THROMBOSIS RESEARCH 60; 87-97,199O 0049-3848/90 $3.00 + .OO Printed in the USA. Copyright (c) 1990 Pergamon Press plc. All rights reserved.
PURIFICATION AND CHARACTERIZATION OF PLATELET FACTOR XI
Sandra Schiffman and Che-Hung Yeh Departments of Medicine and Biochemistry University of Southern California, School of Medicine Los Angeles, California 90033 U.S.A. (Received
28.3.1990;
accepted
in revised form 14.8.1990
by Editor H. Pirkle)
ABSTRACT Factor XI activity and antigen was purified about 300 fold from human platelets through chromatography on Con-A Sepharose, SP-Sephadex C-50, immobilized goat anti-factor XI, and SP-Sephadex. The partially purified platelet factor XI (Pt-XI) could be activated by activated factor XII generated in situ from single chain factor XI in a reaction requiring high molecular weight kininogen (HMWK) and a surface. Native Pt-XI migrated as a molecule of Mr = 245,000 on sodium dodecyl sulfate polyacrylamide gel electrophoresls (SDS-PAGE) as identified by Western blotting. On reduction, Pt-XI appeared to have a M, = 52,000. Neither form was affected by exposure to trypsin. Incubation of Pt-XI with purified factor XII, HMWK, and kaolin produced activated platelet factor XI clotting activity and, concomitantly, the generation over time of a new chain on reduced SDS-PAGE of M, - 44,500. The coagulant activity of the activated form could be neutralized by diisogropyl flurophosphate (DFP). Incubation of the activated mixture with H-DFP followed by reduced SDS-PAGE showed the active site to be associated with a unit of Mr - 44,500. The adsorption domain as defined by adsorption to kaolin was localized to the M, 0 44,500 chain containing the active site. Hence, both active site and adsorption functions, properties of separate chains in plasma factor XI, reside in the same chain of Mr - 44,500 of platelet factor XI.
INTRODUCTIW The interactions among the soluble protein clotting factors have been worked out in considerable detail (l-4), but the B reactions which initiate intrinsic clotting remain a mystery. In the glass test tube it is the interaction of factors XI, XII, prekallikrein, and HMWK with the glass surface to yield activated factor XI. Of these, only factor XI is reauired for normal Zhemostasis (4). Therefore, we have initiated a search for an alternate pathway. Herein, we report the purification, characterization, and
Key words:
Factor XI, platelets, Western blotting. 87
88
PLATELET FACTOR XI
Vol. 60, No. 1
mechanism of activation of a platelet protein which has factor XI procoagulant and antigen activity. It is similar to but different than plasma factor XI.
MATERIALS AND METHODS All chemicals obtained from commercial sources were the best grade available. Bovine trypsin (twice crystallized, dialyzed, salt-free, and lyophilized) and aprotinin (from bovine were from Sigma. lung) Chromatographic media were from Pharmacia. Protein concentration was determined using the Bio-Rad reagent according to the direc ions of the 5H-diisopropyl globulin standard. manufacturer with a bovine gamma flurophosphate (DFP) was from NEN. The chromogenic substrate S-2366 was from described Helena Laboratories. Ellagic acid (Sigma) was prepared as previously (5). Purified reaeents. Factor XI, factor XII, and HMWK were prepared as described previously yielded a single band on SDS-PAGE. Each clotting factor (6, 7, 8). Factor XI had a specific activity of 200 U/mg, factor XII had a specific Corn activity of 30.3 U/mg, and HMWK had a specific activity of 11 U/mg. trypsin inhibitor was prepared by the method of Kirby and McDevitt (9). Assays. Coagulant activitv of factor XI and factor XII was measured in modified activated partial thromboplastin time assays as described previously (10). HMWK was measured in a similar system with commercial Fitznerald trait plasma substrate from George King. Factor XI amidolvtic activiiy was measured as modified from Scott et al (ll), by incubating 50 ~1 of sample with 80 pl of 3.9 mM S-2366 for 30 minutes at room temperature in a 96 well microtiter The reaction was terminated after 30 minutes by adding 100 ~1 of 50% plate. acetic acid, and the plate was read at 405 nM. Radioimmunoassav of factor XI antigen was carried out by a modification of that described by Scott et al Goat antihuman-factor XI antiserum, prepared as described below, at a (11). dilution of l/10,000, was used as the primary antibody; swine antigoat (Boehringer Mannheim) was used as the second antibody, followed by 10% Staphylococcus Protein A (Boehringer Mannheim). Isotope was radiolabeled plasma factor XI (see below). 125I the Factor XI was prepared with (Amersham) by Radiolabeled lactoperoxidase method (12) with slight modification. The final product containing 77 pCi/pg retained its procoagulant activity. Goat antiserum & purified plasma factor XI was raised by five innoculations with purified human factor XI mixed in equal parts with Freund's adjuvant. It reacted with three protein fractions in partially purified platelet factor XI which were separable by gel filtration. Only one of these had procoagulant activity. Therefore, the antiserum was immunoadsorbed by exposure to a column of CNBr-activated Sepharose 4B to which had been coupled the inactive The resulting immunoadsorbed antiserum was monospecific and fractions. activity. This with the fraction having procoagulant only reacted monospecific adsorbed antiserum was then chromatographed on DEAE-Sephadex and the unbound fraction was further adsorbed with immobilized factor XI deficient plasma as described previously (13) and concentrated by dialysis against Sephadex G-25. The antibody used in these experiments had a titer of l/2500. Polvacrylamide & electronhoresis (PAGE). PAGE in the presence of 0.1% sodium dodecvl sulfate (SDS) (Sigma) was carried out on 10% slab gels of 0.75 mm thickness essentially according to the method of Laemmli (14). Gels were either subjected to autoradiography or cut
Vol. 60, No. 1
PLATELET FACTOR XI
into 1.1 mm sections and added to 10 ml PCS scintillation for analysis in a beta spectrometer.
