Flaujeac factor deficiency. Reconstitution with highly purified bovine high molecular weight- and delineation of a new permeability-enhancing peptide released by plasma from bovine high molecular weight-kininogen.

R T Matheson, … , H Kato, K D wuepper

J Clin Invest. 1976;58(6):1395-1406. https://doi.org/10.1172/JCI108595.

Research Article

Flaujeac trait is the functional deficiency of a plasma of the intrinsic , -forming, and plasma fibrinolytic pathways. The Flaujeac factor in man has been isolated and tentatively identified as a kininogen of high molecular weight (HMW). Highly purified bovine HMW-kininogen, but not bovine low molecular weight kininogen, repaired Flaujeac factor deficiency. The two subspecies of this molecule, HMW-kininogen a and HMW-kininogen b, also corrected Flaujeac factor deficiency. When bovine HMW-kininogen was incubated with bovine plasma kallikrein, kinin-free HMW- kininogen, , and a glycopeptide fragment (peptide 1-2; 12,584 daltons) were rapidly released. None of these fragmentation products corrected Flaujeac factor deficiency alone or in mixtures. The function of HMW-kininogen appeared to depend upon the structural integrity of the native molecule. When injected in concentrations of 2 pmol-8 nmol/0.1 ml, peptide 1-2 caused increased vascular permeability in rabbits, rats, or guinea pigs. The enhanced permeability was maximal within 1-2 min and terminated in 5-10 min, differing from that of bradykinin or histamine. Injected together in equimolar amounts, peptide 1-2 and bradykinin produced a synergistic permeability response which was immediate in onset as well as prolonged in duration. Peptide 1-2 is a rapidly acting, highly basic glyco-peptide which mediates increased vascular permeability in a complementary and synergistic manner with bradykinin.

Find the latest version: https://jci.me/108595/pdf Flaujeac Factor Deficiency

RECONSTITUTION WITH HIGHLY PURIFIED BOVINE HIGH MOLECULAR WEIGHT-KININOGEN AND DELINEATION OF A NEW PERMEABILITY- ENHANCING PEPTIDE RELEASED BY PLASMA KALLIKREIN FROM BOVINE HIGH MOLECULAR WEIGHT-KININOGEN

ROBERT T. MATHESON, DONALD R. MILLER, MABIE-JOSt LACOMBE, YONG N. HAN, SADAAIU IWANAGA, HISAo KATO, and KiRK D. WUEPPER From the Department of Dermatology, School of Medicine, University of Oregon Health Sciences Center, Portland, Oregon 97201, Service d'H6matologie, H6'pital Cochin, Paris, France 75674, antd Institute for Protein Research, Osaka University, Osaka, Japan

