In silico prediction of b-cell and t-cell epitopes in plasmodium falciparum merozoite invasion proteins (eba175, rh5 and ripr)
Abstract
emisinin-based combination therapy and insecticide-treated bed nets are not effective in the
long term and there is a need to develop an effective malaria vaccine that will fully combat and
control malaria infections. The RTS,S/AS01 malaria vaccine has achieved limited success,
hence there is still need to identify novel malaria vaccine targets.
This study predicted B-cell epitopes (BCEs) and T-cell epitopes (TCEs) in three merozoite
invasion proteins i.e. Erythrocyte binding antigen-175 (EBA175), Reticulocyte binding-like
homolog 5 (Rh5) and Ripr (Rh5 interacting protein). The approach involved comparing the
prediction of TCEs from predicted BCEs and prediction of TCEs directly from full protein
sequences. Circumsporozoite protein (CSP), the protein from which RTS,S/AS01 vaccine was
developed, was used as a control to determine whether BCE and TCE prediction algorithms
could predict experimental defined BCEs and TCEs. BCEs were predicted using the BCPreds
server while TCEs were predicted using the NetMHCcons (for MHC class I binding
predictions) and NetMHCIIpan 3.0 (for MHC class II binding predictions).
Prediction of TCEs from full length CSP sequence outperformed the prediction of TCEs from
predicted CSP BCEs, hence this approach was applied on EBA175, Rh5 and Ripr. A further
step was included to identify regions of the proteins overlapping TCEs and BCEs using
sequence clustering algorithms. In total, EBA175, Rh5 and Ripr yielded 5, 4 and 5 candidate
cross-reactive TCEs and BCEs respectively, including both conserved and polymorphic
regions across the isolates tested. Of these candidate immunogenic epitopes, (WNEFREKLWE
AMLSEHKNNI, 20mers) in EBA175, 2 in Rh5 (YKNVDYKNVNFLQYHFKELSNYNIANS
IDILQEKEGHLDFVIIPHYTFLDYYKHLSYNSIYHKSSTYGCIAVDAFIKKINETYDKV
KSKCNDIK – 95mers and SCYNNNFCNTNGI RYHYDEYIHKLILSVKS – 30mers) and 1
in Ripr (INCQGMYISLRSVHVHTHNAILQQETLTYIKNLCDGKNNCKFDFDSIKYENKS
LTHYLFFINIQYQCISPLNLQENEMC – 51mers) mapped back to experimentally verified
BCEs.
In silico prediction of BCEs and TCEs minimizes the resources required for laboratory analysis
of pathogen gene products. An immunologist can use these computationally predicted
immunogenic regions to explore the potential of developing effective drugs and vaccines. We
propose that the EBA175-RII, Rh5 and Ripr BCEs and TCEs are immunogenic and recommend
them for experimental lab validation and can be inclusion in the search for an effective malaria
vaccine
Citation
Master of science in BioinformaticsPublisher
University of Nairobi