Pitfalls in characterization of protein interactions using radioiodinated crosslinking reagents. Preparation and testing of a novel photochemical 125I-label transfer reagent.

Much attention has been focused on the study of protein interactions with radioiodinated photo-crosslinking reagents, and pitfalls in using this methodology are discussed. A new photochemical and cleavable heterobifunctional crosslinking reagent, succinimidyl N-14-(2-hydroxybenzoyl)-N-11-(4-azidobenzoyl)-9-oxo-8,11,14-triaza -4,5- dithiatetradecanoate (SHAD) was prepared, and its potential as a label transfer reagent was tested in model systems. SHAD was radioiodinated, and the labeled reagent (125I-SHAD) was converted to an amide (125I-HADM, as a mimicry of conjugation to protein 1) and photolyzed. When compared to the widely used SASD reagent (sulfosuccinimidyl 2-[[(4-azidosalicyl)-amino]ethyl]-1,3- dithiopropionate, Pierce), SHAD has a number of decisive advantages. The amide of 125I-SASD (125I-ASDM) was generated and photolyzed, and it was found that at least 50% of the radioactivity is released from 125I-ASDM after 3 min of irradiation, whereas only approximately 10% is liberated from 125I-HADM under similar conditions. Furthermore, 125I-HADM was photolyzed in the presence of excess amine (mimicry of crosslinking to protein 2), and the product was cleaved by reduction (mimicry of label transfer). The transformations in the course of photolysis were monitored by UV spectroscopy and TLC analysis, and a high degree of reagent cleavage upon reduction was demonstrated. 125I-SHAD was used to crosslink Lys78-plasminogen and fibrin. 125I-SHAD was conjugated to Lys78-plasminogen in the dark. Fibrinogen and thrombin were added, and Lys78-plasminogen was crosslinked to the fibrin clot by exposure to light.(ABSTRACT TRUNCATED AT 250 WORDS)

Fibrin structures during tissue-type plasminogen activator-mediated fibrinolysis studied by laser light scattering: relation to fibrin enhancement of plasminogen activation.

The aim was to relate fibrin structure and the stimulatory effect of fibrin on plasminogen activation during t-PA-mediated fibrinolysis using Lys78-plasminogen as activator substrate. Structural studies were undertaken by static and dynamic laser light scattering, cryo transmission electron microscopy and by the measurement of conversion of fibrin to X-, Y- and D-fragments. The kinetics of plasmin formation were monitored by measurement of the rate of pNA-release from Val-Leu-Lys-pNA. The process of fibrin formation and degradation comprised three phases. In the first phase, protofibrils with an average length of about 10 times that of fibrinogen were formed. The duration of this phase decreased with increasing t-PA concentration. The second phase was characterized by a sudden elongation and lateral aggregation of fibrin fibers, most pronounced at low levels of t-PA, and by formation of fragment X-polymer. The third phase was dominated by fragmentation of fibers and by formation of Y- and D-fragments. Plasmin degraded the fibers from within, resulting in the formation of long loose bundles, which subsequently disintegrated into thin filaments with a length of less than 10 and a mass per length close to one relative to fibrinogen. Plasmin generation at high t-PA concentrations sets in just prior to (and at low t-PA concentrations shortly after) the onset of the rapid second phase of elongation and lateral aggregation of fibrin fibers. The maximal rate of plasmin formation per mol t-PA was the same at all concentrations of activator and was achieved close to the time of the peak level of fragment X-polymer. Plasmin formation ceased after formation of substantial amounts of Y- and D-fragments. At this stage the length was between 300 and 3 and the mass per length close to 1, both relative to fibrinogen. In conclusion our results indicate that (1) formation of short fibrin protofibrils is the minimal requirement for the onset of the stimulatory effect of fibrin on plasminogen activation by t-PA, (2) formation of fragment X protofibrils is sufficient to induce optimal stimulation of plasminogen activation, and (3) plasmin degrades laterally aggregated fibrin fibers from within, resulting in the conversion of the fibers into long loose bundles, which later disintegrate into thin filaments.

Interactions of plasminogen with polymerizing fibrin and its derivatives, monitored with a photoaffinity cross-linker and electron microscopy.

