Binding Exosite Research Articles

Ligand-Receptor Interactions and Structure-Function Relationships in Off-Target Binding of the β3-Adrenergic Agonist Mirabegron to α1A-Adrenergic Receptors.

The β3-adrenoceptor agonist mirabegron is available for the treatment of storage symptoms of overactive bladder, including frequency, urgency, and incontinence. The off-target effects of mirabegron include binding to α1-adrenoceptors, which are central in the treatment of voiding symptoms. Here, we examined the structure-function relationships in the binding of mirabegron to a cryo-electron microscopy structure of α1A. The binding was simulated by docking mirabegron to a 3D structure of a human α1A-adrenoceptor (7YMH) using Autodock Vina. The simulations identified two binding states: slope orientation involving 10 positions and horizontal binding to the receptor surface involving 4 positions. No interactions occurred with positions constituting the α1A binding pocket, including Asp-106, Ser-188, or Phe-312, despite the positioning of the phenylethanolamine moiety in transmembrane regions close to the binding pocket by contact with Phe-288, -289, and Val-107. Contact with the unique positions of α1A included the transmembrane Met-292 during slope binding and exosite Phe-86 during horizontal binding. Exosite binding in slope orientation involved contact of the anilino part, rather than the aminothiazol end, to Ile-178, Ala-103, and Asn-179. In conclusion, contact with Met-292 and Phe-86, which are unique positions of α1A, accounts for mirabegron binding to α1A. Because of its lack of interactions with the binding pocket, mirabegron has lower affinity compared to α1A-blockers and no effects on voiding symptoms.

Dynamic allostery in thrombin-a review.

Thrombin is a serine protease that catalyzes a large number of different reactions including proteolytic cleave of fibrinogen to make the fibrin clot (procoagulant activity), of the protease activated receptors (for cell signaling) and of protein C generating activated protein C (anticoagulant activity). Thrombin has an effector binding site called the anion binding exosite 1 that is allosterically coupled to the active site. In this review, we survey results from thermodynamic characterization of the allosteric coupling as well as hydrogen-deuterium exchange mass spectrometry to reveal which parts of the thrombin structure are changed upon effector binding and/or mutagenesis, and finally NMR spectroscopy to characterize the different timescales of motions elicited by the effectors. We also relate the experimental work to computational network analysis of the thrombin-thrombomodulin complex.

Structures of factor XI and prekallikrein bound to domain 6 of high–molecular weight kininogen reveal alternate domain 6 conformations and exosites

BackgroundHigh–molecular weight kininogen (HK) circulates in plasma as a complex with zymogen prekallikrein (PK). HK is both a substrate and a cofactor for activated plasma kallikrein, and the principal exosite interactions occur between PK N-terminal apple domains and the C-terminal D6 domain of HK. ObjectivesTo determine the structure of the complex formed between PK apple domains and an HKD6 fragment and compare this with the coagulation factor XI (FXI)-HK complex. MethodsWe produced recombinant FXI and PK heavy chains (HCs) spanning all 4 apple domains. We cocrystallized PKHC (and subsequently FXIHC) with a 31-amino acid synthetic peptide spanning HK residues Ser565-Lys595 and determined the crystal structure. We also analyzed the full-length FXI-HK complex in solution using hydrogen deuterium exchange mass spectrometry. ResultsThe 2.3Å PKHC-HK peptide crystal structure revealed that the HKD6 sequence WIPDIQ (Trp569-Gln574) binds to the apple 1 domain and HK FNPISDFPDT (Phe582-Thr591) binds to the apple 2 domain with a flexible intervening sequence resulting in a bent double conformation. A second 3.2Å FXIHC-HK peptide crystal structure revealed a similar interaction with the apple 2 domain but an alternate, straightened conformation of the HK peptide where residues LSFN (Leu579-Asn583) interacts with a unique pocket formed between the apple 2 and 3 domains. HDX-MS of full length FXI-HK complex in solution confirmed interactions with both apple 2 and apple 3. ConclusionsThe alternate conformations and exosite binding of the HKD6 peptide likely reflects the diverging relationship of HK to the functions of PK and FXI.

