Limits...
The structure of a furin-antibody complex explains non-competitive inhibition by steric exclusion of substrate conformers

View Article: PubMed Central - PubMed

ABSTRACT

Proprotein Convertases (PCs) represent highly selective serine proteases that activate their substrates upon proteolytic cleavage. Their inhibition is a promising strategy for the treatment of cancer and infectious diseases. Inhibitory camelid antibodies were developed, targeting the prototypical PC furin. Kinetic analyses of them revealed an enigmatic non-competitive mechanism, affecting the inhibition of large proprotein-like but not small peptidic substrates. Here we present the crystal structures of furin in complex with the antibody Nb14 and of free Nb14 at resolutions of 2.0 Å and 2.3 Å, respectively. Nb14 binds at a site distant to the substrate binding pocket to the P-domain of furin. Interestingly, no major conformational changes were observed upon complex formation, neither for the protease nor for the antibody. Inhibition of furin by Nb14 is instead explained by steric exclusion of specific substrate conformers, explaining why Nb14 inhibits the processing of bulky protein substrates but not of small peptide substrates. This mode of action was further supported by modelling studies with the ternary factor X-furin-antibody complex and a mutation that disrupted the interaction interface between furin and the antibody. The observed binding mode of Nb14 suggests a novel approach for the development of highly specific antibody-based proprotein convertase inhibitors.

No MeSH data available.


Modelling of the furin-factor X enzyme-substrate complex.The furin:Nb14 complex is shown as surface representation and coloured in yellow (furin, catalytic domain), brown (furin, P-domain) and blue (Nb14). For modelling, the catalytic domain of factor X (cartoon representation) was linked to the N-terminus of the P1-P4 tetrapeptide (stick representation, magenta) of the co-crystallised RVKR-CMK inhibitor. (a) The one conformer that fits to the furin:Nb14 complex is coloured in magenta. (b) Conformers that do not fit to the furin:Nb14 complex are shown in grey.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5037460&req=5

f4: Modelling of the furin-factor X enzyme-substrate complex.The furin:Nb14 complex is shown as surface representation and coloured in yellow (furin, catalytic domain), brown (furin, P-domain) and blue (Nb14). For modelling, the catalytic domain of factor X (cartoon representation) was linked to the N-terminus of the P1-P4 tetrapeptide (stick representation, magenta) of the co-crystallised RVKR-CMK inhibitor. (a) The one conformer that fits to the furin:Nb14 complex is coloured in magenta. (b) Conformers that do not fit to the furin:Nb14 complex are shown in grey.

Mentions: Globular folded parts of substrate proteins do however show an additional level of access control to the active site of proteases, as they are largely sterically restricted from the substrate binding cleft. Therefore, the binding of proteinaceous substrates to furin and their cleavage should only be possible if their overall conformation fits to the specific topology of the furin:Nb14 complex. This hypothesis was tested by means of modelling studies with the coagulation factor X (Figs 4 and S4). Proteolysis occurs here C-terminal to the recognition sequence Arg181-[P3]-Lys183-Arg184-↓. The P1-P4 amino acids are estimated to bind in a similar conformation as observed for the RVKR-CMK inhibitor complex. For our modelling studies this peptide stretch was aligned immediately C-terminal to the EGF2-domain of factor Xa (PDB-ID:2GD433), creating an extended factor Xa model, Xa* (catalytic domain including the furin recognition motif). The whole catalytic domain of factor Xa* was hereby treated as a rigid body, varying initially only the relative conformation of the amino acids between Thr178 and Arg181 and later also of the peptide stretch Cys174 to Glu180. Using this approach we modelled 13 different conformers of furin-bound factor Xa* that fit to the substrate binding pocket (Figs 4 and S4a). During modelling we realised that the conformation of Xa* is largely restricted by the deep substrate binding cleft of furin and that the region between Thr178 and Arg181 requires a rather extended conformation to avoid steric clashes of furin and factor Xa. In presence of the Nb14 the conformational space is even more restricted and 12 out of the 13 initially modelled Xa* conformers now show main-chain clashes with the antibody (Figs 4 and S4b). These clashes exclusively occur at sites distant to the active site cleft of furin. Consequently, the modelling studies suggest a mode of action of Nb14 involving conformational restriction of the substrate. Binding and turnover of substrate proteins is still possible if this antibody is bound to furin, but now only a subset of all possible substrate conformers is still allowed to bind to and to be cleaved by the Nb14:furin complex.


