Taxonomic utility of a phylogenetic analysis of phosphoglycerate kinase proteins of Archaea, Bacteria, and Eukaryota: Insights by Bayesian analyses

TMT-Based Quantitative Proteomics Analysis of the Fish-Borne Spoiler Shewanella putrefaciens Subjected to Cold Stress Using LC-MS/MS

Shewanella putrefaciens is a specific spoilage bacterium for fish during cold storage. To better understand the molecular mechanisms of cold stress adaptation of S. putrefaciens, tandem mass tag- (TMT-) based quantitative proteomic analysis was performed to detect the effects of cold stress on protein expression profiles in S. putrefaciens which had been cultivated at 4°C and 30°C, respectively. A total of 266670 peptide spectrum matching numbers were quantified proteins after data analysis. Of the 2292 proteins quantitatively analyzed, a total of 274 were found to be differentially expressed (DE) under cold stress compared with the nonstress control. By integrating the results of Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses, 9 common KEGG terms were found notable for the cold-responsive proteins. Generally, the DE proteins involved in carbohydrate, amino acid, and fatty acid biosynthesis and metabolism were significantly upregulated, leading to a specific energy conservation survival mode. The DE proteins related to DNA repair, transcription, and translation were upregulated, implicating change of gene expression and more protein biosynthesis needed in response to cold stress.

Mechanisms of protein sequence divergence and incompatibility

Alignments of orthologous protein sequences convey a complex picture. Some positions are utterly conserved whilst others have diverged to variable degrees. Amongst the latter, many are non-exchangeable between extant sequences. How do functionally critical and highly conserved residues diverge? Why and how did these exchanges become incompatible within contemporary sequences? Our model is phosphoglycerate kinase (PGK), where lysine 219 is an essential active-site residue completely conserved throughout Eukaryota and Bacteria, and serine is found only in archaeal PGKs. Contemporary sequences tested exhibited complete loss of function upon exchanges at 219. However, a directed evolution experiment revealed that two mutations were sufficient for human PGK to become functional with serine at position 219. These two mutations made position 219 permissive not only for serine and lysine, but also to a range of other amino acids seen in archaeal PGKs. The identified trajectories that enabled exchanges at 219 show marked sign epistasis - a relatively small loss of function with respect to one amino acid (lysine) versus a large gain with another (serine, and other amino acids). Our findings support the view that, as theoretically described, the trajectories underlining the divergence of critical positions are dominated by sign epistatic interactions. Such trajectories are an outcome of rare mutational combinations. Nonetheless, as suggested by the laboratory enabled K219S exchange, given enough time and variability in selection levels, even utterly conserved and functionally essential residues may change.

Orthologs are proteins in different species sharing the same function and structure. However, the mechanisms that underline the divergence of different sequences from a single ancestor remain unclear, particularly because many amino acid exchanges between orthologs result in loss of function (incompatibility). We aimed at disentangling an ancient divergence event within the active-site of a universally spread enzyme that mediates ATP synthesis. Using laboratory evolution experiments, we found that an exchange in a functionally critical active-site residue that is incompatible within contemporary orthologs is enabled by few mutations. These mutations lead to transition sequences in which, unlike the extant sequences, a wide range of amino acids is tolerated. Our experiment reveals the properties of these transition sequences that may resemble the historical ancestral states that underlined this divergence event, and the mechanisms that led to incompatibility within the contemporary orthologs. Our results support theoretical predictions and reshape our understanding of protein structure-function. That a given position is entirely conserved and essential for function does not indicate that it will never exchange, but rather, that the exchange may depend on changes in many other positions.

Uniquely localized intra-molecular amino acid concentrations at the glycolytic enzyme catalytic/active centers of Archaea, Bacteria and Eukaryota are associated with their proposed temporal appearances on earth