A Novel Substrate for Assessing Thrombin Activity in Whole Blood

At sites of vessel injury, thrombin acts as the central mediator of coagulation through catalyzing fibrin clot formation and platelet activation. Currently, thrombin generation is measured using small molecule substrates. These substrates are limited in that they detect both active thrombin and thrombin bound to its inhibitor α2-macroglobulin (α2-M) and have limited utility in whole blood (Ninivaggi et al. Clin Chem 2012). In addition, these substrates contain short thrombin recognition sequences ranging from 3-4 amino acids and are unable to form additional exosite interactions with thrombin, which limit their specificity (Gallwitz et al. Plos One 2012). Furthermore, they are unfeasible for in vivo use because their hydrophobicity allows them to enter cells. However, plasma assays are limited because they ignore the hemostatic contributions of red blood cells, platelets and leukocytes and require anticoagulation and the addition of exogenous coagulation activators (Swystun & Liaw. Blood 2016 & do Amaral et al. Stem Cells Int 2016). Thus, we have developed a fluorescence resonance energy transfer (FRET)-based thrombin sensor to overcome these limitations. We cloned and expressed the construct encoding the FRET-based thrombin substrate in Escherichia coli and purified the substrate using affinity chromatography. We confirmed using Sodium dodecyl sulfate−polyacrylamide gel electrophoresis that the substrate was successfully purified and that thrombin cleavage of the purified substrate changes its FRET efficiency by monitoring its spectrum of emission. To determine whether the FRET-based thrombin substrate can form interactions with thrombin's anion binding exosites and assess its potential for in vivo use, we measured the kinetics parameters of the substrate for thrombin in the presence of exosite inhibitors and thrombin's exosite variants as well as mouse thrombin and plasmin. We then compared the resulting kinetics parameters with commercially available thrombin substrates S2238 and technothrombin. We show that unlike S2238 and technothrombin, our FRET-based thrombin sensor forms strong interactions with thrombin's anion binding exosite I and displays preference for free thrombin compared to thrombomodulin (TM) bound thrombin which implied its preference for procoagulant thrombin rather than anticoagulant TM bound thrombin that activates protein C. Furthermore, our FRET-based thrombin substrate has potential to be used in mouse models because it can be efficiently cleaved by mouse thrombin and is relatively plasmin resistant. To assess the specificity of the FRET-based thrombin substrate for thrombin, we compared the rate of hydrolysis of the substrate by thrombin and several other enzymes in circulation and compared the results to S2238 and technothrombin. Our results indicated that when compared to commercially available thrombin substrates, our FRET-based thrombin substrate is highly specific for thrombin. Since the thrombin inhibitor α2-M is present in circulation, we also compared the rate of hydrolysis of the 3 substrates for α2-M bound thrombin and free thrombin. Unlike S2238 and technothrombin, our FRET-based thrombin substrate is uncleaved by α2-M bound thrombin. The key properties of the FRET-based thrombin substrate in comparison to S22238 and technothrombin are summarized in Table 1. We then measured thrombin generation in plasma using our FRET-based thrombin substrate and showed that it is sensitive to direct oral anticoagulants in ways that are consistent with literature findings where thrombin generation was measured using technothrombin (Sivaraja et al. Plos One 2018 & Douxfils et al. Thromb Res 2012). We also confirmed that the FRET-based thrombin substrate itself is not a coagulation inhibitor since its presence did not alter thrombin generation parameters measured using technothrombin. Lastly, we confirmed that the FRET-based thrombin substrate can measure thrombin generation in whole blood since it successfully produced thrombin generation curves. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Exosite binding modulates the specificity of the immunomodulatory enzyme ScpA, a C5a inactivating bacterial protease

The C5a peptidase from Streptococcus pyogenes (ScpA) is a highly specific enzyme with potential therapeutic value. ScpA is a good model for studying determinants of specificity in the multidomain immunomodulatory enzymes (IMEs), which comprise a large family of bacterial surface proteases. The surface exposed region of ScpA has 5 main domains which includes 3 C-terminal Fn3-like domains (Fn1, Fn2 and Fn3) (Kagawa et al. 2009). Progressive deletion of the Fn3-like domains from the C-ter resulted in loss of enzyme activity and showed an important role for the Fn2 domain in enzyme function. Functional investigation of specific acidic residues on the Fn2 domain identified 3 residues 30–50 Å from the catalytic site (D783, E864 and D889) which impacted to differing degrees on binding and on catalysis, supporting the presence of an exosite on the Fn2. In particular, residue D783 was observed to impact on both substrate binding affinity and the activity of ScpA. A double mutant cycle analysis showed energetic coupling between the targeted ScpA residues and residues in the core portion (residues 1–67) of the C5a substrate. The data supports the presence of a communication network between the active site and the exosite on Fn2. These findings provide a basis for rational engineering of this important enzyme family to enhance stability, activity and/or specificity.

Discovery of an exosite on the SOCS2-SH2 domain that enhances SH2 binding to phosphorylated ligands

Suppressor of cytokine signaling (SOCS)2 protein is a key negative regulator of the growth hormone (GH) and Janus kinase (JAK)-Signal Transducers and Activators of Transcription (STAT) signaling cascade. The central SOCS2-Src homology 2 (SH2) domain is characteristic of the SOCS family proteins and is an important module that facilitates recognition of targets bearing phosphorylated tyrosine (pTyr) residues. Here we identify an exosite on the SOCS2-SH2 domain which, when bound to a non-phosphorylated peptide (F3), enhances SH2 affinity for canonical phosphorylated ligands. Solution of the SOCS2/F3 crystal structure reveals F3 as an α-helix which binds on the opposite side of the SH2 domain to the phosphopeptide binding site. F3:exosite binding appears to stabilise the SOCS2-SH2 domain, resulting in slower dissociation of phosphorylated ligands and consequently, enhances binding affinity. This biophysical enhancement of SH2:pTyr binding affinity translates to increase SOCS2 inhibition of GH signaling.

Protease Domain Exosites Regulate Plasma Recovery and Extravascular Distribution of Factor IX(a) Variants in Hemophilic Mice

