Proc. Nati. Acad. Sci. USA Vol. 74, No. 10, pp. 4636-4640, October 1977 Medical Sciences Role of high-molecular-weight kininogen in surface-binding and activation of coagulation Factor XI and prekallikrein (Hageman factor/contact activation/fibrinolysis) ROGER C. WIGGINS, BONNO N. BOUMA*, CHARLES G. COCHRANE, AND JOHN H. GRIFFIN Department of Immunopathology, Scripps Clinic and Research Foundation, La Jolla, California 92037 Communicated by Frank J. Dixon, August 1, 1977

ABSTRACT In the contact phase of activation of the solubilized lactoperoxidase method (12). The 125I-labeled Factor kinin-forming, intrinsic clotting, and fibrinolytic systems, XI (125I-Factor XI), 5 ,uCi/,ug, retained its procoagulant ac- high-molecular-weight kininogen acts as a cofactor for the ac- tivation of Factor XI, prekallikrein, and Hageman factor. One tivities at 3.6 clotting units/ml. mechanism by which high-molecular-weight kininogen acts as Prekallikrein was isolated from normal human plasma by a cofactor has been studied by using 125f-labeled Factor XI and a method similar to that described for Factor XI (11). This in- prekallikrein in kaolin-activated normal human plasma and volved the chromatography of prekallikrein on DEAE-, QAE-, plasmas deficient in high-molecular-weight kininogen and and SP-Sephadex, and on concanavalin A-Sepharose, and su- Hageman factor. High-molecular-weight kininogen was found crose density gradient ultracentrifugation (B. N. Bouma and to be essential for normal binding and cleavage of both Factor XI and prekallikrein on the kaolin surface. Hageman factor was J. H. Griffin, unpublished data). The protein was more than essential for cleavage but not for binding of Factor XI and 95% homogeneous on NaDodSO4/polyacrylamide gels. The prekallikrein to kaolin. In normal plasma 80% of the activated purified protein had a specific clotting activity of 19 units/mg, Factor XI remained surface-bound, whereas 80% of the kalli- where 1 clotting unit is defined as the amount of activity present krein was not surface-bound. These findings are consistent with in 1 ml of citrated normal human plasma. Prekallikrein was the hypothesis that, in the initial phase of contact activation, radiolabeled with '25iodine by the chloramine-T method (13); high-molecular-weight kininogen links both Factor XI and prekallikrein to the exposed surface where they are activated after radiolabeling 12% of the material appeared as kallikrein. by surface-bound activated Hageman factor. Once activated, Only the heavy chain (45,000 Mr) was radiolabeled by the the Factor XI molecules remain localized at the site of activa- above procedure; therefore the light chain (35,000 Mr) con- tion, in contrast to the kallikrein molecules which are found taining the active site (and which could bind to inhibitors in largely in the surrounding plasma. plasma) was not seen on reduced NaDodSO4/polyacrylamide gels. The 125I-labeled prekallikrein' (125I-prekallikrein), 3 Activation of Hageman factor (Factor XII) upon a negatively ,uCi/,ug, retained its procoagulant activity at 2.3 clotting charged surface initiates intrinsic coagulation, fibrinolysis, and units/ml. the generation of vasoactive peptides (1-5). High-molecular- High Mr