89
cocktail
(Amersham)
Western blotting. et al (15). Protein was blotted according to the method of Mannhalter Samples, either reduced or non-reduced, were subjected to SDS-PAGE in a Hoefer Tall Mighty Small Model SE 280 System using 0.75 mm thick gels at constant Proteins were transferred onto a voltage of 200 V for 80 minutes. in a Hoefer Mighty Small Transphor system using nitrocellulose membrane current (150 mA) for 90 minutes. After transfer, the nitrocellulose constant membrane was quenched with 0.4% gelatin, 0.2% Tween 20, 0.2% NaN3 in phosphate buffered saline pH 7.2 for 4 hours. It was then incubated with 10 ml of diluted to 2% in immunoadsorbed goat antihuman-factor XI antiserum The membrane was then washed concentration in the same buffer for 16 hours. the same buffer without gelatin. three times (20 minutes each) in 10 ml 755 I-factor XI for 4 hours. washed The membrane was then incubated with 3 times, air dried, and set up for autoradiography or cut into 1 or 2 mm strips for counting. All incubations were at room temperature. Purification of Platelet Factor XI (Pt-XI). Pt-XI was purified by sequential chromatography on Con-A Sepharose, Platelets SP-Sephadex, anti-factor XI-Sepharose, and SP-Sephadex as follows: were prepared from forty or fifty units of pooled outdated platelets from the Red Cross by a modification of our previous method (16) in which an American inhibitor cocktail, adapted from Tuszynski et al (17), was added, which yielded a final solution containing 10 mM benzamidine-HCl, 5 mM e-aminocaproic acid (EACA), 2 mM EDTA, 0.02% soybean trypsin inhibitor, and 50 KIU/ml aprotinin. Red blood cells were removed by centrifugation at !jOO g for Platelets were sedimented by centrifugation at 800 g for 15 minutes. with gentle mixing in 15 minutes at 23'C. The precipitate was suspended Tyrode's solution buffered with HEPES [4-(2-hydroxy-ethyl)-l-piperazineethanesulfonic acid], pH 7.35: 0.14 M NaCl, 2.7 mM KCl, 12 mM NaHC03, 1mM Mg?2:6 H20tabZ;Z, ~T;y;;i;?;l~;8 ~s~~~~~,~.~asm”i~~~2~~d ‘;:““;~y ;:; inhl ltors 20 minutes and then gel filtered on a 5.0 x 50 cm column of Sepharose CL-2B which had been equilibrated with above Tyrode's solution, buffered with HEPES, pH 7.35 containing inhibitors. The flow rate was 0.6 ml/minute. Obviously turbid fractions were pooled as g& filtered Dlatelets and frozen at -2O'C. A fraction binding to Con-A Sepharose was prepared from the frozen platelets, as described by Tuszynski et al (17), by thawing and centrifugation at 800 g for 15 minutes. The precipitate was suspended in Tyrode's solution buffered with HEPES, pH 7.35 containing inhibitors as above. Triton X-100 was added to 0.2%. The solubilized & filtered Dlatelet solution was centrifuged overnight in a Beckman Microfuge for 4 minutes and the supernatant dialyzed against Con-A Sepharose column buffer: 1 n&l CaC12, 1 mM MgC12, 1 mM MnC12. and 1 mM ascorbic acid in 0.02 M Tris-HCl, pH 8.0. The dialyzed sample was in the Microfuge for 2 minutes, and the supernatant applied to a centrifuged Con-A SeDharose column (2.5 x 25 cm). The column was washed with about 360 ml of column buffer which resulted in the elution of an unbound protein peak between 60-100 ml. After another 250 ml of buffer, bound protein was eluted with 1 M a-D-methylmannoside in 0.02 M Tris-HCl, pH 8.0. Active fractions eluted in about 35 ml. Pooled active fractions from Con-A Sepharose were dialyzed overnight against 50 mM sodium acetate pH 5.2 containing 1 mM EDTA and 1 mM benzamidineHCl and applied to a 1.5 x 30 cm SP-SeDhadey c-50 column equilibrated with this buffer. Unbound proteins were eluted with the same buffer. Then a linear gradient of 150 ml each of equilibrating buffer and the same buffer containing 0.7 M NaCl was formed which eluted the platelet factor XI. The pooled active fractions from SP-Sephadex were dialyzed against glycol 0.02 M Tris buffer, pH 7.2 containing 0.5 M NaCl , 0.2% polyethylene
90
PLATELET FACTOR XI
Vol. 60, No. 1
and applied to a 0.9 x 3 cm column of BIPIPy.noadsorbed, go& anti*_ XL antlserum coupled to cyanogen-bromide-activated Sepharose 4B. The column was washed with 0.02 M Tris, pH 7.2 buffer containing 0.5 M NaCl, 0.2% PEG. It was eluted with 0.02 M Tris pH 7.2 containing 4 M NaSCN. Samples containing protein were immediately dialyzed against 0.05 M acetate buffer pH 5.2 containing 1 mM benzamidine and 1 mM EDTA and re-chromatographed on a Sephadeq (1.5 x 15 cm) as above. The fractions containing factor XI activity were pooled as DurifiedolateletfactorXT.
RESULTS
Platelet factor XI was purified as described above and summarized in Table 1. The product was purified about 300 fold. It could be identified both by clotting activity and by immunoreactivity. Both assay systems were quite consistent, though comparison suggests that the preparation contains some factor XI antigen lacking activity. Practical evaluation of the size of Pt-XI could be attained only by Western blotting (vide infra) since stained gels required far too much protein.
TABLE 1 .
Purification m
Volume
.
catwn nd Platelet
Protein
Factor 3.L
Total Activity (units)
Recovery
Specific Activity funit-
Solubilized Gel-Filtered Platelets
12.5
Con-A Sepharose
46
24.1
20.7
78.9
0.86
SP-Sephadex-(1)
36
3.8
18.0
68.4
4.74
Anti-XI Affinity
8
0.7
7.2
27.4
10.3
SP-Sephadex-(2)
1
0.27
5.4
20.5
20
400
26.3
100
0.066
Pt-XI was evaluated for ability to undergo activation as defined by Two activators of plasma factor XI were increased coagulant activity. factor XII combined with HMWX and surface activator tested, trypsin and (Tables 2 and 3). Pt-XI readily activated in the presence of factor XII, HMWX and kaolin (Table 2, line 1). Further clotting tests at 75' and 90' were Activation required the unchanged indicating that reaction was complete. complete reaction mixture; no activation occurred in the absence of any of the components (Table 2, lines 3-9). In contrast, both soluble and insoluble trypsin at concentrations which readily activated plasma factor XI failed to activate Pt-XI (Table 3). To determine whether Pt-XI was perhaps interacting with trypsin without undergoing effective activation two evaluations were made. Pt-XI was combined with plasma XI and added to insoluble trypsin which resulted in clotting times
91
PLATELET FACTOR XI
Vol. 60, No. 1
TABLE 2 Activation of Pm
Factor XI (Pt-XI) With Purified ClottinP Factors
15”
Reaction Mixture
Incubation Time 30' 10' 20' ClottinrrTime
§.!E
1)
Pt-XI + XII + HMWK + Kaolin
246
211
182
159
135
2)
Pt-XI + XII + HMWK + Ellagic Acid
262
218
192
173
152
3)
Pt-XI + XII + HMWK + Buffer
255
241
240
238
241
4)
Pt-XI + XII + Buffer + Kaolin
240
234
227
225
225
5)
Pt-XI + XII + Buffer + Buffer
249
243
241
243
241
6)
Pt-XI + Buffer + Buffer + Kaolin
244
246
239
242
240
7)
Pt-XI + Buffer + HMWK + Buffer
255
254
256
249
248
8)
Pt-XI + Buffer + HMWK + Kaolin
248
242
239
243
241
9)
Buffer + XII + HMWK + Kaolin
289
281
282
283
280
10) Plasma-XI + XII + HMWK + Kaolin
232
177
147
1375
128
A mixture of Pt-XI (0.14 U/ml), factor XII (0.19 U/ml), HMWK (0.24 U/ml), and kaolin (0.24 mg/ml) or ellagic acid (8 pg/ml), or buffer was Vortexed and incubated at room temperature. At times shown, 50 pl of the sample were removed, and the clotting activity was measured in the factor XI assay in a polystyrene tube without added kaolin or preincubation.
TABLE 3 Effect of Trvnsin QQ Platelet Factor XJ
II_ Reaction Mixture
Incubation T_& 15' 30' Clottine Time
60'
1)
Plasma-XI + Insoluble Trypsin
241
178
119
88
2)
Platelet-XI + Insoluble Trypsin
246
243
244
241
3)
Plasma-XI + Soluble Trypsin (3.2 pg/ml)
262
176
11.9
86
4)
Platelet-XI + Soluble Trypsin (3.2 pg/ml)
248
244
242
245
5)
Platelet-XI + Soluble Trypsin (19.2 pg/ml)
243
240
237
239
In experiments shown in lines 1 and 2, 500 cl1of plasma- or platelet-factor XI at 2 U/ml activity was combined at room temperature with 300 ~1 insoluble trypsin. In lines 3-5, platelet- and plasma-factor XI at 2 U/ml activity were incubated with soluble trypsin at 3.2 pg/ml (lines 3 & 4) or 19.2 pg/ml (line 5). At times shown, the samples were Vortexed, and 50 ~1 were removed and tested in the factor XI assay in a polystyrene tube without kaolin or preincubation.