Introduction:Among the coagulation protease, factor IX (FIX) is unique in that a substantial pool of extravascular factor exists that contributes to in vivo hemostasis. Extravascular distribution of FIX contributes to poor plasma recovery of infused FIX in hemophilia B (HB). The effect of mutations in the heparin (K126A/K132A) and antithrombin (R150A) binding exosites of FIX and FIXa on the distribution of these variants between the plasma and extravascular binding sites was examined in HB and hemophilia A (HA) mice, respectively.Methods:Recombinant human FIX wild type (18 U/kg) or equimolar FIX K126A/K132A was administered to HB mice (n=8) via tail vein injection with sacrifice 5 min post-injection following retro-orbital blood collection. Equimolar doses of human FIXa wild type (WT), K126A/K132A and R150A were similarly injected into HA mice (n=4-8), with blood collection and sacrifice 5 min post-injection. Plasma was isolated by centrifugation, organs (liver, spleen, kidneys, heart, lung, and brain) harvested, rinsed with PBS, weighed, with a portion made into tissue lysates and remainder frozen. FIX(a) content of tissue lysates was determined by a species-specific human FIX(a) ELISA with variant-specific standard curves. Tissue concentration was determined by total FIX(a) present in lysate (ng) divided by respective tissue weight (mg). Organ distribution was determined by extrapolating FIX(a) tissue concentration to total organ weight. The FIX(a) present in a specific organ was divided by the administered FIX(a) dose to determine the % organ distribution.Results:PK studies in HB mice demonstrated that FIX WT and R150A demonstrated a similar pattern and time course of elimination. In contrast, FIX K126A/K132A demonstrated ~2.4 fold higher plasma recovery relative to FIX WT. Based on plasma concentrations at 5 min post-injection (plasma volume 40 ml/kg), ~12.9% of FIX WT and 30.5% of FIX K126A/K132A was localized to the plasma compartment. Tissue lysates from the liver, spleen, kidney and brain demonstrated that liver had the highest FIX content by far, with ~41% of the FIX WT dose at 5 min post-injection. Other tissues demonstrated markedly less FIX WT content, including kidney (1%), spleen (0.1%) and brain (undetectable). FIX tissue concentration was significantly higher in the liver (0.69 ng/mg tissue) compared to other tissues, followed by kidney (0.07 ng/mg), spleen (0.02 ng/mg) and brain (undetectable). Comparison of FIX K126A/K132A tissue content to WT demonstrated reduced liver localization (25% dose) with concomitant reduction in tissue concentration (0.45 ng/mg). In contrast, FIX K126A/K132A increased localization to the kidney (1.7%) and spleen (0.37%) relative to WT.In protease PK studies, FIXa K126A/K132A and R150A both enhanced plasma recovery (2.2-2.5 fold) in HA mice compared to FIXa WT. Based on plasma concentrations at 5 min post-injection, ~16.1% of FIXa WT and 41.1% of FIXa K126A/K132A dose localized in the plasma compartment. Similar to zymogen, tissue lysates demonstrated that liver had the highest FIXa WT content (28%). Other tissues contained markedly less FIXa, including kidney (2.4%), spleen (0.2%), heart (0.5%) and brain (undetectable). Similarly, FIXa tissue concentration was 4-fold higher in the liver (0.47 ng/mg) compared to other tissues, followed by kidney (0.12 ng/mg), spleen and heart (0.06 ng/mg) and brain (undetectable). Localization of FIXa K126A/K132A to the liver was reduced by nearly half compared to FIXa WT (16%) with reduction in liver tissue concentration (0.29 ng/mg). FIXa K126A/K132A had increased localization to the kidney (3.3%) relative to WT. In contrast, FIXa R150A (not shown) had similar liver distribution (31.9% dose) and tissue concentration (0.53 ng/mg) to FIXa WT, despite increased localization to the plasma (36.2%).Conclusions:Liver is the predominant organ for extravascular FIX binding and the heparin binding exosite contributes to this localization, similar to the collagen IV binding site in the Gla domain. Heparan sulfate and collagen IV co-localize in the basement membrane suggesting synergistic roles. FIXa also localizes to extravascular sites via the heparin-binding exosite, although that binding may be modestly limited by endogenous FIX in the HA mouse. Disruption of the antithrombin binding exosite re-distributes FIXa to the plasma compartment by an independent mechanism, as liver content is unchanged. DisclosuresSheehan: Bayer: Consultancy, Research Funding; Roche: Consultancy, Research Funding; BioMarin: Consultancy, Research Funding.

Exosite Binding in Thrombin: A Global Structural/Dynamic Overview of Complexes with Aptamers and Other Ligands.

Thrombin is the key enzyme of the entire hemostatic process since it is able to exert both procoagulant and anticoagulant functions; therefore, it represents an attractive target for the developments of biomolecules with therapeutic potential. Thrombin can perform its many functional activities because of its ability to recognize a wide variety of substrates, inhibitors, and cofactors. These molecules frequently are bound to positively charged regions on the surface of protein called exosites. In this review, we carried out extensive analyses of the structural determinants of thrombin partnerships by surveying literature data as well as the structural content of the Protein Data Bank (PDB). In particular, we used the information collected on functional, natural, and synthetic molecular ligands to define the anatomy of the exosites and to quantify the interface area between thrombin and exosite ligands. In this framework, we reviewed in detail the specificity of thrombin binding to aptamers, a class of compounds with intriguing pharmaceutical properties. Although these compounds anchor to protein using conservative patterns on its surface, the present analysis highlights some interesting peculiarities. Moreover, the impact of thrombin binding aptamers in the elucidation of the cross-talk between the two distant exosites is illustrated. Collectively, the data and the work here reviewed may provide insights into the design of novel thrombin inhibitors.

A Single Cleavage Reaction in the Proteolytic Activation of Factor V to the Cofactor Is Rate-Limiting in a Step That Is Selectively Dependent on Anion Binding Exosite 2 of Thrombin