Kininogen was isolated from freshly prepared weight (Mr) kininogen is a cofactor for the optimal activation normal human plasma containing acid/citrate/dextrose anti- of surface-bound Hageman factor by kallikrein (6-8) and for coagulant (1 liter contains 13.6 g of citric acid, 25 g of sodium the activation of prekallikrein and Factor XI by surface-bound citrate, and 20 g of dextrose). Plasma (1.4 liters) was chroma- activated Hageman factor (6-8). Prekallikrein is complexed tographed on DEAE-Sephadex exactly as described for the to high Mr kininogen in plasma (9). Recent studies (10; R. C. isolation of Hageman factor (14). The pool of fractions con- Wiggins, B. N. Bouma, C. G. Cochrane, and J. H. Griffin, un- taining activity that corrected the clotting defect of plasma published data) suggest that Factor XI is also complexed to high deficient in high Mr kininogen was brought to 47% saturation Mr kininogen in plasma. In this study we demonstrate that one of (NH4)2SO4 by addition of saturated (NH4)2SO4 at 4°. After mechanism by which high Mr kininogen acts as a cofactor for the mixture was stirred for 1 hr, the precipitate was collected the activation of both Factor XI and prekallikrein in plasma is by centrifugation at 5600 X g for 90 min. The precipitate was to bind both these molecules to the surface where they are ac- dissolved in 100 ml of buffer containing 50 mM acetate, 75 mM tivated by surface-bound activated Hageman factor. We also NaCl, 0.5 mM EDTA, 0.5 mM benzamidine, 25 ,ug of Poly- demonstrate that, while activated Factor XI remains associated brene per ml, 0.01% sodium azide, at pH 5.3, and the sample with the surface, kallikrein dissociates from the surface. was applied to an SP-Sephadex column which was developed exactly as described as step 5 in the isolation of Hageman factor MATERIALS AND METHODS (14). The high Mr kininogen, which was eluted in the terminal Purified proteins region of the peak of activity from this column, was greater than 90% homogeneous as judged on NaDodSO4 gels, and it exhib- Factor XI was isolated from normal human plasma as de- : scribed elsewhere (11) and was determined to be more than 95% ited a specific clotting activity of 12 2 clotting units/mg of homogeneous on sodium dodecyl sulfate (NaDodSO4/poly- protein. acrylamide gels. The purified protein had a specific clotting Deficient plasma activity of 220 units/mg, where 1 clotting unit is defined as the Normal human plasma was made from a pool of blood from 15 amount of activity present in 1 ml of citrated normal human normal adults. factor-deficient was plasma. Factor XI was radiolabeled with '25iodine by the in- Hageman plasma gener- Abbreviations: high Mr kininogen, high-molecular-weight kininogen; The costs of publication of this article were defrayed in part by the 1251-protein, l25I-labeled protein; NaDodSO4, sodium dodecyl sul- payment of page charges. This article must therefore be hereby marked fate. "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate * Present address: Department of Haematology, University Hospital, this fact. Utrecht, The Netherlands. 4636 Downloaded by guest on September 28, 2021 Medical Sciences: Wiggins et al. Proc. Natl. Acad. Sci. USA 74 (1977) 4637

