THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc

Vol. 267, No. 6, Issue of February 25, pp. 3712-3717,1992 Printed in U.S.A.

Domain 3 of Kininogens Contains a Cell-binding Site and a Site That Modifies Thrombin Activation of Platelets*

(Received for publication, July 26, 1991)

Yongping JiangSg, Werner Muller-EsterlTl, and Alvin H. SchmaierS§II** From the $Thrombosis Research Center and the IlHematology Section, Temple University School of Medicine, Philadelphia, Pennsvluania 19140 and the lllnstitute for Phvsiolopical Chemistry and Pathobiochemistry, Johnannes Gutenberg-University a t Mainz; Duesbergweg 6, 0-6500 Mainz, Germany

” -

High and low molecular weight kininogens (HK and LK) are able to bind to platelets to inhibit thrombin binding to and activation of platelets. The heavy chain domain on the kininogens that contains these functions has been determined. Domain 3 (D3) but not domains 1 or 2 , completely inhibited ”‘I-HK binding to platelets (Ki = 24 f 7 nM, n = 4). 12‘I-D3 specifically bound to unstimulated platelets and human umbilical vein endo- thelial cells. On platelets, it was blocked by unlabeled D3 and HK but not prekallikrein, factor XII, Cls , or C1 inhibitor. Further, one monoclonal antibody (HKH13) directed to kininogens’ D3 blocked “‘I-HK and 12’I-D3 binding to platelets. The binding of “‘1- D3 to platelets was fully reversible by addition of 35 molar excess of unlabeled D3. D3 binding to platelets was saturable with an apparent ICd of 39 2 8 nM (n = 4) and 1227 f 404 binding sites/platelet. D3, like HK and LK, inhibited thrombin-induced platelet activation by preventing thrombin binding to platelets. Another monoclonal antibody (HKHlS), directed to D3, which did not block HK binding to platelets, reduced HK’s ability to inhibit “‘I-a-thrombin binding. This result suggests that the region on D3 that inhibits lz‘I-a- thrombin binding to platelets is different from that which directly binds to platelets. These studies indicate that D3 of the kininogens contains both a binding region for platelets and endothelial cells and another region that inhibits thrombin-induced platelet activa- tion.

The plasma kininogens, high and low molecular weight kininogen (HK and LK)’, are multifunctional single gene products that serve as parent proteins for bradykinin and

* This project was supported in part by a grant from the W. W. Smith Charitable Trust and Grant HL35553 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence and reprint requests should be ad- dressed ,Division of Hematology and Oncology, Dept. of Internal Medicine, Simpson Memorial Research Inst., The Universityof Mich- igan Medical Center, 102 Observatory St., Ann Arbor, MI 48109- 0724. Tel.: 313-747-3124; Fax: 313-764-2566.

** Recipient of Research Career Development Award HL01615. The abbreviations used are: HK, high molecular weight kininogen;

LK, low molecular weight kininogen; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; mAb, monoclonal anti- body; D l , Domain 1 of kininogen’s heavy chain; D2, Domain 2 of kininogen’s heavy chain; D3, Domain 3 of kininogen’s heavy chain; HUVEC, human umbilical vein endothelial cells; ICso, 50% inhibitory concentration; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.

contain two domains on their heavy chain that are cysteine protease inhibitors (1-4). The kininogens share the amino- terminal heavy chain portion comprising three domains of cystatin-like structures (domains Dl , D2, and D3), the bra- dykinin sequence, and the first 12 amino acids of their light chains. LK has a truncated carboxyl-terminal light chain, which is due to differential processing of the precursor mRNA (1, 5). The longer light chain of HK is responsible for its procoagulant activity (6, 7 ) , which is dependent upon two characteristics: its ability to form a complex with either prekallikrein or factor XI (8-10) and its ability to interact with negatively charged surfaces in vitro (11, 12). Since HK has a unique region on its light chain that interacts with artificial, negatively charged surfaces (7 , 12, 13), it has been assumed that HK’s ability to interact with platelets, granu- locytes, and endothelial cells (14-18) must be through a domain contained on its light chain. However, recent studies from our laboratory indicate that LK directly binds to plate- lets and competes with HK’s ability to bind to platelets (19). This study indicates that there must be a domain on the heavy chain of the kininogens that allows them to bind to platelets. In this investigation, the specific domain on the heavy chain of the plasma kininogens that contains the region that binds platelets and endothelial cells is identified. Further studies also show that this domain contains another epitope, distinct from its cell-binding site, which has the ability to prevent CY-

thrombin binding to and activation of human platelets.

