Evidence for a plasma inhibitor of the heparin accelerated inhibition of factor Xa by antithrombin III

584

Biochimica et Biophysica Acta, 586 ( 1 9 7 9 ) 5 8 4 - - 5 9 3 © E l s e v i e r / N o r t h - H o l l a n d Biomedical Press

BBA 29006 EVIDENCE F O R A PLASMA INHIBITOR OF THE HEPARIN ACCELERATED INHIBITION OF F A C T O R Xa BY ANTITHROMBIN III

IAN R. M a c G R E G O R , D A V I D A. L A N E * and V I J A Y V. K A K K A R

Thrombosis Research Unit, King's College Hospital, Medical School, Denmark Hill, London SE5 (U.K.) (Received D e c e m b e r 2 7 t h , 1978)

Key words: Heparin; Antithrombin III; Thrombin; Factor Xa; Amidolytic

Summary The ability of heparin fractions of different molecular weight to potentiate the action of antithrombin III against the coagulation factors thrombin and Xa has been examined in purified reaction mixtures and in plasma. Residual thrombin and Xa have been determined by their peptidase activities against the synthetic peptide substrates H-D-Phe-Pip-Arg-pNA and Bz-Ile-Gly-Arg-pNA. High molecular weight heparin fractions were found to have higher anticoagulant activities than low molecular weight heparin when studied with both thrombin and Xa incubation mixtures in purified mixtures and in plasma. The inhibition of thrombin by heparin fractions and antithrombin III was unaffected by other plasma components. However, normal human plasma contained a c o m p o n e n t that inhibited the heparin and antithrombin III inhibition of Xa, particularly when the high molecular weight heparin fraction was used. Experiments using a purified preparation of platelet factor 4 suggested that this platelet-derived heparin-neutralizing protein was not responsible for the inhibition. Introduction An explanation of the anticoagulant effects of heparin has been provided by its interaction with a naturally occurring plasma inhibitor of serine proteases called antithrombin III [1--4]. Heparin has been shown to be able to accelerate the rate of inhibition of thrombin, Xa, IXa, and XIa by antithrombin III [5,6] and may effect this increase b y binding to antithrombin III and inducing a conformational change which facilitates antithrombin III binding to activated serine proteases [7,8]. However, heparin's strongly anionic character allows it * To whom currespondence should be addressed.

585 to bind to many other plasma proteins including factor IX, X, prothrombin, and thrombin. This has suggested an alternative mechanism of heparin action in whcih it binds to thrombin, rendering it more susceptible to neutralisation by antithrombin III [9--11]. The development of small molecular weight substrates for the measurement of thrombin and Xa amidolytic activity has stimulated interest in the heparinantithrombin III-protease reactions because of their sensitivity and specificity [9,12--16]. Use of the substrates H-D-Phe-Pip-Arg-pNA and Bz-Ile-Glu-GlyArg-pNA allows measurements of thrombin and Xa inhibition by heparin and antithrombin III in well~lefined reaction mixtures. In the present study we have used the substrates to study the influence of the molecular weight of heparin upon the antithrombin III protease reactions in purified systems and in plasma. The results of this work suggest the presence of a plasma inhibitor of the heparin accelerated inhibition of Xa by antithrombin III that has greater inhibitory activity towards high than low molecular weight heparin. Materials and Methods Porcine mucosal heparin (Lot CC0025, Riker Inc.) and bovine lung heparin (Lot 328CU, Upjohn Co. Ltd.} were fractionated by gel filtration on Ultrogel AcA44 (LKB). Essentially similar results were obtained using both sources of heparin, and only the results from the mucosal heparin fractions will be presented. Fractions w e r e prepared and their molecular weights estimated as described previously [17]. The Third International Standard heparin of porcine mucosal origin was obtained from NIBSC, Holly Hill, Hampstead, London, U.K. The weight of heparin in fractions was measured by a metachromatic assay using Azure A [18]. This method was previously shown to give results that were in good agreement with uronic acid content measured by a carbozole method and with other dye-binding methods. Bovine thrombin (Parke-Davis) was stored at --20°C as a 100 NIH U/ml stock solution in 0.9% (w/v) sodium chloride containing 1% glycerol and diluted immediately before use. Because of the low specific activity of this thrombin it was further purified by the method of Lundblad on sulphopropylSephadex [19] and some of the heparin-antithrombin III-thrombin reactions repeated. Essentially identical results were obtained using both thrombin preparations. Human antithrombin III, purified by the method of Miller-Andersson et al. [20] and obtained from AB Kabi, Stockholm, was stored at --20°C as a stock solution containing 2.8 antithrombin units/ml in 0.9% (w/v) sodium chloride. Chromatographed bovine factor Xa (Diagen, Thame, Oxon, U.K.) was diluted to 0.6 plasma unit/ml in 1% bovine serum albumin (Sigma, London) immediately before use. Synthetic chromogenic substrates Bz-Ile-Glu-Gly-Arg-pNA, $2222, for the assay of factor Xa activity, and H-D-Phe-Pip-Arg-pNA, $2238, for the assay of thrombin activity were obtained from AB Kabi, London. Full details of the subst.rates are available from the manufacturers, but of relevance to the present work is that $2222 has a 44 times greater specificity for Xa than thrombin,