Localization of the plasminogen binding sites on fibrin has been difficult since these interactions occur on polymerizing fibrin, and studies with fragments can be misleading because of multiple carboxyl-terminal lysines that may bind to plasminogen. A hetero-functional photoaffinity cross-linker was used to study these interactions. Following attachment of the cross-linker to plasminogen in the dark, a clot was formed by addition of fibrinogen or fragment X and thrombin, and then the plasminogen was cross-linked to adjacent parts of fibrin by exposure to light. There was more Glu1-plasminogen bound to fibrin than to fibrinogen and more to fragment X polymer than to fibrin. Electron microscopy of rotary shadowed individual molecules reveals that Glu1-plasminogen appears to be more compact than Lys78-plasminogen or Glu1-plasminogen with 6-aminohexanoic acid. Cross-linked complexes from the dissolved clot observed by electron microscopy reveal plasminogen bound to the end of fibrin or bridging the ends of two fibrin molecules; larger complexes were also observed. Analysis of changes in the appearance of negatively contrasted fibers with plasminogen bound also indicates the probable locations of binding sites, yielding results consistent with the cross-linking studies. The photoaffinity probe was also used to study interactions between plasminogen and fibrin or its derivatives in the course of tissue plasminogen activator-mediated fibrinolysis. Samples cross-linked at various times indicate that complexes with fragment X are particularly dominant during the rapid phase of plasminogen activation. In conclusion, these studies indicate that plasminogen binds to the pocket at the end-to-end junction between two fibrin or fragment X molecules in the protofibril; from this position, it can reach all of the sites that are cleaved during fibrinolysis.

The role of fragment X polymers in the fibrin enhancement of tissue plasminogen activator-catalyzed plasmin formation.

When thrombin-mediated fibrin formation and tissue plasminogen activator (t-PA)-mediated fibrinolysis proceed in dynamic interaction, desA-(desB beta 1-42)-fragment X polymers are shown to be the predominant fibrin derivatives present during the rapid second phase of Glu1- and Lys78-plasminogen activation. To further investigate the effect of this intermediate, a method was developed for the production and purification of fibrinogen-derived desA-(desB beta 1-42)-fragment X, deprived of both COOH-terminal A alpha-chains, but still capable of thrombin-mediated polymerization. DesA-(desB beta 1-42)-fragment X polymer was compared to intact fibrin with regard to its stimulatory effect on Glu1-, Lys78-, and Val443-plasminogen activation, and its binding of Glu1- and Lys78-plasminogen. Pure fragment X polymer gave rise to a biphasic activation pattern like that of fibrin, demonstrating similar kinetics of rapid phase activation. The dissociation constant for the binding of plasminogen to the effector decreases by a factor of 14, and the stoichiometry increases by a factor of 2 upon plasmin-catalyzed cleavage of both native Glu1- to Lys78-plasminogen, and fibrin to fragment X polymer. We conclude that desA-fibrin protofibril formation is sufficient to initiate fibrin enhancement of t-PA-catalyzed plasminogen activation, and that optimal stimulation depends on further plasmin-mediated modification of the fibrin effector to desA-fragment X-related moieties. Optimal stimulation is dependent on the presence of the kringle 1-4 domains of plasminogen and probably results from altered and increased binding of both plasminogen and t-PA to the modified effector.

The oxidative cleavability of protein cross-linking reagents containing organoselenium bridges.

The intensive use of cleavable cross-linking reagents to study macromolecular biological interactions has shown a demand for optimizing these reagents in such a way that the involved macromolecules remain intact. The present work focuses on the development of selenium linkers that are cleavable by mild oxidation. The efficiency of cross-linking and subsequent cross-linker cleavage with a new series of such homo- or heterobifunctional cross-linking reagents have been tested in a simple model system, consisting of albumin and cytochrome c. Resultant, or residual, covalent complex formation is examined by SDS-polyacrylamide gel electrophoresis. From this work it can be concluded that diallyl selenides are readily cleaved by mild oxidation, whereas dialkyl selenides and benzyl alkyl selenides can only be cleaved when the alkyl part of the selenide has an electron-withdrawing group next to the beta-carbon from selenium.