EXPERIMENTAL PROCEDURES~

RESULTS

Inhibition of lZ51-HK Binding to Platelets by Domain 3 of LK-Previous studies showed that both HK and LK specifi- cally bound to platelets (19). To determine the domain on the kininogens’ heavy chain that contained its binding site to platelets, washed platelets were incubated with T - H K and an increasing concentration of the purified, unlabeled do- mains of the kininogens’ heavy chain (Fig. 2). Specific lZ51- HK binding to platelets was completely inhibited by 25-fold molar excess D3, but not D l or D2. D3 inhibited ”‘I-HK binding to platelets with a mean ICso of 50 nM from four experiments (Fig. 2), which calculated to an apparent K , of 15 nM (38).

Binding of 1251-D3 to Platelets and HUVEC-When washed platelets were directly incubated at 37 “C with Iz5I-D3, there was specific lZ51-D3 binding to platelets (Fig. 3A). Both un-

Portions of this paper (including “Experimental Procedures” and Fig. 1) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press.

3712

Domain 3 Contains the Cell-binding Region A

3713

! Ul :]! - 2

u"4 $ 0 0 5 1 0 15 2 0 25 30 35 4 0 4 5

Kininogen Domain (molar excess)

FIG. 2. Competition of '251-HK binding to platelets by kini- nogens' heavy chain domains. Gel-filtered platelets (2.0 X lo8/ ml) in Hepes-Tyrode's buffer were incubated for 20 min with 10 nM I2'I-HK in the presence of 50 p~ Zn" and increasing concentrations (10-430 nM) of unlabeled heavy chain domains (Dl (0), D2 (O), and D3 (0)). Nonspecific binding was determined by adding a 50 molar excess of unlabeled HK. Specific binding was calculated by subtract- ing nonspecific binding from the total "'I-HK binding. The figure presented is the mean S.D. of the data derived from four experi- ments.

labeled D3 (35-fold molar excess) or HK (200-fold molar excess) inhibited lZsI-D3 binding to platelets. Although zinc is an essential cofactor for lZ51-HK or -LK binding to platelets (14, 15, 19), it was not a requirement for "'IT-D3 binding to platelets (data not shown). The specificity of "'II-D3 binding t o platelets was characterized by determining whether a num- ber of related or unrelated proteins blocked its binding to platelets. Binding of lZ51-D3 to platelets was not inhibited by a 50-fold molar excess of Cls, C1 inhibitor, prekallikrein, or factor XII. Only 25-fold molar excess of unlabeled D3 or 250- fold molar excess of HK inhibited lZ5I-D3 binding to platelets. The ICso of increasing concentrations of unlabeled D3 on the binding of lZ51-D3 was 69 nM, a mean of three experiments, with a calculated apparent K; of 15 nM (data not shown). The potential importance of D3 for containing the binding region on the kininogens for all cells was investigated by determining if it would inhibit IZ5I-HK binding to HUVEC (Fig. 3B). Binding of lZ5I-HK to HUVEC was inhibited by a 35-fold molar excess of unlabeled D3.

The binding of lZ5I-D3 to platelets was fully reversible (data not shown). When a 35-fold molar excess of unlabeled D3 was added at 10, 30, and 50 min after binding of "'I-D3 to platelets, the level of the bound radioligand decreased rapidly to the level of nonspecific binding. lZ51-D3 binding to platelets was also saturable. When increasing concentrations of lZ51- D3 were added to platelets in the absence or presence of a 35- fold molar excess of unlabeled D3, a plateau of specific binding was observed between 30 and 40 nM added radioligand (Fig. 4A). When these specific binding data were analyzed by the method of Scatchard (38), a single saturable binding site was found with an apparent Kd of 33 nM and 872 sites/platelet (Fig. 48). In four experiments, the mean & was 39 f 8 nM and 1227 2 404 sites/platelet.