586 while $2238 has a 30 times greater specificity for thrombin than Xa (Kabi Data Sheet). Citrated pooled human platelet-poor plasma was collected from a pool of t w e n t y healthy donors, while plasma anticoagulated by EDTA-theophylline [21] was a pool of eight normal volunteers. Both types of plasma were stored at --20°C. Thrombin and Xa amidolytic activities were measured with an LKB 2086 MkII reaction rate analyzer fitted with diluter (LKB 2075) and kinetic data processor (LKB 2082). The antithrombin-potentiating activity of heparin was determined as follows [22] : 10 ~l plasma or purified antithrombin III, 1 U/ml, containing heparin, 0--24 ~g/ml, was incubated with 1.2 ml of 0.05 mol/1 Tris-HC1, 0.02 mol/1 EDTA, 0.3 mol/1 sodium chloride, pH 8.4, containing 0.4 or 0.5 NIH units of thrombin/ml, for 9 min at 37°C. After incubation 0.5 ml of $2238 at a final concentration of 0.13 mmol/1 in 0.5 mol/1 Tris-HC1, pH 7.5, was added. AA40s was automatically measured, the slope being calculated as change in absorbance units/rain. The anti-Xa-potentiating activity of heparin was determined in a similar manner [23] 50 pl of plasma or purified antithrombin III, 1 U/ml, containing heparin (0--1.2 pg/ml) was added to 1 ml of 0.05 mol/1 Tris-HC1, 0.01 mol/1 EDTA, 0.1 mol/1 sodium chloride, pH 7.9; 0.2 ml chromatographed bovine Xa, (0.6 U/ml in 1% bovine serum albumin) was added and incubated for 120 s at 37°C. 50 pl $2222, 3.4 mmol/1 final concentration in distilled water was then added and AA405 measured over 120 s. Amidolytic activity was calculated by measuring change in absorbance/s, at 405 nm, using a molar extinction coefficient of 10 600 M -1 • cm -1 for p-nitroaniline at this wavelength. Purified platelet factor 4, PF4, a product of the platelet release reaction which has heparin-neutralizing properties [24] was kindly supplied by Dr. Duncan Pepper, Royal Infirmary, Edinburgh, Scotland. Reagents for a radioimmunoassay to platelet factor 4 were supplied b y Dr. E. Workman, Diagnostic Division, A b b o t t , U.S.A. Results To determine if there were differences in the rates of thrombin or Xa neutralization with high (29 000) and low (9700) molecular weight heparin fractions, the following experiments were performed. Preliminary experiments were performed to determine concentrations of high and low molecular weight fractions which gave approximately equal inhibition of thrombin after 10 rain incubation. Then, thrombin activity remaining after various times of incubation with heparin fractions at these concentrations with plasma or antithrombin III was measured, see Fig. 1. In the absence of heparin there was a slight progressive inhibition of thrombin b y plasma antithrombin. In the presence of the heparin fractions the pattern of accelerated thrombin inhibition with increasing incubation time was similar for low and high molecular weight heparin fractions. High and low molecular weight heparin fractions gave almost identical neutralization curves whether antithrombin activity was supplied b y purified antithrombin III or b y plasma. The difference b e t w e e n the slopes of the curves obtained in the purified system and plasma, Fig. l a and b, can be partially