A Kaolin-bound Supernatant A Kaolin-bound Supernatant 4 - 33,000 _48,000 48,000 83,000 8 50r I 2 - + 50000 83,000 4 - 85,000 25k 85,000 +_ B B 5Or

° 4 - 25 6E 4 - n x2- ._YC. 2 _ 50

= - 25 - IA I n- D D - 4 8 _ 50 - 2 O1, 4- ~~~~~25- A 0L 10 i2 30 40 50 10 20 30 40 50 10 20 30 40 50 ~~~~~~~010 20 30 40 50 Gel slice number Gel slice number FIG. 1. The binding and cleavage of 125I-Factor XI in various FIG. 2. The binding and cleavage of 125j-prekallikrein in various plasmas after addition of kaolin and incubation for 2 min at 37°. The plasmas after addition of kaolin and incubation for 2 min at 20°. The kaolin-bound material and supernatant were separated, reduced, and kaolin-bound material and supernatant were separated, reduced, and analyzed on 7.5% NaDodSO4/polyacrylamide gels. (A) Normal plas- analyzed on 7.5% NaDodSO4/polyacrylamide gels. (A) Normal plas- ma; (B) Hageman factor-deflicient plasma; (C) high Mr kininogen- ma; (B) Hageman-factor deficient plasma; (C) high Mr kininogen- deficient plasma; (D) high M, kininogen-deficient plasma plus pu- deficient plasma; (D) high Mr kininogen-deficient plasma plus pu- rified high Mr kininogen. rified high Mr kininogen. ously donated by Gunda Hiatt. High Mr kininogen-deficient 50,000, and 33,000 Mr peaks for Factor XI and the 85,000 and plasma (Fitzgerald trait) was purchased from George King 48,000 Mr peaks for prekallikrein from the kaolin and from the Biomedical. supernatant. A volume correction was made for the aliquot from the supernatant. The proportions of cleaved proteins were Binding studies estimated by calculating the percentage of the total radioac- tivity present in the 50,000 plus 33,000 Mr peaks for Factor XI 125I-Factor XI (0.03 ,ug, 0.15 ,uCi in 2 ,u) or 125I-prekallikrein and the 48,000 Mr peak for prekallikrein. Correspondingly, the (0.14 ,ug, 0.13 ,uCi in 2 ,ul) was added to 30 Al of a 1:1 dilution percentage of each protein that was not cleaved was calculated of plasma in Tris-buffered saline. The plasmas used were nor- as the percentage of total radioactivity in the 83,000 or 85,000 mal human plasma, Hageman factor-deficient plasma, high Mr peak for Factor XI or prekallikrein, respectively. Finally, Mr kininogen-deficient plasma, and high Mr kininogen-defi- in control gels when no kaolin had been added (which gave cient plasma that had been reconstituted by addition of purified profiles identical to those in Figs. 1B and 2B) the proportion high Mr kininogen to a final concentration of 1 clotting unit/ml of radioactivity corresponding to the position of the cleavage of undiluted plasma. After a 15-min incubation of 20°, 30,l fragments was calculated and found to be 5% for 125I-Factor of kaolin suspension (10 mg/ml) in Tris-buffered saline was XI and 12% for 125I-prekallikrein. These corrections were added to the plasma. The mixture was shaken for various times subtracted from the calculated cleavage values to give the at 200 or 370 as indicated, and then centrifuged at 1000 X g for figures presented in Tables 1 and 2. More than 60% of the 2 min at 20°. The supernatant was removed and the kaolin 125iodine added to the plasma was recovered on the poly- pellet was washed three times with 30 l of Tris-buffered saline acrylamide gels. containing 1 mg of bovine serum albumin per ml. NaDodSO4 (10%) (20,ul) containing mercaptoethanol (5%), EDTA (100 mM), and urea (8 M) was added to the kaolin and to a 10-i RESULTS aliquot of the supernatant. These samples were boiled for 7 min Factor XI and then analyzed on 7.5% NaDodSO4/polyacrylamide gels by the method described by Weber et al. (15). After electro- The binding and cleavage of 125I-Factor XI in kaolin-activated phoresis the gels were cut into 1.2-mm slices, and the radioac- normal plasma and in plasmas deficient in Hageman factor and tivity was determined. high Mr kininogen are shown in Fig. 1 and Table 1. Whereas in normal plasma after a 2-min incubation at 370, 87% of the Factor XI was bound to kaolin (Fig. 1A), in the absence of high Calculation of binding and cleavage data Mr kininogen only 14% was bound (Fig. 1C). When high Mr The total bound and free 125I-Factor XI and 125I-prekallikrein kininogen-deficient plasma was reconstituted by addition of was calculated by integrating the radioactivity in the 83,000, purified high Mr kininogen, the binding of 125I-Factor XI to Downloaded by guest on September 28, 2021 4638 Medical Sciences: Wiggins et al. Proc. Nati. Acad. Sci. USA 74 (1977)

Table 1. Binding and cleavage of 125I-Factor XI in various plasmas Kaolin-bound, % Supernatant, % Plasma Total Cleaved Uncleaved Total Cleaved Uncleaved Normal plasma 87 75 12 13 8 5 Hageman factor-deficient plasma 66 <1 66 34 <2 33 High Mr kininogen-deficient plasma 14 <2 13 86 <1 86 High Mr kininogen-deficient plasma + purified high Mr kininogen 79 61 18 21 9 12 The binding and cleavage of 1251-Factor XI in various plasmas after addition of kaolin and incubation for 2 min at 37°. The kaolin-bound material and supernatant were separated, reduced, and analyzed on 7.5% NaDodSO4/polyacrylamide gels. The values given are derived from integration of radiolabeled Factor XI in the 83,000 Mr uncleaved molecule and the 50,000 and 33,000 Mr cleavage fragments shown in Fig. 1.