Inhibition of HK and 03 Binding to Platelets by a Domain 3-specific mAb-Further studies determined whether any anti-kininogen monoclonal antibodies directed to their heavy chain inhibited "'1-HK binding to platelets. After screening 24 mAbs directed to all three domains of kininogens' heavy chain, one mAb, HKH13, which is directed to D3, blocked I2'I-HK binding to unstimulated washed platelets in a con- centration-dependent fashion (Fig. 5A). This interaction was specific because two other mAbs (HKH12 and HKH14) that

$ r oL!- 0 20 4 0 6 0 8 0 1 0 0

Time (min)

B

- : o l ~ r . ~ . l ' m ' ~ . ~ ' ~ ' l r 0 20 4 0 6 0 8 0 100 120 1 4 0 1 6 0

Time (min)

FIG. 3. Binding of D3 to platelets or endothelial cells. A, lZ5I- D3 binding to platelets. Gel-filtered platelets (2.0 X 108/ml) in Hepes- Tyrode's buffer were incubated for 5-90 min a t 37 "C with lZ5I-D3 (30 nM) in the presence of 50 p M Zn2+ without any competitor (0). At each time point, samples were removed and the bound lZ5I-D3 was separated from unbound by centrifugation through an oil gradient (see "Experimental Procedures"). Nonspecific binding was measured concomitantly using replicate incubants containing a 35-fold molar excess of unlabeled D3 (0) or 200-fold molar excess HK (0) in the presence of 50 p~ zinc. The data plotted are the mean of three experiments. B, competition of "'I-HK binding to endothelial cells by unlabeled D3. Confluent monolayers of human umbilical vein endothelial cells in microtiter plates were washed with Hepes-Ty- rode's buffer containing 50 p~ ZnClz, chilled on ice, and incubated with lZ5I-HK (10 nM) in the absence (0) or presence of 30-fold molar excess of unlabeled D3 (W) for the indicated length of time. Nonspe- cific binding was determined by measuring the amount of "'I-HK that binds to the cells in the presence of a 50-fold molar excess of unlabeled HK (0).

were also directed to D3 did not inhibit lZ5I-HK binding to platelets. Six molar excess HKH13 completely inhibited "'1- D3 (30 nM) binding to platelets (Fig. 5B). These studies independently confirm that D3 contains a platelet-binding region.

Effect of 03 on '251-a-Thrombin Binding to Platelets and Thrombin-induced Platelet Actiuation-Since HK and LK inhibited thrombin-induced platelet activation by blocking its binding (19), investigations were performed to determine if D3 could also block the binding of '251-~-thrombin to platelets (Fig. 6A) . Both HK or D3 were able to inhibit '251-cu-thrombin binding to platelets. Further studies were performed to deter- mine if D3 binding to platelets, like HK and LK (19, 40), could also influence thrombin's activation of platelets (Fig. 6B). Washed platelets activated with &-thrombin (0.125 unit/ ml) had an initial change in light transmittance of 50 units/ min. When purified HK (1 PM) was added, platelet aggrega- tion was completely inhibited (5 units/min), as previously reported (19). Similarly, when purified D3 (1 PM) was added, platelet aggregation was also inhibited to a light transmittance value of less than 1 unit/min (Fig. 6B). D3 was also found to block thrombin-induced platelet secretion, independent of aggregation (Table I). D3, like HK and LK (19), decreased

3714

A

Domain 3 Contains the Cell-binding Region

A

B

0 2 0 4 0 6 0 B O D3 Add& (oM)

0.002

0.001

'\ 0.0304 . , . 1 . 1 . I . I : I

0 . 0 2 5 0.050 0 . 0 7 5 0,100 0.125 0 .150

D3 Bound (nM)