The distributions of amino acids at most-conserved sites nearest catalytic/active centers (C/AC) in 4,645 sequences of ten enzymes of the glycolytic Embden-Meyerhof-Parnas pathway in Archaea, Bacteria and Eukaryota are similar to the proposed temporal order of their appearance on Earth. Glycine, isoleucine, leucine, valine, glutamic acid and possibly lysine often described as prebiotic, i.e., existing or occurring before the emergence of life, were localized in positional and conservational defined aggregations in all enzymes of all Domains. The distributions of all 20 biologic amino acids in most-conserved sites nearest their C/ACs were quite different either from distributions in sites less-conserved and further from their C/ACs or from all amino acids regardless of their position or conservation. The major concentrations of glycine, e.g., perhaps the earliest prebiotic amino acid, occupies ≈ 16 % of all the most-conserved sites within a volume of ≈ 7-8 Å radius from their C/ACs and decreases linearly towards the molecule's peripheries. Spatially localized major concentrations of isoleucine, leucine and valine are in the mid-conserved and mid-distant sites from their C/ACs in protein interiors. Lysine and glutamic acid comprise ≈ 25-30 % of all amino acids within an irregular volume bounded by ≈ 24-28 Å radii from their C/ACs at the most-distant least-conserved sites. The unreported characteristics of these amino acids: their spatially and conservationally identified concentrations in Archaea, Bacteria and Eukaryota, suggest some common structural organization of glycolytic enzymes that may be relevant to their evolution and that of other proteins. We discuss our data in relation to enzyme evolution, their reported prebiotic putative temporal appearances on Earth, abundances, biological "cost", neighbor-sequence preferences or "ordering" and some thermodynamic parameters.

Tree preserving embedding

The goal of dimensionality reduction is to embed high-dimensional data in a low-dimensional space while preserving structure in the data relevant to exploratory data analysis such as clusters. However, existing dimensionality reduction methods often either fail to separate clusters due to the crowding problem or can only separate clusters at a single resolution. We develop a new approach to dimensionality reduction: tree preserving embedding. Our approach uses the topological notion of connectedness to separate clusters at all resolutions. We provide a formal guarantee of cluster separation for our approach that holds for finite samples. Our approach requires no parameters and can handle general types of data, making it easy to use in practice and suggesting new strategies for robust data visualization.

Concentration of specific amino acids at the catalytic/active centers of highly-conserved "housekeeping" enzymes of central metabolism in archaea, bacteria and Eukaryota: is there a widely conserved chemical signal of prebiotic assembly?

In alignments of 1969 protein sequences the amino acid glycine and others were found concentrated at most-conserved sites within approximately 15 A of catalytic/active centers (C/AC) of highly conserved kinases, dehydrogenases or lyases of Archaea, Bacteria and Eukaryota. Lysine and glutamic acid were concentrated at least-conserved sites furthest from their C/ACs. Logistic-regression analyses corroborated the "movement" of glycine towards and lysine away from their C/ACs: the odds of a glycine occupying a site were decreased by 19%, while the odds for a lysine were increased by 53%, for every 10 A moving away from the C/AC. Average conservation of MSA consensus sites was highest surrounding the C/AC and directly decreased in transition toward model's peripheries. Findings held with statistical confidence using sequences restricted to individual Domains or enzyme classes or to both. Our data describe variability in the rate of mutation and likelihoods for phylogenetic trees based on protein sequence data and endorse the extension of substitution models by incorporating data on conservation and distance to C/ACs rather than only using cumulative levels. The data support the view that in the most-conserved environment immediately surrounding the C/AC of taxonomically distant and highly conserved essential enzymes of central metabolism there are amino acids whose identity and degree of occupancy is similar to a proposed amino acid set and frequency associated with prebiotic evolution.

A proteomic study of cMyc improvement of CHO culture

Background

The biopharmaceutical industry requires cell lines to have an optimal proliferation rate and a high integral viable cell number resulting in a maximum volumetric recombinant protein product titre. Nutrient feeding has been shown to boost cell number and productivity in fed-batch culture, but cell line engineering is another route one may take to increase these parameters in the bioreactor. The use of CHO-K1 cells with a c-myc plasmid allowing for over-expressing c-Myc (designated cMycCHO) gives a higher integral viable cell number. In this study the differential protein expression in cMycCHO is investigated using two-dimensional gel electrophoresis (2-DE) followed by image analysis to determine the extent of the effect c-Myc has on the cell and the proteins involved to give the new phenotype.

Results

Over 100 proteins that were differentially expressed in cMycCHO cells were detected with high statistical confidence, of which 41 were subsequently identified by tandem mass spectrometry (LC-MS/MS). Further analysis revealed proteins involved in a variety of pathways. Some examples of changes in protein expression include: an increase in nucleolin, involved in proliferation and known to aid in stabilising anti-apoptotic protein mRNA levels, the cytoskeleton and mitochondrial morphology (vimentin), protein biosysnthesis (eIF6) and energy metabolism (ATP synthetase), and a decreased regulation of all proteins, indentified, involved in matrix and cell to cell adhesion.

Conclusion

These results indicate several proteins involved in proliferation and adhesion that could be useful for future approaches to improve proliferation and decrease adhesion of CHO cell lines which are difficult to adapt to suspension culture.

