Prekallikrein - an emerging therapeutic target for Klebsiella pneumoniae infection?†.

Klebsiella pneumoniae is a Gram-negative bacterium that is increasingly difficult to treat due to the emergence of multidrug resistant strains. In a recent article, Ding et al demonstrate that prekallikrein depletion in mice followed by intranasal instillation of K. pneumoniae leads to a reduced bacterial burden and prolonged host survival, together with evidence of reduced distant organ damage. These effects are apparently independent of the role of prekallikrein in the contact system, and are associated with transcriptional changes relevant to innate immunity in the lung, established prior to infection. This study highlights the importance of further investigating the role of prekallikrein and other contact cascade components in host defence to counter K. pneumoniae (and perhaps other pathogens), with an overall aim of identifying potential therapeutic targets relevant to pulmonary infection with such resistant pathogens. © 2020 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Enzymes produced by autoactivation of blood factor XII in buffer: A contribution from the Hematology at Biomaterial Interfaces Research Group.

High-resolution electrophoresis of FXII-derived proteins produced by contact activation of FXII in buffer solutions (i.e. in absence of plasma proteins) with hydrophilic and silanized-glass activators spanning the observable range of water wettability (hydrophilic to hydrophobic), shows no evidence of proteolytic cleavage of FXII into αFXIIa or ßFXIIa. The autoactivation mixture contains only a single-chain protein with a molecular weight of ∼80 kDa, confirming Oscar Ratnoff's previous finding of a single-chain activated form of FXII that he called 'HFea'. Functional assays have shown that these autoactivation products exhibit procoagulant potential (protease activity inducing clotting of blood) or amidolytic potential (cleaves amino bonds in s-2302 chromogen but do not cause coagulation of plasma) or both amidolytic potential and procoagulant potential. Some of these proteins also have the remarkable potential to 'suppress autoactivation' (i.e. suppress creation of enzymes with procoagulant potential). It is thus hypothesized that autoactivation of FXII in the absence of plasma proteins generates not just a single type of activated conformer, as suggested by previous researchers, but rather an ensemble of conformer products with collective activity that varies with activator surface energy used in contact activation of FXII. Furthermore, reaction of αFXIIa with FXII in buffer solution does not produce additional αFXIIa by the putative autoamplification reaction FXIIa + FXII â†’ 2FXIIa as has been proposed in past literature to account for the discrepancy between chromogenic and plasma-coagulation assays for αFXIIa in buffer solution. Instead, net procoagulant activity measured directly by plasma-coagulation assays, decreases systematically with increasing FXII solution concentration. Under the same reaction conditions, chromogenic assay reveals that net amidolytic activity increases with increasing FXII solution concentration. Thus, although autoamplification does not occur it appears that there is some form of "FXII self reaction" that influences products of αFXIIa reaction with FXII. Electrophoretic measurements indicate that no proteolytic cleavage takes in this reaction leading us to conclude that change in activity is most likely due to change(s) in FXII conformation (with related change in enzyme activity).

The relative priority of prekallikrein and factors XI/XIa assembly on cultured endothelial cells.

Investigations determined the relative preference of prekallikrein (PK) or factor XI/XIa (FXI/FXIa) binding to endothelial cells (HUVECs). In microtiter plates, biotinylated high molecular weight kininogen (biotin-HK) or biotin-FXI binding to HUVEC monolayers or their matrix proteins, but not fibronectin-coated plastic microtiter plate wells, was specifically blocked by antibodies to each of the receptors of HK, uPAR, gC1qR, or cytokeratin 1. Fluorescein isothiocyanate (FITC)-PK specifically bound to HUVEC suspensions without added Zn2+, whereas FITC-FXI or -FXIa binding to HUVEC suspensions required 10 microM added Zn2+ to support specific binding. Plasma concentrations of FXI did not block FITC-PK binding to HUVECs in the absence or presence of 10 microM Zn2+. In the absence of HK, the level of FITC-FXI or -FXIa binding was half that seen in its presence. At physiologic concentrations, PK (450 nM) abolished FITC-FXI or -FXIa binding to HUVEC suspensions in the absence or presence of HK in the presence of 10 microM Zn2+. Released Zn2+ from 2-8 x 10(8) collagen-activated platelets/ml supported biotin-FXI binding to HUVEC monolayers, but platelet activation was not necessary to support biotin-PK binding to HUVECs. At physiologic concentrations, PK also abolished FXI binding to HUVECs in the presence of activated platelets, but FXI did not influence PK binding. PK in the presence or absence of HK preferentially bound to HUVECs over FXI or FXIa. Elevated Zn2+ concentrations are required for FXI but not PK binding, but the presence of physiologic concentrations of PK and HK also prevented FXI binding. PK preferential binding to endothelial cells contributes to their anticoagulant nature.

