Fig. 2. (A) Schematic representation of the TbMBAP1 protein of T. brucei. The putative signal sequence is highlighted in red, the catalytically active regions appear in blue and the predicted transmembrane region in grey. The C-terminal amino acids recognized by an anti-peptide antibody appear as a black box. Green lines denote potential N-glycosylation sites. The PFAM domain PF00328 (histidine acid phosphatase) is highlighted in yellow. Numbers indicate amino acid positions. (B) Predicted structure of the TbMBAP1 core domain (amino acids 32-326). The coordinates are based on three homology models that were independently generated with CPHmodels-2.0 (http://www.cbs.dtu.dk/services/CPHmodels/), 3D-JIGSAW-2.0 (http://www.bmm.icnet.uk/servers/3djigsaw/) and SDSC1 (http://cl.sdsc.edu/hm.html). The three sets of coordinates were analysed, optimized and visualized with the Deep View software (http://www.expasy.org/spdbv/). The images show a ribbon representation of the integrated model. Helices appear in yellow and sheets in blue. The two catalytically important histidine residues are shown in red and putative N-glycosylation sites in green. Numbers indicate amino acid positions. (C) Monoclonal antibody mAT502 binds to TbMBAP1. An aliquot of a detergent extract from the total membrane fraction of T. brucei bloodstream forms was immunoprecipitated using the mouse monoclonal antibody mAT502. Samples of the precipitate (lane 3) and the supernatant (lane 2) of the immunoprecipitation as well as the detergent extract (lane 1), each corresponding to the equivalent of 2x10⁶ cells, were subjected to immunoblotting. The blot was probed with rabbit anti-TbMBAP1 and horseradish peroxidase coupled secondary antibodies. The molecular mass of standard proteins in kDa is indicated.