FIG. 4. Concentration-dependent binding of "%D3 to platelets. A, gel-filtered platelets in Hepes-Tyrode's buffer were incubated for 20 min at 37 "C with increasing concentrations of lZ5I- D3 in the presence of 50 pM Zn". Specific binding (A) was calculated by subtracting nonspecific binding (0) (determined by adding a 35- fold molar excess of unlabeled D3) from the level of binding seen in the absence of any competitor (A). B, the specific binding data from Panel A were graphed on a bound/free versus bound plot according to the method of Scatchard (39). The data presented in this figure are from one experiment representive of five.

the ability of a-thrombin to induce secretion of a platelet- dense granule marker, ( ['4C]5-hydroxytryptamine), in con- centration-dependent fashion. The IC6o for HK inhibition of thrombin-induced platelet secretion was between 30 and 50 nM vers'sus an ICs0 between 125 and 200 nM for D3. Alterna- tively, HK (1 PM) had no influence on the ability of SFLLRN (20 p ~ ) , a peptide from the cleaved thrombin receptor (28), to induce platelet aggregation and secretion. In the absence of HK, SFLLRN induced 33 units/ml platelet aggregation and 50% platelet secretion; in the presence of HK, SFLLRN induced 33 units/ml platelet aggregation and 59% secretion (both sets of data are means of two experiments).

Investigations on the Kininogen-Thrombin Interaction- Additional studies were performed to determine if the region on D3 that inhibited thrombin binding to platelets was the same as its cell-binding region. D3-specific mAbs (HKH12 and HKH14) were used to determine if they could nullify HK's ability to block '251-a-thrombin binding to platelets. mAb HKH14 and normal mouse IgG at 20-fold molar excess to HK had no influence on the ability of HK to inhibit 1251- a-thrombin binding to platelets (data not shown). Alterna- tively, mAb HKH12, which did not block HK's ability to bind to platelets (Fig. 5A) , decreased the ability of HK to inhibit '"I-a-thrombin binding to platelets by 50% at a molar ratio equal to that of HK (Fig. 7). This latter result suggested that the epitope on D3 for HKHl2 may be a region on the kini- nogens, independent of its cell-binding area, that functions as a thrombin inhibitory portion.

DISCUSSION

The present studies were undertaken to determine what domain on kininogens' heavy chain participates in cell binding

125

g 100

U 2 75

X 50 5 m

u, 25

0 0 2 4 6 8 1 0

Molar Ratlo 01 mAbs to 1251-HK

B

2 2 T 2 L A 20 Molar Rallo of mAb to 125bD3 8 1 0

FIG. 5. Inhibition of '"I-HK and Iz6I-D3 binding to platelets by monoclonal antibodies directed to the heavy chain of the kininogens. A, gel-filtered platelets (2.0 X 108/ml) in Hepes-Ty- rode's buffer were incubated for 20 min with 10 nM '"1-HK in the presence of 50 p M Zn2' and increasing concentrations (10-85 nM) of mAbs, HKHl2 (n), HKH13 (O), and HKHl4 (0). The graph depicts the percent of specific binding of lZ51-HK on the ordinate. Specific binding at each concentration of added mAb was determined by subtraction of nonspecific binding from total '"1-HK binding. Non- specific binding was determined by the concentration of radioligand associated with the platelet pellet when binding was performed in the presence of a 50-fold molar excess of unlabeled HK. Percent specific binding is normalization between 0-100% of specific Y - H K binding. The figure presented is the mean t- S.D. of the data derived from four independent experiments. B, inhibition of lz5I-D3 binding to platelets by monoclonal antibody HKH13 (0). Gel-filtered platelets (2.0 X 108/ml) in Hepes-Tyrode's buffer were incubated for 20 min with 30 nM '"I-D3 in the presence of 50 p M Zn2' and increasing concentrations (15-240 nM) of mAb HKH13. The graph depicts the percent specific binding of lZ5I-D3 on the ordinate, as described above in the legend to Fig. 5A. The figure presented is the mean f S.D. of the points derived from four experiments.

and what is the effect of this interaction. Kininogens' D3 was found to contain a cell binding region for platelets and endo- thelial cells (Figs. 2, 3, and 5). The parameters of D3 binding to platelets (& = 39 nM; 1227 sites/cell) (Fig. 4) are in agreement with the values obtained for LK binding to plate- lets (19), and the affinity of binding is similar to that of HK's interaction with granulocytes and endothelial cells (16-18).