The course and prerequisites of Lys-plasminogen formation during fibrinolysis.

Plasmin-catalyzed modification of the native plasma zymogen Glu1-plasminogen to its more reactive Lys78 form has been shown to be enhanced in the presence of fibrin. The aim of the present work has been to characterize the influence of fibrinopeptide release, fibrin polymerization, and plasmin cleavage of fibrin on the rate of Lys78-plasminogen formation. 125I-Labeled Glu1- to Lys78-plasminogen conversion was catalyzed by performed Lys78-plasmin, or by plasmin generated during plasminogen activation with tissue plasminogen activator or urokinase. The two forms of plasminogen were quantitated following separation by polyacrylamide gel electrophoresis in acetic acid/urea. Plasmin generated by plasminogen activator was monitored by a fixed-time amidolytic assay. The rate of Lys78-plasminogen formation was correlated, in separate experiments, to the simultaneous, plasmin-catalyzed cleavage of 125I-labeled fibrinogen or fibrin to fragments X, Y, and D. The radiolabeled components were quantitated after separation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The results show that the formation of both bathroxobin-catalyzed des-A-fibrin and thrombin-catalyzed des-AB-fibrin leads to marked stimulation of Lys78-plasminogen formation, whereas inhibition of fibrin polymerization, with Gly-Pro-Arg-Pro, abolishes the stimulatory effect. The rate of Lys78-plasminogen formation varies markedly in the course of fibrinolysis. The apparent second-order rate constant of the reaction undergoes a transient increase upon transformation of fibrin to des-A(B) fragment X polymer and decreases about 10-fold to the level observed during fibrinogenolysis upon further degradation to soluble fragments Y and D.(ABSTRACT TRUNCATED AT 250 WORDS)

Protein crosslinking reagents containing a selenoethylene linker are cleaved by mild oxidation.

A homobifunctional cleavable crosslinking reagent containing a selenoethylene group in the linker, and related reagents, have been synthesized and tested in a model system involving formation of a complex between albumin and cytochrome c. Functionally, complex formation was suggested by albumin inhibition of the ascorbate reduction of cytochrome c. Structurally, complex formation was demonstrated by crosslinking and subsequent separation of crosslinked complex from non-crosslinked proteins by SDS-polyacrylamide gel electrophoresis. The crosslinks were found to be cleavable by mild oxidation with low concentrations of periodate or with N-chlorobenzenesulfonamide immobilized on polystyrene beads (Iodo-Beads).

Trinitrobenzoylated poly(D-lysine) as a stimulator of interactions between plasminogen, plasmin, and tissue-type plasminogen activator.

Trinitrobenzyl alkylation of poly(D-lysine) provides a novel powerful stimulator of tissue-type plasminogen activator. Its stimulatory effect on plasminogen activation is far greater than that of the original poly(D-lysine), and even surpasses that of fibrin. Its effect on plasmin-catalysed modification of both tissue-type plasminogen activator (t-PA) and native (Glu-1-) plasminogen are also investigated. Cleavage of one-chain t-PA to its two-chain form is monitored by measuring the increase in amidolytic activity which accompanies this transformation. Presupposing apparent first-order reaction kinetics, a theory is developed by which the rate constant, kcat/Km = 1.0 X 10(6) M-1 X s-1 of plasmin cleavage of one-chain t-PA can be calculated. Plasmin-catalysed transformation of 125I-labelled Glu-1- to Lys-77-plasminogen is quantified following separation by polyacrylamide gel electrophoresis at pH 3.2. A rate constant, kcat/Km = 4.4 X 10(3) M-1 X s-1 is obtained for the reaction between plasmin and Glu-1-plasminogen in the presence of 1 mM trans-4-(aminomethyl)cyclohexane-1-carboxylic acid. Both of the above plasmin-catalysed reactions are strongly enhanced by trinitrobenzoylated poly(D-lysine). The mechanism of action of this stimulator is elucidated by studying its binding to both activator and plasmin(ogen), and by direct comparison of the results with measurements of plasminogen activation kinetics in the presence of the stimulator. Binding studies are performed exploiting the observation that an insoluble yellow complex is formed between plasminogen and modified poly(D-lysine). Protein-polymer interactions are also studied with solubilised components in an aqueous two-phase partition system containing dextran and poly(ethylene glycol). The rate enhancement of plasminogen activation is found to be closely correlated to the association of plasminogen to the stimulator. It is proposed that the stimulator effects of this simple polymer on the enzymatic activities of both plasminogen activator and plasmin are brought about by association of the proteinase and its substrate to a common matrix. Similarities between the action of the artificial and the natural stimulator (fibrin) are stressed. These properties of trinitrobenzoylated poly(D-lysine) makes it useful as a model for the study of the regulatory mechanism of the fibrinolytic process at the molecular level.

