Modelling and epitope prediction of Toxoplasma gondii P22 antigen

COSTA, J. G.; FACCENDINI, P. L.; LAGIER, C. M.; MARCIPAR, I. S.

Córdoba

Congreso; 2do Congreso Argentino de bioinformática y biología computacional; 2011

Asociación Argentina de Bioinformática y Biología Computacional

Modelling and epitope prediction of Toxoplasma gondii P22 antigen Juan Gabriel Costa1, Pablo Faccendini2, Claudia Marina Lagier2, Iván Sergio Marcipar1 Juan Gabriel Costa1, Pablo Faccendini2, Claudia Marina Lagier2, Iván Sergio Marcipar1 1Laboratorio de Tecnología Inmunológica, Facultad de Bioquímica y Ciencias Biológicas, Universidad Nacional del Litoral, Santa Fe, Argentina 2Departamente de Química, Facultad de Bioquímica y Farmacia, Universidad Nacional de Rosario,  Rosario, Argentina Background: P22 is a membrane protein from Toxoplasma gondii which has been proposed as a useful diagnostic tool to assess specific antibodies in acute toxoplasmasis. The identification of its antigenic regions would allow the rational design of vaccines and antigens to be used in immunochemical assays. To identify these regions we have modeled the 3D structure and we have used the model to predict conformational and continuous B epitopes. Methods and Results: The model was obtained by homology modeling using Modeller software. As template we used the X-ray crystallography structure from sporosag protein. This molecule is expressed by T. gondii and although it has low identity with P22 antigen (29%), it is highly evolutionarily related to P22 [1]. Several models where obtained and the best of them was selected based on online Verify3D (http://nihserver.mbi.ucla.edu/Verify_3D/) and ANOLEA (http://protein.bio.puc.cl/cardex/servers/anolea/index.html) servers results. As evaluation criterion, the ratio from aminoacids which overcome the value of 0,15 from the internal score of Verify3D and display ANOLEA values below 0 E/kT were considered. The model which had the best ratio in both analysis was selected. The refined model shown in Figure 1 was obtained using UCSF Chimera software and  LOBO (online: http://protein.cribi.unipd.it/lobo/). Based on the 3D structure, conformational B epitopes were predicted using online Discotope software (http://www.cbs.dtu.dk/services/DiscoTope/). The best antigenic performance were predicted in the residues corresponding to the position  34 to 40, 52, 63, 65, 66, 69, 77, 80, 83 a 86, 95, 96, 98, 162, 163, 165, 166, 169 y 182 to 186. Continous B epitopes were also assessed using online AAPPred software (http://www.bioinf.ru/aappred/), being the main antigenic regions predicted between the residues 30 to 45, 93 to 99  and 115 to 145. Conclusion: Experimental approaches are time consuming, expensive and they may lead to not reliable results. The bioinformatics approach can provide a valuable source of information which complements the experimental results. Using free softwares we have predicted the tridimensional structure of T. gondii acute phase protein P22 and the putatives epitopes. References: Crawford J, Lamb E, Wasmuth J, Grujic O, Grigg M,   Boulanger M. Structural and functional characterization of SporoSAG: a SAG2-related surface antigen from Toxoplasma gondii.  J.Biol.Chem. 2010, 285: 12063-12070