Although in previous studies zinc ions are important cofac- tors for kininogens' interaction with cells (14-19), zinc ions were found not to be required for binding of D3 to platelets. The requirement for zinc in HK binding to cells has been assumed to be based upon the presence of its anionic surface binding region on its domain 5 (12). However zinc ion is also a requirement for LK binding to platelets (19), and since LK does not directly bind to zinc (19), the requirement for this ion may be with the platelets themselves. Zinc ion could be important for the expression of the putative kininogen recep- tor on platelets. Recent studies indicate the 50 pM zinc ion is an absolute requirement for full expression of the human growth hormone receptor (41). The finding that there is no requirement for zinc for D3 binding to platelets indicates that

Domain 3 Contains the Cell-binding Region 3715

A 60 7

c iL,,,:, m N c

10 5,

0 0 2 5 50 75 1 0 0 1 2 5

l i m e ( m i n )

TABLE I H K and D3 inhibition of a-thrombin-induced platelet secretion

Gel-filtered ["C]5-hydroxytryptamine radiolabeled platelets (2 X 10*/ml) were incubated in the absence or presence of each concentra- tion of the HK or D3 indicated in the first column. The reaction was initiated by the addition of 0.125 unit/ml human a-thrombin. The experiments were performed under the conditions described under "Experimental Procedures." The data are the means of two full experiments on different samples of washed platelets with all of the concentrations of kininogen indicated. The numbers shown in the second and third columns represent percent secretion of [14C]5- hydroxytryptamine measured after 1 min after the introduction of the thrombin.

'i Dose HK treatment D3 treatment

HK Added 11s Addrd

W 0 0.015 0.03 0.05 0.125 0.2 0.3 0.4 0.6

73 53 46

3 1 5 3 0.3 0

%secretion 73 70 61 57 51 33 20 12 10

D l Added 11. Added I ,

FIG. 6. D3 inhibition of thrombin binding and thrombin- induced platelet activation. A , D3 inhibition of radiolabeled thrombin binding to platelets. Gel-filtered platelets (2 X 10*/ml) in Hepes-Tyrode's buffer containing 50 p~ ZnCIP and 2 mM CaC12 were incubated for 5-120 min at 37 "C with 1 nM '251-thrombin alone (0) or in the presence of 200 nM HK (0) or 200 nM D3 (0). The data plotted are the mean of two independent experiments. B, HK and D3 inhibition of thrombin-induced platelet activation. Gel-filtered plate- lets (2.0 X 10*/ml) in Hepes-Tyrode's buffer were treated with 1 p~ HK or D3 for 1 min before the introduction of human a-thrombin (0.125 units/ml), both indicated by the arrows. The platelet aggrega- tion was measured for 5 min after the introduction of the agonist. CONTROL platelets received an identical volume of buffer instead of HK or D3. The figure is a experiment representative of three per- formed with different platelet donors and different batches of HK and D3 that were carefully dialyzed to remove any trace of proteolytic inhibitors added during their preparation. Ilu, a-thrombin.

the binding of zinc to kininogens is not necessary for their binding to cells. Further, it also suggests that the smaller D3 may be able to insert itself into the putative kininogen recep- tor(s) without the latter having the conformational changes necessary to accept the bulkier molecules of HK and LK. This hypothesis will be tested when the kininogen receptor(s) on cells is isolated.