PLATELET FACTOR XI
92
Vol. 60, No. 1
of the mixture identical to those of plasma factor XI tested alone. Hence, there was no evidence of competition. Furthermore, insoluble trypsin exposed to Pt-XI for one hour, removed, washed 5 times and added to plasma XI was unchanged; tryptic activity was unaffected by exposure to Pt-XI. Nor was there any change in size of Pt-XI as indicated by Western blotting after 30 minutes exposure to trypsin which readily cleaved plasma XI. Hence, Pt-XI appears to be at best very insensitive to tryptic cleavage. Since the data of Table 2, lines 1 and 3, indicated that kaolin played a critical role in the activation of Pt-XI, it was of interest to evaluate what another pseudo-surface could do in this reaction system. Therefore, ellagic acid, which is effective with plasma factor XI (5), was tested. As shown in line 2, Table 2, ellagic acid effectively promoted the activation of Pt-XI. Reaction mixtures of Pt-XI combined with potential activators, such as those described in Table 2, were evaluated for generation of amidolytic activity. The only reaction mixture which showed increasing amidolytic activity was that containing factors XII, HMWK, and kaolin (Table 4, line 1).
TABLE 4 Activation of Platelet Factor u
as Reflected $y Amidolvtic Assay
Reaction Mixture 15"
Incubation m 15' Absorbance 405~
30’
1)
Pt-XI + XII + HMWK + Kaolin
0.179
0.529
1.002
2)
Pt-XI + XII + HMWK + Buffer
0.065
0.065
0.052
3)
Pt-XI + XII + Buffer + Kaolin
0.059
0.058
0.064
4)
Pt-XI + XII + Buffer + Buffer
0.054
0.060
0.060
5)
Pt-XI + Buffer + Buffer + Kaolin
0.061
0.066
0.068
6)
Pt-XI + Buffer + HMWK + Buffer
0.061
0.061
0.055
7)
Pt-XI + Buffer + HMWK + Kaolin
0.066
0.065
0.059
8)
Buffer + XII + HMWK + Kaolin
0.055
0.055
0.055
Incubation mixtures of 210 ~1 containing 25 ~1 platelet factor XI at 1.2 U/ml, 50 pl of HMWK at 1 U/ml, 100 ~1 of factor XII at 0.4 U/ml, 10 al 5 mg/ml kaolin, or Tris/HSA as required, were combined. At the times shown, 50 /.bl were removed and added to 80 ~1 of 3.9 mM S-2366 in a 92 well microtiter After 30 minutes at room temperature, 100 ~1 of 50% acetic acid was plate. added, and the absorbance at 405 runwas read.
The size of Pt-XI and its activated products was evaluated on SDS-PAGE The results using Western blotting and autoradiography as described above. are shown in Figure 1. Pt-XI was dramatically bigger than plasma factor XI, having an apparent molecular weight of 245,000 (lane 8). By contrast, plasma factor XI displayed a M, - 165,000 (lane 10). Reduced Pt-XI migrated as a
PLATELET FACTOR XI
Vol. 60, No. 1
93
single entity of Mr = 52,000 (lane 4), slightly larger than the heavy chain of trypsin-activated plasma factor XI at M, = 49,000 (lane 3). Pt-XI which had been exposed to trypsin under comparable conditions was unchanged (data not shown). However, Pt-XI exposed to factor XII, HMWK, and kaolin for 30 minutes, which had activated factor XI activity (see Table 2), migrated as two discrete chains; one of IQ = 52,000 and the other of Mr = 44,500 (lane 5). On continued reaction with factor XII, HMWK, and kaolin for 60 minutes, Pt-XI was apparently entirely converted to the smaller chain, M, = 44,500 (lane 6). Hence, activation of platelet factor XI with clotting proteins leads to conversion over time from M, = 52,000 to 44,500 (lanes 4, 5, and 6).
1234
FIG. 1
Immunoblot
of Platelet-
56
7
8
and Plasma-Factor
910
XI.
Ten ~1 samples were prepared for SDS-PAGE followed by Western blotting as described in Materials and Methods. Lanes 1-6 contained reduced lanes 7-10 contained n.on-reduced samples; samples. Lane 1 contained prestained protein standards of M 200 K, 92.5 K, 68 K, 43 K, 25.7 K, and 18.4 K. After SDS-PAGh a:: transfer, they were manually marked with 1251-factor XI and subjected to autoradiography. Lanes 2 through 10 contained the following proteins: lanes 2 and 10 - 0.1 pg factor XI; lanes 3 and 9 - 0.1 pg trypsin-activated factor XI (6); lanes 4 and 8 - 0.13 pg Pt-XI; lanes 5, 6, and 7 - 0.13 pg Pt-XI activated by coagulant proteins for 30 minutes (lane 5). or 60 minutes (lanes 6 and 7) as described in Table 2. line 1.
Activated Pt-XI (Pt-XI,), prepared as described in the legend to Table 2, could be reacted with DFP which neutralized its procoagulant activity. For example, the clotting time of a fully activated mixture diluted l/10 rose from 136" to 252" on the addition of DFP indicating neutralization yf 98% of the clotting activity. When concentrated Pt-XI, was reacted with H-DFP, and the product subjected to SDS-PAGE, the resulting gel could be cut into lanes and subdivided into 1.1 mm sections to be counted in a beta spectrometer as described in Materials and Methods. As shown in Figure 2, DFP
94
PLATELET FACTOR XI
Vol. 60, No. 1
reacted with a M, = 44,500 band. Thus, the active site region appeared to included in the new, lower molecular weight band arising on activation.
be
To rule out the remote possibility that the labeled material defined in Fig. 2 at M, = 44,500 could be derived from the only other potential protease added to the system, factor XII, a parallel experiment was prepared lacking only Pt-XI but containing factor XII. It was neutralized with CTI, reacted with DFP, and subjected to SDS-PAGE followed by slicing and counting. No DFP-binding protein was detected. Hence, the labeled fraction defined in Fig. 2 could not be attributed to factor XII. localize the surface adsorption sites of Pt-XI, Pt-XI, was labeled with '$-DFP as above exposed to kaolin, the kaolin washed 5 times, and the were radioactivity of the kaolin and of the kaolin-adsorbed fraction determined. About 35% of the label was associated with the kaolin. This was consistent with the conclusion that the adsorption site is included in the region which contains the DFP-binding active site.
co
1.0
0 T; 0.8
I
0.6
& 0.4 0.2 0
Gel Slice Number
FIG. 2
SDS-polyacrylamide gel electrop oresis platelet factor XI inhibited with !3 H-DFP.
of
reduced,
activated
200 ~1 of a mixture containing Pt-XI at 1.5 U/ml was activated with factor XII, HMWK, and kaolin for 60 minutes as described in the legend to Table 2, line 1. Then 5 pl of corn trypsin inhibitor (1 mg/ml) were added, and the combination was incubated for 10 minutes at room temperature. They the mix was reacted for H-DFP (20 @i), dialyzed 120 minutes with 20 ~1 of 3.3 mM SDS-polyacrylamide gel and 37.5 ~1 subjected to overnight, electrophoresis as described in Materials and Methods followed by slicing and counting in a beta spectrometer.
The surface binding portion of Pt-XI, was more explicitly demonstrated by adsorbing Pt-XI, with kaolin, washing the kaolin, and eluting with 10% SDS.