Human factor V (hFV) is a procofactor with a domain organization of A1-A2-B-A3-C1-C2. It requires proteolytic activation before it can assemble within the prothrombinase complex and function as a cofactor. Activation by thrombin results from cleavages following Arg709, Arg1018 and Arg1545 to release the B domain in two fragments and form heterodimeric factor Va (FVa, A1-A2/A3-C1-C2). While A1-A2/A3-C1-C2 is the prototypic active cofactor, variable cofactor function is attributed to partially cleaved species. Several studies have proposed that meizothrombin (mIIa), an intermediate produced during prothrombin activation, is also an efficient activator of membrane-bound hFV and possibly responsible for FVa formation early in the clotting process. However, this is inconsistent with the proposed importance of anion binding exosite 2 (ABE2) for hFV activation by thrombin which is occluded by the covalently attached propiece in mIIa. We have studied the kinetics of hFV cleavage using a quantitative two-color immunoblotting approach to simultaneously and unambiguously identify intermediates and products containing A1-A2 and/or A3-C1-C2. We have also compared hFV cleavage by thrombin by cleavage with a stable mIIa variant and by mIIa lacking both the 4-carboxyglutamic acid and kringle 1 domains (mIIa-desF1) in the presence of membranes. Addition of 0.1 nM thrombin yielded quantitative cleavage of hFV (50 nM) in ~6 minutes, producing products consistent with cleavage at Arg709 and/or Arg1018. However, the C-terminal intermediate beginning at 1019 but uncleaved at Arg1545 persisted for more than 90 minutes. A3-C1-C2, arising from cleavage at Arg1545, appeared at a ~20-fold slower rate than the A1-A2 species. Thus, for the reaction with thrombin, cleavage at Arg1545 represents the rate limiting step in the conversion of hFV to heterodimeric FVa. Similar rates of hFV consumption and appearance of species cleaved at Arg709 and/or Arg1018 were seen with mIIa. However, cleavage at Arg1545 to produce A3-C1-C2 was substantially slower, by a factor of ~100, than A1-A2 formation. Equivalently large differences in the rates of A1-A2 and A3-C1-C2 formation were evident in the absence of membranes except the rate of all cleavages by mIIa was further reduced. Because the results obtained with mIIa were replicated with mIIa-desF1, the data imply that it is covalent occlusion of ABE2 in these mIIa variants that underlies their selectively impaired ability to cleave at Arg1545 and produce heterodimeric Va but does not affect cleavage at Arg709 or Arg1018. The binding of Xa to activated FV, essential for membrane-dependent prothrombinase assembly, is a prerequisite for cofactor function. We correlated hFV cleavage with the development of cofactor function by binding measurements using Xa inactivated with a fluorescent probe covalently tethered to the active site with a peptidyl chloromethylketone. For hFV cleavage by thrombin, Xa binding and the assembly of prothrombinase proceeded unexpectedly slowly and correlated with cleavage at Arg1545, widely separated in time from rapid cleavage at Arg709 and/or Arg1018. Therefore, slow cleavage at Arg1545 dictates both the formation of heterodimeric FVa and the ability of the product to bind Xa with high affinity to form prothrombinase. Only a small fraction of Xa binding developed within an hour in hFV cleavage reactions initiated with mIIa in comparison to the same concentration of thrombin. Our findings indicate that ABE2 in thrombin variants plays an essential role in the rate-limiting cleavage at Arg1545 required for the formation of heterodimeric FVa and for the assembly of prothrombinase. Occlusion of ABE2 by the covalently linked propiece in mIIa variants greatly impairs their ability to activate hFV to the active cofactor by selectively decreasing the rate of cleavage at Arg1545. The surprising and near absolute need for cleavage at Arg1545 to produce cofactor is consistent with new biochemical and structural evidence implicating an essential role played by an acidic sequence immediately N-terminal to Arg1545 in concert with a basic sequence C-terminal to Arg1018 within the B domain in restricting the binding of Xa to the incompletely cleaved procofactor and preventing cofactor function. DisclosuresKrishnaswamy:Novo Nordisk: Consultancy; Janssen: Consultancy, Research Funding; Portola: Research Funding.

Serine protease dynamics revealed by NMR analysis of the thrombin-thrombomodulin complex

Serine proteases catalyze a multi-step covalent catalytic mechanism of peptide bond cleavage. It has long been assumed that serine proteases including thrombin carry-out catalysis without significant conformational rearrangement of their stable two-β-barrel structure. We present nuclear magnetic resonance (NMR) and hydrogen deuterium exchange mass spectrometry (HDX-MS) experiments on the thrombin-thrombomodulin (TM) complex. Thrombin promotes procoagulative fibrinogen cleavage when fibrinogen engages both the anion binding exosite 1 (ABE1) and the active site. It is thought that TM promotes cleavage of protein C by engaging ABE1 in a similar manner as fibrinogen. Thus, the thrombin-TM complex may represent the catalytically active, ABE1-engaged thrombin. Compared to apo- and active site inhibited-thrombin, we show that thrombin-TM has reduced μs-ms dynamics in the substrate binding (S1) pocket consistent with its known acceleration of protein C binding. Thrombin-TM has increased μs-ms dynamics in a β-strand connecting the TM binding site to the catalytic aspartate. Finally, thrombin-TM had doublet peaks indicative of dynamics that are slow on the NMR timescale in residues along the interface between the two β-barrels. Such dynamics may be responsible for facilitating the N-terminal product release and water molecule entry that are required for hydrolysis of the acyl-enzyme intermediate.

Exosite binding drives substrate affinity for the activation of coagulation factor X by the intrinsic Xase complex

Factor X activation by the intrinsic Xase complex, composed of factor IXa bound to factor VIIIa on membranes, is essential for the amplified blood coagulation response. The biological significance of this step is evident from bleeding arising from deficiencies in factors VIIIa or IXa in hemophilia. Here, we assess the mechanism(s) that enforce the distinctive specificity of intrinsic Xase for its biological substrate. Active-site function of IXa was assessed with a tripeptidyl substrate (PF-3688). The reversible S1 site binder, 4-aminobenzamidine (pAB), acted as a classical competitive inhibitor of PF-3688 cleavage by Xase. In contrast, pAB acted as a noncompetitive inhibitor of factor X activation. This disconnect between peptidyl substrate and protein substrate cleavage indicates a major role for interactions between factor X and extended sites on Xase in determining substrate affinity. Accordingly, an uncleavable factor X variant, not predicted to engage the active site of IXa within Xase, acted as a classical competitive inhibitor of factor X activation. Fluorescence studies confirmed the binding of factor X to Xase assembled with IXa with a covalently blocked active site. Our findings suggest that the recognition of factor X by the intrinsic Xase complex occurs through a multistep "dock-and-lock" pathway in which the initial interaction between factor X and intrinsic Xase occurs at exosites distant from the active site, followed by active-site docking and bond cleavage.