Our studies indicate that the intact kininogens inhibit a- thrombin activation of platelets by being a noncompetitive inhibitor of a-thrombin binding to platelets (19). The kini- nogens are probably not binding to a-thrombin's specific binding site on platelets because the kininogens did not in- hibit '251-PPACK-thrombin binding to platelets (19). Vu et al. (28) have reported that the binding of a-thrombin to its receptor results in the receptor's cleavage at the amino ter- minus. The new amino terminus then functions as the agonist for platelet activation. HK does not inhibit SFLLRN, a peptide from the amino terminus of the a-thrombin receptor (28), from inducing platelet aggregation and secretion, sug- gesting that the kininogens do not inhibit a-thrombin's acti- vation of platelets by blocking its cleaved receptor's agonist function. If kininogens inhibited a-thrombin's activation of

"'IF 1" €u , ,

- z 5 : u)

g a -

c -

- m D c - -

0 2 Molar Rallo 01 mAb to HK Added

FIG. 7. Influence of monoclonal antibody HKHlP on HK's ability to inhibit '2SI-a-thrombin binding to platelets. Two hundred nM of HK in Hepes-Tyrode's buffer were incubated with increasing concentrations of mAb HKHl2 (0-2 molar excess of HKH12 to HK) at 37 "C for 5 min. Afterwards, gel-filtered platelets were added, and the reaction was initiated by the introduction of lz5I- a-thrombin (12sI-IZu) (1 nM) followed by incubation at 37 "C for 20 min. The total inhibition (100%) was calculated as the concentration of '251-a-thrombin bound to platelets in the presence of a 200-fold molar excess of HK alone. No inhibition (0%) was determined by the concentration of '251-a-thrombin bound to platelets in the absence of HK. The ordinate of the graph depicts the percent inhibition of l Z 5 I -

a-thrombin binding to platelets by HK. The abscissa represents the molar ratio of mAb HKH12 to HK. Eachpoint in the figure represents the mean k S.D. from four experiments of the values of inhibition of HK's ability to block '"1-a-thrombin binding in the absence or presence of mAb HKH12.

platelets by blocking the action of the new amino terminus of the receptor, then there should have been normal a-thrombin binding to platelets in their presence. Last, there is no evi- dence to show that kininogens inhibit a-thrombin's action on platelets by directly interfering with its proteolytic activity (19). Thus, the influence of the kininogens on thrombin- induced platelet activation is simply blocking '251-a-thrombin binding and not inhibiting either the cleavage of the thrombin receptor or the ability of the new amino terminus of the receptor to activate platelets. D3 inhibits a-thrombin's plate- let binding and activation similar to HK and LK (Fig. 6 and Table I). The present studies also reveal that D3 is a less efficient inhibitor of a-thrombin-induced platelet secretion than HK. The IC5o for HK's inhibition of secretion was between 30 and 50 nM, but for D3 it was between 125 and 200 nM. The larger, bulkier molecule of HK serves as a better noncompetitive inhibitor of a-thrombin binding than D3. Alternatively, HK could be a better inhibitor than D3 of a - thrombin-induced platelet activation because there may be

3716 Domain 3 Contains the Cell-binding Region

more than one binding site on HK that supports its binding to platelets.

D3 of the kininogens appears to be multifunctional. It is well established that it contains a cysteine protease inhibitory region (3,4). The finding that D3 also contains a cell-binding region indicates another function for this part of the kinino- gens. Studies with mAb HKHl2 suggest that the inhibitory region on D3 of a-thrombin binding to platelets may be different from its cell-binding region. HKH12, a mAb directed to D3, does not block HK binding to platelets (Fig. 5A) but does reduce by 50% HK's ability to inhibit '251-~-thrombin binding to platelets (Fig. 7). This result suggests that the epitopes on D3 involved in cysteine protease inhibition, cell binding, and inhibition of thrombin binding may be different. It is as yet unclear how HKH12 actually influences HK's ability to inhibit '251-~-thrombin binding to platelets. HKH12 could be directed to a specific epitope on D3 in intact HK that overlaps the thrombin receptor on platelets. Alterna- tively, the binding of HKH12 to HK could induce a confor- mational change in the native protein such that its region that blocks the a-thrombin binding is moved.