kaolin increased to 79% (Fig. 1D). In the absence of Hageman DISCUSSION factor, binding was slightly reduced at 66% (Fig. 1B). High Mr Human plasma deficient in high Mr kininogen exhibits gross kininogen, but not Hageman factor, is therefore necessary for abnormalities in contact activation reactions (16-20). The the efficient binding of Factor XI to the kaolin surface. functional role of high Mr kininogen as a cofactor in the acti- Most of the Factor XI bound to kaolin (86%) and in the su- vation of Hageman factor by kallikrein and in the activation in pernatant (62%) was cleaved in normal plasma. However, of Factor XI or prekallikrein by surface-bound activated both Hageman factor-deficient and high Mr kininogen-defi- Hageman factor was subsequently demonstrated in studies cient plasma minimal cleavage of Factor XI occurred. When using purified proteins (6-8). In the present study the role of the high Mr kininogen-deficient plasma was reconstituted with high Mr kininogen in the binding of 125I-Factor XI and 125I_ purified high Mr kininogen, cleavage of the bound Factor XI prekallikrein to kaolin in the milieu of plasma was investigated. occurred normally. These results suggest that both high Mr In addition, the limited proteolytic cleavage that is coincident are kininogen and Hageman factor required for efficient with activation of the two molecules (11, 21) was studied in cleavage of Factor XI in plasma. normal plasma and in plasmas deficient in high Mr kininogen Experiments conducted over shorter incubation periods at and Hageman factor. Previous studies utilizing 125I-Hageman 200 and longer incubation periods at 370 gave substantially factor added to normal plasma and plasmas deficient in coag- was similar results, showing that high Mr kininogen required ulation factors have provided new insight into the binding and for normal binding and cleavage levels to be attained, and that proteolytic activation of Hageman factor during contact acti- once bound, the 125I-Factor XI remained attached to the ka- vation (22). olin. The experiments with 125I-Factor XI demonstrate that high Mr kininogen is required for normal extensive binding of Factor Prekallikrein XI to kaolin in plasma. The fact that the addition of purified The binding and cleavage of 125I-prekallikrein in kaolin-acti- high Mr kininogen to the kininogen-deficient plasma recon- vated normal plasma and plasmas deficient in Hageman factor stituted the binding of 125I-Factor XI to kaolin shows that the and high Mr kininogen are shown in Fig. 2 and Table 2. After binding abnormality is directly related to high Mr kininogen a 2-min incubation at 200 in normal plasma most of the pre- and not to an inhibitor present in the deficient plasma. Since kallikrein was cleaved both on the kaolin (64%) and in the su- in Hageman factor-deficient plasma the binding of 125I-Factor pernatant (62%) (Fig. 2A). In the absence of high Mr kininogen XI to kaolin is extensive although there is no proteolytic (Fig. 2C) and Hageman factor (Fig. 2B) cleavage was minimal cleavage of Factor XI, we conclude that high Mr kininogen (i.e., less than 0.5%), whereas upon reconstitution of the high mediates the binding of unactivated Factor XI to kaolin. This Mr kininogen-deficient plasma with purified high Mr kinino- suggests that the formation of surface-bound complexes of gen, cleavage occurred normally (Fig. 2D). Therefore, both Factor XI and high Mr kininogen may occur before the acti- high Mr kininogen and Hageman factor are required for the vation of Factor XI and, indeed, may be a prerequisite for ef- efficient cleavage of prekallikrein in plasma. ficient activation of Factor XI. Fig. 3 summarizes four hy- In high Mr kininogen-deficient plasma only 2% of the pothesized reactions that involve Factor XI and high Mr kini- 125I-prekallikrein was bound to kaolin in contrast to 14% bound nogen. The first reaction, a reversible equilibrium for which in normal plasma. Reconstitution of high Mr kininogen-defi- the equilibrium constant is unknown, describes complex for- cient plasma with purified high Mr kininogen increased the mation in plasma between the two molecules. The reaction is binding to normal levels. supported by studies of these two molecules in plasma analyzed Experiments conducted over longer incubation periods up by Sephadex G-200 chromatography (ref. 10; R. C. Wiggins, to 10 min at 37° gave substantially similar results. B. N. Bouma, C. G. Cochrane, and J. H. Griffin, unpublished