A B S T R A C T Flaujeac trait is the functional defi- permeability was imailximllal withinl 1-2 mIi aInd ter- ciency of a plasma protein of the intrinsic coagula- minated in 5-10 immi, differinig friom that of brady- tion, kinin-forming, and plasma fibrinolytic pathways. kinin or histaimine. Injected together in e(Iuiimolatr The Flaujeac factor in man has been isolated and amounts, peptide 1-2 aInld bradykinin produtced a tentatively identified as a kininogen of high molecular synergistic perne-ability response which was iimmedi- weight (HMW). ate in onset as well aIs prolonged in dtiration. Highly purified bovine HMW-kininogen, but not Peptide 1-2 is ai ralpi(dly acting, highly basic glyco- bovine low molecular weight kininogen, repaired peptide which me(diattes iniereatse(d vascular permiiea- Flaujeac factor deficiency. The two subspecies of this bility in a complemiientalry aind synergistic mlanner with molecule, HMW-kininogen a and HMW-kininogen b, bradykinin. also corrected Flaujeac factor deficiency. When bovine HMW-kininogen was incubated with bovine plasma INTRODUCTION kallikrein, kinin-free HMW-kininogen, bradykinin, and a glycopeptide fragment (peptide 1-2; 12,584 The deficiency of a plasma protein which partici- daltons) were rapidly released. None of these frag- pates in the intrinsic coagulation, kinin-forming, and mentation products corrected Flaujeac factor defi- plasma fibrinolytic pathways was reported recently by ciency alone or in mixtures. The function of HMW- three laboratories (3-5). They were recognized in- kininogen appeared to depend upon the structural dependently and identified by eponym.*s as Fitzgerald integrity of the native molecule. (3, 6), Flaujeac (4, 7-9), and Williams (5, 10) traits. When injected in concentrations of 2 pmol-8 nmol/ We have exchanged reagents with Dr. Waldmann (3, 6) 0.1 ml, peptide 1-2 caused increased vascular perme- and Dr. Colman (5, 10) and have reported that mutual ability in rabbits, rats, or guinea pigs. The enhanced correction of the coagulation assays did not take place when mixtures were made of all possible combinations of the Fitzgerald, Flaujeac, or Williams plasma (8,9) This work was presented in part at the Western Section suggesting the deficiency of a common plasma protein. meeting of the American Federation for Clinical Research, Normal plasma corrected the activated partial thrombo- 6 February 1976, Carmel, Calif. Preliminary reports have plastin time (APTT)' in each instance, excluding the been published (1, 2). Dr. Matheson was supported by a National Research Service possibility of a cireculating which might Individual Award of the U. S. Public Health Service be responsible for the functional abnormalities. (F32-HL-05170-01). Dr. Wuepper was recipient of a U. S. Public Health Service Research Career Development Award IAbbreviations used in this paper: APTT, activated partial AM-40086-06. thromboplastin time; HMW, high molecular weight; LMW, Receivedfor publication 29 March 1976 and in revised form low molecular weight; PF/dil, permeability factor of dilu- 16 August 1976. tion. The Journal of Clinical Investigation Volume 58 December 1976-1395-1406 1395 mol wt and LMW-kininogen of 49,000 mol wt have been successfully purified from bovine plasma by Yano et al. (16, 17), and Komiya et al. (18-20). As depicted in Fig. 1, bovine HMW-kininogen is a 76,000 mol wt glycoprotein with a masked N-terminal residue and C-terminal leucine. Two HMW-kinino- gen subspecies, a and b, which have the same num- BOVINE HMW KININOGEN (a) BOVINE HMW KININOGEN (b) ber and composition of tryptic peptides have been recovered. They differ only by a single proteolytic PLASMA KALLIKREIN cleavage at the C-terminal end of the bradykinin moiety without loss of bradykinin or any other frag- I ments (19). The vasoactive peptide bradykinin and a recently KININ PEPTIDE 1-2 described glycopeptide, peptide 1-2, are located within an intrachain disulfide loop (19, 21-23). The PLASMA KALLIKREIN N-terminal portion of the peptide 1-2 is connected with the C-terminus of bradykinin within the loop in the HMW-kininogen molecule (22). FRAGMENT FRAE+_ Upon incubation of bovine plasma kallikrein with FRAGMENT FRAMENT 2 bovine HMW-kininogen a or HMW-kininogen b, FIGURE 1 Fragmentation of bovine HMW-kininogen with bradykinin, kinin-free HMW-kininogen, and the glyco- bovine plasma kallikrein. (Redrawn from Kato, H., et al.) peptide fragment (peptide 1-2, 12,584 daltons) are (22). See text for details. rapidly and simultaneously released in equimolar amounts as shown schematically in Fig. 1 (21, 22). Donaldson and her co-workers recognized a fourth Upon prolonged incubation with bovine plasma kalli- example of this deficiency (11, 12) by employing krein, peptide 1-2 is subsequently cleaved into a Fitzgerald plasma as the reference source. A fifth glycopeptide (fragment 1; 8,000 daltons) (21,23) and a case has been identified, with Flaujeac plasma as the "histidine-rich peptide" (fragment 2; 4,584 daltons) source of deficient plasma (13). (24). Bovine HMW-kininogen from which both the We have previously isolated and characterized bradykinin and the peptide 1-2 moieties have been the Flaujeac factor, shown a deficiency of kinino- enzymatically removed by plasma kallikrein has been gen antigen in the proposita's plasma, and inhibited designated "kinin-free HMW-kininogen." Further the function of the factor with antiserum to human study of this residual of the parent molecule has kininogen (4,9). Of the two general forms of human demonstrated it to be a 66,000 mol wt glycoprotein kininogen in human plasma, we recognized that consisting of two polypeptide chains (heavy (H) and partially purified high molecular weight (HMW) kinino- light (L) chains) linked by a single disulfide bridge gen, but not low molecular weight (LMW) kininogen, (Fig. 1) (22). reconstituted the Flaujeac plasma (4,9). This observa- Amino acid composition and carbohydrate content tion has already been confirmed with plasma from of all the fragments have been determined (21), another patient with the same functional deficiency and the sum of the total amino acid residues of (11). This activity has been retained by human kinin-free HMW-kininogen, peptide 1-2, and brady- HMW-kininogen purified more than 5,000-fold.2 In kinin coincides with that of the whole bovine HMW- this collaborative study, we have tested very highly kininogen molecule. By contrast, bovine LMW-kinino- purified bovine and their fragments, sub- gen is not susceptible to plasma kallikrein, and in- stantial segments of which have been subjected to cubation of LMW-kininogen with a glandular kalli- amino acid sequence analysis, for their ability to krein releases only bradykinin with no counterpart reconstitute Flaujeac factor deficient plasma. for peptide 1-2 (22). The existence of kininogens of high molecular We have now tested bovine kininogens and their weight was first described by Jacobsen (14) and native cleavage products to learn whether they are Jacobsen and Kriz (15). In contrast to LMW-kininogen, corrective in human HMW-kininogen deficient plasma the HMW-kininogen showed a preferential suscepti- and what molecular role these play in coagula- bility to plasma kallikrein. Attempts to purify HMW- tion, , and induction of increased vascular kininogen revealed it to be quite labile and recover- permeability. Moreover, it has been possible to study able only in low yield. A HMW-kininogen of 76,000 the two molecular forms of bovine HMW-kininogen, 2Matheson, R. T., D. R. Miller, and K. D. Wuepper. a and b. In addition, we now report that the bovine Unpublished observations. peptide 1-2 fragment is a mediator of increased 1396 Matheson, Miller, Lacombe, Han, Iwanaga, Kato, and Wuepper vascular permeability in its own right. Permeability model L5-65 with an SW-S0 rotor (Beckman Instruments was studied in several animal species which were Inc., Palo Alto, Calif.) at 45,000 rpm for 16 h at 5°C. Frac- tions of 15 drops were collected and assayed for protein all responsive. The duration of effectiveness of peptide content and permeability activity. 