Redox-sensitive bZIP68 plays a role in balancing stress tolerance with growth in Arabidopsis.

Perturbation of the cellular redox state by stress conditions is sensed by redox-sensitive proteins so that the cell can physiologically respond to stressors. However, the mechanisms linking sensing to response remain poorly understood in plants. Here we report that the transcription factor bZIP68 underwent in vivo oxidation in Arabidopsis cells under oxidative stress which is dependent on its redox-sensitive Cys320 residue. bZIP68 is primarily localized to the nucleus under normal growth conditions in Arabidopsis seedlings. Oxidative stress reduces its accumulation in the nucleus and increases its cytosolic localization. Chromatin immunoprecipitation followed by sequencing (ChIP-seq) revealed that bZIP68 primarily binds to promoter regions containing the core G-box (CACGTG) or G-box-like motif of the genes involved in abiotic and biotic stress responses, photosynthesis, biosynthetic processes, and transcriptional regulation. The bzip68 mutant displayed slower growth under normal conditions but enhanced tolerance to oxidative stress. The results from the ChIP-seq and phenotypic and transcriptome comparison between the bzip68 mutant and wildtype indicate that bZIP68 normally suppresses expression of stress tolerance genes and promotes expression of growth-related genes, whereas its inactivation enhances stress tolerance but suppresses growth. bZIP68 might balance stress tolerance with growth through the extent of its oxidative inactivation according to the environment.

A global characterization and identification of multifunctional enzymes.

Multi-functional enzymes are enzymes that perform multiple physiological functions. Characterization and identification of multi-functional enzymes are critical for communication and cooperation between different functions and pathways within a complex cellular system or between cells. In present study, we collected literature-reported 6,799 multi-functional enzymes and systematically characterized them in structural, functional, and evolutionary aspects. It was found that four physiochemical properties, that is, charge, polarizability, hydrophobicity, and solvent accessibility, are important for characterization of multi-functional enzymes. Accordingly, a combinational model of support vector machine and random forest model was constructed, based on which 6,956 potential novel multi-functional enzymes were successfully identified from the ENZYME database. Moreover, it was observed that multi-functional enzymes are non-evenly distributed in species, and that Bacteria have relatively more multi-functional enzymes than Archaebacteria and Eukaryota. Comparative analysis indicated that the multi-functional enzymes experienced a fluctuation of gene gain and loss during the evolution from S. cerevisiae to H. sapiens. Further pathway analyses indicated that a majority of multi-functional enzymes were well preserved in catalyzing several essential cellular processes, for example, metabolisms of carbohydrates, nucleotides, and amino acids. What's more, a database of known multi-functional enzymes and a server for novel multi-functional enzyme prediction were also constructed for free access at http://bioinf.xmu.edu.cn/databases/MFEs/index.htm.

Prediction of functional class of proteins and peptides irrespective of sequence homology by support vector machines.

Various computational methods have been used for the prediction of protein and peptide function based on their sequences. A particular challenge is to derive functional properties from sequences that show low or no homology to proteins of known function. Recently, a machine learning method, support vector machines (SVM), have been explored for predicting functional class of proteins and peptides from amino acid sequence derived properties independent of sequence similarity, which have shown promising potential for a wide spectrum of protein and peptide classes including some of the low- and non-homologous proteins. This method can thus be explored as a potential tool to complement alignment-based, clustering-based, and structure-based methods for predicting protein function. This article reviews the strategies, current progresses, and underlying difficulties in using SVM for predicting the functional class of proteins. The relevant software and web-servers are described. The reported prediction performances in the application of these methods are also presented.