Anticoagulant and antithrombotic effects of heparin and low molecular weight heparin fragments in rabbits

1982

TliROMEiOSISRESE.ARCH 25; 475-485, 0049-3838/82/060475-11$03.00/O

Copyright

Printed in the L'SX. (c) 1982 Pergamon Press Ltd. -All rights reserved.

X~TICOAGULA~T A:u'D ANTITHROMBOTIC LOW >fOLECULAR WEIGHT HEPARIS

E.

Research

and

Holmer,

C.

Development,

Mattsson

EFFECTS OF HEPARIS FRXG:iESTS IS RABBITS

and

KabiVitrum

S. AB,

ASD

?;ilsson Stockholn,

Sweden

(Received 21.12.1981; in revised form 26.1.1982. Accepted by Editor U. Abildgaard)

ABSTRACT Heparin and heparin fragments of different molecular weight and with different anti-factor X,/XPTT activity ratios were studied with respect to their ability to IXhibit thrombus formation in an animal model. It is concluded that: a) Neither anti-factor Xa nor the APTT activity alone is a good reflector of the antithrombotic activity. fragments must have b) Anti-factor X, active a minimum molecular weight in order to elicit good antithrombotic activity. c) High affinity for antithrombin III is important for good antithrombotic activity. d) A heparin fragment of molecular weight 4 000 has the same antithrombotic activity as heparin but less effect on the clotting time.

INTRODUCTION The clinical use of heparin for the prevention and treatment of thrombosis has a long tradition and its effectiveness is well documented. It is now generally agreed that the major portion of its anticoagulant activity is due to its ability to complex with antithrombin III (AT-III) and accelerate the velocity at which this inhibitor neutralizes serine proteases in the coagulation cascade. The relationship between its anticoagulant and antithrombotic properties is still not fully understood, however. Heparin preparations are polydisperse with molecular weights ranging from 5 000 to 30 000 with an average of about 12-15 000 Further, heparin is heterogeneous with respect to its a(1). bility to bind AT-III - about 30-505: of the molecules in clinically used preparations have a high affinity for AT-III and disKey

words:

Anticoagulant, ments.

Antithrombotic,

475

Meparin,

Heparin

frag-

play

high anticoagulant activity, while thy r*j: of th2 ma:erir: Lnactivs (2-L). I: :-:a5 522:
MATERIALS

AND

METHODS

Heparin fragments Heparin fragments of different molecular weights were obtainacid degradation of porcine mucosal hepaed by controlled nitrous essentially as previously descririn (KabiVitrum AB, Stockholm), The molecular weights were determined by analytical gel bed (10). filtration on Sephadex G-75 (Pharmacia Fine Chemicals, Uppsala, using reference samples with a known degree of polymeriSweden), The calibration of the reference samples was kindly perzation. Swedish University of Agricultural Sciformed by Dr. L. Thunberg, The average molecular weights of the Sweden (10). ences, Uppsala, a molefragments were found to be 2 100, 3 300 and 4 000 assuming A fragment with low cular weight of 600 per disaccharide unit. activity (LA fragment) was prepared from the fragment MW 4 000 by The specific affinity chromatography on immobilized AT-III (4). anticoagulant activity of heparin and the fragments was determined assay utilizing the chromoby an APTT assay (4) and by an anti-X, genic substrate S-2222 (Kabi Diagnostica, Stockholm, Sweden) (11). Standard curves were made up from dilutions of the 3rd InternaThe heparin concentrational Heparin Standard (NIBSC, London). using the carbation of solutions of the fragments was determined zole-H2SO4 method (16). Ex vivo anticoagulant activities New Zeeland white rabbits of both sexes weighing 2-2.5 kg fed on a standard diet (Ewos pellets, SBdertZlje, Sweden) were given a bolus injection of heparin or LMW heparin fragments in a marginal

v ei 3 ear other-disa

;;fy;;~e~r;m23~

before and 1 ml blood ear and

blood

was

immediately

the plasma activities.

obtained Activated

III to than

used

x r 2acn

centrifuged

at

analysed Cephaloplaj

1

with tix

(anti-X,). anti-factor

the citrate

,3Oi;

:-: g t

rease~:

iIf .

.

:
11,3t u9-

Iaardifrasnent. after adninistra-

and

TEND?;

a

plasma steeper

to

obtained

a concentration standard curve-

with

rabbit

of

with

atropine

plasma

(0.1

0.5 xor2

and

artery (102). for

10 APTT (_\cti;l(“) to

of

the The

minutes a;?d and anti-Xa ; Dade Di tls-

U/ml. This reproducible

was don2 results

only.