1-2 differed from that of bradykinin or histamine, Assay methods. The kaolin APTT has been described and an equimolar mixture of pepetide 1-2 and brady- (9). The ability of highly purified bovine HMW-kininogen, kinin resulted in a prolonged, complementary, and LMW-kininogen, and cleavage products of HMW-kininogen synergistic reaction of increased vascular permeability. to correct the prolonged APTT of Flaujeac plasma was assessed by the addition of S0-,ul samples of serial dilu- These studies establish a third permeability-enhancing tions of the purified protein solutions to 50 ,ul of the kaolin/ activity of the kinin-forming system in mammalian cephalin mixture and SOul Flaujeac plasma. After incuba- species in addition to bradykinin and permeability tion for 30 min at 37°C, the mixture was recalcified with factor of dilution (PF/dil). 50 ,ul of 0.05 M CaCl2 and the clotting time was recorded. The ability of the same highly purified proteins to correct the fibrinolytic defect of Flaujeac plasma was measured MATERIALS with a kaolin-activated euglobulin lysis assay as described Reagents and chemicals. Synthetic bradykinin (BRS 640, (9). Sandoz Ltd., Basel, Switzerland), histamine dihydrochloride Vascular permeability and PFIdil formation. Bioassay (Fisher Scientific Co., Pittsburgh, Pa.), chlorpheniramine of vascular permeability followed a modification of the maleate, (Schering Corp., Bloomfield, N. J.) and ristocetin method of Miles and Wilhelm (28). Albino guinea pigs (Abbott Laboratories, Chicago, Ill.) were purchased from their (400-600 g), rabbits (2,000-2,500 g), and rats (300-500 g) suppliers. Rabbit brain "cephalin", a platelet substitute in were injected intravenously with 50-70 mg/kg body weight coagulation assays, was obtained from Sigma Chemical Co. of Evans blue dye in a 5.0% solution in saline. 0.05- (St. Louis, Mo.). or 0.10-ml test solutions were then injected intracutane- Human plasmas. Flaujeac factor deficient plasma was ously. The animal was sacrificed 30 min after the injections obtained from the proposita, Mme. Flaujeac. Nine parts were begun and the diameter of blueing on the reflected blood was mixed with one part 3.8% citrate or acid-citrate skin was measured by transillumination. Three to eight ani- dextrose, and stored at -70°C until used. Normal pooled mals were used for each experiment, and duplicate injec- plasma from 28 donors was treated identically. tions were performed from rump to neck or neck to rump Biological reagents. Human , 90%o coagulable to offset regional variability that occurs in some animals. (Grade L, AB Kabi, Stockholm, Sweden), soluble collagen, Arithmetic means were calculated by measuring the largest and adenosine diphosphate (Sigma Chemical Company) diameter of blueing and its perpendicular, and the results were purchased from their suppliers. Bovine , expressed as mean permeability diameter. Measurements topical, and epinephrine hydrochloride were obtained from were performed in a single blind manner, and unless other- Parke, Davis and Co., Detroit, Mich. wise stated, the results were those found in rabbits. Purified bovine proteins. HMW-kininogen, LMW-kinino- Tests for PF/dil formation were performed as described gen, kinin-free HMW-kininogen, and peptide 1-2. Highly elsewhere (29). 20-,ul samples of plasma combined with purified HMW-kininogen was prepared from fresh bovine varying amounts of the purified bovine peptides were diluted plasma by the previous method (18), and LMW-kininogen to 2.0 ml with sterile saline in a new glass tube, covered was purified according to the revised method by precipita- with parafilm, (American Can Co., Greenwich, Conn.) and tion with zinc acetate (25). Bovine plasma kallikrein was inverted 40 times each min on a Lab Tek sample mixer (Miles prepared as previously described (26, 27). The specific Laboratories Inc., Miles Research Products, Elkhart, Ind.) for activity of the kallikrein was 23.5 TAME U/,Lg protein. 15 min. 0.1-ml sample volumes were then assayed for Preparation and isolation of peptide fragments produced permeability in the rabbit as above. from bovine HMW-kininogen by highly purified bovine The time course of permeability was determined by in- plasma kallikrein was accomplished as described (21, 22). jecting a series of single doses at time intervals before dye Briefly, bovine HMW-kininogen was incubated with bovine injection. 1 min after blueing, the animals were sacrificed plasma kallikrein (0.1%, wt:wt) in 0.2 M ammonium bi- and mean permeability diameters were measured. Equimolar carbonate buffer pH 8.0, at 37°C (21), and the digest amounts of histamine, bradykinin, and a mixture of peptide applied to a column of Sephadex G-75 (Superfine, Pharmacia 1-2 and bradykinin were injected simultaneously in each Fine Chemicals, Inc., Piscataway, N.J.). The kinin-free animal for comparative studies. HMW-kininogen, which did not release kinin upon incuba- The effect of chlorpheniramine maleate on permeability tion with plasma kallikrein, was eluted in the void volume was evaluated by injecting peptide 1-2, 0.1-1.0 ug (8 fraction, and one major peak with the absorbance at 280 pmol-0.8 nmol), or histamine dihydrochloride, 0.01-1.0 nm (peptide 1-2) and bradykinin were separated. The iso- ,ug (60 pM-3 nM histamine base), in blued rabbits. After lated peptide 1-2 and kinin-free kininogen showed single 30 min, 10 mg chlorpheniramine maleate was administered bands, respectively, on sodium dodecyl sulfate-gel electro- intravenously, and the intracutaneous injections were re- phoresis (21). peated at different sites. 30 min after the second series of The highly purified HMW-kininogen, LMW-kininogen, injections the animals were sacrificed and permeability diam- kinin-free HMW-kininogen, and peptide 1-2 were dialyzed eters were measured as above. against distilled water and lyophilized. The proteins were Disc-gel electrophoresis. Polyacrylamide-gel electro- solubilized 1 mg/ml in sterile normal saline and stored in phoresis was performed in 7% acrylamide (Eastman Kodak aliquots at -70°C until used. Co., Rochester, N. Y.) according to the method described Ultracentrifugation. Preparative ultracentrifugation was by Davis (30). Gels were stained for protein with Coomas- carried out in linear gradients of 10-37% sucrose dissolved sie brilliant Blue in 10%o trichloroacetic acid. Unstained in 0.15 M NaCl buffered with Tris-HCl 0.05 M, containing gels were cut into 1.3-mm segments and the proteins 0.001 M EDTA. 200-,g samples in 0.2 ml were layered eluted into 0.2 ml of 0.05 M Tris buffer in 0.15 M NaCl onto the gradients and spun in a Beckman ultracentrifuge for 48 h at 4°C. Bovine High Molecular Weight Kininogen Permeability Enhancing Peptide 1397 15r LMW-KININOGEN (.b -N 5X0 Rm -0418 Rm 0.462 It k. ,w.x.5-01 cz:. K HMW-KININOGEN -- -4t t 100

