Computational protocol: Characterization of indole-3-pyruvic acid pathway-mediated biosynthesis of auxin in Neurospora crassa

Similar protocols

Protocol publication

[…] Homologous protein sequences of tryptophan aminotransferase (ARO-8), indole-3-pyruvate decarboxylase (CFP), indole-3-acetaldehyde (IAAld) dehydrogenases and flavin monooxygenase (MOX-2) from different fungal species were obtained by BLAST [] analysis and used in both bioinformatics as well as in experimental studies. First, Neurospora crassa protein sequences for the enzymes mentioned above were determined using the Broad Institute database ( and FungiDB ( Then, homologous sequences in other species were checked and retrieved by bidirectional BLAST using the Broad Institute database, FungiDB and the National Centre for Biotechnology Information (NCBI) [, ]. Only full-length sequences of the above-mentioned enzymes from different fungal species were considered for further analysis. Sequences were aligned using MUltiple Sequence Comparison by Log-Expectation (MUSCLE) [, ], and the alignments were visualized, edited, and presented using GeneDoc []. In a similar way, homologous sequences of tryptophan-2-monooxygenase (LAO-2), indole-3-acetamide hydrolase (NAA-2), aromatic-L-amino acid decarboxylase (aad-1) and glutamate decarboxylase (gdc-3) were obtained only for N. crassa and were used only for experimentation along with other genes. Throughout this study, we used the names of the genes and their corresponding enzymes per N. crassa conventions. However, we also found that some genes were annotated differently in different databases. For this reason, a description of the gene identity with its annotations is presented in , showing annotations for both N. crassa as well as the general nomenclature for enzymes.For the structural analysis, we implemented two independent methods to establish the consistency and reliability of each method, as well as the structure. We used the Ab initio method from the I-TASSER ( online-server [] and the "SWISS-MODEL Repository portal" ( for homology modeling []. For the Ab initio method, corresponding enzyme sequences from N. crassa were submitted to the I-TASSER server, and the best model was chosen based on the C-scores. To choose a proper homologous structural template, individual enzymes were analyzed using BLAST against the PDB database ( prior to homology modeling. Suitable models were chosen for homology modeling based on the BLAST scores. We then predicted and cross-checked the cofactor binding properties in our predicted models by assessing the "structure-based function prediction" using the COFACTOR online server ( [] for ARO-8, CFP, CBS-3 and MOX-2. COFACTOR makes predictions about the ligands and their binding sites on the structures. The reliability of ligand binding is defined by the BS-score. A score >1 reflects significant ligand binding in the structure. Ligand-binding sites present on both N. crassa enzymes and template structures (those used for homology modeling) were identified by alignments with known proteins. Finally, I-TASSER-predicted enzyme structures were aligned with template structures from the PDB. This structural alignment helps to obtain information about key residues inside enzymatic catalytic sites. [...] All the predicted structures generated by I-TASSER or SWISS-MODEL were evaluated in the PDBsum database using PROCHECK analysis []. As two independent approaches (Ab initio and homology modeling) were used for structure prediction, we further checked the consistency and reliability of the models by structural alignment of the same enzyme predicted by different methods. [...] All the molecular visualization and editing was performed using PyMol (The PyMOL Molecular Graphics System, Version Schrödinger, LLC.) and YASARA ( Images were processed using POV-Ray ( […]

Pipeline specifications

Databases PDBsum
Applications Drug design, Protein structure analysis
Organisms Escherichia coli, Neurospora crassa, Fungi
Chemicals Tryptophan