The finding that the kininogens and their cell-binding domain inhibit thrombin binding to platelets suggest an anti- thrombin function for this protein. This notion is especially relevant since the kininogens' influence on thrombin's acti- vation of platelets occurs at physiological concentrations (19, 40). The presence of HK requires higher concentrations of y- thrombin to produce platelet aggregation in plasma (40). HK also has the ability to decrease the heparin-induced enhance- ment of the rate of inactivation of thrombin by antithrombin (42). Since recent investigations suggest that thrombin is a better activator of factor X1 than factor XI1 (43, 44), HK, which serves as the platelet-binding site for factor XI (45), could also be functioning to prevent factor XI activation by thrombin. In physiologic states, this activity could prevent excessive amplification of thrombin formation and resultant thrombus formation. These antithrombin activities are sup- ported by further indirect influences of the kininogens on limiting thrombosis. Bradykinin, when liberated from the kininogens, is a potent stimulator of prostacyclin production (46) and an inducer of tissue-type plasminogen activator secretion from endothelial cells in uiuo (47). All of these activities could summate into an antithrombin, profibrinolytic activity for kininogen. It has to be clarified whether these influences compound in vivo to decrease the risk for throm- bosis.

Acknowledgment-We are indebted to Jorg Kaufmann, Hoffmann- La Roche, Basel, Switzerland for providing the monoclonal antibodies to the heavy chain of the kininogens.

REFERENCES

3.

4.

5.

6.

7.

8.

9.

10. 11.

12.

13.

14. 15.

16.

17.

18.

20. 19.

21.

22.

23.

25. 24.

26.

27.

28.

29.

30.

31.

32.

33.

Okbuko, I., Kurachi, K., Takasawa, T., Shiokawa, H., and Sasaki, M. 11984)

Bradford, H., Schmaier, A. H., and Colman, R. W. (1990) Biochem. J . 270,

Kakizuka, A., Ingi, T., Murai, T., Nakanishi, S. (1990) J. Biol. Chem. 265,

Griffin, J. H., and Cochrane, C. G. (1977) Proc. Natl. Acad. Sci. U. S. A.

Nakanishi, S. (1985) J . Biol. Chem. 260,8610-8617

Biochemistry 23, 5691-5697

83-90

10102-10108

74.46?.fi-A6dn Thompson, R. E., Mandle, R., Jr., and Kaplan, A. P. (1978) J. Exp. Med.

Mandle, R., Jr., Colman, R. W., and Kaplan, A. P. (1976) Proc. Natl. Acad.

Thompson, R. E., Mandle, R., Jr., and Kaplan, A. P. (1977) J. Clin. Inuest.

_-,_ _-_- 147,488-499

Sci. U. S. A. 73,4179-4183

60.1376-1380 Scott; C. F., and Colman, R. W. (1980) J. Clin. Inuest. 65,413-421 Griffin, J. H., and Cochrane, C. G. (1976) Proc. Natl. Acad. Sci. U. S. A.

Retzios, A. D., Rosenfeld, R., and Schiffman, S. (1987) J. Biol. Chem. 262,

Scott, C. F., Silver, L. D., Schapira, M., and Colman, R. W. (1984) J . Clin.

Greengard, J. S., and Griffin, J. H. (1984) Biochemistry 23, 6863-6869 Gustafson, E. J., Shutsky, D., Knight, L. C., and Schmaier, A. H. (1986) J .

73,2554-2558

3074-3081

Inuest. 73,954-962

Clin. Inuent. 78. 310-318 . .~ ~ - Gustafson, E. J., Schmaier, A. H., Wachtfogel, Y. T., Kaufman, N., Kucich,

Van Iwaarden. F.. deGroot. P. G.. and Bouma. B. N. (1988) J. Bid. Chem.

. " ~ ~

U., and Colman, R. W. (1989) J. Clin. Inuest. 84,28-35 , ,

263,4698-4703

J. Biol. Chem. 263, 16327-16333

. .

Schmaier, A. H., Kuo, A., Lundberg, D., Murray, S., and Cines, D. B. (1988)

Meloni, F. J., and Schmaier, A. H. (1991) J. Biol. Chem. 266,6786-6794 Schmaier, A. H., Claypool, W., and Colman, R. W. (1980) Blood 56,1013-

l n l s Colman, R. W., Bagdasarian, A., Talamo, R. C., Scott, C . F., Seavey, M.,

Guimares J. A,, Pierce, J. V., and Kaplan, A. P. (1975) J. Clin. Inuest.

