The Mechanism of the Inhibition of Plasmin Activity by E-Aminocaproic Acid*

SUMMARY The streptokinase-induced conversion of human plasminogen to plasmin is inhibited by concentrations of e-amino-caproic acid which stimulates the esterolytic activity of plasmin on the synthetic substrate, tosyl-L-arginine methyl ester. The inhibitory effect of e-aminocaproic acid decreases as its concentration is decreased and is eliminated upon incubation of plasminogen with streptokinase, regardless of the presence of e-aminocaproic acid. At increased levels, e-aminocaproic acid further functions as a competitive inhibitor of plasmin activity on tosyl-L-arginine methyl ester with a K1 of 0.32 M. In the presence of concentrations of c-aminocaproic acid (0.05 M) sufficient to nearly saturate its inhibitory effect on the plasminogen to plasmin conversion, the sedimentation coefficient (S%,u) of plasminogen decreases from a native value of 5.0 GX 0.1 S to 3.8 f 0.1 S without a decrease in molecular weight suggesting a gross conformational change in plasminogen induced by c-aminocaproic acid. This conformational alteration is


SUMMARY
The streptokinase-induced conversion of human plasminogen to plasmin is inhibited by concentrations of e-aminocaproic acid which stimulates the esterolytic activity of plasmin on the synthetic substrate, tosyl-L-arginine methyl ester. The inhibitory effect of e-aminocaproic acid decreases as its concentration is decreased and is eliminated upon incubation of plasminogen with streptokinase, regardless of the presence of e-aminocaproic acid. At increased levels, e-aminocaproic acid further functions as a competitive inhibitor of plasmin activity on tosyl-L-arginine methyl ester with a K1 of 0.32 M. In the presence of concentrations of c-aminocaproic acid (0.05 M) sufficient to nearly saturate its inhibitory effect on the plasminogen to plasmin conversion, the sedimentation coefficient (S%,u) of plasminogen decreases from a native value of 5.0 GX 0.1 S to 3.8 f 0.1 S without a decrease in molecular weight suggesting a gross conformational change in plasminogen induced by c-aminocaproic acid. This conformational alteration is also evidenced in circular dichroism measurements.
The effect of c-aminocaproic acid on the conformation of plasminogen is readily reversible and restoration of the native structure is apparent after dialysis.
A mechanism for the inhibition of the plasminogen to plasmin conversion by e-aminocaproic acid is postulated involving the formation of a plasminogen-e-aminocaproic acid complex, which due to the altered conformation of plasminogen, is not acted upon by streptokinase.
This inactive complex is rapidly reversible, yielding a fully streptokinase reactive plasminogen.
Specific activators isolated from sev-* These studies were supported by a grant-in-aid from the Indiana Heart Association and Grant HE-13423 from the National Institute of Health.
f To whom inquiries should be addressed. 1 In this report plasminogen refers to human plasminogen. When plasminogens from other species are ment.ioned their sources will be specified. era1 sources are known to induce plasmin formation from plasminogen.
These activators can be isolated from either bacterial sources (streptokinase), or human origin (urokinase and plasma activator), and are also present in a variety of animal tissues (pig heart activator).
Robbins and coworkers have demonstrated that the activation of human plasminogen to plasmin by urokinase or trace amounts of streptokinase takes place by the urokinase-or streptokinaseinduced cleavage of a single arginyl-valine bond in the plasminogen molecule (1,2). Human plasminogen consists of a single ltolypeptide chain and activation of the molecule to l)lasmill results in a two chain structure stabilized by disulfide bridges (1,2). The two chains have molecular weights of 25,700 and 57,200 daltons and are called the light, and heavy chains, respectively (3).
Several investigators have noted that compounds such as c-aminocaproic acid and p-aminomethylbenzoic acid are potent antifibrinolytic agents. Although there are numerous papers published on these compounds, the mechanism of their antifibrinolytic activity is not clear. Alkjaersig, Fletcher, and Sherry (4) reported that eAcp2 acts as an inhibitor of the plasminogen-plasmin conversion, thus manifesting its antifibrinolytic activity.