587

35C

300~

30C

• 25

:

250

W 2oc z_ rn~ Z =

15C

~-~

"1(2:0

z

II ::::::~

I~'1

~5

L©

25

5

TO

II 9

25

INCUBATION TIME (MIN)

5

tF

?.5

INCUBATION TIME (MIN)

Fig. 1. Antithrombin-potentiating ability of two Riker hepazin fractions (R 2, M r 9700 and RT, M r 29 000) s t u d i e d using the substrate S 2 2 3 8 . (a) Purified h u m a n a n t i t h r o m b i n I I I , 1 . 7 • 1 0 -8 m o i f l final c o n c e n t r a t i o n w a s i n c u b a t e d for various l e n g t h s o f t i m e w i t h b o v i n e t h r o m b i n , 1 . 1 2 • 1 0 - 8 m o l ] l final c o n c e n t r a t i o n and residual t h r o m b i n d e t e r m i n e d . • =, R 2 final c o n c e n t r a t i o n 0 . 2 4 1 ~tg/ml; • •, R7 final c o n c e n t r a t i o n 0 . 0 5 4 # g / m l . ( b ) N o r m a l p o o l o f h u m a n p l a t e l e t p o o r p l a s m a giving 1.7 - 1 0 -8 m o l ] l final c o n c e n t r a t i o n o f a n t i t h r o m b i n III w a s i n c u b a t e d for various l e n g t h s o f t i m e w i t h b o v i n e t h r o m b i n , 0 . 8 9 • 1 0 - 8 t o o l / 1 final c o n c e n t r a t i o n and residual t h r o m b i n d e t e r m i n e d . • . . - - i R 2 final c o n c e n t r a tion 0.335 #g/ml; •-e R 7 final c o n c e n t r a t i o n 0 . 0 6 4 # g / m l .

explained by the difference in thrombin and heparin concentrations of the reaction mixtures, see legend to Fig. 1. The slightly higher thrombin concentration used in the purified system resulted in less inhibition by antithrombin III and the heparin fractions. However, it is possible that there may also be small kinetic differences involved in the neutralization of thrombin in plasma rather than buffered solutions. Molar concentrations of the heparin fractions which gave equivalent antithrombin effects were in the ratio o f 15.6 of low molecular weight fraction to 1 of high molecular weight fraction.