C‐terminal residues of activated protein C light chain contribute to its anticoagulant and cytoprotective activities

BackgroundActivated protein C (APC) is an important homeostatic blood coagulation protease that conveys anticoagulant and cytoprotective activities. Proteolytic inactivation of factors Va and VIIIa facilitated by cofactor protein S is responsible for APC's anticoagulant effects, whereas cytoprotective effects of APC involve primarily the endothelial protein C receptor (EPCR), protease activated receptor (PAR)1 and PAR3. ObjectiveTo date, several binding exosites in the protease domain of APC have been identified that contribute to APC's interaction with its substrates but potential contributions of the C‐terminus of the light chain have not been studied in detail. MethodsSite‐directed Ala‐scanning mutagenesis of six positively charged residues within G142‐L155 was used to characterize their contributions to APC's anticoagulant and cytoprotective activities. Results and ConclusionsK151 was involved in protein S dependent‐anticoagulant activity of APC with some contribution of K150. 3D structural analysis supported that these two residues were exposed in an extended protein S binding site on one face of APC. Both K150 and K151 were important for PAR1 and PAR3 cleavage by APC, suggesting that this region may also mediate interactions with PARs. Accordingly, APC's cytoprotective activity as determined by endothelial barrier protection was impaired by Ala substitutions of these residues. Thus, both K150 and K151 are involved in APC's anticoagulant and cytoprotective activities. The differential contribution of K150 relative to K151 for protein S‐dependent anticoagulant activity and PAR cleavage highlights that binding exosites for protein S binding and for PAR cleavage in the C‐terminal region of APC's light chain overlap.

Antithrombin- and Heparin-Binding Exosites Regulate Extravascular Binding and Clearance of Factor IXa

Introduction: Administration of increasing doses of FIX to hemophilia B (HB) mice reveals that the extravascular compartment is capable of binding several-fold more FIX than circulates in the plasma (Cooley et al., Blood 2019). FIX binds rapidly and reversibly to vascular endothelium and the extravascular matrix, in part mediated by interaction of the Gla domain with collagen IV (Col IV). Anticoagulant heparan sulphate and Col IV localize predominantly to the basement membrane and supramolecular structure analysis suggests that heparan sulphate chains are integrated into Col IV-containing networks. Previous work has demonstrated the contribution of the heparin and antithrombin (AT) binding exosites on the protease domain to the regulation of FIXa activity. We now compare the contribution of these protease exosites to in vivo recovery and clearance of the FIX(a) protease and zymogen in hemophilia A (HA) or HB mice. Methods: Recombinant human FIX variants expressed in HEK293 cells with substitutions in the AT- (R150A) and heparin-binding (K126A/K132A) exosites (chymotrypsinogen #) were purified to homogeneity and activated to FIXa with FXIa. FIX(a) pharmacokinetics (PK) in HA or HB mice were determined with injection of a weight-based dose into the tail vein. Serial retro-orbital blood samples were collected, plasma isolated and FIX(a) content determined with a human FIX specific ELISA. FIX(a) concentrations were plotted versus time and fit to the equation: [IX]t = [IX]max*(C1*e(-k1*t)+ C2*e(-k2*t)), where k1and k2are rate constants for the initial and terminal elimination phases, respectively. As the 4-parameter fit resulted in some large error estimates, the two phases were fit separately to estimate the respective parameters. Results: PK for clearance of predicted 30% plasma levels of human FIXa (5 min to 3 hr) or FIX (5 min to 25 hr) were examined in HA (Fig 1) or HB mice (Fig 2), respectively. HA mice were employed for FIXa to avoid in vivoclot formation triggered by active protease. Notably, FIXa plasma clearance was relatively prolonged with ~50% of recovered antigen levels present at 25-30 min post-injection. Similar to FIX, the pattern of FIXa plasma clearance suggested a 2-compartment model with initial and terminal phases of elimination. FIXa WT demonstrated approximately ~23% recovery of predicted levels at 5 min in HA mice, similar to the zymogen (21%) in HB mice (Table I). FIXa R150A (reduced AT affinity) demonstrated markedly enhance recovery (43%) in HA mice relative to WT, while FIX R150A (24%) in HB mice which was similar to WT. FIXa K126A/132A (reduced heparin affinity) demonstrated markedly enhanced recovery (55%) in HA mice, slightly enhanced relative to the similarly increased zymogen recovery (48%) in HB mice. FIXa WT-GGACK (active site inhibited) demonstrated similar recovery to the unmodified FIXa WT with a significantly faster rate of clearance (2.2 fold) in the initial phase (k1) and a significantly slower rate of clearance (3.6-fold) in the terminal phase (k2). Overall, the initial rate (k1) was not significantly different among the active FIXa variants in HA mice or among zymogen variants in HB mice. However, comparison of individual proteases variants to their respective zymogens demonstrated that the initial rate (k1) of plasma clearance was 2-3 fold faster for the proteases, while the terminal rate (k2) was 5-9 fold faster (FIXa R150 had the smallest relative increase). Conclusions: The 2-phase elimination pattern for FIXa suggests that significant amounts of protease may exist in a non-circulating extravascular pool, similar to the zymogen. Further, clearance of FIXa antigen from mouse plasma is prolonged, consistent with the half-life of FIXa activity in human plasma (~41 min). FIXa variants with reduced affinity for AT (R150A) or heparin (K126A/K132A) showed enhanced recovery compared to FIXa WT, suggesting that these protease exosites make independent contributions to the extravascular binding of FIXa. The heparin exosite makes a similar contribution to zymogen binding to extravascular sites, but the AT exosite does not, likely due to being unavailable in the zymogen. Active site inhibited FIXa (GGACK) demonstrates an accelerated initial elimination phase relative to FIXa WT, suggesting re-distribution of the protein is conformation-dependent. The subsequent terminal elimination phase was slowed, consistent with an inability to be inhibited and cleared by AT Disclosures Sheehan: Pfizer: Research Funding; Bioverativ: Consultancy; Genetech: Consultancy.