Other theories have been presented such as ~Acp acting as an antiplasmin (5), and papers reporting that tAcp was antifibrinolytic due to its inducing a conformational alteration in the structure of the substrate, fibrin, have been published (6,7). More recently, it has been demonstrated that ~Acp has no effect on the activation of human plasminogen (8).
Due to these apparent inconsistencies, we have undertaken a study of the mechanism of the inhibition of plasmin activity 1)~ ehcp. Our results conclusively demonstrate that this inhibitiol~ is complex and the mechanism proposed involves both inhibition at the level of the conversion of plasminogen to plasmin and the inhibition of the proteolytic activity of plasmin.

Materials
PurQkation of Plamlinogen-Human plasminogen was prepared in one step from Cohn III, prepared from age outdated c&rated human plasma, by an affinity chromatography technique utilizing L-lysine bound to Sepharose 4B (Pharmacia) to selec- In order to purify plasminogen from human plasma using this technique, 5 ml of Sepharose-L-lysine were packed into an ll-mm (diameter) column and equilibrated with 0.3 M phosphate buffer, pH 7.5. Approximately 50 ml of Cohn III extract were passed through the column and eluted with 0.3 M phosllhate buffer, pH 7.5, until a steady base-line, indicating no further absorbance, was obtained.
At this point, a solution of 0.1 M phosphate buffer-O.2 nr ~Acp, $1 7.5, was percolated through the column and a sharp peak was immediately obtained.
The yield of plasminogen is 85% under these conditions.
No det,ectable plasmin activity was found in this plasminogen preparation. Fig. 1 shows a typical elution profile. Other Proteins-Streptokinasc (Varidase) was obtained from Lederle Laboratories through a local drug outlet in vials containiug 20,000 units of activity.
Reagents--EAcp, was purchased from Calbiochem and tosyl-AlIe was purchased from Cycle Chemical Company.
All other reagent)s were the best commercially available.

Methods
PZa,smin Assays-Since these assays were done under a variety of conditions they are described in appropriate sections of the manuscript.
All assay components were prepared in 0.1 RI Tris.hydrochloride, 1~11 8.0, and all assays were performed at 30".
In general, the assays consisted of converting plasminogen to plasmin with streptokinase and following the action of plasmin on tosyl-AMe.
Analysis of the amounts of tosyl-AMe cleaved by plasmin were performed essentially as described by Hestrin (10) with the following minor modifications. ilfter a given time of reaction a 0.2.ml aliquot of the reaction mixture was added to a solution containing 0.2 ml of 4 N sodium hydroxide aud 0.2 ml of 2 N hydroxylamine hydrochloride. These conditions were suffcient to immediately stop the enzymatic reaction.
The reaction was allowed to proceed for 30 min, and 0.2 ml of a solution of 4 N hydrochloric acid containing 6 g of trichloroacetic acid was added followed by addition of 0.2 ml of water.
Following t,his, 4 ml of a solution containing 0.11 JI ferric chloride in 0.004 M hydrochloric acid were added, and the absorbances of these solutions lvere determined on a Gilford model 240 spectrophotomct'er at 525 nm against an appropriate blank. This Iprocedure allowed ~1s to determine the final concentration of tosyl-AMe from a standard curve. Initial concentrations of tosyl-AMe were obtained in the same fashion by preparing incubation mixtures in the absence of any enzymes. The rate of reaction of plasmin &h tosyl-AMe was shown to be linear at the times of incubat'ion used in these studies.
Ultracentrijuge Studies~ -Sedimentation coefficients of plasminogen in 0.1 M l)hosph:rte, $1 7.5, and plasminogen in 0.1 M phosl)hate-0.05 M ~Acp, pII 7.5, were measured in a Spinco model E analytical ultraccntrifuge using absorption optics at 280 nm. Protein concentrations were approximately 0.2 mg per ml. Sedimentation coefficient's were calculated in the usual manner and corrected to the density and viscosity of water at 20" (11).
Circular Dichroism &u&es-These were performed lvith a Cary 60 spectropolarimeter circular dichroism apparatus using I-, 5-, and lo-mm cells. These cells were interchanged during a run so that the optical density of the protein did not exceed 1.0 at any wave length.