..m

60

o

~

~

2c

m

W x")2

S

T 0

60 INCUBATION TIME (SEC)

b

•

120

40

°1 T O

60 INCUBATION TIME (SEC)

120

Fig. 2. A n t i - X a - p o t e n t i a t i n g ability o f f r a c t i o n s 112 and 117 s t u d i e d using t h e substrate $ 2 2 2 2 t o m e a s u r e residual Xa. (a) Purified h u m a n a n t i t h r o m b i n III 8 . 2 • 1 0 -8 m o l ~ w a s i n c u b a t e d for various l e n g t h s o f t i m e w i t h b o v i n e X a , 2.1 • 1 0 - 8 m o l / l final c o n c e n t r a t i o n and residual X a d e t e r m i n e d . • -', R2 final c o n c e n t r a t i o n 0 . 1 1 7 p g / m l ; • e , R 7 final c o n c e n t r a t i o n 0 . 0 2 6 ~ g / m l . (b) N o r m a l p o o l o f h u m a n p l a t e l e t p o o r p l a s m a giving 8 . 2 • 1 0 -8 m o l f l final c o n c e n t r a t i o n o f a n t i t h r o m b i n III w a s i n c u b a t e d for various l e n g t h s o f t i m e w i t h b o v i n e X a , 2.1 • 1 0 - 8 m o l f l final c o n c e n t r a t i o n and residual X a d e t e r mined. • m, R 2 final c o n c e n t r a t i o n 0 . 1 1 7 # g / m l ; • -', R 7 final c o n c e n t r a t i o n 0 . 0 2 6 # g / m l .

588 When the incubation t i m e of heparin, antithrombin III and Xa was varied, a superimposable pattern of Xa inhibition with changing incubation time was seen with high and low molecular weight heparin fractions, Fig. 2a, when the molar concentration ratio of low to high molecular weight heparin fractions was again 15.6 to 1. When the same molar concentration ratio of low to high molecular weight heparin was used in the anti-Xa system with plasma-supplying antithrombin III activity, a very different result was obtained to that in the putified antithrombin III system (Fig. 2b). While the time course of Xa inhibition was again approximately linear, the reaction curves with low and high molecular weight material were no longer superimposable. Where the low molecular weight fraction showed almost the same pattern of Xa inhibition b o t h in purified and plasma systems, the high molecular weight fraction exhibited a greatly reduced capacity to inhibit Xa in the plasma system. To study the possible reasons for this effect the following mixing experiments were performed. Concentrations of high and low molecular weight fractions were chosen to give equal antithrombin III potentiating effects, both in the antithrombin and anti-Xa assays when purified antithrombin III supplied the inhibitory activity. Mixtures of plasma and purified antithrombin III were then made up to give various ratios of plasma to purified antithrombin III while maintaining overall antithrombin III content (as measured by progressive antithrombin activity) of the mixtures constant. Fig. 3a shows that the rate of thrombin inhibition b y high and low molecular weight fractions remained the same over the whole range of mixtures, from 100% purified antithrombin III to 100% plasma. In the anti-Xa assay system, Fig. 3b, while low molecular weight heparin activity remained fairly constant with an increasing percentage of plasma the activities of the high molecular weight heparin fractions were progressively reduced as the percentage of plasma was icnreased. These results therefore indicated that a factor present in plasma was more readily inhibiting high molec-

50 4OO

40 ~-

/

<

300

30 Z nn~

20(2

~

~oc

~

0

50 1(30 PLASMA PURIFIED ATIII (o&)

20

T 0

510 PLASMA PURIFIED ATIII (°/o)

m

fl00

F i g . 3. E f f e c t o f P l a s m a u p o n the h e p a r i n , a n t i t h r o m b i n III, p r o t e a s e r e a c t i o n s t u d i e d i n i n c u b a t i o n m i x t u r e s c o n t a i n i n g v a r i o u s a m o u n t s o f p l a s m a , b u t w i t h c o n s t a n t a n t i t h r o m b i n III c o n c e n t r a t i o n . (a) A n t i t h r o m b i n - p o t e n t i a t i n g ability o f f r a c t i o n s R 2 and R 7 u s i n g s u b s t r a t e $ 2 2 3 8 t o m e a s u r e residual t h r o m b i n . F i n a l c o n c e n t r a t i o n s w e r e : a n t i t h r o m b i n I I I 1.7 • 10 -8 t o o l / l ; b o v i n e t h r o m b i n , 1 . 1 2 1 • 10 -8 tool/1. • •, R2, 0.241 ~g/ml; o-e R 7 0.054 /~glml. (b) Anti-Xa-potentiating ability of fractions R 2 and R 7 using substrate $ 2 2 2 to measure residual Xa. Final concentrations were: antithrombin In 8.2 • 10-8 molfl; bovine Xa, 2.1 • 10 -8 mol/l; •-------I R 2 0.117 ~tg/ml; • ~-o, R 7 0.026 /~g/ml.