Anthraquinones from Cassiae semen as thrombin inhibitors: in vitro and in silico studies

Thrombin inhibitor therapy is one of the most effective therapeutic strategies for the prevention and treatment of cardiovascular and thrombotic diseases. Although several marketed direct thrombin inhibitors (DTIs) have been widely used in clinic, the potentially serious complications of these DTIs prompted the researchers to find more DTIs with improved safety profiles. Herein, we report that natural anthraquinones in Cassiae semen (the seed of Cassia obtusifolia L. or C. tora L.), including obtusifolin, obtusin, aurantio-obtusin and chryso-obtusin, display strong to moderate inhibition on human thrombin, with the IC50 values ranging from 9.08 μM to 27.88 μM. Further investigation on the inhibition kinetics demonstrates that these anthraquinones are mixed inhibitors against thrombin-mediated Z-GGRAMC acetate hydrolysis, while obtusifolin and aurantio-obtusin show strong thrombin inhibition capacity, with the Ki values of 9.63 μM and 10.30 μM, respectively. Docking simulations demonstrate that both obtusifolin and aurantio-obtusin can simultaneously bind on the catalytic cavity and the two anion binding exosites (ABE1 and ABE2), while the hydroxyl group at the C-7 site and the methoxyl group at the C-8 site can create key interactions with the amino acids surrounding the catalytic cavity via hydrogen bonding. All these findings suggest that obtusifolin and aurantio-obtusin are strong thrombin inhibitors possessing a unique anthraquinone skeleton, and could be used as lead compounds for the development of new thrombin inhibitors with improved properties.

Natural constituents of St. John's Wort inhibit the proteolytic activity of human thrombin

Thrombin, a multifunctional serine protease responsible for the proteolytic hydrolysis of soluble fibrinogen, plays a pivotal role in the blood coagulation cascade. Currently, thrombin inhibitor therapy has been recognized as an effective therapeutic strategy for the prevention and treatment of thrombotic diseases. In this study, the inhibitory effects of natural constituents in St. John's Wort against human thrombin are carefully investigated by a fluorescence-based biochemical assay. The results clearly demonstrate that most of naphthodianthrones, flavonoids and biflavones exhibit strong to moderate inhibition on human thrombin. Among all tested compounds, hypericin shows the most potent inhibitory capability against thrombin, with the IC50 value of 3.00 μM. Further investigation on inhibition kinetics demonstrates that hypericin is a potent and reversible inhibitor against thrombin-mediated Z-GGRAMC acetate hydrolysis, with the Ki value of 2.58 μM. Inhibition kinetic analyses demonstrate that hypericin inhibits thrombin-mediated Z-GGRAMC acetate hydrolysis in a mixed manner, which agrees well with the results from docking simulations that hypericin can bind on both catalytic cavity and anion binding exosites. All these findings suggest that hypericin is a natural thrombin inhibitor with a unique dianthrone skeleton, which can be used as a good candidate to develop novel thrombin inhibitors with improved properties.

Molecular Dynamics Simulations Reveal a Possible Membrane‐Involved Microkinetic Mechanism for Long‐Lasting Action of Salmeterol, a β2‐Adrenergic Receptor Agonist

ObjectiveThe exact mechanisms underlying the longer duration of action and capacity to “reassert” airway smooth muscle relaxation in vitro despite repeated washing, of several long‐acting β2‐adrenergic receptor (β2‐AR) agonists such as salmeterol remain unresolved. Recent experimental studies suggest that the “microkinetic model” in which membrane acts as a depot enriching local drug concentration around the receptor adequately explains the long‐lasting effects. Specifically, the 700‐fold higher potency of salmeterol to β2‐AR in comparison to the structurally related agonists was found to be majorly due to its interactions with the membrane. The ‘reduction of dimensionality’ effect whereby ligands move on or within the membrane by 2D lateral diffusion could likely produce further enhancement of drug association rates and corresponding affinity values. These studies also suggested that the “exosite” described before probably reflects the phospholipid membrane rather than specific residues on the receptor. Most recently, a crystal structure of salmeterol bound to the active β2‐AR revealed an exosite for its aryloxyalkyl tail at the extracellular vestibule of the receptor. However, the association and dissociation pathways through which salmeterol access and leave the binding site is unknown. In addition, the possibility of lipophilic salmeterol reaching the binding site through transmembrane helices from the surrounding membrane environment is compelling and warrants thorough investigation, especially with the availability of the bound crystal structure. Here, we present an extensive computational study investigating the plausible lipid‐mediated association and dissociation pathways of salmeterol.MethodsThe most favorable bilayer location, orientation, and conformation of salmeterol were investigated using a 1‐palmitoyl‐2‐oleoyl‐sn‐glycero‐3‐phosphocholine (POPC) model membrane with 10% cholesterol by steered molecular dynamics (MD) and Umbrella Sampling simulations, using NAMD2.12 software. The dissociation and association pathways of salmeterol from the bound β2‐AR crystal structure (PDB ID 6CSY) were investigated by: 1) steered MD, 2) temperature accelerated MD, and 3) funnel metadynamics simulations. The protein, POPC lipids, and salmeterol were parameterized using the CHARMM36 and CGenFF force fields respectively.ResultsThe results thus far, reveal that the dissociation of salmeterol from its bound form proceeds through the extracellular side of the receptor, with a transient stay in the aqueous bulk before finally partitioning into the membrane (near the β2‐AR) in its energetically favorable bilayer depth (~15 Å from the bilayer center) determined by our partitioning studies. The specific interactions of the ligand with the receptor residues along the binding pathways and associated free energies were analyzed.ConclusionsComputational results suggest that the ‘microkinetic” mechanism attributed to the high membrane affinity of salmeterol may play a concerted role with its exosite binding towards the unique pharmacological properties.Support or Funding InformationNIH – R15GM131293This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

γ-Secretase Studied by Atomistic Molecular Dynamics Simulations: Global Dynamics, Enzyme Activation, Water Distribution and Lipid Binding.