Fibrin and plasminogen structures essential to stimulation of plasmin formation by tissue-type plasminogen activator.

Plasminogen activation catalysed by tissue-type plasminogen activator (t-PA) has been examined in the course of concomitant fibrin formation and degradation. Plasmin generation has been measured by the spectrophotometric method of Petersen et al. (Biochem. J. 225 (1985) 149-158), modified so as to allow for light scattering caused by polymerized fibrin. Glu1-, Lys77- and Val442-plasminogen are activated in the presence of fibrinogen, des A- and des AB-fibrin and the rate of plasmin formation is found to be greatly enhanced by both des A- and des AB-fibrin polymer. Plasmin formation from Glu1- and Lys77-plasminogen yields a sigmoidal curve, whereas a linear increase is obtained with Val442-plasminogen. The rate of plasmin formation from Glu1- and Lys77-plasminogen declines in parallel with decreasing turbidity of the fibrin polymer effector. In order to study the effect of polymerization, this has been inhibited by the synthetic polymerization site analogue Gly-Pro-Arg-Pro, by fibrinogen fragment D1 or by prior methylene blue-dependent photooxidation of the fibrinogen used. Inhibition of polymerization by Gly-Pro-Arg-Pro reduces plasmin generation to the low rate observed in the presence of fibrinogen. Antipolymerization with fragment D1 or photooxidation has the same effect on Glu1-plasminogen activation, but only partially reduces and delays the stimulatory effect on Lys77- and Val442-plasminogen activation. The results suggest that protofibril formation (and probably also gelation) of fibrin following fibrinopeptide release is essential to its stimulatory effect. The gradual increase and subsequent decline in the rate of plasmin formation from Glu1- or Lys77-plasminogen during fibrinolysis may be explained by sequential exposure, modification and destruction of different t-PA and plasminogen binding sites in fibrin polymer.

Zymogen-activation kinetics. Modulatory effects of trans-4-(aminomethyl)cyclohexane-1-carboxylic acid and poly-D-lysine on plasminogen activation.

The kinetics of plasminogen activation catalysed by urokinase and tissue-type plasminogen activator were investigated. Kinetic measurements are performed by means of a specific chromogenic peptide substrate for plasmin, D-valyl-L-leucyl-L-lysine 4-nitroanilide. Two methods are proposed for the analysis of the resulting progress curve of nitroaniline formation in terms of zymogen-activation kinetics: a graphical transformation of the parabolic curve and transformation of the curve for nitroaniline production into a linear progress curve by the addition of a specific inhibitor of plasmin, bovine pancreatic trypsin inhibitor. The two methods give similar results, suggesting that the reaction between activator and plasminogen is a simple second-order reaction at least at plasminogen concentrations up to about 10 microM. The kinetics of both Glu1-plasminogen (residues 1-790) and Lys77-plasminogen (residues 77-790) activation were investigated. The results confirm previous observations showing that trans-4-(aminomethyl)cyclohexane-1-carboxylic acid at relatively low concentrations enhances the activation rate of Glu1-plasminogen but not that of Lys77-plasminogen. At higher concentrations both Glu1- and Lys77-plasminogen activation are inhibited. The concentration interval for the inhibition of urokinase-catalysed reactions is shown to be very different from that of the tissue-plasminogen activator system. Evidence is presented indicating that binding to the active site of urokinase (KD = 2.0 mM) is responsible for the inhibition of the urokinase system, binding to the active site of tissue-plasminogen activator is approx. 100-fold weaker, and inhibition of the tissue-plasminogen activator system, when monitored by plasmin activity, is mainly due to plasmin inhibition. Poly-D-lysine (Mr 160 000) causes a marked enhancement of plasminogen activation catalysed by tissue-plasminogen activator but not by urokinase. Bell-shaped curves of enhancement as a function of the logarithm of poly-D-lysine concentration are obtained for both Glu1- and Lys77-plasminogen activation, with a maximal effect at about 10 mg/litre. The enhancement of Glu1-plasminogen activation exerted by trans-4-(aminomethyl)cyclohexane-1-carboxylic acid is additive to that of poly-D-lysine, whereas poly-D-lysine-induced enhancement of Lys77-plasminogen activation is abolished by trans-4-(aminomethyl)cyclohexane-1-carboxylic acid. Analogies are drawn up between the effector functions of poly-D-lysine and fibrin on the catalytic activity of tissue-plasminogen activator.

