The initial mapping of the

The initial mapping of the epitopes for new anti-PAR mAbs was performed by immobilizing the antibody in question on a CM5 sensor chip™ (GE Helthcare) with conventional amide chemistry (EDC/NHS). First, the kinetics rate constants (k and k) for the mAb·uPAR interactions were determined at low surface densities to minimize possible confounding effects from mass transport limitations. Second, the domain reactivity of these mAbs was established by measuring the binding to intact uPAR, uPAR DI, and uPAR DIIDIII. Third, epitope binning was performed by measuring whether a second anti-uPAR mAb can bind to the uPAR captured by the immobilized mAb. Fourth, a complete single-site epitope mapping was performed by measuring the kinetics between the immobilized mAb and a library of purified single site uPAR mutants produced in S2 RGFP966 [3,6]. Those anti-uPAR mAbs with unique and relevant functional and/or structural properties were then selected for further X-ray crystallography studies to provide a high-resolution structure of the topography of the mAb uPAR binding interface and define which of the uPAR conformations is selected by that particular mAb. The power of this approach is illustrated by the apparent convergent evolution of the three-dimensional hot-spot configuration of the binding interface between the native biologic ligand (vitronectin) and the surrogate ligand (mAb 8B12) – see Fig. 5 in the original publication [5]. This relationship is not evident from direct sequence alignments of the primary structures of the mAbs in question and the natural ligand vitronectin. The summary of the topographic epitope landscape on human uPAR is illustrated below in Fig. 1 and this information provides an important working tool for selecting those particular mAbs that are optimal for a given biological experiment and the knowledge of the corresponding hot-spot residues on uPAR may in some case also provide the ideal negative control uPAR mutant for such intervention studies. The general utility of this “toolbox”, which is providing data on epitope-mapped anti-uPAR mAbs, is evident from the following example. How do you choose the optimal reagent for functional blocking of uPAR-mediated effects on cell adhesion? If mAbs from epitope bin 1 (i.e. R3 or R21) are selected as intervention agents they will indeed abrogate uPAR-mediated adhesion on vitronectin in conditions with very low levels of uPA [3,12,13], but this effect will critically depend on the level of uPAR-occupancy with uPA as these mAbs will not bind uPA·uPAR complexes [5]. Another confounding factor in such studies is the observation that uPA-binding as such increases cell migration [4,14]. These complicating factors are, nonetheless, minimized if anti-uPAR mAbs from bin 6 are selected as intervention agents, as mAbs from this particular epitope bin (i.e. 8B12 or 19.10) inhibit vitronectin binding and uPAR-mediated cell adhesion even under conditions when the receptor is completely saturated with uPA [5].
Materials and methods
Acknowledgment This work was supported by National Science Foundation of China (30811130467), The Danish National Research Foundation (26-331-6) (Centre for Proteases and Cancer), and the Italian Association for Cancer Research (AIRC) (IG 10494 and IG 14466).
Specifications table
Value of the data
Data, experimental design, materials and methods
Conflicts of interest
Acknowledgments The work was partly supported by the Special Fund for Agro-scientific Research in the Public Interest of the People\'s Republic of China (Grant no. 201403075) and the Major Science and Technology Program of Hainan Province (ZDZX2013010).
Data To identify Brakeless target genes in the early embryo, RNA was isolated from embryos derived from brakeless (bks) germline clones, which lack the maternal contribution of Brakeless. This was compared to RNA from germline clone embryos generated with the unmutagenized FRT chromosome on which the bks allele was induced [8]. The RNA was converted to cDNA and hybridized to an Affymetrix array. Mis-regulated genes that change their expression more than 1.5 fold were identified (supplementary material Table 1). We compared our gene list to an RNA-seq dataset that distinguishes maternal from zygotic transcripts in Drosophila embryos using polymorphisms [7]. The Brakeless-regulated genes were categorized as being maternally, zygotically, or maternally and zygotically (matzyg) derived (supplementary material Table 1). They were also subjected to functional annotation analysis using DAVID [3], which groups genes into clusters based on co-association with gene ontology (GO) terms (supplementary material Table 1). As shown in Fig. 1, up-regulated and down-regulated gene fall into distinct GO clusters.