Cytological characterization of YpsB, a novel component of the Bacillus subtilis divisome

Cell division in bacteria is carried out by an elaborate molecular machine composed of more than a dozen proteins and known as the divisome. Here we describe the characterization of a new divisome protein in Bacillus subtilis called YpsB. Sequence comparisons and phylogentic analysis demonstrated that YpsB is a paralog of the division site selection protein DivIVA. YpsB is present in several gram-positive bacteria and likely originated from the duplication of a DivIVA-like gene in the last common ancestor of bacteria of the orders Bacillales and Lactobacillales. We used green fluorescent protein microscopy to determine that YpsB localizes to the divisome. Similarly to that for DivIVA, the recruitment of YpsB to the divisome requires late division proteins and occurs significantly after Z-ring formation. In contrast to DivIVA, however, YpsB is not retained at the newly formed cell poles after septation. Deletion analysis suggests that the N terminus of YpsB is required to target the protein to the divisome. The high similarity between the N termini of YpsB and DivIVA suggests that the same region is involved in the targeting of DivIVA. YpsB is not essential for septum formation and does not appear to play a role in septum positioning. However, a ypsB deletion has a synthetic effect when combined with a mutation in the cell division gene ftsA. Thus, we conclude that YpsB is a novel B. subtilis cell division protein whose function has diverged from that of its paralog DivIVA.

Benchmarking protein classification algorithms via supervised cross-validation

Development and testing of protein classification algorithms are hampered by the fact that the protein universe is characterized by groups vastly different in the number of members, in average protein size, similarity within group, etc. Datasets based on traditional cross-validation (k-fold, leave-one-out, etc.) may not give reliable estimates on how an algorithm will generalize to novel, distantly related subtypes of the known protein classes. Supervised cross-validation, i.e., selection of test and train sets according to the known subtypes within a database has been successfully used earlier in conjunction with the SCOP database. Our goal was to extend this principle to other databases and to design standardized benchmark datasets for protein classification. Hierarchical classification trees of protein categories provide a simple and general framework for designing supervised cross-validation strategies for protein classification. Benchmark datasets can be designed at various levels of the concept hierarchy using a simple graph-theoretic distance. A combination of supervised and random sampling was selected to construct reduced size model datasets, suitable for algorithm comparison. Over 3000 new classification tasks were added to our recently established protein classification benchmark collection that currently includes protein sequence (including protein domains and entire proteins), protein structure and reading frame DNA sequence data. We carried out an extensive evaluation based on various machine-learning algorithms such as nearest neighbor, support vector machines, artificial neural networks, random forests and logistic regression, used in conjunction with comparison algorithms, BLAST, Smith-Waterman, Needleman-Wunsch, as well as 3D comparison methods DALI and PRIDE. The resulting datasets provide lower, and in our opinion more realistic estimates of the classifier performance than do random cross-validation schemes. A combination of supervised and random sampling was used to construct model datasets, suitable for algorithm comparison.

Revised minimal standards for description of new species of the class Mollicutes (division Tenericutes)

Minimal standards for novel species of the class Mollicutes (trivial term, mollicutes), last published in 1995, require revision. The International Committee on Systematics of Prokaryotes Subcommittee on the Taxonomy of Mollicutes proposes herein revised standards that reflect recent advances in molecular systematics and the species concept for prokaryotes. The mandatory requirements are: (i) deposition of the type strain into two recognized culture collections, preferably located in different countries; (ii) deposition of the 16S rRNA gene sequence into a public database, and a phylogenetic analysis of the relationships among the 16S rRNA gene sequences of the novel species and its neighbours; (iii) deposition of antiserum against the type strain into a recognized collection; (iv) demonstration, by using the combination of 16S rRNA gene sequence analyses, serological analyses and supplementary phenotypic data, that the type strain differs significantly from all previously named species; and (v) assignment to an order, a family and a genus in the class, with an appropriate specific epithet. The 16S rRNA gene sequence provides the primary basis for assignment to hierarchical rank, and may also constitute evidence of species novelty, but serological and supplementary phenotypic data must be presented to substantiate this. Serological methods have been documented to be congruent with DNA-DNA hybridization data and with 16S rRNA gene placements. The novel species must be tested serologically to the greatest extent that the investigators deem feasible against all neighbouring species whose 16S rRNA gene sequences show >0.94 similarity. The investigator is responsible for justifying which characters are most meaningful for assignment to the part of the mollicute phylogenetic tree in which a novel species is located, and for providing the means by which novel species can be identified by other investigators. The publication of the description should appear in a journal having wide circulation. If the journal is not the International Journal of Systematic and Evolutionary Microbiology, copies of the publication must be submitted to that journal so that the name may be considered for inclusion in a Validation List as required by the International Code of Bacteriological Nomenclature (the Bacteriological Code). Updated informal descriptions of the class Mollicutes and some of its constituent higher taxa are available as supplementary material in IJSEM Online.