PLATELET FACTOR XI
Vol. 60, No. 1
95
The eluate as well as the initial Pt-XI, and the non-adsorbed fraction, comprising about l/3 of the total Pt-XI,, were subjected to SDS-PAGE followed by Western blotting. As shown in Figure 3, the eluate, which must contain the adsorption domain, migrated with a Mr - 44,500. This was the identical position attained by the Pt-XI, and the non-adsorbed fraction. Hence, in this activated mixture there is only one size of immunoreactive molecule which contains m the surface binding domain and the active site. There was no evidence of non-productive tryptic cleavage products.
1.8 1.6 44,500
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 5
10
15
20
25
30
35
Gel Slice
FIG. 3
Immunoblot of SDS-eluate of Pt-XI, adsorbed to kaolin SDS-PAGE.
subjected to
800 pl of fully activated Pt-XI containing 1.2 U/ml Pt-XI was added to 2 mg kaolin, incubated for 1 hr. The supernatant was removed and the kaolin washed 5 times with buffer, then eluted with 100 pl of 10% SDS for 10 min. The kaolin eluate was run on 10% SDS-PAGE and then subjected to Western blotting. The membrane was sliced in 2 mm sections and counted in a gamma spectrometer.
DISCUSSION Since its earliest recognition, factor XI has been studied primarily as it occurs in plasma. However, it is also present in platelets, as first identified by Tuszynski et al (17). They observed platelet factor XI at Mr - 220,000, which upon reduction was converted to a M, - 53,000 form. Herein we report some unique properties of platelet factor XI. Pt-XI shares both procoagulant activity and antigenic recognition with plasma factor XI. From the observation of procoagulant activity associated with Pt-XI one infers that activation may proceed via the same mechanism as in plasma. We demonstrate herein for the first time that indeed platelet factor XI, like plasma factor XI, can be activated by purified plasma
96
PLATELET FACTOR XI
Vol. 60, No. 1
factor XII and high molecular weight kininogen in the presence of an activating surface. This combination results in the cleavage over time of reduced platelet factor XI having M = 52,000 to the final form of = 44,500. Further experiments will 6e required to establish a definitive Mr comparison of the relative rates of reaction of the plasma and platelet forms. It is clear that platelet and plasma factor XI are dissimilar. Pt-XI is significantly larger, manifesting, In contrast, plasma factor XI is smaller at Mr = 245,000. reduction the differences are amplified. Pt-XI appears chains of M, = 52,000, whereas the plasma form yields two chains of M, = 80,000.
physically quite in our hands, a On Mr = 160,000. to yield several larger identical
On activation the products are distinct. Plasma XI is cleaved into two dissimilar chains of about 48 and 33 kD (6) which have separate functions. In The lighter bears the active site and the heavier the adsorption domain. contrast, as shown above, Pt-XI is degraded over time from 52 kD to a product of M, - 44,500 on reduction. This one chain carries both the active site which binds DFP and the adsorption domain of Pt-XI,. It is interesting to consider this form of activated factor XI as a possible intermediate in physiologic clotting. Surprisingly, whereas platelet factor XI is activated by purified clotting factors, it is not activated by six times the amount of trypsin which effectively activates plasma factor XI. This is consistent with the absence of any evidence of cleavage of Pt-XI under conditions readily cleaving the plasma form. A comparison of the sizes of plasma and platelet factor XI, suggests that the plasma form may contain additional information, perhaps relating to other plasma-specific reactions. It will be very revealing to subject Pt-XI to sequence analysis, as has been previously reported for plasma factor XB (18), to attempt to localize the foci of biologic activity in Pt-XI,.
ACKNOWLEDGMENT We wish to thank Cyndia Hanna for her expert help in preparing this manuscript. This work was supported in part by grant HL 17402 from the National Heart, Lung, and Blood Institute, National Institutes of Health.
REFERENCES 1.
LAMMLE, B. and GRIFFIN, J.H. Formation of the fibrin clot: The balance of procoagulant and inhibitory factors. Clinics b Haematoloav. u, 281342, 1985.
2.
KAPLAN, A.P. and SILVERBERG, M. The coagulation-kinin pathway of plasma. Blood._a, 1-15, 1987.
3.
MANN, K.G., JENNY, R.J. and KRISHNASWANY, S. Cofactor proteins in assembly and expression of blood clotting enzyme complexes. Am Biochem, z; 915-56, 1988.
4.
COLMAN, R.W. Initiation of blood coagulation at artificial surfaces. &Q H I&& U m, 253-67, 1987.
human
the Rev
Vol. 60, No. 1
PLATELET FACTOR XI
97
5.
SCHIFFMAN, S., PECCI, R. and LEE, P. Contact activation of factor XI: Evidence that the primary role of contact activation cofactor (CAC) is to facilitate the activation of factor XII. Thromb Res lo. 319-23, 1977.
6.
MANNHALTER, C., SCHIFFMAN, S. and JACOBS, A. Trypsin activation factor XI. -_--7 J Biol Chem 255 2667-9, 1980.
7.
JACOBS, A., MANNHALTER, C., MARGALIT, R. and SCHIFFMAN, activation of factor XI. & J Haematol. 49, 77-86, 1981.
8.
SCHIFFMAN, S., MANNHALTER, C. and TYNER, K.D. Human high molecular weight kininogen. Effects of cleavage by kallikrein on protein structure and procoagulant activity. J Biol Chem, 255, 6433-8 1980.
9.
KIRBY, E.P. and McDEVITT, Blood a, 652-9, 1983. A
10.
SCHIFFMAN, S. and LEE, P. Preparation, characterization, and activation of a highly purified factor XI. Evidence that a hitherto unrecognized plasma activity participates in the interaction of factors XI and XII. & J Haematol, 21, 101-14, 1974.
11.
R.W. SCOTT, C.F., SINHA, D., SEAMAN, F.S., WALSH, P.N. AND COLMAN, Amidolytic assay of human factor XI in plasma: Comparison with a coagulant assay and a new rapid radioimmunoassay. Blood . 63, 42-50. 1984.
12.
;rfyLk25
P.J. The binding of
factor XII
S.
to
of human
Contact
kaolin.
J.L. and JOHANSSON, B.G. Enzymatic iodination of polypeptides I to high specific activity. Biochim Bionhvs Acta. 2jl, 363-9,
1971. 13.
RIMON, A., SCHIFFMAN, S., FEINSTEIN, D.I. and RAPAPORT, S.I. activity and factor XI antigen in homozygous and heterozygous deficiency. Blood. 48, 165-74, 1976.
Factor XI factor XI
14.
LAEMMLI, U.K. Cleavage of structural proteins during assembly head of bacteriophage T4. Nature. 227. 680-5, 1970.
15.
MANNHALTER, C., HELLSTERN, P. and DEUTSCH, defective factor XI cross-reacting material patient. Blood. 70, 31-7, 1987.
16.
SCHIFFMAN, S., RIMON, A., and RAPAPORT, S.I. Factor XI and platelets: Evidence that platelets contain only minimal factor XI activity and antigen. & J Haematol. 35, 429-36, 1977.
17.
TSUZYNSKI, G.P., BEVACQUA, S.J., SCHMAIER, A.H., COLMAN, R.W., and WALSH, P.N. Factor XI antigen and activity in human platelets. Blood . 59, 1148-56, 1982.
18.