γ-secretase, an intramembrane-cleaving aspartyl protease is involved in the cleavage of a large number of intramembrane proteins. The most prominent substrate is the amyloid precursor protein, whose proteolytic processing leads to the production of different amyloid Aβ peptides. These peptides are known to form toxic aggregates and may play a key role in Alzheimer's disease (AD). Recently, the three-dimensional structure of γ-secretase has been determined via Cryo-EM, elucidating the spatial geometry of this enzyme complex in different functional states. We have used molecular dynamics (MD) simulations to study the global dynamics and conformational transitions of γ-secretase, as well as the water and lipid distributions in and around the transmembrane domains in atomic detail. Simulations were performed on the full enzyme complex and on the membrane embedded parts alone. The simulations revealed global motions compatible with the experimental enzyme structures and indicated little dependence of the dynamics of the transmembrane domains on the soluble extracellular subunits. During the simulation on the membrane spanning part a transition between an inactive conformation (with catalytic residues far apart) toward a putatively active form (with catalytic residues in close proximity) has been observed. This conformational change is associated with a distinct rearrangement of transmembrane helices, a global compaction of the catalytically active presenilin subunit a change in the water structure near the active site and a rigidification of the protein fold. The observed conformational rearrangement allows the interpretation of the effect of several mutations on the activity of γ-secretase. A number of long-lived lipid binding sites could be identified on the membrane spanning surface of γ-secretase which may coincide with association regions of hydrophobic membrane helices to form putative substrate binding exosites.

Role of Protease Exosites for In Vivo Hemostatic Activity and Clearance of FIX

Introduction:FIX(a) binds rapidly and reversibly to vascular endothelium and extravascular matrix, in part mediated by the interaction of specific Gla domain residues with collagen IV. Both anticoagulant heparan sulphate and collagen IV localize predominantly to the basement membrane and supramolecular structure analysis suggests that heparan sulphate chains are integrated into collagen IV-containing networks. Previous work has demonstrated the contribution of the heparin and antithrombin (AT)-binding exosites in the protease domain to the regulation of human FIXa. Here we evaluate the contribution of these protease exosites to FIX hemostatic function and clearance in the hemophilia B (HB) mouse.Methods:Recombinant human FIX variants expressed in HEK293 cells with substitutions in the heparin- (K126A/K132A or R170A) or AT-binding (R150A) exosites (chymotrypsinogen #) were purified to homogeneity. In vivohemostasis was assessed in a saphenous vein bleeding model in HB mice. Anesthetized adult mice were injected with a weight-based dose predicted to achieve 30% FIX plasma concentrations and the exposed left saphenous vein was punctured with a 29 gauge needle. Clot formation at the injury site was repeatedly disrupted over 30 min to determine the average time to clot formation and total #clots/30 min. FIX pharmacokinetics (PK) in HB mice were determined with a weight-based dose injected into the tail vein. Serial retro-orbital blood samples were collected, plasma isolated and FIX content determined with a human FIX specific ELISA. FIX concentrations were plotted versus time and initially fit to the equation: [IX]t = [IX]max*(C1*e(-k1*t)+ C2*e(-k2*t)), where k1 and k2 are rate constants for the initial and terminal elimination phases, respectively. As the 4-parameter fit resulted in some large error estimates, the two phases were fit separately to estimate the respective parameters.Results:Previous characterization of APTT-based coagulant activities for these FIX variants (% WT ± SEM) were as follows: FIX WT 100 ± 1.4, R150A 73.6 ± 1.4, K126A/K132A 20.6 ± 9.2, and R170A 609.2 ± 37. In the saphenous vein bleeding assay, HB mice had no clot formation over 30 min, while wild type mice demonstrated 11 clots/30 min. When HB mice were injected with increasing concentrations of FIX WT (20%-100%), a dose dependent increase was observed from 4.3 to 16.5 clots/30 min (Table 1). When injected with sufficient FIX to achieve 30% plasma levels, the AT-binding exosite variant FIX R150A and heparin binding site variant K126A/K132A demonstrated similar hemostatic activity to FIX WT in the HB mouse. FIX R170A (30%) trended towards enhanced activity relative to WT, while 10% FIX R170A levels demonstrated similar activity to 30% FIX WT (not shown).The PK (5 min to 25 hr) of FIX was examined in HB and wild type mice. FIX WT demonstrated approximately ~30% recovery (initial peak) of predicted levels. FIX K126A/K132A had markedly enhanced recovery (2.2-fold higher than FIX WT), while FIX R150A and R170A demonstrated similar recovery to FIX WT (Table 1, Figure 1). FIX K126A/K132A demonstrated ~70% of predicted recovery. A rapid decrease in FIX levels was observed over the ensuing 4-6 hours with similar k1 values for all proteins. In contrast, terminal phase k2 values trended lower and predicted half-lives longer for the heparin-binding exosite mutants (FIX K126A/K132A and R170A). Consistent with occupation of extravascular sites by endogenous FIX, PK studies in wild type mice demonstrated ~20% enhancement of FIX recovery relative to HB mice (not shown).Conclusions:In contrast to in vitrocoagulant and thrombin generation activity, FIX K126A/K132A demonstrated similar in vivohemostatic activity to FIX WT, markedly enhanced recovery and modestly slower terminal clearance. The enhanced in vivo hemostatic activity of this variant relative to coagulant activity likely results from the markedly enhanced plasma FIX recovery. In contrast, FIX R150 demonstrates similar in vivo hemostatic activity and zymogen clearance to FIX WT. FIX R170A demonstrated ~3-fold enhanced hemostatic potency and similar zymogen clearance compared to FIX WT. This hemostatic potency is consistent with the enhanced coagulant activity of this FIX Padua (R170L) related variant. In addition to the collagen IV binding site, the FIX(a) heparin binding exosite is likely to play a substantial role in FIX localization to the extravascular compartment. [Display omitted] DisclosuresSheehan:Pfizer: Research Funding; Bayer: Consultancy, Research Funding.