Factor XI, but not prekallikrein, blocks high molecular weight kininogen binding to human umbilical vein endothelial cells.

Previous studies on the interaction of high molecular weight kininogen (HK) with endothelial cells have reported a large number of binding sites (106-107 sites/cell) with differing relative affinities (KD = 7-130 nm) and have implicated various receptors or receptor complexes. In this study, we examined the binding of HK to human umbilical vein endothelial cells (HUVEC) with a novel assay system utilizing HUVEC immobilized on microcarrier beads, which eliminates the detection of the high affinity binding sites found nonspecifically in conventional microtiter well assays. We report that HK binds to 8.5 x 104 high affinity (KD = 21 nm) sites per HUVEC, i.e. 10-100-fold fewer than previously reported. Although HK binding is unaffected by the presence of a physiological concentration of prekallikrein, factor XI abrogates HK binding to HUVEC in a concentration-dependent manner. Disruption of the naturally occurring complex between factor XI and HK by the addition of a 31-amino acid peptide mimicking the factor XI-binding site on HK restored HK binding to HUVEC. Furthermore, HK inhibited thrombin-stimulated von Willebrand factor release by HUVEC but not thrombin receptor activation peptide (SFLLRN-amide)-stimulated von Willebrand factor release. Factor XI restored the ability of thrombin to stimulate von Willebrand factor release in the presence of low HK concentrations. These results suggest that free HK, or HK in complex with prekallikrein but not in complex with factor XI, interacts with the endothelium and can maintain endothelial cell quiescence by preventing endothelial stimulation by thrombin.

Kallikreins when activating bradykinin B2 receptor induce its redistribution on plasma membrane.

The bradykinin (BK) B2 receptor (R) is directly activated by kallikreins and other serine proteases independent of BK release. Both the Galpha(i) and Galpha(q) proteins are involved, shown by the release of arachidonic acid and [Ca2+]i elevation. Site-directed mutagenesis of the receptor and the lack of heterogeneous desensitization of the human B2R by the BK and kallikrein emphasize among others the differences between activation by the proteases and the peptide. To characterize further the mechanism thereby kallikreins activate and desensitize the B2R we investigated the distribution of the human B2R tagged with the green fluorescent protein (B2-GFP(Ct)) on the plasma membrane of stably transfected Chinese hamster ovary (CHO) cells. We visualized the movement of B2-GFP(Ct) R with confocal fluorescence microscopy after activation by BK or a by serine protease. Continued exposure of the cells to BK led to B2R internalization within 15-20 min. Porcine pancreatic and human recombinant tissue kallikreins induced a rapid definite redistribution of receptors on the plasma membrane within 5 min, prior to internalization. These effects of kallikrein were blocked by the B2R antagonist HOE 140 and by the kallikrein inhibitor, aprotinin. The B2R was also activated by endoproteinases LysC and ArgC and trypsin, but these enzymes did not induce redistribution, only internalization. In control experiments, kallikrein had no effect on cells transfected to stably express the angiotensin-converting enzyme-green fluorescent protein (GFP). Thus, kallikreins when activating the BK B2R also trigger its redistribution on plasma membrane.

Activation of the kinin-forming cascade on the surface of endothelial cells.