589 ular weight than low molecular weight heparin and that this effect was seen in the anti-Xa system but not in the antithrombin system. It was initially considered possible that this could be due to the higher specific activity of the high molecular weight heparin. This would result in molar concentrations being much less than with the low molecular weight fractions, resulting in more readily measurable neutralization. In these anti-Xa mixing experiments, the low molecular weight to high molecular weight fraction to antithrombin III concentration ratio was 13.3 : 1 : 97. However, in the antithrombin assay the ratios were 15.6 : 1 : 15.5. That is, the molar concentration of purified antithrombin II! or plasma antithrombin III compared to the heparin fractions was more than six times greater in the anti-Xa than in the antithrombin assays. Therefore the possibility arises that heparin-neutralizing activities contributed by platelet release products or plasma were responsible for the difference between the antithrombin and anti-Xa activities of the heparin fractions. Several types of experiments were performed to explore this possibility. Firstly, experiments were performed using the purified antithrombin III/Xa system in which low and high molecular weight heparins were added in quantities that produced equivalent Xa inhibition. Increasing quantities of purified PF4 were then preincubated in the reaction mixtures before the addition of substrate, and resultant amidolytic activity finally determined (Fig. 4). The quantities of PF4 required to neutralize the low and high molecular weight heparins to 50% of their initial activity, suggested that PF4 neutralizes heparin fractions on a weight rather than molar basis. The second approach was to produce a similar heparin to plasma (molar) concentration ratio in the anti-Xa reaction mixture for the low molecular weight heparin as was used for the high molecular weight experiments. To achieve this it was necessary to decrease the amount of low molecular weight heparin added and increase the incubation time of heparin, antithrombin III (or plasma) and Xa to produce a similar amidolytic activity with the substrate. The results in Fig. 5 demonstrate no inhibition of the activity of the low molecular weight heparin with increasing plasma concentration even though the weight of low molecular weight heparin added was about one-third of that of the high molecular heparin.

6C

50

.~

~~

40

T

0011

0O45

018

F~4 ([Jg / ml)

Fig. 4. Neutralisaflon of heparin anti-Xa-potentiating activity b y h u m a n platelet factor 4, final concentrations were: a n t i t h r o m b i n III 4 . 9 . 10 -8 rnol/1; bovine Xa 2 . 1 . 10-8 mol/1, e---------a RI 0.068 ~g/nfl; o - - - - - - - - o R~ 0,014/~g/ml.

590 2501

20C 5C

4C

S

~

~© ,4 z

3C

~d

I%0

2

2c

z

~

~- v

l

0

50 PLASMA

1 O0

T

100

0

PURIFIED ATIII (%)

50 PLASMA

PURIFIED

1QO ATIII

(°/s)

F i g . 5. Mixing e x p e r i m e n t s w i t h a n t i t h r o m b i n III and p l a s m a p e r f o r m e d as in Fig. 3b e x c e p t that the final c o n c e n t r a t i o n s w e r e : a n t i t h r o m b i n III 8.2 • 1 0 - 8 m o l f l ; b o v i n e Xa, 2 . 1 • 1 0 - 8 m o l f l ; • • , R 2, 0 . 0 0 9 p g / m l . I n c u b a t i o n t i m e w a s altered t o give similar residual X a a c t i v i t y . F i g . 6. M i x i n g e x p e r i m e n t s w i t h a n t i t h r o m b i n III and p l a s m a p e r f o r m e d as i n F i g . 3a, e x c e p t t h a t the final c o n c e n t r a t i o n s w e r e : a n t i t h r o m b i n III 1 . 5 9 - 1 0 - 7 m o l f l ; b o v i n e t h r o m b i n 1 . 1 2 • 1 0 - 8 m o l / 1 ; • •, R2, 0.205/lg/ml; o-e R 7 0 . 0 4 5 m g f l . I n c u b a t i o n t i m e w a s altered t o give similar residual t h r o m b i n activity.