The ellipticity [0] values were recorded directly from the instrument and converted to molecular ellipticity [0] expressed in degree cm2 per dmole of amino acid according to the relationship where Ma is the mean residue weight of the protein, 1, the path length in the sample solution in centimeters, and C is the protein concentration in grams per ml.

Polyacrylamide
Gel Electrophoresis--These experiments n-ere performed at $19.5 (12), pl-I 4.3 (13), pH 3.2 in 6.25 M urea (13), and in sodium dodecyl sulfate (14). These results are consistent with the observations that human plasminogen consists of multiple molecular forms (4,9,15). In agreement with these facts, plasminogen gave only one band when examined by polyacrylamide gel electrophoresis in sodium dodecyl sulfate.
The molecular weight of on the activity of plasmin (Fig. 4), the inhibition observed must be due to an inhibition of the conversion of pla.sminogen to plasmin.  of e&p on the conversion of plasminogen to plasmin in the following manner. Experiments were conducted identical with those in Fig. 3 except that the eRcp concentration was varied and thr data is plotted in Fig. 5. This data shows the percentage of inhibition of plasmin activity plotted against the concentration of ~Acp and includes an inhibition of the conversion of plasminogen to plasmin and a direct stimulation of plasmin at these &cl) concentrations.
The concentration of eAcp was not high enough to consider the inhibition of plasmin directly by e9cp. Since the data in Fig. 4 was obtained at the same initial concentration of tosyl-AMe as the data in Fig. 5, we have added the clcrzje in Fig. 4 to that in Fig. 5 to obtain the inhibition data of ehcp on the conversion of plasminogen to plasmin. This corrected data is also presented in At this time 0.02 ml of tosyl-AMe (100 pmoles per ml) was added to initiate the plasmin reaction and allowed to proceed for 10 min. The amount of tosyl-AMe cleaved was analyzed as previously described.
Parallel experiments were performed with 0.01 ml of Tris buffer added i n place of ~Acp at each streptokinase concentration.
The results of these experiments are presented in Table I. Clearly, the prior incubation abolishes the inhibitory effect of ~Acp at the streptokinnse concentrations indicated.

Inhibitory
Properties of Plasminogen to Plasmin Conversion-~Acp was incubated with plasminogen for various times and the extent of inhibition of the conversion of plasminogen to plasmin was noted in order to determine whether the time factor was important in e&p exhibiting its inhibitory effect. These experiments were carried out as follows in the indicated orders of addition: 0.1 ml of l)lasminogen and 0.01 ml of 0.5 M ~hp were incubated for various times. Following this 0.110 ml of Tris buffer and 0.02 ml of tosyl-AMe (100 pmoles per ml) were added. The reaction was initiated by addition of 0.01 ml of streptokinase and tosyl-A?\ie cleaved in a 0.2.ml aliquot was determined as de- The results are presented in Table II. Clearly, the time of incubation of plasminogen with ~Acp is not an important factor and the inhibition observed is attained very rapidly.
Sedimentation Coejicient of Plasminogen in Presence and Absence of &q-The effect of 0.05 M ~Acp on the sedimentation coefficient (s~O,w) of pl asminogen is given in Table III. There is a decrease in the sio,W of plasminogen upon addition of ~Acp and a return to the native value is evident upon dialysis against buffers spectrum of plasmiuogen in &cl) appears almost devoid of any helical structure. These results support the inferences made from analysis of the .s$+ values of plasminogen in eAcp in that a gross conformational alteration takes place resulting in a more random polypeptide chain. DISCUSSION The studies presented here show that the mechanism of the inhibition of plasmin activity by ~Acp is complex and involves both inhibition of the conversion of plasminogen to plasmin and stimulation of plasmin at low eAcp concentrations and direct inhibition of plasmin activity at high eAcp concentrations.
A schematic mechanism for these effects can be illustrated which is in accord with the data presented in this manuscript. This scheme is as follows: In this diagrammatic representation, under normal conditions, plasminogen (Pg) is activated by streptokinase (SK) to form plasmin (pm).
