P35 and P22 Toxoplasma gondii antigens abbreviate regions to diagnose acquired toxoplasmosis during pregnancy: toward single-sample assays.

BACKGROUND: P35 and P22 Toxoplasma gondii proteins are recognized by specific IgG at the early infection stage, making them ideal for acute toxoplasmosis pregnancy control. Both proteins have been studied to discriminate between acute and chronic toxoplasmosis. However, results were hardly comparable because different protein obtainment procedures led to different antigens, the reference panels used were not optimally typified, and avidity tests were either not performed or narrowly examined. METHODS: We bioinformatically predicted P35 and P22 regions with the highest density of epitopes, and expressed them in pET32/BL21DE3 alternative expression system, obtaining the soluble proteins rP35a and rP22a. We assessed their diagnostic performance using pregnant woman serum samples typified as: not infected, NI (IgG-, IgM-), typical-chronic, TC (IgM-, IgG+), presumably acute, A (IgG+, IgM+, low-avidity IgG), and recently chronic, RC (IgG+, IgM+, high-avidity IgG). RESULTS: rP35a performed better than rP22a to differentiate A from RC, the areas under the curve (AUC) being 0.911 and 0.818, respectively. They, however, performed similarly to differentiate A from TC+RC (AUC: 0.915 and 0.907, respectively). rP35a and rP22a evaluation by avidity ELISA to discriminate A from RC rendered AUC values of 0.974 and 0.921, respectively. The indirect ELISA and avidity ELISA results analyzed in tandem were consistent with those obtained using commercial kits. CONCLUSIONS: rP35a and rP22a features suggest that, with complementary use, they could replace parasite lysate for toxoplasmosis infection screening and for acute toxoplasmosis diagnosis. Our proposal should be validated by a longitudinal study and may lead to a reliable toxoplasmosis pregnancy control, performing tests in only one serum sample.

Evaluation and comparison of the ability of online available prediction programs to predict true linear B-cell epitopes.

This work deals with the use of predictors to identify useful B-cell linear epitopes to develop immunoassays. Experimental techniques to meet this goal are quite expensive and time consuming. Therefore, we tested 5 free, online prediction methods (AAPPred, ABCpred, BcePred, BepiPred and Antigenic) widely used for predicting linear epitopes, using the primary structure of the protein as the only input. We chose a set of 65 experimentally well documented epitopes obtained by the most reliable experimental techniques as our true positive set. To compare the quality of the predictor methods we used their positive predictive value (PPV), i.e. the proportion of the predicted epitopes that are true, experimentally confirmed epitopes, in relation to all the epitopes predicted. We conclude that AAPPred and ABCpred yield the best results as compared with the other programs and with a random prediction procedure. Our results also indicate that considering the consensual epitopes predicted by several programs does not improve the PPV.