Weight matrix based identification of terpene synthases conserved motifs in Arabidopsis thaliana proteome

Abstract

Terpenes comprise the most diverse collection of natural products. Out of more than 30,000 individual terpenoids identified, at least half are synthesized by plants. A relatively small, but quantitatively significant, number of terpenoids are involved in primary plant metabolism. However, the vast majorities are classified as secondary metabolites; compounds not required for plant growth and development but presumed to have an ecological function in communication or defense and are widely used in industrial applications. Terpene hydrocarbon scaffolds are generated by the action of the mechanistically intriguing family of mono-, sesqui-, and diterpene synthases collectively termed as terpene synthases, that catalyze multistep reactions with diphosphorylated substrates of 10 (geranyl diphosphate), 15 (farnesyl diphosphate) or 20 (geranylgeranyl diphosphate) carbons. In the studied work, we performed a computational study on proteome wide identification of terpene synthase motifs in Arabidopsis thaliana proteome on the basis of weight matrix approach. We have developed an optimal weight matrix for the identification of terpene synthase motifs in the plant’s proteome. Weight matrix was constructed by aligning orthologous sequences of known terpene synthases originated from diverse plant species viz., Abies grandis, Nicotiana tobaccum etc. Sequences of detected domains & motifs were retrieved through SwissProtKB/NCBI on the basis of specific conservation IDs of Prosite, Pfam, Interpro, Prodom, COG, TIGR databases, while position specific scoring matrices were made through MEME, MotifSampler, PossuMsearch tools. Weight matrix based search of conserved motifs in the proteome of A. thaliana was done through ESA, Lahead and Simple algorithm based search tools of PossuMsearch biosuite in Linux system. Prediction was first validated by using positive control data set and optimized the method to reach prediction accuracy upto >90%. After tool performance evaluation, prediction was made on whole proteome at specific threshold/score value. Significant results were found in A. thaliana with motif similarity ranges from 80% to 100%. This proteome wide search model paves the path to identify more terpene synthases genes in A. thaliana, as well as in other plant systems.