Experimental thrombus model The antithrombotic effect of the tested and compared to that of heparir. as originally described by Wessler et premeditated s ingborg, barbitone

i?.

zinctes 20 and 30 15, were collect2t frowith l/LO of sztiuf

5, 10, samples mixed was

ZT’iks 352: c332

a -3.5 the procedure vas that w a s ,ujed In the APTT assay in the manufacturers data sheet, apart from the fact The concentration of CaCl,, was to 0.05 x. i-creased subs traactivities were determined usic; t h 2 chromogenic (Kabi Diagnostica, Stockholm, Sweden). Th2 previous 1:~ method (11) was modified by adding purified humar, AT-

to tile obtain were

were

a n i ma : 3

ately :iol, opposite

nos tics) 3utli::ed chat :h2 ant i-Ii\, :e S-22?? described

;oefj; r for

stared,

3i s

its).

t:>ses

a t

LXW in al.

mg/kg)

heparin fragments a rabbit thrombus (17). The animals

and

Hypnorm(K)!

Sweden (2:).

Leo,

was model, ;_ere Hel-

(0 .1 ml/kg) and then anaesthetized with pentoThe procedure for the model and the interpretation of the results are summarized in Fig. I. The size of the formed thrombi are scored on a scale from 6 to 0 according to I!.!e s s 1 cl r e t a 1 . In doubtful cases the ‘iiisher score was alwa>.s give:,.

EVALUATION

of= ANTITHR~M~~~TIC

AND HEPARINFRAGMENTS

I

15.30.45 _or 60 min.

30 sec.

\ t

t

f 1-v.

”

mjeCtKm

.

[

I

MODEL

IO ml” W-

glassy isOlStu3

injectton of activated human 1.5ml/kg plasma

IV.

of hepartn or LMW fragmer~t

ACTIVITY OF HEPARIN

IN A RABBIT THROMBUS

of a two cm. long segment of the jugular vet” j&$&

remove the segment arid empty Its content into a Petri dish.The amount Of ClcItform& S Scored on a xale graded

rhrombus

thrombi J two or more

1 few maCrOsCOp,C

large thmmbi

strands

FIG. t Evaluation of the antithrombotic of heparin and heparin fragments rabbit thrombus model.

from 0 to 4

effect in

the

The initlatlan zf t;hrcm.:j,, formatiz7; glass-activated !hu m 1 ?, p I 3 s ma vias oerfcrmsd - . i, ; ,.'"
: j-G 1 ,27. _71 _ t_ t. L-..1s i :A_rit ;5 a ::2 3 ,: ; i n !1t e j after 7 32 II 2 j

t J

c: n e

a n d

c w 0

i,alf

on ,$?TT). i;fter drun ti in-j. antithronbotic effect was scorec at two different points by using both jugular veins, By performing the same experi ment after 45 arid 60 minues the duration of the antithronbotic effect of heparin and the fra,nment YW 4 009 could be estimated from time-response curve. RESULTS Heparin fragments Three heparin fragments of different molecular weights were The prepared by nitrous acid degradation of commercial heparin. specific activity (anti -Xa) of the isolated fragments was found to be lower than that of the parent heparin and also to decrease When designing this with increasing degrees of depolymerization. however, it was considered important to inject equal astudy, mounts (by weight) of polysaccharide when equal activity (antiThis can only Xa> doses of different fragments were administered. be done with fragments having the same specific activity. To a portion of each fragment was further fractionated achieve this, The high by affinity chromatography on matrix-bound AT-III (4). obtained - around 1 000 anti-X, IU/mg (8) activity preparations were then mixed with the corresponding fragments in appropriate A fragment with proportions to obtain equal specific activities. low affinity for AT-III and no, or low, anticoagulant activity was also isolated upon affinity chromatography of fragment FlW activities of the pa4 000. The molecular weights and specific rent heparin and of the fragments obtained are shown in Table 1.