Activation of the plasma kallikrein-kinin forming cascade takes place upon incubation with human umbilical vein endothelial cells. The mechanism by which initiation occurs is uncertain. Zinc-dependent binding of plasma proteins to gC1qR, cytokeratin 1, and perhaps u-PAR is requisite for activation to take place. We demonstrate here that during a 2 hour incubation time plasma deficient in either factor XII or high molecular weight kininogen (HK) fails to activate, as compared to normal plasma, but with more prolonged incubation, factor XII-deficient plasma gradually activates while HK-deficient plasma does not. Our data support both factor XII-dependent (rapid) and factor XII-independent (slow) mechanisms; the latter may require a cell-derived protease to activate prekallikrein and the presence of zinc ions and HK.

Factor XII-dependent contact activation on endothelial cells and binding proteins gC1qR and cytokeratin 1.

Although proteins of the kinin-forming pathway are bound along the surface of endothelial cells, the mechanism of activation of this proteolytic cascade is unclear. Endothelial cell surface proteins, gC1qR and cytokeratin 1, are capable of binding Factor XII and high molecular weight kininogen (HK) in a zinc-dependent reaction thus we considered the possibility that these proteins might catalyze initiation of the cascade. Incubation of Factor XII, prekallikrein, and HK with gC1qR or cytokeratin 1 leads to a zinc-dependent and Factor XII-dependent conversion of prekallikrein to kallikrein. We also demonstrate that normal plasma is capable of activating upon interaction with the cells whereas plasma deficient in Factor XII, prekallikrein and HK do not activate. Normal plasma activation was inhibitable by antibody to gC1qR and cytokeratin 1. Thus, gC1qR and cytokeratin 1, represent potential initiating surfaces for activation of the plasma kinin-forming cascade and may do so as a result of their expression along cell surfaces.

Human bradykinin B(2) receptor is activated by kallikrein and other serine proteases.

Bradykinin (BK) and kallidin (Lys-BK), liberated from kininogens by kallikreins, are ligands of the BK B(2) receptor. We investigated whether kallikreins, besides releasing peptide agonist, could also activate the receptor directly. We studied the effect of porcine and human recombinant tissue kallikrein and plasma kallikrein on [Ca(2+)](i) mobilization and [(3)H]arachidonic acid release from cultured cells stably transfected to express human BK B(2) receptor (CHO/B(2), MDCK/B(2), HEK/B(2)), and endothelial cells were used as control cells. As with BK, the actions of kallikrein were blocked by the B(2) antagonist, HOE 140. Kallikrein was inactive on cells lacking B(2) receptor. Kallikrein and BK desensitized the receptor homologously but there was no cross-desensitization. Furthermore, 50 nM human cathepsin G and 50 nM trypsin also activated the receptor; this also was blocked by HOE 140. Experiments excluded a putative kinin release by proteases. [(3)H]AA release by BK was reduced by 40% by added kininase I (carboxypeptidase M); however, receptor activation by tissue kallikrein, trypsin, or cathepsin G was not affected. Prokallikrein and inhibited kallikrein were inactive, suggesting cleavage of a peptide bond in the receptor. Kallikreins were active on mutated B(2) receptor missing the 19 N-terminal amino acids, suggesting a type of activation different from that of thrombin receptor. Paradoxically, tissue kallikreins decreased the [(3)H]BK binding to the receptor with a low K(D) (3 nM) and inhibited it 78%. Thus, kallikreins and some other proteases activate human BK B(2) receptor directly, independent of BK release. The BK B(2) receptor may belong to a new group of serine protease-activated receptors.

Platelet-bound prekallikrein promotes pro-urokinase-induced clot lysis: a mechanism for targeting the factor XII dependent intrinsic pathway of fibrinolysis.