FUJIKAWA, K., CHUNG, D.W., HENDRICKSON, L.E. and DAVIE, E.W. Amino acid sequence of human factor XI, a blood coagulation factor with four tandem repeats that are highly homologous with plasma prekallikrein. Biochemistrv. 2, 2417-24, 1986.
of
the
E. Identification of a in a factor XI deficient
PURIFICATION AND CHARACTERIZATION OF PLATELET FACTOR XI
Sandra Schiffman and Che-Hung Yeh Departments of Medicine and Biochemistry University of Southern California, School of Medicine Los Angeles, California 90033 U.S.A. (Received
28.3.1990;
accepted
in revised form 14.8.1990
by Editor H. Pirkle)
ABSTRACT Factor XI activity and antigen was purified about 300 fold from human platelets through chromatography on Con-A Sepharose, SP-Sephadex C-50, immobilized goat anti-factor XI, and SP-Sephadex. The partially purified platelet factor XI (Pt-XI) could be activated by activated factor XII generated in situ from single chain factor XI in a reaction requiring high molecular weight kininogen (HMWK) and a surface. Native Pt-XI migrated as a molecule of Mr = 245,000 on sodium dodecyl sulfate polyacrylamide gel electrophoresls (SDS-PAGE) as identified by Western blotting. On reduction, Pt-XI appeared to have a M, = 52,000. Neither form was affected by exposure to trypsin. Incubation of Pt-XI with purified factor XII, HMWK, and kaolin produced activated platelet factor XI clotting activity and, concomitantly, the generation over time of a new chain on reduced SDS-PAGE of M, - 44,500. The coagulant activity of the activated form could be neutralized by diisogropyl flurophosphate (DFP). Incubation of the activated mixture with H-DFP followed by reduced SDS-PAGE showed the active site to be associated with a unit of Mr - 44,500. The adsorption domain as defined by adsorption to kaolin was localized to the M, 0 44,500 chain containing the active site. Hence, both active site and adsorption functions, properties of separate chains in plasma factor XI, reside in the same chain of Mr - 44,500 of platelet factor XI.
INTRODUCTIW The interactions among the soluble protein clotting factors have been worked out in considerable detail (l-4), but the B reactions which initiate intrinsic clotting remain a mystery. In the glass test tube it is the interaction of factors XI, XII, prekallikrein, and HMWK with the glass surface to yield activated factor XI. Of these, only factor XI is reauired for normal Zhemostasis (4). Therefore, we have initiated a search for an alternate pathway. Herein, we report the purification, characterization, and
Key words:
Factor XI, platelets, Western blotting. 87
88
PLATELET FACTOR XI
Vol. 60, No. 1
mechanism of activation of a platelet protein which has factor XI procoagulant and antigen activity. It is similar to but different than plasma factor XI.
MATERIALS AND METHODS All chemicals obtained from commercial sources were the best grade available. Bovine trypsin (twice crystallized, dialyzed, salt-free, and lyophilized) and aprotinin (from bovine were from Sigma. lung) Chromatographic media were from Pharmacia. Protein concentration was determined using the Bio-Rad reagent according to the direc ions of the 5H-diisopropyl globulin standard. manufacturer with a bovine gamma flurophosphate (DFP) was from NEN. The chromogenic substrate S-2366 was from described Helena Laboratories. Ellagic acid (Sigma) was prepared as previously (5). Purified reaeents. Factor XI, factor XII, and HMWK were prepared as described previously yielded a single band on SDS-PAGE. Each clotting factor (6, 7, 8). Factor XI had a specific activity of 200 U/mg, factor XII had a specific Corn activity of 30.3 U/mg, and HMWK had a specific activity of 11 U/mg. trypsin inhibitor was prepared by the method of Kirby and McDevitt (9). Assays. Coagulant activitv of factor XI and factor XII was measured in modified activated partial thromboplastin time assays as described previously (10). HMWK was measured in a similar system with commercial Fitznerald trait plasma substrate from George King. Factor XI amidolvtic activiiy was measured as modified from Scott et al (ll), by incubating 50 ~1 of sample with 80 pl of 3.9 mM S-2366 for 30 minutes at room temperature in a 96 well microtiter The reaction was terminated after 30 minutes by adding 100 ~1 of 50% plate. acetic acid, and the plate was read at 405 nM. Radioimmunoassav of factor XI antigen was carried out by a modification of that described by Scott et al Goat antihuman-factor XI antiserum, prepared as described below, at a (11). dilution of l/10,000, was used as the primary antibody; swine antigoat (Boehringer Mannheim) was used as the second antibody, followed by 10% Staphylococcus Protein A (Boehringer Mannheim). Isotope was radiolabeled plasma factor XI (see below). 125I the Factor XI was prepared with (Amersham) by Radiolabeled lactoperoxidase method (12) with slight modification. The final product containing 77 pCi/pg retained its procoagulant activity. Goat antiserum & purified plasma factor XI was raised by five innoculations with purified human factor XI mixed in equal parts with Freund's adjuvant. It reacted with three protein fractions in partially purified platelet factor XI which were separable by gel filtration. Only one of these had procoagulant activity. Therefore, the antiserum was immunoadsorbed by exposure to a column of CNBr-activated Sepharose 4B to which had been coupled the inactive The resulting immunoadsorbed antiserum was monospecific and fractions. activity. This with the fraction having procoagulant only reacted monospecific adsorbed antiserum was then chromatographed on DEAE-Sephadex and the unbound fraction was further adsorbed with immobilized factor XI deficient plasma as described previously (13) and concentrated by dialysis against Sephadex G-25. The antibody used in these experiments had a titer of l/2500. Polvacrylamide & electronhoresis (PAGE). PAGE in the presence of 0.1% sodium dodecvl sulfate (SDS) (Sigma) was carried out on 10% slab gels of 0.75 mm thickness essentially according to the method of Laemmli (14). Gels were either subjected to autoradiography or cut
Vol. 60, No. 1
PLATELET FACTOR XI
into 1.1 mm sections and added to 10 ml PCS scintillation for analysis in a beta spectrometer.
