In silico approach to identify non-synonymous SNPs with highest predicted deleterious effect on protein function in human obesity related gene, neuronal growth regulator 1 (NEGR1)

Abstract

Neuronal growth regulator 1 (NEGR1) is a candidate gene for human obesity, which encodes the neural cell adhesion and growth molecule. The aim of the current study was to recognize the non-synonymous SNPs (nsSNPs) with the highest predicted deleterious effect on protein function of the NEGR1 gene. We have used five computational tools, namely, PolyPhen, SIFT, PROVEAN, MutPred and M-CAP, to predict the deleterious and pathogenic nsSNPs of the NEGR1 gene. Homology modeling approach was used to model the native and mutant NEGR1 protein models. Furthermore, structural validation was performed by the PROCHECK server to interpret the stability of the predicted models. We have predicted four potential deleterious nsSNPs, i.e., rs145524630 (Ala70Thr), rs267598710 (Pro168Leu), rs373419972 (Arg239Cys) and rs375352213 (Leu158Phe), which might be involved in causing obesity phenotypes. The predicted mutant models showed higher root mean square deviation and free energy values under the PyMoL and SWISS-PDB viewer, respectively. Additionally, the FTSite server predicted one nsSNP, i.e., rs145524630 (Ala70Thr) out of four identified nsSNPs found in the NEGR1 protein-binding site. There were four potential deleterious and pathogenic nsSNPs, i.e., rs145524630, rs267598710, rs373419972 and rs375352213, identified from the above-mentioned tools. In future, further functional in vitro and in vivo analysis could lead to better knowledge about these nsSNPs on the influence of the NEGR1 gene in causing human obesity. Hence, the present computational examination suggest that predicated nsSNPs may feasibly be a drug target and play an important role in contributing to human obesity.

Keywords: Computational tools; Homology modeling; NEGR1; Obesity; nsSNPs.

Fig. 1

Screening of deleterious nsSNPs in the NEGR1 gene: this bar graph depicts screening and prediction of potential deleterious nsSNPs by five different computational tools as follows: PolyPhen (predicted deleterious nsSNPs = 9), SIFT (predicted deleterious nsSNPs = 12), PROVEAN (predicted deleterious nsSNPs = 31), MutPred (predicted deleterious nsSNPs = 21) and M-CAP (predicted pathogenic nsSNPs = 21)

Fig. 2

Homology-modeled 3D structures and model—template alignment of native and mutant NEGR1 protein. a Native protein model, b mutant protein model, rs145524630 (Ala70Thr), c mutant protein model, rs267598710 (Pro168Leu), d mutant protein model, rs373419972 (Arg239Cys), e mutant protein model, rs375352213 (Leu158Phe), f PDB ID 1QZ1.1.A (Rattus norvegicus) used as template for NEGR1 protein modeling

Fig. 3

Superimposed structures of mutants and native model of NEGR1 protein. a Native vs. rs145524630 (Ala70Thr), b native vs. rs267598710 (Pro168Leu), c native vs. rs373419972 (Arg239Cys), d native vs. rs375352213 (Leu158Phe). In the figure, green color signifies the native model, whereas red color is the mutant model

Fig. 4

Ramachandran plot analysis: a Ramachandran plot of the native model. b Ramachandran plot of the rs145524630 (Ala70Thr) model. c Ramachandran plot of the rs267598710 (Pro168Leu) model. d Ramachandran plot of the rs373419972 (Arg239Cys) model. e Ramachandran plot of the rs375352213 (Leu158Phe) model

Fig. 5

Ligand-binding sites of the NEGR1 protein. This figure represents three ligand-binding sites (mesh loops). The first ligand-binding site (mesh loop 1) is shown in cyan. The second ligand-binding site (mesh loop 2) is shown in magenta. The third ligand-binding site (mesh loop 3) is shown in red. The nsSNP rs145524630 (Ala70Thr) is present in the second ligand-binding site of the NEGR1 protein