Clots formed from platelet rich plasma were found to be lysed more readily by low concentrations of pro-urokinase (pro-UK) than clots formed from platelet poor plasma. This was not a non-specific effect since the reverse occurred with tissue plasminogen activator. A mechanical explanation due to platelet-mediated clot retraction was excluded by experiments in which retraction was inhibited with cytochalasin B. Therefore, a platelet-mediated enzymatic mechanism was postulated to explain the promotion of fibrinolysis. Casein autography of isolated platelets revealed a approximately 90 kDa band of activity which comigrated with plasma prekallikrein (PK)/kallikrein, a known activator of pro-UK. Furthermore, treatment of platelets with plasma PK activator (PPA), consisting essentially of factor XIIa, induced activation of pro-UK and of chromogenic substrate for kallikrein (S-2302). This activity corresponded to approximately 40-200 pM kallikrein per 10(8) washed and gel filtered platelets per ml. The activation of pro-UK by PPA-pretreated platelets was dose-dependent and inhibited by soybean trypsin inhibitor but not by bdellin, a specific inhibitor of plasmin, nor by the corn inhibitor of factor XIIa. Kinetic analysis of pro-UK activation by kallikrein showed promotion of the reaction by platelets. The KM of the reaction was reduced by platelets by approximately 7-fold, while the kcat was essentially unchanged. In conclusion, PK was shown to be tightly associated with platelets where it can be activated by factor XIIa during clotting. The activation of pro-UK by platelet-bound kallikrein provides an explanation for the observed platelet mediated promotion of pro-UK-induced clot lysis.(ABSTRACT TRUNCATED AT 250 WORDS)

Structure, kinetics, and function of human and rhesus plasma prekallikreins are similar.

To determine if rhesus monkeys (Macaca mulatta) could serve as a model for studying the role of the contact system in the pathophysiology of human infections, we compared structural, kinetic, and functional characteristics of plasma prekallikrein and its activation products in rhesus and humans. Three prekallikrein variants (85-, 89- and 93-kDa) were revealed in rhesus plasma as compared with the two variants (85- and 88-kDa) in human plasma by immunoblotting with the monoclonal antibody MAb 13G11. The prekallikrein concentration in rhesus plasma was 1.5-fold that in human plasma as determined by computerized immunoblot analyses (CIBA) and amidolytic activity. The electrophoretic mobility of prekallikrein from plasma of both species increased after deglycosylation. Inhibition of prekallikrein activation by MAb 13G11 was 55% (rhesus plasma) and 76% (human plasma), with similar inhibition curves. Immunoblots of activated rhesus plasma showed prekallikrein, complexes of kallikrein with C1 inhibitor, alpha 2-macroglobulin and approximately 60-kDa inhibitor(s) (viz. antithrombin III), and 45-kDa fragments, like those in activated human plasma. Concentrations and molecular masses of factor XII and high molecular weight kininogen were similar in rhesus and human plasma. The activated partial thromboplastin time (APTT) and prothrombin time were 20.1 +/- 1.6 and 9.7 +/- 0.3 s for rhesus and 32.0 +/- 5.6 and 12 +/- 0.5 s for human plasma. Human and rhesus APTTs were similar when prekallikrein concentrations in human and rhesus plasma became alike by adding human purified prekallikrein.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of heparin on the activation of factor XII and the contact system in plasma.

We examined in purified systems and in human plasma whether heparin serves as a contact system activating compound. Purified human factor XII zymogen was not activated by heparin through an autoactivation mechanism, but was activated in the presence of purified prekallikrein. Zn2+ (12 microM) did not support autoactivation by heparin. The activation of factor XII and the contact system by heparin in plasma anticoagulated with citrate or with hirudin (not chelating ions) was examined by the cleavage of 125I-labeled factor XII and high molecular weight kininogen (HK). Heparin at 1.6 and 16 USP U/ml was not able to produce activation, in contrast to dextran sulfate (20 micrograms/ml) which supported activation of both factor XII and HK. This study indicates that heparinized plasma does not support activation of the contact system mediated through activation of factor XII. It is not expected that heparin anticoagulant therapy will contribute to activation of the contact system.

Molecular interaction of bovine kininogen and its derivatives with papain.

The molecular interaction of bovine kininogen and its derivatives with papain was investigated. High-molecular-weight kininogen (HMWK) or low-molecular-weight kininogen (LMWK) and inactive papain treated with N-[N-(L-3-trans-carboxyoxiran-2-carbonyl)-L-leucyl]agmatine (E-64) formed, respectively, a complex, which was dissociable on sodium dodecyl sulfate polyacrylamide-gel electrophoresis (SDS-PAGE). The densitometric determination of the bands separated on SDS-PAGE and amino acid analysis of the samples extracted from the electrophoresis gel revealed that the complex between kininogen and papain is formed in a molar ratio of one to one. Moreover, analysis of the inhibition of the caseinolytic activity of papain by these kininogens indicated that HMWK, LMWK, and kinin-free derivatives obtained from both kininogens inhibit active papain with a stoichiometry of 1:1. On the other hand, the papain activity was inhibited by two kinds of cyanogen bromide fragments isolated from the heavy chain of HMWK. These two fragments with Ki values of 38 and 0.64 nM corresponded, respectively, to residue Nos. 47 to 243 and Nos. 244-360 of the HMWK heavy chain. These results suggest that in the intact HMWK and LMWK, one of the two potential reactive sites interacts with papain to form a complex and that the other reactive site becomes active only after separation of the two sites.