50 IJ,l 0J 0.5 1.0 5 10 50 100 10 20 30 40 50 Microgroms Protein Added (0 Segment Number FIGURE 2 Correction of Flaujeac plasma by purified bovine FIGURE 4 Disc gel electrophoresis of purified bovine HMW-kininogen. Normal human APTT = 65 s. A dose-related HMW-kininogen. Two zones of correction migrated with response was given by 0.1-20 ,g of HMW-kininogen. the two subspecies of HMW-kininogen identified in a parallel LMW-kininogen gave no correction. gel. Rm, relative mobility

Smooth muscle contraction. The ability of peptide 1-2 ability to cause aggregation of human blood platelets in to contract smooth muscle was measured by bioassay on concentrations of 0.1-10 ,g/0.45 ml platelet-rich-normal the rat uterus or guinea pig ileum. Peptide 1-2, 1-100 Ag, plasma (platelet Aggregometer, Chrono-log Corp., Broomall, was added to a 10-ml muscle bath containing a rat uterine Pa.). The ability of peptide 1-2 to alter the platelet horn or guinea pig ileum suspended in oxygenated Tyrode's aggregation response patterns of known aggregating agents solution at 37°C. Bradykinin or histamine served as positive was assessed by incubating 1.0-,ug amounts of peptide controls for each preparation respectively. 1-2 with adenosine diphosphate 2.0/,M/50 Al, soluble col- Granulocyte chemotaxis. The ability to attract human lagen 16 Ag/50 /A, epinephrine hydrochloride 2.5 ,mol/50 granulocytes was measured in a radioassay by employing ,lI, or ristocetin 0.6 mg/50 ul for 10 min at 27°C, after 5'Cr-labeled granulocytes in a double micropore filter system which 50 ,A of each mixture was added to 0.45-ml volumes (31). Gey's balanced salt solution and endotoxin-stimulated of platelet-rich-plasma and aggregometry performed. In serum served as negative and positive controls, respectively. separate experiments, 1 ug of peptide 1-2 was incubated Peptide 1-2 was assayed for chemotactic activity in con- with 0.45 ml platelet-rich-plasma for 10 min at 27°C before centrations of 0.05-50 ,ug/ml. Each concentration of peptide the addition of the same aggregating agents. 1-2 was tested in triplicate on two separate occasions. Platelet aggregometry. Peptide 1-2 was tested for its RESULTS Correction of abnormal kaolin-activated partial Polyacrylamide Gel Electrophoresis thromboplastin time of Flaujeac plasma with highly purified bovine HMW-kininogen. Freshly prepared bovine HMW-kininogen significantly corrected the prolonged kaolin APTT of Flaujeac plasma in a dose- dependent manner in the range of 0.1-10.0 ,g (Fig. 2). By contrast, LMW-kininogen, in added of o -.. ..: amounts 1.0-100 ,ug failed to affect the prolonged APTT. Upon completion of this study polyacrylamide _;,.0 gel electrophoresis of the purified bovine HMW- kininogen confirmed the earlier observation (19) that two proteins were present (Fig. 3). These two proteins differ by a single proteolytic cleavage contained by HMW-kinonogen a and not by HMW-kininogen b. These results suggested that the proteins were in their ( I) ( 2) ( 3 ) native state and had not undergone degradation. Upon incubation with plasma kallikrein and electro- ( I ) HMW Kininogen phoresis, both subspecies converted to the more acidic fragment, kinin-free HMW-kininogen (Fig. 3) (21, 22). (2) kinin-free HMW Kininogen The two species of HMW-kininogen were resolved by polyacrylamide gel electrophoresis. The gel was (3) LMW Kininogen cut into 1.3-mm segments and eluted into a buffer FIGURE 3 Acrylamide gel electrophoresis of bovine HMW- solution. The amount of Flaujeac factor present in kininogen (showing subspecies a and b), kinin-free HMW- each sample relative to that in pooled normal human kininogen, and LMW-kininogen. plasma as determined by coagulation assay was ex- 1398 Matheson, Miller, Lacombe, Han, Iwanaga, Kato, and Wuepper A this solution, and that bovine kinin-free HMW-kinino- a A PEPTIDE 12 -% 500 -o gen did not correct the abnormal APTT of Flaujeac -.---- BRADYKININ Il k plasma. The second smaller protein peak in Fig. 6 0-- (above) has not yet been identified. It was not as- I-FREE --0_0----- O KININ _____ HiW~~~~~~~-1.-KININOGEN sociated with any corrective activity and may have 50 HMW-KININOGEN ---Q_ represented an aggregate of HMW-kininogen. Reconstitution of fibrinolytic activity of Flaujeac D- plasma with bovine HMW-kininogen. In the fibrino- 0I .... lytic assay, Flaujeac plasma differed from normal in Ql 05 10 5 l0 50 100 Microgroms Protein Added that lysis times were as long as the reagent control. This was corrected by mixing with normal human FIGURE 5 Failure of bovine HMW-kininogen cleavage plasma (1:1) (Table I). The addition of HMW-kinino- products to correct Flaujeac plasma. Kinin-free HMW- kininogen gave correction 0.6 of 1% as effective as HMW- gen to Flaujeac plasma reconstituted its ability to kininogen. Peptide 1-2 and bradykinin gave no correction. give kaolin-activated lysis in a dose-dependent man- ner. This was not due to in the HMW-kinino- gen preparation as shown by a control in which pressed as "Relative % Factor." Two zones of correc- plasma was excluded. LMW-kininogen did not correct tion of the APTT of Flaujeac factor deficient plasma Flaujeac deficient plasma (Table I). were recognized at relative mobilities of 0.42 and 0.46 Failure of bovine HMW-kininogen cleavage prod- (Fig. 4). These mobilities corresponded exactly with the two subspecies of HMW-kininogen identified in a O., 16 parallel gel which was stained for protein. Since the stained peaks and recovered activity were of equal I0 a 4 magnitude, HMW-kininogen a and HMW-kininogen b I I I I appeared equipotent. 9 ,0 Failure ofbovine HMW-kininogen cleavage products 12 to correct the prolonged APTT of Flaujeac plasma. I I HMW-KININOGEN The results of attempted reconstitution of the pro- IUIn% 0 longed APTT of Flaujeac plasma by the three cleavage 0 products of bovine HMW-kininogen are shown in 0.04r I 8 I. I Fig. 5. Compared to native bovine HMW-kininogen, I II-I kinin-free HMW-kininogen gave partial correction but 0.03F 6 ^ peptide 1-2 and bradykinin were completely inef- fectual. In a separate experiment, an equimolar mix- 0.02 1 ture of the three fragments did not contribute more than the bovine kinin-free HMW-kininogen alone. It I --0% may be seen that 0.6 ,ug of native HMW-kininogen 0.01 F was as effective as 100 ,u g of kinin-free HMW- 94 kininogen (Fig. 5). Attempts were then made to deter- -k0.05 10 % mine whether the kinin-free HMW-kininogen retained .. partial corrective ability or whether residual native 6 t HMW-kininogen in the preparation was responsible k' 0.04 KININ -FREE for the correction. HMW- KININOGEN Samples of bovine HMW-kininogen or bovine kinin- 0.03 free HMW-kininogen were applied to gradients of sucrose and sedimented overnight. Fractions were col- 0.021 lected and tested for correction of the APTT with Flaujeac plasma. Corrective ability was determined relative to that of pooled normal human plasma and 0.01 expressed as "Relative Activity." The fractions which corrected Flaujeac plasma sedimented identically 5 50 75 100 (Fig. 6). However, the bulk of the protein in the (Top) PERCENT OF GRADIENT (Bottom) native HMW-kininogen differed from the lighter FIGURE 6 Sucrose gradient ultracentrifugation of bovine sedimenting kinin-free HMW-kininogen fragment, HMW-kininogen and kinin-free HMW-kininogen. Flaujeac indicating that a small amount of undegraded native factor activity (O --- 0) sediemented identically with native HMW-kininogen was responsible for the activity of HMW-kininogen. Bovine High Molecular Weight Kininogen Permeability Enhancing Peptide 1399 TABLE I cleavage products released from bovine HMW-kinino- Reconstitution of Kaolin-Activated Euglobulin gen by plasma kallikrein revealed the peptide 1-2 Lysis with Bovine HMW-Kininogen to be an effective mediator of increased vascular Euglobulin lysis permeability in the rabbit (Fig. 7). Serial dilutions of peptide 1-2 in sterile normal saline were injected Plasma source Additions Time intracutaneously in rabbits previously blued with in- min travenous Evans blue due. Dilutions of bradykinin and histamine dihydrochloride were used for compara- Exp. 1 Exp. 2 tive studies and served as positive controls. Sterile Saline >120. 80.0 normal saline, a negative control, produced no more Normal 6.2 6.1 than 1-1.5 mm diameter blueing due to the trauma of injection. In the range of 2 pmol-0.8 nmol Flaujeac 110. 70.2 (0.025-10 ,ug/0.1 ml injection) peptide 1-2 produced Flaujeac + normal - 7.7 9.2 permeability in a dose-dependent manner which was (1: 1) intermediate in effect between that produced by bradykinin or histamine base in rabbit skin (Fig. 7). Flaujeac HMW-Kininogen 13.9 13.3 After the demonstration of its permeability activity, 25 ug a sample of peptide 1-2 was layered onto a gradient Flaujeac HMW-Kininogen 34.1 21.8 of sucrose and sedimented for 16 h. Fractions collected 2.5 ,ug from the gradient were diluted 1:20 in sterile normal Flaujeac HMW-Kininogen 47.9 23.6 saline for determination of protein content and per- 0.25 ug meability activity. A single protein peak sedimented in a range consistent with the known molecular weight Flaujeac HMW-Kininogen 63.9 47.2 of peptide 1-2, viz. 12,584 daltons (Fig. 8, top). 0.025 ,ug As shown in Fig. 8 (bottom) the vascular permeability - HMW-Kininogen NT* 81.2 25 Ag TABLE II Flaujeac LMW-Kininogen NT 78.1 Failure of Cleavage Products of Bovine HMW-Kininogen 25 ,ug to Correct the Kaolin-Activated Euglobulin Lysis of Flaujeac Plasma LMW-Kininogen NT 45.6 25 ug Euglobulin lysis Plasma * NT, not tested. source Additions Time min ucts to correct fibrinolysis. Bovine kinin-free HMW- Normal KF-HMW-Kininogen* 25 Mug 7.5 kininogen, peptide 1-2, or synthetic bradykinin were tested singly and in a mixture containing equimolar Normal Fragment 1-2 10 ug 8.2 amounts of these fragments and found to be ineffec- Normal Bradykinin 5 ug 10.0 tive in reconstituting the fibrinolytic defect in Flaujeac plasma (Table II). In addition they did not enhance KF-HMW-Kininogen 10 Mug lysis by normal plasma (Table II). Normal Fragment 1-2 2 Ag 10.0 Correction of absent PFIdil formation in Flaujeac plasma with bovine HMW-kininogen. Flaujeac Bradykinin 0.2,ug plasma, which failed to form the PF/dil at a 1:100 Flaujeac KF-HMW-Kininogen 25 ,ug 58.9 dilution in rabbits was reconstituted in a dose-de- pendent manner by bovine HMW-kininogen (Table Flaujeac Fragment 1-2 10 ,ug 56.8 III). LMW-kininogen demonstrated no correction. Flaujeac Bradykinin 5 Ag 61.0 Although peptide 1-2 and bradykinin did not re- KF-HMW-Kininogen 10 Mg constitute the absent PF/dil fonnation of Flaujeac plasma, these fragments combined in equimolar Flaujeac Fragment 1-2 2 Mig 54.7 amounts with kinin-free HMW-kininogen caused sub- Bradykinin 0.2 Mg stantial correction, a reaction to be explained below. Biologic activity of a fragment of bovine HMW- * KF-HMW-Kininogen, kinin-free high molecular weight kininogen, peptide 1-2. Further investigation of the kininogen.

