.atr/tem8 Vwa Domain And Pa Interaction

Anthrax toxin is a major virulence factor of Bacillus anthracis. The toxin consists of three protein subunits: protective antigen (PA), lethal factor (LF), and edema factor (EF). The anthrax toxin receptors, tumor endothelium marker-8(TEM8) and capillary morphogenesis protein-2 (CMG2), both contain an extracellular vWA domain shared 60% identical residues. Inhibition of PA-receptor binding could effectively cut off anthrax intoxication way, and is especially important for anthrax therapy at late stage when the toxin has been release and antibiotics intervention doesn't work. Crystal structures of PA,CMG2 and PA-CMG2 have been analyzed and interactions between PA and CMG2 has been well described. Receptor-like agonists, such as mammalian cell expressed vWA domain of CMG2 (sCMG2), have been explored for their potency as anti-toxin reagents.In this work, A high resolution (1.7 ?) crystal structure of ATR/TEM8 vWA domain was analyzed by X-ray diffraction. Our crystal structure contained six TEM8 vWA molecules in the asymmetric unit (also unit cell). In this hexamer, the MIDAS ligand binding area from each monomer was blocked which indicates that such a hexamer is an inactive oligomeric form. Moreover, TEM8 was determined to exist as a monomer in solution by analytical ultracentrifuge and gel filtration (data not shown). Thus, the observed TEM8 hexamer is most likely an artifact of crystal packing.Consistent with their sequence homology, the structure of the TEM8 extracellular domain is very similar to that of CMG2 and the integrin A domain. It adopts a classicalα/βopen sheet fold that has also been called the dinucleotide-binding fold, Rossmann fold, or doubly wound fold. It should be noted that an acetate ions were found correspond to an analogous ion in the CMG2 structure, which acted as a mimic ligand of the side chain of D683 in PA and occupied the MIDAS coordination site. Besides this ligand, S52, S54,T118 and two water molecular was found to coordinate with the metal ion Mg²⁺, indicating that the TEM8 extracellular domain contained a conserved Mg²⁺-coordinated MIDAS motif that assumed an integrin-like open conformation.Based on this structure and the PA-CMG2 complex structure (PDB 1T6B), a model structure of PA-TEM8 complex was generated by molecular replacement after molecular superimposing with tether. After being applied with Solvation and Energy Minimization programs by Discovery with CHARMm forcefield , models of complex of PA and both receptors in solution were gain. By comparing the models, serveral conserved interactions between receptors and PA were found: PA D683 N682 interacted with receptors MIDAS motif residues S52,S54 and T118, besides Y119 and K51 in receptor could bind to residues in PA tightly. Also, there existed edsignificant differences in our complex models, including interactions between unconserved residues at position 56, 88,113,115,156(based on TEM8's sequence) and PA residues, respectively. Also a salt bridge of E344 existed in PA-CMG2 model bingding with PA residue E344 was observed replaced by attenuated electricity potential complement with PA R659 in PA–TEM8 Model.A homologue-replacement scan mutantion of the unconserved residues homologue to that at the PA-CMG2 interaction interface was introduced to TEM8 vWA domain, After expressing in E.coli and purification of the soluble parts, the protection ability from LeTx attacking of the mutants was evaluated using J774A.1 cells. Of all the soluble single-residue-replacement mutants, L56A showed most excellent performance (about 3 times less of wild type's IC50), and E155G also showed an increment of protection ability (about 1/2 of wild type's IC50), compared to the wild type of TEM8. When combined E155G and L56A, the viability was elevated(about 6 folds less of IC50) In contrast, the H57N, R88Q, L157V, and F158P point mutants had decreased protective ability. SPR assay was also used to assay their affinity to PA, and the results were consistent with those from cell assays.Also, a SDS-resistant pore formation assay was performed to analyze these mutations'influence on the pH threshold for conversion of a PA prepore to pore. All single point mutations did not cause a significant change in the pH threshold of pore formation. However, a multiple mutation (154-158) showed a striking change in the pH threshold of pore formation, with a value as approaching to that of CMG2.In addition, a mathematic model supporting the toxin neutralization assay and receptor-like antagonists based Schild plot assay was deduced. With this model, the results of cell assay (measured as IC50s in protection assay, and EC50s in Schild plot assay) were related with the kinetic parameter KD (equilibrium dissociation constant). This model to applied to evaluate the affinity of TEM8 mutants L56A and E155G+L56A, while the SPR result was consistent with this model at some extent.At last, the potent of TEM8 and its mutant L56A as an anti-toxin for PA was tested on rats. Unlike previous report, TEM8 expressed in this work exhibited a much better performance, with total protection at 7.5:1-3:1 (mol ratio, receptor:PA), whereas CMG2's protection was much consistent at 3:1.The L56A, with total protection between 3:1 and 1:1, seems perform even better than CMG2, in spite of its KD and cell protection ability inferior to that of CMG2. this disaccord was ascribed to that CMG2 may binds to other ligands in vivo while TEM8 and its mutant would not, which implied a possible side effect of CMG2 as anti-toxin。However, further evidence still needed.In conclusion, the crystal structure of the TEM8 vWA domain at 1.7 ? resolution was determined. In addition, models of PA-receptor complex in solution were constructed. The structure and models provide fundament for further dynamic simulation about the conform alternation and molecular behavior in solution, which would provide further details of interaction and help us to design affinity-elevated receptor mutants as anti-toxins or PA variants selectively bind to TEM8 as tumor targeting vector. Meanwhile, residues at position 56 88 155 were found to differentiate receptors'bind to PA, and homologue-replacment mutations to 56 and/or 155 in TME8 would all elevate the affinity, while R88Q mutation could attenuate the binding. Furthermore tem8 mutant L56A exhibited a slight superior protection ability on rats, which would provide an alternative choice for CMG2 considering possible side-effect implied in our work. These findings point out potent hot spot for designing more efficient receptor-like antitoxins. Moreover, the mathematic model deduced in this work for relating the results of cell assay to affinity parameters (dissociation constant, KD), would reduce the errors induced by discrimination of cell numbers and receptor-expressing states of different experimental repeats, and provide a more precise quantification for evaluating receptor-like inhibitors.L56A,a bacteria produced TEM8 mutant presented in this work, could as effectively inhibit toxication on rats as sCMG2, and would be provided as a candidate as anti-toxin drugs.