Next, blood was collected in an anticoagulant mixture of EDTA and theophylline recommended to produce minimal in vitro release of platelet constituents, including PF4 [13]. The inhibition of high and low molecular weight heparin activities in the Xa reaction were then compared in incubation mixtures containing purified antithrombin III, normal citrated plasma and the EDTA-theophylline plasma. The results of this study are illustrated in Table I. Although the concentration of PF4 was of the order of 20 times less in EDTA plasma than in citrated plasma there was minimal alteration in the plasma inhibition of the high molecular weight heparin interaction with antithrombin III and Xa. TABLE I A n t i - X a - p o t e n t i a t i n g a c t i v i t y o f R i k e r f r a c t i o n s R'I (M r 8 7 0 0 ) and R'5 (M r 23 4 0 0 ) in n o r m a l c i t r a t e d h u m a n p l a s m a , in p l a s m a f r o m b l o o d t a k e n i n t o E D T A - t h e o p h y l l i n e a n d in p u r i f i e d h u m a n a n t i t h r o m b i n III s o l u t i o n . ( T h e s e heparins w e r e p r e p a r e d in d i f f e r e n t c h r o m a t o g r a p h i c e x p e r i m e n t s t o t h o s e f r a c t i o n s R 2 and R 7 d e s c r i b e d e l s e w h e r e in the p a p e r . ) Plasma or p u r i f i e d a n t i t h r o m b i n III c o n c e n t r a t i o n , 4 . 9 1 0 - 8 m o l f l ; b o v i n e Xa, 2.1 • 1 0 -8 m o l f l . R e s i d u a l Xa a c t i v i t y ( n k a t fl)

In a n t i t h r o m b i n III In p l a s m a f r o m E D T A - t h e o p h y l l i n e b l o o d In p l a s m a f r o m c i t r a t e d b l o o d

R1 (68.2 ng/ml)

t RS (13.6 ng/ml)

24.8 25.2 25.2

27.6 45.6 46.0

PF4 concn. (ng/ml plasma)

13 300

591

Finally, the conditions of the antithrombin reaction system were adjusted such that the relative concentrations of heparin fractions and plasma were similar to those in the anti-Xa assay. For mixing experiments with purified antithrombin III and plasma, similar in design to those illustrated in Fig. 3, this entailed increasing the total antithrombin III content of the reaction mixtures. The results in Fig. 6 demonstrate no plasma inhibitory activity towards the heparin fractions. Discussion From considerations of the interactions between heparin, antithrombin III and the serine proteases of the coagulation system, it has now been established that the presence of heparin accelerates the neutralisation of these proteases by antithrombin III. Our experiments using peptide substrates are in agreement with these conclusions and further suggest that in purified reaction mixtures the molecular weight of heparin influences the antithrombin III-protease reaction only insofar as the specific activity of heparin varies with molecular weight. That is, in both anti-Xa and antithrombin systems approximately 15 times more low than high molecular heparin was required on a molar basis to produce equal potentiation of antithrombin III. These experiments were performed with heparin fractions whose approximate mean molecular weights were 9700 and 29 000, respectively. Intermediate quantities were generally required for equal potentiation of antithrombin III with heparin fractions of intermediate mean molecular weights, and also for unfractionated heparins. A structural hypothesis for the increase in specific activity with increasing molecular weight has recently been advanced [25]. It is based upon the probability of finding on each molecule a dodecasaccharide with a sequence required for binding to antithrombin III. As the molecular weight of heparin increases so there is an increased likelihood of the correct sequence being present on the molecule. Our experiments comparing amidolytic activities in purified systems with plasma suggest that the heparin-antithrombin III-thrombin reaction proceeds in a similar way regardless of whether the incubation media contained other plasma proteins. No inhibitory activities, other than those of heparin and antithrombin III, were detected when purified antithrombin III was progressively replaced by plasma containing the same amount of antithrombin III inhibitory activity. However, the experiments performed with the heparin-antithrombin III-Xa incubation systems suggest the presence of plasma inhibition of the action of heparin in the antithrombin III and Xa reaction, particularly when high molecular weight heparin is used. The most potent heparin-neutralizing activity, platelet factor 4, so far isolated from human tissue, has been purified from the contents of the platelet release reaction [24] and can be detected in normal citrated plasma by radioimmunoassay to the protein [26]. Several types of experiments were designed to see whether PF4 was responsible for the observed plasma inhibitory activity. The results of these experiments, illustrated in Figs. 4 and 5 and Table I, suggest that purified PF4 does not show the same selectivity towards the neutralization of high molecular heparin as the plasma inhibitor. Also, when the amount of PF4 in plasma was reduced from 300 to