The plasmin possesses esterase activity on tosyl-AMe (8). In Fig. 2, in the absence of ~Acp and without prior incubation of plasminogen with streptokinase, the enzymatic activity of plasmin increases as the streptokinase concentration is increased. This effect can best be explained by considering plasminogen as the substrate for streptokinase action. Then, as the streptokinaee concentration is increased there is a more rapid conversion of plasminogen into plasmin and thus a higher rate of reaction of plasmin with tosyl-AMe.
Consistent with this hypothesis, when plasminogen is incubated with different concentrations of streptokinase for various times prior to addition of tosyl-AMe, there is no dependence of plasmin activity on streptokinase concentration given a sufficient time of incubation. This occurs since all the plasminogen present will eventually be converted into plasmin.
We have shown this to be the case during the course of these studies and others have published confirmatory data (16).
In considering the inhibition data in Fig. 3 at 0.02 M ~Acp and no prior incubation of plasminogen with streptokinase, all one Inhibition of Plasmin Activity Vol. ,346, x0. 14 can conclude at this point is that plasmin activity is competitively inhibited by ~hcp. This inhibition can occur at the level of the conversion of plnsminogerl to plasmin or ~Xcp can be a direct inhibitor of plasmin. This question is resolved by coupling the datn of Fig. 3 with that of Fig. 4. In Fig. 4 we conclusivcly demonrtratc that tAcp concentrations up to 0.1 M have a stimulatory effect on the activity of plasmin. Thus, the inhibition of the plasminogen to plasmin conversion should in fact be greater than what is observed by performing experiments as in Fig. 3. We have made a rough correction for this stirnulatory effect as illustrated in Fig. 5. Clearly, the net inhibition of plaemin activity which occurs at ~Acp concentrations to 0.1 IVY is due only to the inhibition of the plasrninogen to plasmin conversion by ~Acp. These effects are illustrated in the scheme prcPentcd above. It can be seen that plasminogen can react with tAcp and foErn a l)lasrninogell.Ehcp complex. This complex is refractive to activation by streptokinase. We feel that this complex can be rapidly converted to I&~sminogen and thus be activated by streptokinase for the following reason. As shown in Table I, the inhibitory effect of 0.02 M ~Acp on the conversion of plasminopen to plasmin can be abolishctl by incubation of plasminogen, 0.02 hf tAcp, and streptokinasc prior to addition of tosyl-AMe.
What must bc happening in this case is that strcptokinase, in an irreversible manner, reacts with the free plasminogen in the equilibrium, plasminogen + ~Acp 4 &asminogen .&cl), to form plasmin.
Thus, the equilibrium is pulled toward free plasminogen.
Given enough time of incubation before tosyl-AMc addition, the plasminogen ~Acp complex will completely dissociate into free plasminogen which will react with the streptokinase present to produce plasmin.
Since, under these conditions all the plasrninogen present will be converted into plasmin, no inhibition occurs when tosyl-AMe is added, regardless of t,he presence of ~Acp. On the other hand, inhibition at 0.02 M ~Acp is only seen when all components of the assay mixture are added together in a definite order (Fig. 3-referred to as "no preincubatjon").
In this case the rate of plasmin reactivity with tosyl-AMe will be slower in the presence of .~Acp than in its absence. All the data collected in this study are consistent with these views.
;\t this point KC feel that we call perhaps explain some discrcp-:ulciea which appear in the literature concerning the activation of plasminofen b\-streptokiiiase.
For example, Murarnatu et al. (16) proposed that t,hcrc wcrc two mechanisms of activation of plasrninogen by strcptokinasc.
The first mechanism, the details of which are not important for discussion here, occurred at low streptokinase concentrations and was based on the fact that. when plasminogen mas activated ai low strel)tokinase concent,rations (10 units), the activity of plasmin incrcnsetl with time.
In other words, incubation of streptokinase with plasminogcn before substrate addition was necessary to obtain full plasmin activity. The second mechanism of activation of plnsminogen by streptokinase, according to these authors, occurred at high strcptokinase concentrations and was based on the fact that at high streptokinase concentrations (900 units) a much shorter incubation time of st'reptokinase with plasminogen was necessary before substrate addition to obtain full plasminogcn activity. Although their mechanism is consistent with the data obtained, a rnore simple rate effect can also explain the data. According to our scheme, it is not necessary to propose two mechanisms of action.