1400 Matheson, Miller, Lacombe, Han, lwanaga, Kato, and Wuepper TABLE III found in the rabbit (Fig. 9). Peptide 1-2 appeared Reconstitution of the PFIdil Phenomenon slightly more potent in the rat skin. with Bovine HMW-Kininogen Synergism of peptide 1-2 and bradykinin. Since Mean diameter bradykinin and peptide 1-2 are released simul- Plasma taneously in equimolar quantities from bovine HMW- source Additions Blueing kininogen by plasma kallikrein, the permeability ac- mm tivities of equimolar combinations of bradykinin and peptide 1-2 were determined and compared to those - Saline 2 f* of bradykinin or peptide 1-2 alone in the same ani- Normal 8 mal. The combination of bradykinin and peptide 1-2 produced a synergistic increase in vascular per- Flaujeac 1 f meability activity (Fig. 10). The addition of an equi- Flaujeac HMW-Kininogen 10 ,ug 8 molar amount of peptide 1-2 to bradykinin produced permeability equal to that expected from a 10-20- Flaujeac HMW-Kininogen ltg1 6 fold increase in bradykinin alone. The synergistic Flaujeac HMW-Kininogen 0.1 Zg 4 increase in permeability persisted even with the addi- tion of amounts of peptide 1-2 which were less than Flaujeac HMW-Kininogen 0.01 jig 3 f one-tenth of the threshold permeability dose of the HMW-Kininogen 10 ,ug 4 peptide alone (2 pmol). Time course of permeability reactions. The dura- HMW-Kininogen 1 ug 2 f tion of increased vascular permeability was determined Flaujeac LMW-Kininogen 10 ,ug 2f by injecting the Evans blue dye after the completion of a series of timed sequential intracutaneous injec- * f, Faint blueing. tions. The animal was sacrificed immediately after blueing and the diameters of dye leakage were meas- ured. A representative time-course study is shown in activity sedimented identically with the peptide 1-2 Fig. 11. The onset and duration of permeability due and was clearly distinct from that induced by in- to peptide 1-2 was distinct from that of either hista- creasing concentrations of sucrose alone. mine or bradykinin. The enhanced permeability of When assayed in the skin of the rat or guinea peptide 1-2 was maximal within 1-2 min and ter- pig, peptide 1-2 produced increased vascular perme- minated in 5-10 min. Histamine induced permea- ability in a dose-dependent manner similar to that bility peaked within 5-10 min and was of 18-20 min

16r

- 12

E"I

4

* 0.15 M NoCI

lo-13 to-It 10-11lo-too-10* O-I K0r Moles Injected FIGURE 7 Comparison of bovine peptide 1-2 (O 0), bradykinin (EO [O), or histamine (A - A) permeability responses in the rabbit. Brackets denote the range of response; symbols represent the mean of 3 to 8 determinations at the stated dosage levels.