592 13 ng/ml by changing the anticoagulant in which blood was collected, a minimal reduction in the plasma inhibitory activity was observed. These experiments rather suggest that the plasma component responsible for the inhibition of high molecular weight heparin is not PF4. Further evidence for a different plasma inhibitor was obtained by the finding that high molecular weight heparin was not inhibited in the antithrombin III-thrombin reaction when the conditions were similar to those of the antithrombin III-Xa reaction. It would seem that the inhibitor is primarily directed against the high molecular weight, antithrombin III and Xa reaction. The identity of this inhibitor has yet to be established. One possibility is that another plasma protease inhibitor such as ~ 2 macroglobulin may bind to factor Xa and protect it from binding by antithrombin III, without impairing its ability to cleave the small substrate used. However, such a sequence of events might be anticipated in similarly designed thrombin experiments, making an explanation of the differences between the factor Xa and thrombin reactions difficult. Alternatively, binding of heparin to other plasma proteins may influence the heparin-antithrombin III-protease reactions. It has recently been shown that low density lipoprotein binds strongly and specifically to high molecular weight heparin fractions [27]. Indeed, preliminary experiments have demonstrated that purified low density lipoprotein is a strong inhibitor of the high molecular weight heparin-antithrombin III-factor Xa reaction (unpublished observations). There has been one other preliminary report suggesting an inhibitor of antithrombin III neutralization of factor Xa in plasma [28]. In this study, the ability of different dilutions of normal plasma to inhibit factor Xa-induced clotting was examined, but no suggestions were offered as to the nature of the inhibitor. The differing inhibitory activities of high molecular weight heparin fractions in plasma against Xa and thrombin provide a partial explanation for discrepancies in specific activity determinations of heparins made in plasma with clotting assays. It has been found that thrombin clotting time: anti-Xa and kaolin cephalin clotting time: anti-Xa specific activity ratios increase with increasing molecular weight of heparin fractions prepared by gel filtration and ionexchange chromatography [17,29]. The clotting activities of these fractions were determined against the Third International Heparin Standard, which is an unfractionated heparin of mean molecular weight in between the molecular weights of the fractions studied here. The plasma inhibitor of the heparin-antithrombin III-Xa reaction inhibits such unfractionated heparins to give a net anti-Xa activity in between that of the high and low molecular weight fractions. If therefore, the anti-Xa activities of heparin fractions are compared with that of the International Standards the high molecular weight heparin will have a lower and the low molecular weight heparin a higher apparent activity. The specific activities in thrombin assay systems will be unaffected by the presence of the plasma inhibitor, resulting in a molecular weight dependence of the thrombin clotting time: anti-Xa specific activity ratio. Presumably the contribution of the heparin-antithrombin III-Xa reaction is of little quantitative importance in the kaolin cephalin clotting time assay system, so that the kaolin cephalin clotting time: anti-Xa ratio also increases with increasing molecular weight.