Surface activation of factor XII (Hageman factor)--critical role of high molecular weight kininogen and another potentiator).

When factor XII was adsorbed to kaolin it slowly became activated and converted prekallikrein to kallikrein. In the presence of HMW-kininogen the rate of activation of factor XII and consequently that of prekallikrein was markedly enhanced. The enhancing effect of HMW-kininogen was a dose-dependent phenomenon. In order to enhance the activation of factor XII on a surface the HMW-kininogen molecule had to be intact. Cleavage of HMW-kininogen by kallikrein decreased the enhancing effect of HMW-kininogen, there being an inverse relation between the bradykinin-generated and the capacity to enhance factor XII activation. Another 'potentiator' of factor XII activation was isolated from proteins adsorbed to aluminum hydroxide. This potentiator further increased the activation of factor XII, also in a dose-dependent fashion. It was postulated that factor XII is slowly converted into its active form by exposure to negatively charged surfaces; that this process is enhanced by kallikrein and further accelerated by HMW-kininogen and the 'potentiator'; and that these enhancing substances probably act by opening active sites on the factor XII molecule.

Activation and function of human Hageman factor. The role of high molecular weight kininogen and prekallikrein.

The activation and function of surface-bound Hageman factor in human plasma are dependent upon both high molecular weight (HMW) kininogen and prekallikrein. HMW kininogen does not affect the binding of Hageman factor to surfaces, but it enhances the function of surface-bound Hageman factor as assessed by its ability to activate prekallikrein and Factor XI. The initial conversion of prekallikrein to kallikrein by the surface-bound Hageman factor in the presence of HMW kininogen is followed by a rapid enzymatic activation of Hageman factor by kallikrein. The latter interaction is also facilitated by HMW kininogen. Kallikrein therefore functions as an activator of Hageman factor by a positive feedback mechanism and generates most of the activated Hageman factor during brief exposure of plasma to activating surfaces. HMW kininogen is a cofactor in the enzymatic activation of Hageman factor by kallikrein and it also augments the function of the activated Hageman factor generated. The stoichiometry of the Hagman factor interaction with HMW kininogen suggests that it enhances the activity of the active site of Hageman factor. Since HMW kininogen and prekallikrein circulate as a complex, HMW kininogen may also place the prekallikrein in an optimal position for its reciprocal interaction with Hageman factor to proceed. The surface appears to play a passive role upon which bound Hageman factor and the prekallikrein-HMW kininogen complex can interact.

The relationship of structure and function in human Hageman factor. The association of enzymatic and binding activities with separate regions of the molecule.

Three regions of the human Hageman factor molecule termed the c, d, and e regions have been defined. Division of the molecule into these three regions is based on the analysis of fragments obtained by enzymatic cleavage during fluid-phase activation. The three regions have the following properties: (a) the c region has a mol wt of 40,000, has the capacity to bind to negatively charged surfaces, and does not have detectable enzymatic activity; (b) the e region possess a mol wt of 28,000 has enzymatic activity, and does not bind to negatively charged surfaces; (c) the d region has a mol wt of 12,000, is located between the c and e fragments but has not been detected as a freely existing polypeptide, and can bind firmly to negatively charged surfaces. The preparation of antibodies specific for the c and e regions is described as well as their use in defining the electrophoretic characteristics of the cde, cd, de, c, and e polypeptide fragments of Hageman factor. Evidence is given showing that the e region, but not the c or d, is released from a negatively charged surface when bound Hageman factor is exposed to proteolytic enzymes or whole plasma and that when this occurs in the presence of normal plasma, the e fragment becomes bound to C1 esterase inhibitor.