Bovine High Molecular Weight Kininogen Permeability Enhancing Peptide 1401 (Fig. 12), showing a decrease in permeability equiva- 0 0.14 lent to that of a 10-fold decrease in the dose of frag- 0.12 ment injected. By conitrast, histamine-induced perme- t 0.10 ability was completely inhibited in the same animal ,.08 by chlorpheniramine maleate (Fig. 12). ea 0.06 Further investigation of peptide 1-2. Attempts were made to delineate further biologic activities of 0.04 peptide 1-2. In concentrations up to 10 ,g/ml, t 0.02 peptide 1-2 did not cause smooth muscle contrac- I tion in a rat uterine horn responsive to 5 ng/ml of bradykinin. Nor did equal concentrations of the frag- ment contract a guinea pig ileum responsive to hista- 8 0-0 Peptide 1-2 mnine. I$ 1. J\\ *---O Sucrose With 51Cr-labeled humilan granulocytes in chemotac- I.. tic assays, peptide 1-2 (0.05-50,tg/ml) did not cauise -Z! 6 q) increased directional movement of cells whein tested F. I I I~~~~~~~~~~~~~~~~~~~~~ tj in three separate experimenits. In standard aggregome- (Z.) 4 0 0 U try determinations peptide 1-2 (0.1-10 ,g/0.45 ml ,tz did not cause of IZ-. \O platelet-rich-plasma) aggregation plate- -CA t3 2 0-1°O lets, nor did it alter response patterns to adenosine q) _ ll. diphosphate, ristocetin, epinephrine, or collagen when (Z preincubated with the platelet-rich-plasma or with the aggregating agents themselves. 25 50 75 100 (Top) Percent of Gradient (Bottom) DISCUSSION FIGURE 8 Sucrose gradient ultracentrifugation of bovine peptide 1-2. Protein (0 0) and permeability activity In the data reported here, bovine HMW-kininogen, (O 0) from the same gradient. The permeability activitY a monomeric plasma glycoprotein of mol wt 76,000 of a gradient of sucrose (U --- *) is shown. significantly corrected the activated partial thrombo- plastin time of Flaujeac plasma in a dose-dependent duration. Injected together in equimolar amounts, manner, while LMW-kininogen did not. The presenice peptide 1-2 and bradykinin produced permeability of two bovine HMW-kininogen subspecies, a and b, which was maximal within 5 min, was enhanced was confirmed, and thev were found to be equi- over that of bradykinin alone, and was significantly potent in their ability to reconstitute this plasmia. prolonged, lasting 30-40 min. To date, it has not yet been determined whether the Chlorpheniramine effect on peptide 1-2 induced two subspecies represent a postsynthetic modification permeability. To further compare the permeability of the H.MW-kininogen molecule or an artifact which induced bv peptide 1-2 with that of histamine, the results during the purification process. These stub- susceptibility of permeability by peptide 1-2 to inhibi- species have been shown to have the same number tion by a histamine antagonist was evaluated. Pep- and composition of trvptic peptides (19) and appear to tide 1-2 induced permeability was partiallv inhibited differ only in that a single proteolytic cleavage has by 10 mg of intravenous chlorpheniramine maleate taken place at the C-terminal end of the bradykinin moiety. This limited proteolvtic cleavage within an intrachain disulfide loop in H'MW-kininogen leads to an alteration in charge characteristics (Figs. 1, 3, 4). c-c The data reported here also showed bovine HMW- kininogen significantly repaired the fibrinolytic and PF-dil forming defects of Flaujeac plasma in a dose- RJ b dependent fashion while LMW-kininogen did not. 0 O ~ X;neo P 1 In addition to analysis of native bovine HMW- M kininogen, we studied the fragmentation products of ° OD5LOO0.5NOCI HMW-kininogen released by plasma kallikreini. When 10. IO-:2 10- §-' ,0,9 Moles Injected highly purified bovine HMW-kininogen was inctubated FIGURE 9 Permeability responses to bovine peptide 1-2 with bovine plasma kallikrein (0.1% wt:wt) equimolar in the rat (O 0), rabbit (0 0), or guinea pig amounts of bradvkinin, kinin-free HMW-kininogen (A A). (21), and peptide 1-2 were rapidlyT released (22); 1402 Matheson, Miller, Lacombe, Han, Iwanaga, Kato, and Wuepper 18j

16

14

12 Fj

.* 10

6

I0 9 Moles Injected

FIGURE 10 Synergistic permeability activity of bradykinin and bovine peptide 1-2 (A A) in the rabbit. The addition of an equimolar amount of peptide 1-2 to bradykinin pro- duced permeability equal to that of a 10-fold increase in bradykinin alone. the reaction went to completion in 30 min. Peptide The inability of the three fragmentation products 1-2 was so designated since upon prolonged incuba- of bovine HMW-kininogen, alone or in equimolar tion with plasma kallikrein, it was subsequently combination to correct Flaujeac factor deficient plasma cleaved into a glycopeptide (fragment 1) (28), and a suggested that the function of this molecule may be "histidine-rich peptide" (Fragment 2) (24) in a reaction which required 18 h to complete (22). Bradykinin is positioned in an intrachain disulfide loop (19) within the kininogen. Within the same disulfide loop the 14 N-terminus of peptide 1-2 is linked to the C- terminal residue of bradykinin (21). Thus, plasma kalli- 12 Peptide 1-2 O-o krein produces three peptide bond cleavages within 10 Histomine A-A bovine HMW-kininogen to generate kinin-free HMW- Brodykinin o-a kininogen, peptide 1-2, and bradykinin. Amino acid Peptide 1-2 + Brodykinin A-A 8 composition studies support the contention that the kinin-free HMW-kininogen, bradykinin, and peptide 1-2 comprise the entire bovine HMW-kininogen molecule (21). Bovine kinin-free HMW-kininogen appeared to give partial correction of Flaujeac plasma but peptide 1-2 2 and bradykinin were ineffectual. The apparent ac- tivity of the kinin-free HMW-kininogen was shown I0 20 30 40 (Fig. 6) to be due to a small amount of undegraded Age of Lesion When Dye Injected (min) native HMW-kininogen (less than 1%) in the kinin- FIGURE 11 Time course of permeability produced by equi- free HMW-kininogen solution. Equimolar combina- molar doses of bovine peptide 1-2 (O 0), histamine tions of the three fragments gave no more correction base (A A), bradykinin (El El), and bradykinin than bovine kinin-free HMW-kininogen alone. + peptide 1-2 (A A) in the rabbit. Bovine High Molecular Weight Kininogen Permeability Enhancing Peptide 1403 8 r (21, 22) to learn whether this fragment had biologi- p Before Peptide 1-2 cal activity. There is no counterpart for peptide 6 1-2 in bovine LMW-kininogen (22). The complete 0 amino acid sequence of bovine peptide 1-2 has been 4 o0 0 *After reported (23, 24). It is a highly basic 12,584 dalton glycopeptide rich in histidine and glycine. In this 2 investigation, the isolated bovine peptide 1-2 was 0 - 0 0.15 M NUCI found to produce increased vascular permeability when injected intracutaneously in the rabbit, rat, or guinea