89
cocktail
(Amersham)
Western blotting. et al (15). Protein was blotted according to the method of Mannhalter Samples, either reduced or non-reduced, were subjected to SDS-PAGE in a Hoefer Tall Mighty Small Model SE 280 System using 0.75 mm thick gels at constant Proteins were transferred onto a voltage of 200 V for 80 minutes. in a Hoefer Mighty Small Transphor system using nitrocellulose membrane current (150 mA) for 90 minutes. After transfer, the nitrocellulose constant membrane was quenched with 0.4% gelatin, 0.2% Tween 20, 0.2% NaN3 in phosphate buffered saline pH 7.2 for 4 hours. It was then incubated with 10 ml of diluted to 2% in immunoadsorbed goat antihuman-factor XI antiserum The membrane was then washed concentration in the same buffer for 16 hours. the same buffer without gelatin. three times (20 minutes each) in 10 ml 755 I-factor XI for 4 hours. washed The membrane was then incubated with 3 times, air dried, and set up for autoradiography or cut into 1 or 2 mm strips for counting. All incubations were at room temperature. Purification of Platelet Factor XI (Pt-XI). Pt-XI was purified by sequential chromatography on Con-A Sepharose, Platelets SP-Sephadex, anti-factor XI-Sepharose, and SP-Sephadex as follows: were prepared from forty or fifty units of pooled outdated platelets from the Red Cross by a modification of our previous method (16) in which an American inhibitor cocktail, adapted from Tuszynski et al (17), was added, which yielded a final solution containing 10 mM benzamidine-HCl, 5 mM e-aminocaproic acid (EACA), 2 mM EDTA, 0.02% soybean trypsin inhibitor, and 50 KIU/ml aprotinin. Red blood cells were removed by centrifugation at !jOO g for Platelets were sedimented by centrifugation at 800 g for 15 minutes. with gentle mixing in 15 minutes at 23'C. The precipitate was suspended Tyrode's solution buffered with HEPES [4-(2-hydroxy-ethyl)-l-piperazineethanesulfonic acid], pH 7.35: 0.14 M NaCl, 2.7 mM KCl, 12 mM NaHC03, 1mM Mg?2:6 H20tabZ;Z, ~T;y;;i;?;l~;8 ~s~~~~~,~.~asm”i~~~2~~d ‘;:““;~y ;:; inhl ltors 20 minutes and then gel filtered on a 5.0 x 50 cm column of Sepharose CL-2B which had been equilibrated with above Tyrode's solution, buffered with HEPES, pH 7.35 containing inhibitors. The flow rate was 0.6 ml/minute. Obviously turbid fractions were pooled as g& filtered Dlatelets and frozen at -2O'C. A fraction binding to Con-A Sepharose was prepared from the frozen platelets, as described by Tuszynski et al (17), by thawing and centrifugation at 800 g for 15 minutes. The precipitate was suspended in Tyrode's solution buffered with HEPES, pH 7.35 containing inhibitors as above. Triton X-100 was added to 0.2%. The solubilized & filtered Dlatelet solution was centrifuged overnight in a Beckman Microfuge for 4 minutes and the supernatant dialyzed against Con-A Sepharose column buffer: 1 n&l CaC12, 1 mM MgC12, 1 mM MnC12. and 1 mM ascorbic acid in 0.02 M Tris-HCl, pH 8.0. The dialyzed sample was in the Microfuge for 2 minutes, and the supernatant applied to a centrifuged Con-A SeDharose column (2.5 x 25 cm). The column was washed with about 360 ml of column buffer which resulted in the elution of an unbound protein peak between 60-100 ml. After another 250 ml of buffer, bound protein was eluted with 1 M a-D-methylmannoside in 0.02 M Tris-HCl, pH 8.0. Active fractions eluted in about 35 ml. Pooled active fractions from Con-A Sepharose were dialyzed overnight against 50 mM sodium acetate pH 5.2 containing 1 mM EDTA and 1 mM benzamidineHCl and applied to a 1.5 x 30 cm SP-SeDhadey c-50 column equilibrated with this buffer. Unbound proteins were eluted with the same buffer. Then a linear gradient of 150 ml each of equilibrating buffer and the same buffer containing 0.7 M NaCl was formed which eluted the platelet factor XI. The pooled active fractions from SP-Sephadex were dialyzed against glycol 0.02 M Tris buffer, pH 7.2 containing 0.5 M NaCl , 0.2% polyethylene
90
PLATELET FACTOR XI
Vol. 60, No. 1
and applied to a 0.9 x 3 cm column of BIPIPy.noadsorbed, go& anti*_ XL antlserum coupled to cyanogen-bromide-activated Sepharose 4B. The column was washed with 0.02 M Tris, pH 7.2 buffer containing 0.5 M NaCl, 0.2% PEG. It was eluted with 0.02 M Tris pH 7.2 containing 4 M NaSCN. Samples containing protein were immediately dialyzed against 0.05 M acetate buffer pH 5.2 containing 1 mM benzamidine and 1 mM EDTA and re-chromatographed on a Sephadeq (1.5 x 15 cm) as above. The fractions containing factor XI activity were pooled as DurifiedolateletfactorXT.
RESULTS
Platelet factor XI was purified as described above and summarized in Table 1. The product was purified about 300 fold. It could be identified both by clotting activity and by immunoreactivity. Both assay systems were quite consistent, though comparison suggests that the preparation contains some factor XI antigen lacking activity. Practical evaluation of the size of Pt-XI could be attained only by Western blotting (vide infra) since stained gels required far too much protein.
TABLE 1 .
Purification m
Volume
.
catwn nd Platelet
Protein
Factor 3.L
Total Activity (units)
Recovery
Specific Activity funit-
Solubilized Gel-Filtered Platelets
12.5
Con-A Sepharose
46
24.1
20.7
78.9
0.86
SP-Sephadex-(1)
36
3.8
18.0
68.4
4.74
Anti-XI Affinity
8
0.7
7.2
27.4
10.3
SP-Sephadex-(2)
1
0.27
5.4
20.5
20
400
26.3
100
0.066
Pt-XI was evaluated for ability to undergo activation as defined by Two activators of plasma factor XI were increased coagulant activity. factor XII combined with HMWX and surface activator tested, trypsin and (Tables 2 and 3). Pt-XI readily activated in the presence of factor XII, HMWX and kaolin (Table 2, line 1). Further clotting tests at 75' and 90' were Activation required the unchanged indicating that reaction was complete. complete reaction mixture; no activation occurred in the absence of any of the components (Table 2, lines 3-9). In contrast, both soluble and insoluble trypsin at concentrations which readily activated plasma factor XI failed to activate Pt-XI (Table 3). To determine whether Pt-XI was perhaps interacting with trypsin without undergoing effective activation two evaluations were made. Pt-XI was combined with plasma XI and added to insoluble trypsin which resulted in clotting times
91
PLATELET FACTOR XI
Vol. 60, No. 1
TABLE 2 Activation of Pm
Factor XI (Pt-XI) With Purified ClottinP Factors
15”
Reaction Mixture
Incubation Time 30' 10' 20' ClottinrrTime
§.!E
1)
Pt-XI + XII + HMWK + Kaolin
246
211
182
159
135
2)
Pt-XI + XII + HMWK + Ellagic Acid
262
218
192
173
152
3)
Pt-XI + XII + HMWK + Buffer
255
241
240
238
241
4)
Pt-XI + XII + Buffer + Kaolin
240
234
227
225
225
5)
Pt-XI + XII + Buffer + Buffer
249
243
241
243
241
6)
Pt-XI + Buffer + Buffer + Kaolin
244
246
239
242
240
7)
Pt-XI + Buffer + HMWK + Buffer
255
254
256
249
248
8)
Pt-XI + Buffer + HMWK + Kaolin
248
242
239
243
241
9)
Buffer + XII + HMWK + Kaolin
289
281
282
283
280
10) Plasma-XI + XII + HMWK + Kaolin
232
177
147
1375
128
A mixture of Pt-XI (0.14 U/ml), factor XII (0.19 U/ml), HMWK (0.24 U/ml), and kaolin (0.24 mg/ml) or ellagic acid (8 pg/ml), or buffer was Vortexed and incubated at room temperature. At times shown, 50 pl of the sample were removed, and the clotting activity was measured in the factor XI assay in a polystyrene tube without added kaolin or preincubation.
TABLE 3 Effect of Trvnsin QQ Platelet Factor XJ
II_ Reaction Mixture
Incubation T_& 15' 30' Clottine Time
60'
1)
Plasma-XI + Insoluble Trypsin
241
178
119
88
2)
Platelet-XI + Insoluble Trypsin
246
243
244
241
3)
Plasma-XI + Soluble Trypsin (3.2 pg/ml)
262
176
11.9
86
4)
Platelet-XI + Soluble Trypsin (3.2 pg/ml)
248
244
242
245
5)
Platelet-XI + Soluble Trypsin (19.2 pg/ml)
243
240
237
239
In experiments shown in lines 1 and 2, 500 cl1of plasma- or platelet-factor XI at 2 U/ml activity was combined at room temperature with 300 ~1 insoluble trypsin. In lines 3-5, platelet- and plasma-factor XI at 2 U/ml activity were incubated with soluble trypsin at 3.2 pg/ml (lines 3 & 4) or 19.2 pg/ml (line 5). At times shown, the samples were Vortexed, and 50 ~1 were removed and tested in the factor XI assay in a polystyrene tube without kaolin or preincubation.