593

Acknowledgements D.A.L. was supported by an MRC Programme Grant during this work. The purified PF4 was kindly supplied by Dr. Pepper, Scotland. We would like to thank Dr. M.F. Scully for his advice and criticism. References 1 Brinkhouse, K., Smith, H.P., Warner, E.D. and Seegers, W.H. (1939) Am. J. Physiol. 125, 683--691 2 Seegers, W.H., Cole, E.R., Harmison, C.R. and Monkhouse, F.C. (1964) Can. J. Biochem. 42, 359-366 3 Yin, E.J., Wessler, S. and Stoll, P.J. (1971) J. Biol. Chem. 246, 3694--3719 4 Abildgaard, U. (1968) Scand. J. Clin. Lab. Invest. 21, 89--91 5 Rosenberg, R.D. and Damus, P.S. (1973) J. Biol. Chem. 248, 6490--6505 6 Damus, P.S., Hicks, M. and Rosenberg, R.D. (1973) Nature 246,355--357 7 Einarsson, R. and Anderson, L.O. (1977) Biochim. Biophys. Acta 490,104--111 8 Villanueva, G.B. and Danishefsky (1977) Biochem. Biophys. Res. Commun. 74,803--809 9 Smith, G.F. (1977) Biochem. Biophys. Res. Commun. 77,111--117 10 Machovich, R., Blarko, G. and Palor, L. (1975) Biochim. Biophys. Acta 397,193--200 11 Machovich, R., Staub, M. and Patthy, L. (1978) Eur. J. Biochem. 83,473---477 12 Sturzbecher, J. and Markwardt, F. (1977) Thromb. Res. 11,835--846 13 Bjork, I. and Nordenman, B. (1976) Eur. J. Biochem. 6 8 , 5 0 7 ~ 5 1 1 14 Odegard, O.R., Abildgaard, U., Lie, M. and Miller-Andersson, M. (1977) Thromb. Res. 11,205--216 15 Teien, A.N., Lie, M. and Abildgaard, U. (1976) Thromb. Res. 8,413---416 16 Borsodi, A. and Machovich, R. (1979) Biochim. Biophys. Acta 566,385--389 17 Lane, D.A., MacGregor, I.R., Michalski, R. and Kakker, V.V. (1978) Thromb. Res. 12,257--271 18 Lam, L.H., Silbert, J.E. and Rosenberg, R.D. (1976) Biochem. Biophys. Res. Commun. 69,570---574 19 Lundblad, R.L. (1971) Biochemistry 10, 2501--2506 20 Miller-Andersson, M.I., Bord, H. and Andersson, L.O. (1974) Thromb. Res. 5,439--542 21 Bolton, A.E., Ludlam, C.A., Moore, S., Pepper, D.S. and Cash, J.D. (1976) Br. J. Haematol. 32, 72-81 22 Seully, M.F. and Kakkar, V.V. (1977) Clin. Chim. Acta 79,595--602 23 Seully, M.F. and Kakkar, V.V. (1977) Thromb. Haemostasis 38, 218 24 Moore, S. and Pepper, D.S. (1975) Biochim. Biophys. Acta 379,370--384 25 Laurent, T.C., Tengblad, A., Thunberg, L., Hook. M. and Lindahl, U. (1978) Biochem. J. 175, 691-701 26 Dawes, J., Smith, R.C. and Pepper, D.S. (1978) Thromb. Res. 12,851--861 27 Pan, Y.T., Kruski, A.W. and Elbein, A.D. (1978) Archic. Biochem. Biophys. 189,231--240 28 Yin, E.T., Eirenkramer, L. and Butler, J.V. (1975) Advances in Experimental Medicine and Biology (Bradshaw, R.A. and Wessler, S., eds.), Vol. 52, pp. 239--242, Plenium Press, New York