Sequence of formation of molecular forms of plasminogen and plasmin-inhibitor complexes in plasma activated by urokinase or tissue-type plasminogen activator.

The pathway of plasminogen transformation was studied in plasma, particularly in relation to fibrin formation and the subsequent stimulation of plasminogen activation. Plasminogen was activated by urokinase (low fibrin-affinity) or tissue-type plasminogen activator (high fibrin-affinity). Formation of 125I-labelled free and inhibitor-bound plasminogen derivatives was quantified after their separation by acetic acid/urea/polyacrylamide-gel electrophoresis. In plasma activator converted Glu-plasminogen (residues 1-790) into Glu-plasmin, which was complexed to alpha 2-plasmin inhibitor. When this inhibitor was saturated, Glu-plasmin was autocatalytically converted into Lys-plasmin (residues 77-790). No plasmin-catalysed Lys-plasminogen formation was observed. Upon fibrin formation, activation initially followed the same Glu-plasminogen-into-Glu-plasmin conversion pathway, and stimulation of plasminogen activation was only observed with tissue-type plasminogen activator. In agreement with the emergence of novel effector function, on early plasmin cleavage of fibrin [Suenson, Lützen & Thorsen (1984) Eur. J. Biochem. 140, 513-522] the fibrin-binding of Glu-plasminogen increased when solid-phase fibrin showed evident signs of degradation. This was associated with the formation of considerable amounts of the more easily activatable Lys-plasminogen, most of which was fibrin-bound. At the same time the rate of plasmin formation with urokinase increased over that in unclotted plasma and the rate of plasmin formation with tissue-type plasminogen activator accelerated. Altogether these processes favoured enhanced fibrin degradation. The rates of Lys-plasminogen and plasmin formation abruptly decreased after lysis of fibrin, probably owing to a compromised effector function on further fibrin degradation.

Initial plasmin-degradation of fibrin as the basis of a positive feed-back mechanism in fibrinolysis.

This study deals with the effect of fibrin on the transformation of Glu-plasminogen to Glu-plasmin during fibrinolysis. It focuses particularly on changes in fibrin effector function caused by plasmin-catalysed fibrin degradation. Conversion of 125I-labelled Glu-plasminogen to Glu-plasmin was catalysed by urokinase or tissue plasminogen activator, in the presence of different preparations of progressively degraded fibrin. Plasmin catalysis of Glu-plasminogen and the fibrin (derivative) effector was inhibited by aprotinin. The presence of intact fibrin enhanced the rate of Glu-plasmin formation catalysed by tissue plasminogen activator, but not by urokinase. The presence of initially plasmin-cleaved fibrin, however, increased the rates of Glu-plasmin formation with both activators, as compared to those found with intact fibrin. The rate enhancements induced by initial plasmin degradation of the fibrin effector were associated with an increase in its affinity to both Glu-plasminogen and tissue plasminogen activator, suggesting causal relationships. The weak binding of urokinase was unaffected by fibrin degradation, indicating that effector function was solely exerted on the Glu-plasminogen moiety of urokinase-activated systems. Further degradation of fibrin decreased the stimulating effect on Glu-plasmin formation. This decrease occurred at an earlier stage of degradation with tissue plasminogen activator than with urokinase, indicating that greater integrity of the fibrin effector is necessary for its optimal interaction with the tissue plasminogen activator than with Glu-plasminogen. Concentrations of tranexamic acid that saturate low-affinity lysine-binding sites nearly completely dissociated the binding of Glu-plasminogen to degraded fibrin, but not to intact fibrin. In analogy with the binding of lysine analogues to these sites, the conformation of Glu-plasminogen may be altered by binding to degraded fibrin, thus giving rise to the increased activation rate.