The structure of a furin-antibody complex explains non-competitive inhibition by steric exclusion of substrate conformers
Modelling of the furin-factor X enzyme-substrate complex.The furin:Nb14 complex is shown as surface representation and coloured in yellow (furin, catalytic domain), brown (furin, P-domain) and blue (Nb14). For modelling, the catalytic domain of factor X (cartoon representation) was linked to the N-terminus of the P1-P4 tetrapeptide (stick representation, magenta) of the co-crystallised RVKR-CMK inhibitor. (a) The one conformer that fits to the furin:Nb14 complex is coloured in magenta. (b) Conformers that do not fit to the furin:Nb14 complex are shown in grey.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5037460&req=5

f4: Modelling of the furin-factor X enzyme-substrate complex.The furin:Nb14 complex is shown as surface representation and coloured in yellow (furin, catalytic domain), brown (furin, P-domain) and blue (Nb14). For modelling, the catalytic domain of factor X (cartoon representation) was linked to the N-terminus of the P1-P4 tetrapeptide (stick representation, magenta) of the co-crystallised RVKR-CMK inhibitor. (a) The one conformer that fits to the furin:Nb14 complex is coloured in magenta. (b) Conformers that do not fit to the furin:Nb14 complex are shown in grey.
Mentions: Globular folded parts of substrate proteins do however show an additional level of access control to the active site of proteases, as they are largely sterically restricted from the substrate binding cleft. Therefore, the binding of proteinaceous substrates to furin and their cleavage should only be possible if their overall conformation fits to the specific topology of the furin:Nb14 complex. This hypothesis was tested by means of modelling studies with the coagulation factor X (Figs 4 and S4). Proteolysis occurs here C-terminal to the recognition sequence Arg181-[P3]-Lys183-Arg184-↓. The P1-P4 amino acids are estimated to bind in a similar conformation as observed for the RVKR-CMK inhibitor complex. For our modelling studies this peptide stretch was aligned immediately C-terminal to the EGF2-domain of factor Xa (PDB-ID:2GD433), creating an extended factor Xa model, Xa* (catalytic domain including the furin recognition motif). The whole catalytic domain of factor Xa* was hereby treated as a rigid body, varying initially only the relative conformation of the amino acids between Thr178 and Arg181 and later also of the peptide stretch Cys174 to Glu180. Using this approach we modelled 13 different conformers of furin-bound factor Xa* that fit to the substrate binding pocket (Figs 4 and S4a). During modelling we realised that the conformation of Xa* is largely restricted by the deep substrate binding cleft of furin and that the region between Thr178 and Arg181 requires a rather extended conformation to avoid steric clashes of furin and factor Xa. In presence of the Nb14 the conformational space is even more restricted and 12 out of the 13 initially modelled Xa* conformers now show main-chain clashes with the antibody (Figs 4 and S4b). These clashes exclusively occur at sites distant to the active site cleft of furin. Consequently, the modelling studies suggest a mode of action of Nb14 involving conformational restriction of the substrate. Binding and turnover of substrate proteins is still possible if this antibody is bound to furin, but now only a subset of all possible substrate conformers is still allowed to bind to and to be cleaved by the Nb14:furin complex.

View Article: PubMed Central - PubMed

ABSTRACT

Proprotein Convertases (PCs) represent highly selective serine proteases that activate their substrates upon proteolytic cleavage. Their inhibition is a promising strategy for the treatment of cancer and infectious diseases. Inhibitory camelid antibodies were developed, targeting the prototypical PC furin. Kinetic analyses of them revealed an enigmatic non-competitive mechanism, affecting the inhibition of large proprotein-like but not small peptidic substrates. Here we present the crystal structures of furin in complex with the antibody Nb14 and of free Nb14 at resolutions of 2.0 Å and 2.3 Å, respectively. Nb14 binds at a site distant to the substrate binding pocket to the P-domain of furin. Interestingly, no major conformational changes were observed upon complex formation, neither for the protease nor for the antibody. Inhibition of furin by Nb14 is instead explained by steric exclusion of specific substrate conformers, explaining why Nb14 inhibits the processing of bulky protein substrates but not of small peptide substrates. This mode of action was further supported by modelling studies with the ternary factor X-furin-antibody complex and a mutation that disrupted the interaction interface between furin and the antibody. The observed binding mode of Nb14 suggests a novel approach for the development of highly specific antibody-based proprotein convertase inhibitors.

No MeSH data available.