Table 2. Binding and cleavage of 1251-prekallikrein in various plasmas Kaolin-bound, % Supernatant, % Plasma Total Cleaved Uncleaved Total Cleaved Uncleaved Normal plasma 14 9 5 86 53 33 Hageman factor-deficient plasma 12 <1 12 88 <1 88 High Mr kininogen-deficient plasma 2 <1 2 98 <1 98 High Mr kininogen-deficient plasma + purified high Mr kininogen 12 7 5 88 57 31 The binding and cleavage of 125I-prekallikrein in various plasmas after addition of kaolin and incubation for 2 min at 200. The kaolin-bound- material and supernatant were separated, reduced, and analyzed in 7.5% NaDodSO4/polyacrylamide gels. The values given are derived from integration of reduced radiolabeled prekallikrein in the 85,000 Mr uncleaved molecule and the 48,000 Mr cleavage fragment shown in Fig. 2. Downloaded by guest on September 28, 2021 Medical Sciences: Wiggins et al. Proc. Nati. Acad. Sci. USA 74 (1977) 4639

FACTOR XI t HMWK COMPLEX

FACTOR XI >r4NN

FACTOR XI : HMWK-SURFACE

|SURFACE-BOUND ACTIVATED HAGEMAN FACTOR

FACTOR XI FACTOR XI a HMWK-SURFACE + a HMWK- SURFACE FIG. 3. Schematic role for high Mr kininogen (HMWK) in surface-binding and activation of Factor XI in plasma.

data). In the second reaction, supported by data described in be eluted from the surface (S. D. Revak, R. C. Wiggins, B. N. this study, the bimolecular complex is bound to a kaolin surface Bouma, C. G. Cochrane, and J. H. Griffin, unpublished data), in a manner such that the high Mr kininogen actually bridges suggests that the equilibrium is in favor of association rather or links Factor XI to the surface. Although more complex mo- than dissociation. Data from previous studies (23, 24) in which lecular mechanisms than this one are possible, the simplest contact product activity was measured also support this con- description consistent with the data has been chosen for this clusion. hypothesis. The third reaction is the activation of bound Factor The studies using 125I-prekallikrein demonstrate that high XI by surface-bound activated Hageman factor, as suggested Mr kininogen enhances the binding of prekallikrein to kaolin by our earlier report (6). The fourth reaction depicts the asso- in plasma by more than 7-fold. However, in sharp contrast to ciation-dissociation equilibrium between activated Factor XI the binding of four-fifths of the Factor XI molecules to kaolin, and the surface. The evidence from Table 1 in this study, only one-seventh of the available prekallikrein or kallikrein showing that 80% of the activated Factor XI remained bound molecules are bound. The binding data for 125I-prekallikrein to the surface, and from studies using repetitive washing of in plasma resemble those found for 125I-Factor XI in that glass-bound Factor XI, which showed that Factor XI could not Hageman factor was not required for binding and purified high

PREKALL I KRE I N | HMWK COMPLEX

PREKALL I KRE I N + SURFACE HMWK

PREKALLIKREIN : HMWK-SURFACE

RECIPROCAL ACTIVATION SURFACE-BOUND OF HAGEMAN FACTOR ACTIVATED HAGEMAN FACTOR

4 KALLIKREIN KALL I KRE I N t HMWK-SURFACE + HMWK- SURFACE FIG. 4. Schematic role for high Mr kininogen (HMWK) in surface-binding and activation of prekallikrein in plasma. Downloaded by guest on September 28, 2021 4640 Medical Sciences: Wiggins et al. Proc. Natl. Acad. Sci. USA 74 (1977)

