoreductase, rubiscoactivase and glutamine synthetase. To understand that the differences in response of ATP synthase proteins were ultimately associated with extracellular ATP signaling response, ozone sensitive cv. Nakasennari was treated with 1 mM ATP. Unifoliate leaves of ATP treated plants have significantly reduced injury symptoms (20%) under ozone stress compare to control plants (47%). This confirms that extracellular ATP signaling in ATP synthase dependent manner is playing a pivotal role in inducing ozone stress tolerance and preventing injuries in soybean cultivars.

The marine ecosystem is able to provide enormous biomolecule diversity that could be used for treatment of various diseases. In this highly competitive environment, organisms need chemical barriers to reduce or avoid microorganism contamination. Among the molecules that protect these animals the antimicrobial peptides (AMPs) are included. In the present study, crude extracts of coral coral specimens Carijoa riisei, Muriceopsis sulphurea, Neospongodes atlantica, Palythoa caribeorum, Phyllogorgia dilatata and Plexaurella grandiflora were challenged against multiple Grampositive and -negative bacteria showing different activities. P. dilatata crude extract showed the antibacterial activity, and was ammonium-sulfate (0-40%) fractionated, being able to control the growth of K. pneumoniae, S. flexineri and S. aureus. Rich-fraction was further purified by using Amicon® Ultra Centrifugal 10 kDa associated with reversed-phase HPLC chromatography (C18), producing the peptide named Pd-AMP1. Pd-AMP1 was able to inhibit S. aureus development. Mass spectrometry analyses showed a monoisotopic mass of 5372.66 Da and N-terminal sequence showed no significant match with databank. In this view, the prospecting of protein biomolecules and biotechnological potential from marine animals is still little explored and may serve as an alternative to common antibiotics.

A 64-kDa hemagglutinin from a Phaseolus vulgaris cultivar, the northeast red bean, was purified by a protocol composed of three chromatographic steps involving affinity chromatography on Affi-gel blue gel, cation exchange chromatography on SP-Sepharose and FPLC-gel filtration on Superdex 75. The purified hemagglutinin appeared as a single 32-kDa band in SDS-PAGE indicating its dimeric nature. The N-terminal amino acid sequence of the hemagglutinin resembled the sequences of lectins and hemagglutinins from a number of Phaseolus species. The hemagglutinin manifested moderate thermostability and pH stability. It retained full activity up to 65 °C and in the pH range 2 – 12. It did not interact with simple sugars such as glucose, mannose and galactose. The hemagglutinin exerted immunostimulatory effects by upregulating the expression of cytokines like interferon-γ and tumor necrosis factor-α. It also exhibited antiproliferative activity on a number of tumor cells including MCF7 (breast cancer), HepG2 (liver cancer), CNE1 and CNE2 (nasopharyngeal cancer) cells, with stronger activity toward MCF7 and CNE1 cells. The hemagglutinin induced phophatidylserine externalization, mitochondrial depolarization and DNA condensation in MCF7 cells, indicating initiation of apoptosis. However, at high hemagglutinin concentrations, severe damage to the MCF7 cells was detected.

Hyriopsis cumingii (Lea, Unionidae), a freshwater bivalve species widely distributed in China and commercially exploited for freshwater pearl production, was chosen as the reference model to investigate the protein signature in the organic scaffold matching calcium carbonate crystallization mode. This study takes advantage of different calcium carbonate habits production by the organism: aragonite in shell and pearl and vaterite in alternative pearl formation. Amino acid global composition and proteomics analysis have been undertaken to study the amino acid imbalance with respect to biominerals and microstructures. Forty peptides sequences were obtained by proteomics, of which ten are shared by all the different samples, nine are laced with aragonite; another nine with vaterite and twelve are related to pearls. Bioinformatics analysis allowed the peptides to be matched to the deduced protein sequences from EST databases and allowed functional assignment (e.g. scaffolding, strain strength, chitin binding or carbonic anhydrase function) to the proteins found in the different materials. Such panel of motifs tailored in vaterite and aragonite habits produced in a freshwater mollusk gives food for thought about organic control of the biomineralization processes.

New methods for reliable quantitative analysis of biological network data are in high demand in today’s bioinformatics and systems biology. Here we demonstrate the applicability of the co-citation, developed earlier for the analysis of scientific literature for finding functionally similar nodes in protein-protein interaction networks in several model organisms. We prove the power of our approach in a novel way: the predicted closely related enzymes are compared to the closeness of their enzyme commission (EC) numbers, therefore we can numerically evaluate our prediction method. We have found clear correspondence between related enzymatic functions and high co-citation of proteins in interaction networks.