Secondary-site binding of Glu-plasmin, Lys-plasmin and miniplasmin to fibrin.

Active-site-inhibited plasmin was prepared by inhibition with d-valyl-l-phenylalanyl-l-lysylchloromethane or by bovine pancreatic trypsin inhibitor (Kunitz inhibitor). Active-site-inhibited Glu-plasmin binds far more strongly to fibrin than Glu-plasminogen [native human plasminogen with N-terminal glutamic acid (residues 1-790)]. This binding is decreased by alpha(2)-plasmin inhibitor and tranexamic acid, and is, in the latter case, related to saturation of a strong lysine-binding site. In contrast, alpha(2)-plasmin inhibitor and tranexamic acid have only weak effects on the binding of Glu-plasminogen to fibrin. This demonstrates that its strong lysine-binding site is of minor importance to its binding to fibrin. Active-site-inhibited Lys-plasmin and Lys-plasminogen (Glu-plasminogen lacking the N-terminal residues Glu(1)-Lys(76), Glu(1)-Arg(67) or Glu(1)-Lys(77))display binding to fibrin similar to that of active-site inhibited Glu-plasmin. In addition, alpha(2)-plasmin inhibitor or tranexamic acid similarly decrease their binding to fibrin. Glu-plasminogen and active-site-inhibited Glu-plasmin have the same gross conformation, and conversion into their respective Lys- forms produces a similar marked change in conformation [Violand, Sodetz & Castellino (1975) Arch. Biochem. Biophys.170, 300-305]. Our results indicate that this change is not essential to the degree of binding to fibrin or to the effect of alpha(2)-plasmin inhibitor and tranexamic acid on this binding. The conversion of miniplasminogen (Glu-plasminogen lacking the N-terminal residues Glu(1)-Val(441)) into active-site-inhibited miniplasmin makes no difference to the degree of binding to fibrin, which is similarly decreased by the addition of tranexamic acid and unaffected by alpha(2)-plasmin inhibitor. Active-site-inhibited Glu-plasmin, Lys-plasmin and miniplasmin have lower fibrin-binding values in a plasma system than in a purified system. Results with miniplasmin(ogen) indicate that plasma proteins other than alpha(2)-plasmin inhibitor and histidine-rich glycoprotein decrease the binding of plasmin(ogen) to fibrin.

Relationship between plasma prolactin concentration and pituitary function in patients with a pituitary adenoma.

The influence of hyperprolactinaemia on endocrine functions in forty-two consecutive patients with untreated pituitary tumours was studied. Patients with acromegaly, Cushing's disease and Nelson's syndrome were excluded. Sixteen patients (eleven men and five women) had a pituitary adenoma with suprasellar extension and twenty-six (eleven men and fifteen women) had a small intrasellar tumour. Basal plasma prolactin concentration was measured in all. Thyroid function was assessed by plasma thyroxine (T4) and TSH concentrations, adrenocortical function and growth hormone (GH) secretion by the maximum plasma cortisol, adrenocorticotrophin (ACTH) and GH concentrations, respectively, during insulin-induced hypoglycaemia (tITT). Gonadal function was studied by measuring plasma concentrations of luteinizing hormone (LH), follicle stimulating hormone (FSH), oestradiol-17 beta and in men, testosterone. On the basis of computer assisted tomography of the sella turcica, the tumour volume was calculated. The basal plasma prolactin concentration was elevated in 69% of the patients. Decreased GH secretion was the most frequent pituitary dysfunction (78%) followed in men by gonadal insufficiency (77%), adrenocortical insufficiency (31%) and thyroid insufficiency (21%). There was no difference between patients with elevated and normal plasma prolactin concentration as to the tumour volume and any of the endocrine variables.