If one considers the action of streptokinase on plasminogen to be as stated above then at low streptokinaee concentrations the rate of activation of plasminogeri Tao plasmin will be slower than nt higli streptokinase concentrations.
In addition, Muramatu et (11. feel that two mechanisms of RCtion for streptokinase activation of plasminogen must exist since ~Acp did not inhibit plasmin activity at high concentrations of streptokinase (800 units), whereas inhibition did occur at low concentrations of streptokinase (10 units).
In these experiments incubation of plasminogen, ~Acp, and streptokinase before sul)strate addition was performed.
Bgain, we feel, based on our scheme, that it is not necessary to Ijropose a complicated esplanation. Clearly at 800 units of streptokinase, according to our scheme, all the plasminogcn I)rrsent is converted to plasnlin regardless of the formation of a plasminogen.tAcp inactive cornplex. At these high streptokinase concentrations t,he rate of breakdown of this complex is very rapid.
On the other hand, at 10 units of streptokinase, the rate of plasmirrogen. ~Acp complcs breakdown is suflicicnlly slow to require much larger I)rior illcanbation times. We have lengthened the prior incubation times with 10 units of strcptokinasc in an effort to demonstrntr this point, but the rate of complex dissociation was so slow that in-activ&ioii of plasmin occurred, severc,l,v complicating our rrsults. IIowever, wc have tlernonst~rated that at several strt,ptokinase conccntrntions above 100 units the inhibition by low t.%cp can be abolished, supporting our contention.
Quite a different picture of the inhibition of plasmin nctivit> occurs at high ~hcp concentrations.
Here, two effects are noted. (a) inhibition of plasmin artivit,y at the level of the conversion of plasminogen to plasmin still occurs and (b) at high concentrations &cp is a competitive inhibitor of plasmin activity with a K1 of 0.32 M. This effect is indicated in the scheme presented above. These results require no prolonged analysis since L&I) is a substrate analogue of L-lysine and L&sine methyl ester is a substrate of plasmin.
Therefore, tL2cp probably binds at the substr:ltc binding site of plasmin thereby producing csompetitire illhibitiotl kinetics.
The k'l of ~Acp toward plasmin is very high illtlicnting that ~Acp is not strongly bound to the enzyme. 1Ve art I)rcsently in the process of testing rnorc substrate analogues as inhibitors of plasmin and a report on these studies will shortly appear.
Although the inhibitory cffrc,t, of high collccntratiolls OC t.\cap on the enzymatic activit,y of plasmin appears to 1)~ rraao~rabl~ straightforward, the reason for the inhibitory effect of Ion-CO~Icentrations of .&cp on t'he c~onversion of 1)l:isminogen to I)l:lsmill st,ill requires explanation.
Re f(lel that thi:: inhibition is tluc to ~Acp causing a freely reversible conformat ional change in the plasrninogen molecule.
This conformational alteration ~"odnces a plasminogen which is not capable of being acted upon by strep tokinase.
This conformational alteration is clearly evidenced 1)) analysis of the S&~ values of plasminogen in the presence alld absence of 0.05 &CC ~Acp given in Table III. This conformational alteration is also evidenced in circular dichroism studies. X11:1lysis of the data in Fig. 6 shows that plasminogen in the :Lbscll(~e of ~Acp possesses some, but' not a great deal of, helical structure, as evidenced by the trough at 220 nm. This trough disappears upon addition of 0.05 M ~Acp and there is a decrease of the molecular ellipticity.
These conditions indicate that there is :I considerable loss of structure of plasminog~n upon addition OF ~.Zcp. The native structure reappears upon removing the .&p by tlinlysis suggesting a freely reversible conformational transition. With regard to the stimulatory effect of low concentrations of ~Acp on the activity of plasmin, this is due to a direct effect on plasmin and not on t'he plasminogen to plasmin conversion. Oul