Mr kininogen corrected the binding abnormality found in high 5. Nossel, H. L., Rubin, H., Drillings M. & Hsieh, R. (1968) J. Clin. Mr kininogen-deficient plasma. Hence, high Mr kininogen Invest. 47, 1172-1180. mediates the efficient binding of prekallikrein to kaolin. Fig. 6. Griffin, J. H. & Cochrane, C. G. (1976) Proc. Natl. Acad. Sci. USA 4 summarizes four reactions that involve high Mr kininogen and 73,2554-2558. 7. Meier, H. L., Webster, M. E., Liu, C. Y., Colman, R. W. & prekallikrein. The first reaction depicts a reversible complex Kaplan, A. P. (1976) Fed. Proc. 35,692 (abstr.). formation in plasma between the two molecules for which the 8. Liu, C. Y., Bagdasarian, A., Meier, H., Scot, C. F., Pierce, J., equilibrium constant in unknown (9, 25, 26). The second re- Kaplan, A. P. & Colman, R. W. (1976) Fed. Proc. 35, 692 action involves the binding of this bimolecular complex to ka- (abstr.). olin; the notation is meant to imply that the kininogen molecule 9. Mandle, R. J., Colman, R. W. & Kaplan, A. P. (1976) Proc. Natl. bridges or links prekallikrein to kaolin in plasma, as suggested Acad. Sci. USA 73,4179-4183. above for Factor XI. In the third reaction surface-bound acti- 10. Thompson, R. E., Mandle, R. J. & Kaplan, A. P. (1977) Thromb. vated Hageman factor cleaves prekallikrein to give the active Haemostasis 38, 13 (abstr.). enzyme kallikrein, as suggested previously (6). The fourth re- 11. Bouma, B. N. & Griffin, J. H. (1977) J. Biol. Chem., in press. action is 12. David, G. S. & Reisfeld, R. A. (1974) Biochemistry 13, 1014- the dissociation of surface-bound kallikrein. The final 1021. equilibrium in favor of dissociation is consistent with the fact 13. McConahey, P. J. & Dixon, F. J. (1966) Int. Arch. Allergy Appl. that 80% of the kallikrein is free in solution, and repetitive Immunol. 29, 185-189. washing studies showed that kallikrein could be eluted from 14. Griffin, J. H. & Cochrane, C. G. (1976) in Methods in Enzy- a glass surface (S. D. Revak, R. C. Wiggins, B. N. Bouma, C. G. mology, ed. Lorand, L. (Academic Press, New York), Vol. XLV, Cochrane, and J. H. Griffin, unpublished data). However, this part B, pp. 56-65. does not preclude the possibility of activation of prekallikrein 15. Weber, K., Pringle, J. R. & Osborn, M. (1972) in Methods in in solution, a reaction that is readily catalyzed by a 28,000 Mr Enzymology, eds. Hirs, C. H. W. & Timasheff, S. N. (Academic form of activated Hageman factor (27, 28). Press, New York), Vol. XXVI, part C, pp. 3-27. In normal plasma after contact activation 16. Wuepper, K. D., Miller, D. R. & Lacombe, M. J. (1975) J. Clin. the observations Invest. 56, 1663-1672. summarized in Tables 1 and 2 show that 5 out of 6 molecules 17. Saito, H., Ratnoff, 0. D., Waldmann, R. & Abraham, J. P. (1975) of kallikrein are in solution whereas 9 out of 10 molecules of J. Clin. Invest. 55, 1082-1089. activated Factor XI are surface-bound. If this distribution of 18. Donaldson, V. H., Glueck, H. I., Miller, M. A., Movat, H. Z. & activated enzymes were to occur upon contact activation in vivo Habal, F. (1976) J. Lab. Clin. Med. 87,327-337. after injury and exposure of damaged connective tissue (29), 19. Colman, R. W., Bagdasarian, A., Talamo, R. C., Scott, C. F., collagen (30-32), or basement membrane structures (33), it Seavey, M., Guimaraes, J. A., Pierce, J. V., Kaplan, A. P. & would suggest that coagulation reactions might tend to remain Weinstein, L. (1975) J. Clin. Invest. 