A Protein Classification Benchmark collection for machine learning

Protein classification by machine learning algorithms is now widely used in structural and functional annotation of proteins. The Protein Classification Benchmark collection () was created in order to provide standard datasets on which the performance of machine learning methods can be compared. It is primarily meant for method developers and users interested in comparing methods under standardized conditions. The collection contains datasets of sequences and structures, and each set is subdivided into positive/negative, training/test sets in several ways. There is a total of 6405 classification tasks, 3297 on protein sequences, 3095 on protein structures and 10 on protein coding regions in DNA. Typical tasks include the classification of structural domains in the SCOP and CATH databases based on their sequences or structures, as well as various functional and taxonomic classification problems. In the case of hierarchical classification schemes, the classification tasks can be defined at various levels of the hierarchy (such as classes, folds, superfamilies, etc.). For each dataset there are distance matrices available that contain all vs. all comparison of the data, based on various sequence or structure comparison methods, as well as a set of classification performance measures computed with various classifier algorithms.

Application of a simple likelihood ratio approximant to protein sequence classification

MOTIVATION

Likelihood ratio approximants (LRA) have been widely used for model comparison in statistics. The present study was undertaken in order to explore their utility as a scoring (ranking) function in the classification of protein sequences.

RESULTS

We used a simple LRA-based on the maximal similarity (or minimal distance) scores of the two top ranking sequence classes. The scoring methods (Smith-Waterman, BLAST, local alignment kernel and compression based distances) were compared on datasets designed to test sequence similarities between proteins distantly related in terms of structure or evolution. It was found that LRA-based scoring can significantly outperform simple scoring methods.

Initial transcriptome and proteome analyses of low culture temperature-induced expression in CHO cells producing erythropoietin

Low culture temperature is known to enhance the specific productivity of Chinese hamster ovary (CHO) cells expressing erythropoietin (EPO) (LGE10-9-27). Genomic and proteomic approaches were taken to better understand the intracellular responses of these CHO cells resulting from use of low culture temperature (33 degrees C). For transcriptome analysis, commercially available rat and mouse cDNA microarrays were used. The data obtained from the rat and mouse cDNA chips were only somewhat informative in understanding the gene expression profile of CHO cells because of their different sequence homologies with CHO transcriptomes. Overall, transcriptome analysis revealed that low culture temperature could lead to changes in gene expression in various cellular processes such as metabolism, transport, and signaling pathways. Proteome analysis was carried out using 2-D PAGE. Based on spot intensity, 60 high intensity protein spots, from a total of more than 800, were chosen for MS analysis. Forty of the 60 protein spots, which represent 26 different kinds of proteins, were identified by MALDI-TOF-MS and validated by MS/MS. Compared to the reference temperature (37 degrees C), the expression levels of seven proteins (PDI, vimentin, NDK B, ERp57, RIKEN cDNA, phosphoglycerate kinase, and heat shock cognate 71 kDa protein) were increased over twofold at 33 degrees C and those of two proteins (HSP90-beta and EF2) were decreased over twofold at 33 degrees C. Taken together, the results demonstrate the potential of combined analysis of transcriptome and proteome analyses as a tool for the systematic comprehension of cellular mechanisms in CHO cells.

Application of compression-based distance measures to protein sequence classification: a methodological study

MOTIVATION

Distance measures built on the notion of text compression have been used for the comparison and classification of entire genomes and mitochondrial genomes. The present study was undertaken in order to explore their utility in the classification of protein sequences.

RESULTS

We constructed compression-based distance measures (CBMs) using the Lempel-Zlv and the PPMZ compression algorithms and compared their performance with that of the Smith-Waterman algorithm and BLAST, using nearest neighbour or support vector machine classification schemes. The datasets included a subset of the SCOP protein structure database to test distant protein similarities, a 3-phosphoglycerate-kinase sequences selected from archaean, bacterial and eukaryotic species as well as low and high-complexity sequence segments of the human proteome, CBMs values show a dependence on the length and the complexity of the sequences compared. In classification tasks CBMs performed especially well on distantly related proteins where the performance of a combined measure, constructed from a CBM and a BLAST score, approached or even slightly exceeded that of the Smith-Waterman algorithm and two hidden Markov model-based algorithms.