PLATELET FACTOR XI
92
Vol. 60, No. 1
of the mixture identical to those of plasma factor XI tested alone. Hence, there was no evidence of competition. Furthermore, insoluble trypsin exposed to Pt-XI for one hour, removed, washed 5 times and added to plasma XI was unchanged; tryptic activity was unaffected by exposure to Pt-XI. Nor was there any change in size of Pt-XI as indicated by Western blotting after 30 minutes exposure to trypsin which readily cleaved plasma XI. Hence, Pt-XI appears to be at best very insensitive to tryptic cleavage. Since the data of Table 2, lines 1 and 3, indicated that kaolin played a critical role in the activation of Pt-XI, it was of interest to evaluate what another pseudo-surface could do in this reaction system. Therefore, ellagic acid, which is effective with plasma factor XI (5), was tested. As shown in line 2, Table 2, ellagic acid effectively promoted the activation of Pt-XI. Reaction mixtures of Pt-XI combined with potential activators, such as those described in Table 2, were evaluated for generation of amidolytic activity. The only reaction mixture which showed increasing amidolytic activity was that containing factors XII, HMWK, and kaolin (Table 4, line 1).
TABLE 4 Activation of Platelet Factor u
as Reflected $y Amidolvtic Assay
Reaction Mixture 15"
Incubation m 15' Absorbance 405~
30’
1)
Pt-XI + XII + HMWK + Kaolin
0.179
0.529
1.002
2)
Pt-XI + XII + HMWK + Buffer
0.065
0.065
0.052
3)
Pt-XI + XII + Buffer + Kaolin
0.059
0.058
0.064
4)
Pt-XI + XII + Buffer + Buffer
0.054
0.060
0.060
5)
Pt-XI + Buffer + Buffer + Kaolin
0.061
0.066
0.068
6)
Pt-XI + Buffer + HMWK + Buffer
0.061
0.061
0.055
7)
Pt-XI + Buffer + HMWK + Kaolin
0.066
0.065
0.059
8)
Buffer + XII + HMWK + Kaolin
0.055
0.055
0.055
Incubation mixtures of 210 ~1 containing 25 ~1 platelet factor XI at 1.2 U/ml, 50 pl of HMWK at 1 U/ml, 100 ~1 of factor XII at 0.4 U/ml, 10 al 5 mg/ml kaolin, or Tris/HSA as required, were combined. At the times shown, 50 /.bl were removed and added to 80 ~1 of 3.9 mM S-2366 in a 92 well microtiter After 30 minutes at room temperature, 100 ~1 of 50% acetic acid was plate. added, and the absorbance at 405 runwas read.
The size of Pt-XI and its activated products was evaluated on SDS-PAGE The results using Western blotting and autoradiography as described above. are shown in Figure 1. Pt-XI was dramatically bigger than plasma factor XI, having an apparent molecular weight of 245,000 (lane 8). By contrast, plasma factor XI displayed a M, - 165,000 (lane 10). Reduced Pt-XI migrated as a
PLATELET FACTOR XI
Vol. 60, No. 1
93
single entity of Mr = 52,000 (lane 4), slightly larger than the heavy chain of trypsin-activated plasma factor XI at M, = 49,000 (lane 3). Pt-XI which had been exposed to trypsin under comparable conditions was unchanged (data not shown). However, Pt-XI exposed to factor XII, HMWK, and kaolin for 30 minutes, which had activated factor XI activity (see Table 2), migrated as two discrete chains; one of IQ = 52,000 and the other of Mr = 44,500 (lane 5). On continued reaction with factor XII, HMWK, and kaolin for 60 minutes, Pt-XI was apparently entirely converted to the smaller chain, M, = 44,500 (lane 6). Hence, activation of platelet factor XI with clotting proteins leads to conversion over time from M, = 52,000 to 44,500 (lanes 4, 5, and 6).
1234
FIG. 1
Immunoblot
of Platelet-
56
7
8
and Plasma-Factor
910
XI.
Ten ~1 samples were prepared for SDS-PAGE followed by Western blotting as described in Materials and Methods. Lanes 1-6 contained reduced lanes 7-10 contained n.on-reduced samples; samples. Lane 1 contained prestained protein standards of M 200 K, 92.5 K, 68 K, 43 K, 25.7 K, and 18.4 K. After SDS-PAGh a:: transfer, they were manually marked with 1251-factor XI and subjected to autoradiography. Lanes 2 through 10 contained the following proteins: lanes 2 and 10 - 0.1 pg factor XI; lanes 3 and 9 - 0.1 pg trypsin-activated factor XI (6); lanes 4 and 8 - 0.13 pg Pt-XI; lanes 5, 6, and 7 - 0.13 pg Pt-XI activated by coagulant proteins for 30 minutes (lane 5). or 60 minutes (lanes 6 and 7) as described in Table 2. line 1.
Activated Pt-XI (Pt-XI,), prepared as described in the legend to Table 2, could be reacted with DFP which neutralized its procoagulant activity. For example, the clotting time of a fully activated mixture diluted l/10 rose from 136" to 252" on the addition of DFP indicating neutralization yf 98% of the clotting activity. When concentrated Pt-XI, was reacted with H-DFP, and the product subjected to SDS-PAGE, the resulting gel could be cut into lanes and subdivided into 1.1 mm sections to be counted in a beta spectrometer as described in Materials and Methods. As shown in Figure 2, DFP
94
PLATELET FACTOR XI
Vol. 60, No. 1
reacted with a M, = 44,500 band. Thus, the active site region appeared to included in the new, lower molecular weight band arising on activation.
be
To rule out the remote possibility that the labeled material defined in Fig. 2 at M, = 44,500 could be derived from the only other potential protease added to the system, factor XII, a parallel experiment was prepared lacking only Pt-XI but containing factor XII. It was neutralized with CTI, reacted with DFP, and subjected to SDS-PAGE followed by slicing and counting. No DFP-binding protein was detected. Hence, the labeled fraction defined in Fig. 2 could not be attributed to factor XII. localize the surface adsorption sites of Pt-XI, Pt-XI, was labeled with '$-DFP as above exposed to kaolin, the kaolin washed 5 times, and the were radioactivity of the kaolin and of the kaolin-adsorbed fraction determined. About 35% of the label was associated with the kaolin. This was consistent with the conclusion that the adsorption site is included in the region which contains the DFP-binding active site.
co
1.0
0 T; 0.8
I
0.6
& 0.4 0.2 0
Gel Slice Number
FIG. 2
SDS-polyacrylamide gel electrop oresis platelet factor XI inhibited with !3 H-DFP.
of
reduced,
activated
200 ~1 of a mixture containing Pt-XI at 1.5 U/ml was activated with factor XII, HMWK, and kaolin for 60 minutes as described in the legend to Table 2, line 1. Then 5 pl of corn trypsin inhibitor (1 mg/ml) were added, and the combination was incubated for 10 minutes at room temperature. They the mix was reacted for H-DFP (20 @i), dialyzed 120 minutes with 20 ~1 of 3.3 mM SDS-polyacrylamide gel and 37.5 ~1 subjected to overnight, electrophoresis as described in Materials and Methods followed by slicing and counting in a beta spectrometer.
The surface binding portion of Pt-XI, was more explicitly demonstrated by adsorbing Pt-XI, with kaolin, washing the kaolin, and eluting with 10% SDS.