56, 1650-1662. localized while kallikrein-dependent reactions would occur in 20., Schiffman, S. & Lee, P. (1975) J. Clin. Invest. 56, 1082-1092. solution away from the site of This must be 21. Bouma, B. N. & Griffin, J. H. (1977) Thromb. Haemostasis 38, injury. implication 136 (abstr.). tempered by the possible differences in specific enzyme ac- 22. Revak, S. D., Cochrane, C. G. & Griffin, J. H. (1977) J. Clin. tivities or susceptibility to inhibition of soluble versus surface- Invest. 59, 1167-1175. bound enzymes. 23. Waaler, B. A. (1959) Scand. J. Clin. Lab. Invest. Suppl. II, Harpel (34) showed that partially purified kallikrein would 37. bind to collagen and that, after brief incubation of particulate 24. Nossel, H. L. (1964) The Contact Phase of Blood Coagulation collagen with undiluted plasma, kallikrein activity could -be (Blackwell, Oxford). identified on the washed collagen. Assuming that collagen and 25. Nagasawa, S. & Nakayasu, T. (1973) J. Biochem. (Tokyo) 74, kaolin both act as surfaces with negative changes, the binding 401-403. of kallikrein to collagen observed by Harpel could represent 26. Wendel, U., Vogt, W. & Seidel, G. (1972) Hoppe-Seyler's Z. the limited binding that occurred in the absence of high Mr Physiol. Chem. 353, 1591-1606. 27. Kaplan, A. P. & Austen, K. F. (1970) J. Immunol. 195, 802- kininogen observed in the present study. 811. The skillful technical work of Gregory Beretta and Alice Kleiss is 28. Wuepper, K. D., Tucker, E. S. & Cochrane, C. G. (1970) J. Im- gratefully acknowledged. This manuscript is no. 1151 from the De- munol. 105, 1307-1311. partment of Immunopathology, Scripps Clinic and Research Foun- 29. Cochrane, C. G. & Wuepper, K. D. (1971) J. Exp. Med. 134, dation, 10666 N. Torrey Pines Road, La Jolla, CA 92037. This work 986-1004. was supported by Training Grant HL 07195 and National Institutes 30. Soltay, M. J., Movat, H. Z. & Ozge-Anwar, A. H. (1971) Proc. Soc. of Health Grants HL-16411, HL-21544, and AI-07007, The Council Exp. Biol. Med. 138,952-958. of Tobacco Research, and John H. Griffin's Research and Career De- 31. Niewiarowski, S., Bankowski, E. & Rogowicka, I. (1965) Thromb. velopment Award. B.N.B. received support from The Netherlands Diath. Haemorrh. 14,387-400. Organization for the Advancement of Pure Research (Z.W.O.). 32. Wilner, G. D., Nossel, H. C. & LeRoy, E. C. (1968) J. Clin. Invest. 47,2608-2615. 1. Margolis, J. (1958) J. Physiol. (London) 144, 1-22. 33. Cochrane, C. G., Revak, S. D., Aiken, B. S. & Wuepper, R. D. 2. Niewiarowski, S. & Prou-Wartelle, 0. (1959) Throm. Diath. (1972) in Inflammation, Mechanisms, and Control, eds. Lepow, Haemorrh. 3, 593-603. I. M. & Ward, P. A. (Academic Press, New York), pp. 119- 3. Ratnoff, 0. D. (1966) Prog. Hematol. 5,204-245. 138. 4. Miles, A. A. (1964) Ann. N.Y. Acad. Sci. 116,855-865. 34. Harpel, P. C. (1972) J. Clin. Invest. 51, 1813-1822. Downloaded by guest on September 28, 2021