e ~ A Before pig. It produced increased permeability in the range of approximately 10 pmol- 10 nmol which was dose Histamine Bose A 6;- dependent and intermediate in potency between bradykinin or histamine in rabbit skin. 4 Permeability induced by bovine peptide 1-2 was further distinguished from that of bradykinin or hista- 2 mine in the rabbit by a more rapid onset of maxi- !. After A -- mal activity and shorter duration of effect. The per- A A *0.15 M A-~~~A A A NoCI meability caused by peptide 1-2 was only partially 10o1- lo-lo 10-9 Moles Injected reduced by pretreatment with an antihistamine, whereas that of histamine was completely inhibited. FIGURE 12 Chlorpheniramine effect on perneability caused More significantly, and simulating the simultaneous by bovine peptide 1-2. release of peptide 1-2 and bradykinin from bovine HMW-kininogen, intracutaneous injection of equi- dependent upon their combined presence in the native molar combinations of bovine peptide 1-2 and brady- molecule. Since peptide 1-2 is a highly basic frag- kinin produced a synergistic permeability response ment, it might subserve a strong binding function, which was rapid in onset and of longer duration than possibly in association with negatively charged sur- that of bradykinin alone. Preliminary experiments faces, and exert conformational effects upon one or both in this laboratory have suggested that the 4,584 of the other two components of the contact activation dalton fragment 2 moiety (released from peptide 1-2 system; namely, Hageman factor and Fletcher factor upon prolonged incubation with plasma bovine kalli- (). Additional experiments are planned to krein) has permeability properties similar to those of test this hypothesis. peptide 1-2. Fragment 1 did not have significant Coincident with the presentation of this work in permeability increasing properties. abstract form (1), McGregor et al. presented data on This is the first demonstration that a polypeptide the reconstitution of a similar functional deficiency by fragment released by plasma kallikrein from HMW- using bovine proteins (32). They studied a fixed kininogen increases vascular permeability and po- quantity of bovine HMW-kininogen, 100 ,g, in the ac- tentiates the action of bradykinin. Prostaglandins of tivated partial thromboplastin time test and also the E series have been reported to potentiate brady- learned that it was corrective for this plasma. Similarly kinin's effects (33, 34), and bradykinin potentiating 100 ,ug bovine kinin-free HMW-kininogen gave a cor- factors have been isolated from the venom of rep- rection which was not statistically different from that of tiles. The mechanism of action of these venom HMW-kininogen. These findings, which differ from potentiating factors is by interference with a dipep- those reported here by more than two orders of magni- tide hydrolase (kininase II, converting en- tude, are explained by examination of the dose- zyme, carboxydipeptidase) (35) which is present in response relationship (Figs. 2 and 5). We have found serum and is abundant on the membranes of endo- that addition of 20-100 ,ug of bovine HMW-kininogen thelial cells throughout the body (36) and has also did not give shorter coagulation times than 20 ,ug, been recognized in the proximal tubule of the kidney and that 100 gg was even slightly inhibitory. A thorough (37). Work is now underway in our laboratories to analysis of the dose response is thus necessary to explain the mechanism by which synergy takes place compare the two molecular forms of kininogen. between peptide 1-2 and bradykinin. Our data support the contention that the partially Except for bradykinin, the fragmentation products corrective activity of the bovine kinin-free HMW- produced from human HMW-kininogen by plasma kininogen was due to the presence of trace amounts kallikrein have not been characterized. Such investi- of intact HMW-kininogen in the preparation. gation will require highly purified human HMW- Particular attention was given to the newly- kininogen. The ability of bovine HMW-kininogen to described bovine glycopeptide fragment (peptide 1-2) repair the defects of Flaujeac plasma suggests a degree 1404 Matheson, Miller, Lacombe, Han, Iwanaga, Kato, and Wuepper of functional and structural similarity with human M. Seavey, J. A. Guimaraes, J. V. Pierce, and A. P. HMW-kininogen. However, until human HMW- Kaplan. 1975. Williams trait. Human kininogen deficiency with diminished levels of plasminogen proactivator and kininogen is better characterized and its fragmentation prekallikrein associated with abnormalities of the products identified, the data presented here must Hageman factor-dependent pathways. J. Clin. Invest. remain suggestive for the human kinin-forming system. 56: 1650-1662. These observations do serve to substantiate further 11. Donaldson, V. H., H. I. Glueck, M. A. Miller, H. Z. Movat, and F. Habal, 1976. Kininogen deficiency in a functional role for a kininogen, HMW-kininogen, Fitzgerald trait: Role of high molecular weight kininogen in the plasma pathways of coagulation and fibrinolysis in clotting and fibrinolysis. J. Lab. Clin. Med. 87: 327- and in the permeability reactions observed with diluted 337. plasma. The new basic permeability agent, peptide 12. Donaldson, V. H., H. I. Glueck, H. Z. Movat, F. M. consideration in set- Habal, and M. A. Miller. 1975. Deficiency of plasma 1-2, will require careful those kininogen in Fitzgerald trait. Clin. Res. 23: 522A. (Abstr.) tings in which plasma kallikrein may be activated 13. Lutcher, C. L. 1976. Reid trait: A new expression of in vivo and simultaneously release bradykinin and pep- high molecular weight kininogen (HMW-kininogen) tide 1-2 from HMW-kininogen. More work is required deficiency. Clin. Res 24: 47A. (Abstr.) to explain the synergy between bradykinin and 14. Jacobson, S. 1966. Separation of two different substrates for plasma kinin-forming enzymes. Nature. (Lond.). peptide 1-2. 210: 98-99. 15. Jacobsen, S., and M. Kriz. 1967. Some data on two ACKNOWLEDGMENTS purified kininogens from human plasma. Br.J. Pharmacol. 29: 25-36. This work was supported by U. S. Public Health Service 16. Yano, M., S. Nagasawa, K. Horiuchi, and T. Suzuki. grants HL-18618 and AM-05300, the Medical Research 1967. Separation of a new substrate, kininogen-I, for Foundation of Oregon, and the Harlow Foundation. plasma kallikrein in bovine plasma.J. Biochem. (Tokyo). 