PLATELET FACTOR XI
Vol. 60, No. 1
95
The eluate as well as the initial Pt-XI, and the non-adsorbed fraction, comprising about l/3 of the total Pt-XI,, were subjected to SDS-PAGE followed by Western blotting. As shown in Figure 3, the eluate, which must contain the adsorption domain, migrated with a Mr - 44,500. This was the identical position attained by the Pt-XI, and the non-adsorbed fraction. Hence, in this activated mixture there is only one size of immunoreactive molecule which contains m the surface binding domain and the active site. There was no evidence of non-productive tryptic cleavage products.
1.8 1.6 44,500
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 5
10
15
20
25
30
35
Gel Slice
FIG. 3
Immunoblot of SDS-eluate of Pt-XI, adsorbed to kaolin SDS-PAGE.
subjected to
800 pl of fully activated Pt-XI containing 1.2 U/ml Pt-XI was added to 2 mg kaolin, incubated for 1 hr. The supernatant was removed and the kaolin washed 5 times with buffer, then eluted with 100 pl of 10% SDS for 10 min. The kaolin eluate was run on 10% SDS-PAGE and then subjected to Western blotting. The membrane was sliced in 2 mm sections and counted in a gamma spectrometer.
DISCUSSION Since its earliest recognition, factor XI has been studied primarily as it occurs in plasma. However, it is also present in platelets, as first identified by Tuszynski et al (17). They observed platelet factor XI at Mr - 220,000, which upon reduction was converted to a M, - 53,000 form. Herein we report some unique properties of platelet factor XI. Pt-XI shares both procoagulant activity and antigenic recognition with plasma factor XI. From the observation of procoagulant activity associated with Pt-XI one infers that activation may proceed via the same mechanism as in plasma. We demonstrate herein for the first time that indeed platelet factor XI, like plasma factor XI, can be activated by purified plasma
96
PLATELET FACTOR XI
Vol. 60, No. 1
factor XII and high molecular weight kininogen in the presence of an activating surface. This combination results in the cleavage over time of reduced platelet factor XI having M = 52,000 to the final form of = 44,500. Further experiments will 6e required to establish a definitive Mr comparison of the relative rates of reaction of the plasma and platelet forms. It is clear that platelet and plasma factor XI are dissimilar. Pt-XI is significantly larger, manifesting, In contrast, plasma factor XI is smaller at Mr = 245,000. reduction the differences are amplified. Pt-XI appears chains of M, = 52,000, whereas the plasma form yields two chains of M, = 80,000.
physically quite in our hands, a On Mr = 160,000. to yield several larger identical
On activation the products are distinct. Plasma XI is cleaved into two dissimilar chains of about 48 and 33 kD (6) which have separate functions. In The lighter bears the active site and the heavier the adsorption domain. contrast, as shown above, Pt-XI is degraded over time from 52 kD to a product of M, - 44,500 on reduction. This one chain carries both the active site which binds DFP and the adsorption domain of Pt-XI,. It is interesting to consider this form of activated factor XI as a possible intermediate in physiologic clotting. Surprisingly, whereas platelet factor XI is activated by purified clotting factors, it is not activated by six times the amount of trypsin which effectively activates plasma factor XI. This is consistent with the absence of any evidence of cleavage of Pt-XI under conditions readily cleaving the plasma form. A comparison of the sizes of plasma and platelet factor XI, suggests that the plasma form may contain additional information, perhaps relating to other plasma-specific reactions. It will be very revealing to subject Pt-XI to sequence analysis, as has been previously reported for plasma factor XB (18), to attempt to localize the foci of biologic activity in Pt-XI,.
ACKNOWLEDGMENT We wish to thank Cyndia Hanna for her expert help in preparing this manuscript. This work was supported in part by grant HL 17402 from the National Heart, Lung, and Blood Institute, National Institutes of Health.
REFERENCES 1.
LAMMLE, B. and GRIFFIN, J.H. Formation of the fibrin clot: The balance of procoagulant and inhibitory factors. Clinics b Haematoloav. u, 281342, 1985.
2.
KAPLAN, A.P. and SILVERBERG, M. The coagulation-kinin pathway of plasma. Blood._a, 1-15, 1987.
3.
MANN, K.G., JENNY, R.J. and KRISHNASWANY, S. Cofactor proteins in assembly and expression of blood clotting enzyme complexes. Am Biochem, z; 915-56, 1988.
4.
COLMAN, R.W. Initiation of blood coagulation at artificial surfaces. &Q H I&& U m, 253-67, 1987.
human
the Rev
Vol. 60, No. 1
PLATELET FACTOR XI
97
5.
SCHIFFMAN, S., PECCI, R. and LEE, P. Contact activation of factor XI: Evidence that the primary role of contact activation cofactor (CAC) is to facilitate the activation of factor XII. Thromb Res lo. 319-23, 1977.
6.
MANNHALTER, C., SCHIFFMAN, S. and JACOBS, A. Trypsin activation factor XI. -_--7 J Biol Chem 255 2667-9, 1980.
7.
JACOBS, A., MANNHALTER, C., MARGALIT, R. and SCHIFFMAN, activation of factor XI. & J Haematol. 49, 77-86, 1981.
8.
SCHIFFMAN, S., MANNHALTER, C. and TYNER, K.D. Human high molecular weight kininogen. Effects of cleavage by kallikrein on protein structure and procoagulant activity. J Biol Chem, 255, 6433-8 1980.
9.
KIRBY, E.P. and McDEVITT, Blood a, 652-9, 1983. A
10.
SCHIFFMAN, S. and LEE, P. Preparation, characterization, and activation of a highly purified factor XI. Evidence that a hitherto unrecognized plasma activity participates in the interaction of factors XI and XII. & J Haematol, 21, 101-14, 1974.
11.
R.W. SCOTT, C.F., SINHA, D., SEAMAN, F.S., WALSH, P.N. AND COLMAN, Amidolytic assay of human factor XI in plasma: Comparison with a coagulant assay and a new rapid radioimmunoassay. Blood . 63, 42-50. 1984.
12.
;rfyLk25
P.J. The binding of
factor XII
S.
to
of human
Contact
kaolin.
J.L. and JOHANSSON, B.G. Enzymatic iodination of polypeptides I to high specific activity. Biochim Bionhvs Acta. 2jl, 363-9,
1971. 13.
RIMON, A., SCHIFFMAN, S., FEINSTEIN, D.I. and RAPAPORT, S.I. activity and factor XI antigen in homozygous and heterozygous deficiency. Blood. 48, 165-74, 1976.
Factor XI factor XI
14.
LAEMMLI, U.K. Cleavage of structural proteins during assembly head of bacteriophage T4. Nature. 227. 680-5, 1970.
15.
MANNHALTER, C., HELLSTERN, P. and DEUTSCH, defective factor XI cross-reacting material patient. Blood. 70, 31-7, 1987.
16.
SCHIFFMAN, S., RIMON, A., and RAPAPORT, S.I. Factor XI and platelets: Evidence that platelets contain only minimal factor XI activity and antigen. & J Haematol. 35, 429-36, 1977.
17.
TSUZYNSKI, G.P., BEVACQUA, S.J., SCHMAIER, A.H., COLMAN, R.W., and WALSH, P.N. Factor XI antigen and activity in human platelets. Blood . 59, 1148-56, 1982.
18.
FUJIKAWA, K., CHUNG, D.W., HENDRICKSON, L.E. and DAVIE, E.W. Amino acid sequence of human factor XI, a blood coagulation factor with four tandem repeats that are highly homologous with plasma prekallikrein. Biochemistrv. 2, 2417-24, 1986.
of
the
E. Identification of a in a factor XI deficient