Anti-CUB1 or Anti-Spacer Antibodies That Increase ADAMTS13 Activity Act By Allosterically Enhancing Metalloprotease Domain Function

BackgroundADAMTS13 circulates in a folded conformation, which is mediated by interactions between the C-terminal CUB domains and its central Spacer domain. Binding of ADAMTS13 to the VWF D4-CK domains disrupts the CUB-Spacer interaction, inducing a structural change that extends ADAMTS13 into an open conformation that enhances catalytic efficiency ~2-fold. This mechanism supports a model in which ADAMTS13 unfolding induces exposure of an exosite in the Spacer domain that interacts with the VWF A2 domain, increasing the affinity between the two molecules, and, therefore, the rate of proteolysis. The D4-CK-mediated conformational activation of ADAMTS13 can be mimicked in vitro with the use of antibodies that disrupt the CUB-Spacer interaction, such as the previously published anti-CUB antibody, Ab17G2. We recently generated a novel, activating antibody against the Spacer domain (Ab3E4).AimTo characterize the mechanism by which the Ab17G2 and Ab3E4 enhance the catalytic efficiency of ADAMTS13.MethodsThe effects of the Ab17G2 and Ab3E4 on the activity of ADAMTS13 were studied using FRETS-VWF73. The effects of the Ab17G2 and Ab3E4 on the kinetics of VWF96 (VWF G1573-R1668) proteolysis were characterized using an in-house assay. ELISA was used to investigate conformational changes in ADAMTS13 induced by the Ab17G2 and Ab3E4.ResultsBoth Ab17G2 and Ab3E4 enhanced FRETS-VWF73 proteolysis by ~1.7-fold. This result was reproduced using the VWF96 substrate; the Ab17G2 and Ab3E4 enhanced the catalytic efficiency (kcat/Km) of ADAMTS13 by ~1.8- and ~2.0-fold, respectively. The activation was dependent on the conformational extension of ADAMTS13, since the antibodies could not enhance the activity of an ADAMTS13 variant that lacks the TSP2-CUB2 domains (MDTCS). Surprisingly, ADAMTS13 activation was not mediated through exposure of the Spacer or Cys-rich domain exosites as previously proposed, as the Ab17G2 and Ab3E4 efficiently enhanced proteolysis of VWF96 variants in which the Spacer/Cys-rich exosite binding sites were disrupted. Kinetic analysis of VWF96 proteolysis showed that the Ab17G2- and Ab3E4-induced activation of ADAMTS13 is primarily manifest through a ~1.5- to ~2-fold increase in enzyme turnover (kcat). Thus, contrary to the current model, this suggests that the conformational extension of ADAMTS13 influences the functionality of the active site, and not substrate binding affinity (Km). Incubating ADAMTS13 with either Ab17G2 or Ab3E4 exposed a cryptic epitope in the metalloprotease domain that was specifically detected by ELISA, further corroborating that the antibodies induce a conformational change in ADAMTS13 affecting the M domain.ConclusionAntibodies can be used as tools for understanding the structure/function of enzymes. Using activating antibodies against the Spacer and CUB1 domains of ADAMTS13, we show for the first time that the activation of ADAMTS13 following its unfolding is not a result of exposure of a functional exosite in Spacer/Cys-rich domain that increases affinity to VWF. Rather, our data are consistent with an allosteric activation mechanism upon the metalloprotease domain. We propose that ADAMTS13 unfolding causes a conformational change in the active site that further activates the enzyme. We are currently investigating whether the D4-CK-induced enhancement of ADAMTS13 proteolytic activity is also mediated by conformational changes in the active site. DisclosuresVanhoorelbeke:Ablynx: Consultancy; Shire: Consultancy.

Monoclonal antibodies against metzincin targets.

The metzincin clan of metalloproteinases includes the MMP, disintegrin and metalloproteinase (ADAM) and ADAM with thrombospondin motifs families, which cleave extracellular targets in a wide range of (patho)physiological processes. Antibodies constitute a powerful tool to modulate the activity of these enzymes for both therapeutic and research purposes. In this review, we give an overview of monoclonal antibodies (mAbs) that have been tested in preclinical disease models, human trials and important studies of metzincin structure and function. Initial attempts to develop therapeutic small molecule inhibitors against MMPs were hampered by structural similarities between metzincin active sites and, consequently, off-target effects. Therefore, more recently, mAbs have been developed that do not bind to the active site but bind to surface-exposed loops that are poorly conserved in closely related family members. Inhibition of protease activity by these mAbs occurs through a variety of mechanisms, including (i) barring access to the active site, (ii) disruption of exosite binding, and (iii) prevention of protease activation. These different modes of inhibition are discussed in the context of the antibodies' potency, selectivity and, importantly, the effects in models of disease and clinical trials. In addition, various innovative strategies that were used to generate anti-metzincin mAbs are discussed. LINKED ARTICLES: This article is part of a themed section on Translating the Matrix. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v176.1/issuetoc.