62: 504-506. 17. Yano, M., S. Nagasawa, and T. Suzuki. 1971. Partial REFERENCES purification and some properties of high molecular 1. Wuepper, K. D., D. R. Miller, M. J. LaCombe, H. Kato, S. weight kininogen, bovine kininogen-I. J. Biochem. Iwanaga, and Y. Han. 1976. Flaujeac factor deficiency: (Tokyo). 69: 471-481. Reconstitution with highly purified bovine HMW-kinino- 18. Komiya, M., H. Kato, and T. Suzuki. 1974. Bovine gen. Clin. Res. 24: 110A. (Abstr.) plasma kininogens. I. Further purification of high 2. Matheson, R. T., D. R. Miller, Y. N. Han, S. Iwanaga, molecular weight kininogen and its physicochemical H. Kato, and K. D. Wuepper. 1976. Delineation of a new properties. J. Biochem. (Tokyo). 76: 811-822. perneability-enhancing peptide released by plasma kalli- 19. Komiya, M., H. Kato, and T. Suzuki. 1974. Bovine plasma krein from HMW-kininogen (Flaujeac factor). Clin. Res. kininogens. II. Microheterogeneities of high molecular 24: 314A. (Abstr.) weight kininogens and their structural relationships. 3. Waldmann, R., and J. P. Abraham. 1974. Fitzgerald factor: J. Biochem. (Tokyo). 76: 823-832. A heretofor unrecognized coagulation factor. Blood. 20. Komiya, M., H. Kato, and T. Suzuki. 1974. Bovine plasma 44: 934. (Abstr.) kininogens. III. Structural comparison of high molecular 4. Wuepper, K. D., D. R. Miller, and M. J. LaCombe. 1975. weight and low molecular weight kininogens.J. Biochem. Flaujeac trait: Deficiency of kininogen in man. Fed. (Tokyo). 76: 833-845. Proc. 34: 859. (Abstr.) 21. Han, Y. N., M. Komiya, H. Kato, S. Iwanaga, and 5. Colman, R. W., A. Bagdasarian, R. C. Talamo, M. Seavey, T. Suzuki. 1975. Primary structure of bovine high C. F. Scott, and A. P. Kaplan. 1975. Williams trait. molecular weight kininogen: Chemical compositions Combined deficiency of plasma plasminogen proacti- of kinin-free kininogen and peptide fragments, released vator, kininogen and a new procoagulant factor. Fed. by plasma kallikrein. FEBS (Fed. Eur. Biochem. Soc.) Proc. 34: 859. (Abstr.) Lett. 57: 254-258. 6. Saito, H., 0. D. Ratnoff, R. Waldmann, and J. P. Abraham. 22. Kato, H., Y. N. Han, S. Iwanaga, T. Suzuki, and M. 1975. Fitzgerald trait. Deficiency of a hitherto unrecog- Komiya. 1975. Bovine plasma HMW and LMW kinino- nized agent, Fitzgerald factor, participating in surface- gens: Isolation and characterization of the polypeptide mediated reactions of clotting, fibrinolysis, generation fragments produced by plasma and tissue . of , and the property of diluted plasma enhanc- In Kinins-Pl*rmacodynamics and Biological Roles. ing vascular permeability (PF/dil). J. Clin. Invest. 55: F. Sicuteri, and N. Back, editors. Plenum Publishing 1082- 1089. Corporation. New York. 135-150. 7. Lacombe, M-J. 1975. Deficit constitutionnel en un 23. Han, Y. N., H. Kato, S. Iwanaga, and T. Suzuki. 1976. nouveau facteur de la coagulation intervenant au niveau Bovine plasma high molecular weight kininogen: The de contact: le facteur "Flaujeac". C. R. Hebd. Seances amino acid sequence of fragment 1 (glycopeptide) re- Acad. Sci. 280: 1039-1041. leased by the action of plasma kallikrein and its loca- 8. Lacombe, M-J., B. Varet, and J-P. Levy. 1975. A hitherto tion in the precursor protein. FEBS (Fed. Eur. Biochem. undescribed plasma factor acting at the contact phase Soc.) Lett. 63: 197-200. of blood coagulation (Falujeac factor): Case report and 24. Han, Y. N., M. Komiya, S. Iwanaga, and T. Suzuki. coagulation studies. Blood. 46: 761-768. 1975. Studies on the primary structure of bovine high- 9. Wuepper, K. D., D. R. Miller, and M. J. Lacombe. 1975. molecular weight kininogen. Amino acid sequence of a Flaujeac trait. Deficiency of human plasma kininogen. fragment ("Histidine-rich peptide") released by plasma J. Clin. Invest. 56: 1663-1672. kallikrein.J. Biochem. (Tokyo) 77: 55-68. 10. Colman, R. W., A. Bagdasarian, R. C. Talamo, C. F. Scott, 25. Yano, M., H. Kato, S. Nagasawa, and T. Suzuki. 1967. Bovine High Molecular Weight Kininogen Permeability Enhancing Peptide 1405 An improved method for the purification of kininogen- 32. McGregor, R. K., A. G. Scicli, R. Waldmann, J. P. II from bovine plasma. J. Biochem. (Tokyo). 62: 386- Abraham, and 0. A. Carretero. 1976. Effect of bovine 388. high molecular weight kininogen and its fragments on 26. Takahashi, H., S. Nagasawa, and T. Suzuki. 1972. Studies Fitzgerald trait plasma. Clin. Res. 24: 108A. (Abstr.). on Prekallikrein of bovine plasma. I. Purification and 33. Willaims, T. J., and J. Morley. 1973. Prostaglandins as properties. J. Biochem. (Tokyo). 71: 471-483. potentiators of increased vascular permeability in in- 27. Komiya, M., S. Nagasawa, and T. Suzuki. 1972. Bovine flammation. Nature (Lond.) 246: 215-217. prekallikrein activator with functional activity as Hage- 34. Moncada, S., S. H. Ferreira, and J. R. Vane. 1973. man factor. J. Biochem. (Tokyo). 72: 1205-1218. Prostaglandins, aspirin-like drugs and the oedema of 28. Miles, A. A., and D. L. Wilhelm. 1955. Enzyme- inflammation. Nature (Lond.) 246: 217-219. like globulins from serum reproducing the vascular 35. Ferreira, S. H., L. J. Greene, V. A. Alabaster, Y. S. phenomena of inflammation. I. An activable permea- Bakhle, and J. R. Vane. 1970. Activity of various bility factor and its inhibitor in guinea-pig serum. fractions of bradykinin potentiating factor against angio- Br. J. Exp. Pathol. 36: 71-81. tensin I converting enzyme. Nature (Lond.) 225: 379- 29. Wuepper, K. D. 1973. Prekallikrein deficiency in man. 380. J. Exp. Med. 138: 1345-1355. 36. Ryan, J. W., U. S. Ryan, D. R. Schultz, C. Whitaker, 30. Davis, B. J. 1964. Disc electrophoresis-II. Method A. Chung, and F. E. Dorer. 1975. Subcellular localiza- and application to human serum proteins. Ann. N. Y. tion of pulmonary angiotensin-converting enzyme Acad. Sci. 121: 404-427. (kininase II). Biochem. J. (Tokyo) 146: 497-499. 31. Gallin, J. I., R. A. Clark, and H. R. Kimball. 1973. 37. Caldwell, P. R. B., B. C. Seegal, K. C. Hsu, M. Das, Granulocyte chemotaxis: An improved in vitro assay and R. L. Soffer. 1976. Angiotensin-converting enzyme: employing 5'Cr-labeled granulocytes. J. Immunol. 110: Vascular endothelial localization. Science (Wash. D. C.). 233-240. 191: 1050-1051.

1406 Matheson, Miller, Lacombe, Han, Iwanaga, Kato, and Wuepper