Schmaier, A. H., Silver, L., Adams, L. A,, Fischer, C. G., Munoz, C. P.,

Schmaier, A. H., Bradford, H., Silver, L. D., Farber, A,, Scott, C., Schutsky,

Laemmli, U. K. (1970) Noture 227,680-685 Salvesen, G., Parkes. C., Abrahamson, M., Grubb, A., and Barrett, A. J.

Muller-Esterl, W., Johnson, D. A., Salvesen, G., and Barrett, A. J. (1987)

Schmaier, A. H., Schutsky, D., Farber, A., Silver, L. D., Bradford, H. N.,

Vu, T.-K. H., Hung, D. T., Wheaton, V. I., and Coughlin, S. R. (1991) Cell

_"_" 56,1650-1662

Vroman, L., and Colman, R. W., (1984) Thromb. Res. 33,51-67

D., and Colman, R. W. (1986) J . Clin. Inuest. 77, 1565-1573

(1986) Biochem. J . 234,429-434

Methods Enzymol. 163,240-256

and Colman, R. W. (1987) J. Bbl. Chem. 262,1405-1411

64. 105i"l#6R Schmaier, A. H., Smith, P. M, Purdon, A. D., White, J. G., and Colman, R.

Scott, C., Liu, C. Y., and Colman, R. W. (1979) Eur. J . Biochem. 100,77-

- -, . . . . - . . -

W. (1986) Blood 67, 119-130

Schmaier, A. H., Smith P. M., and Colman, R. W. (1985) J. Clin. Inuest.

Vogel, R., Assfalg-Machleidt, I., Esterl, A,, Machleidt, W., and Muller-

Kaufmann, J. (1990) Ph.D. thesis, Universlty of Munich, Germany 34. Jiang, Y.-P., Kaufmann, J., Meloni, F. J., Muller-Esterl, W., and Schmaier,

35. Fraker, P. J., and Speck, S. C . , Jr. (1978) Biochem. Biophys. Res. Cornmun.

83

75,242-250

Esterl, W. (1988) J. B d . Chem. 263, 12661-12668

A. H. (1991) Thromb. Haemostcrsis 65,1248

36. Schmaier, A. H., Zuckerberg, A., Silverman, C., Kuchibhotla, J., Tuszynski,

37. Canellas, P. F., and Karu, A. E. (1981) J. Immunol. Methods 47,375-385 38. Muller, R. (1983) Methods Enzymol. 92, 589-601 39. Scatchard, C. (1949) Ann. N. Y. Acad. Sci. 51,660-672 40. Puri, R., Zhou, F., Hu, C.-J., Colman, R. F., and Colman, R. W. (1991)

41. Cunningham, B. C., Bass, S., Fuh, G., and Wells, J. A. (1990) Science 250,

42. Bjork, I., Olson, S. T., Sheffer, R. G., and Shore, J. D. (1989) Biochemistry

43. Naito, K., and Fujikawa, K. (1991) J . Biol. Chem. 266,7353-7358

45. Greengard, J. S., Heeb, M. J., Eisdal, E., Walsh, P. N., and Griffin, J. H. 44. Gailani, D., and Broze, G. J., Jr. (1991) Science 253, 909-912

46. Hong, S. L. (1980) Thromb. Res. 18, 787-795 (1986) Biochemistry 25, 3884-3890

00,849-857

G. P., and Colman, R. W. (1983) J. Clin. Inuest. 71,1477-1489

Blood 77,500-507

1709-1712

28, 1213-1221

47. Smit.h. D.. Gilhert. M.. and Owen. W. G. (1985) Blood 66. 835-839

1. Takagaki, Y., Kitamura, N., and Nakanishi, S. (1985) J. Biol. Chem. 260,

2. Kitarnura, N., Kitagawa, N., Fukushima, D., Takagaki, Y., Miyata, T., and .. """", ~~ , ~ "" ~. ," . 8601-8609

Domain 3 Contains the Cell-binding Region 3717

S L K D 3 02 Dl

110- p 8 4 -

24- c '9