TABLE

1

Specific anticoagulant and heparin fragments.

activities

Substance

Average MW

Specific APTT ILJlmg

Heparin

12

165

000

of

heparin

Activity anti-X, IU/mg

Ratio anti-X, APTT

175

1

Fragment

2 100

2

150

75

Fragment

3 300

20

145

7

Fragment

4 000

30

155

5

LA

3 800

4

fragment


As crease

previously shown the anti-Xa/APTT with decreasing MW (4).

ratio

was

found

to

in-

Ex

EFFECTS 3F HZ’tiIN

So.6

V31.25,

viva

anticoagulant

i 7 ‘3

FRAQiZNTS

activitiss

30 and 60 IU,“kg of intravexousi:: with heparin and hepaiin fragments. .A dose 3f 75i3 ICikg was alss used for the The plasma anticoagulant activities were 2 190 fragment. Eeparir, and the diffetime. measured after certain periods of rent fragments produced equal anti-Sa responses - no significant Therefore, the curves in Fig. 2a - she:;differences were found. ing plasma anti-Xa the txo dosages - represent activities after the means heparin and the different + SD of values found for fragments together, from the logaritmic timeIt was calculated activity curve in Fig. 25 that a dose of 30 IV/‘kg of heparin or any of the fragments produced an Lnitiel anti-S, activity 3f @.:;cj LUI ml. Xabbits

were

inject22

.lJ 5

10

IS

20

25

30

5

minutes

FIG. 2a Anti-factor X, activities in ma after intravenous injection and heparin fragments. Doses and 60 IV/kg A.

10

15

minutes

20

2.5

30

FIG. 2b rabbit plasof heparin ?? 30 IV/kg

This activity disappears from the plasma with a t1/2 of 35 minutes during the time period studied. Corresponding values for 60 IV/kg were 0,.90 IU/ml anti 40 minutes. Injection of 0.4 mg/kg of LA fragment - corresponding to 60 Iv/kg of active fragment caused no anti-Xa activity in the rabbit plasma. APTT activities found in rabbit plasma are shown in Fig. 3. In accordance with the low in vitro APTT activity of the fragments, there was only a slight prolongation of APTT after bolus injections of 30 and 60 IU/kg and 250 IU/kg for the smallest frag ment, and at these dose levels no definite dose dependence was observed. On the other hand, at equivalent doses, heparin showed a marked dose-dependent prolongation of APTT. Injection of 0.4 mg/kg of the LA fragment also resulted in a small prolongation of the APTT. The same was also observed in rabbits injected with saline, suggesting that this effect reflects variability in the animal model and/or the test method rather than a fragment-induced prolongation.

, .,;_._:,

Hepar in

4

Fragment

+

. 10 iU/‘kg ). 30

1.5

m.w. 2100 ??

A

1.4

a 61)

:;3 . ;

60 250

IUi’kg

1.3 1.2 1.1 1.0 -

5

10

15 20 25

-

30

4 +.

1.5

??

30

k

1.4

A

60

Q

1.3

._y? 1.2 5 1.1 a, L 1.0

7

10 15 20 25

4 fragment

m.w. 3300

Fragment

5

30

minutes

minutes

1.5

W/kg ”

:

1.4

:

1.3

.;

1.2

m.w. 4000 ?? 30 NJ/kg A 60 ” - --c LA fragment

Tii 1.1 z L- 1.0

5 minutes FIG. 3 APTT response (mean + S.D. ma after intravenous -injection parin fragments in varying

minutes n = 6) of doses.

in rabbit heparin

plasand he-

Antithrombotic activity The antithrombotic activities of heparin and the fragments were studied in a stasis model using altogether about 80 rabbits. A minimum of six rabbits were used for most doses. Every tenth rabbit served as a control and received an equivalent volum2 of saline instead of heparin. The thrombi produced were examined and scored on a scale ranging from 4 to 0 as described in Fig. 1. Scores of 0 and 1 can be considered to represent a good antithrom botic effect, a score of 2 is a moderate effect and 3 and 4 stand for little or no effect at all. The individual results found in this study are presented in Table 2. To facilitate comparison of the antithrombotic potency of the different drugs a mean antithrombotic score was calculated for each dose. For all the drugs studied there is a good dose-response relationship, i.e. increasing the dose increases the antithrombotic effect. The lowest dose of heparin that will produce a good to moderate effect for 15 minutes is within the range of lo-30 IU/kg.

EFFSGTS OF HEPAXIN F?ACXE?(‘TS

TXELE

7

antithrombotic effect of he?arin 2nd Scores 0: _ the S = mean antiat different doses. ‘neparin fragments Small figures represent the result throm’botic score. of each individual experiment. Upper and lower dose figand AiPTT activities respectively. ures refer to anti-X a About 7/3 of the control rabbits were scored as k and the remaining as 3.

60 60

s=O 0.0.0.00.0

-t-

._ _ 30

--_.

s= 3.7

s&g;1 2,*.4 1' 2.2.2 2 4 -+-

s= 1.3 !.I,!.,,?.2

In order to obtain similar protection using fragments XW 2 100 and NW 3 300, the doses have to be increased about lo-fold and 4-fold, The NW 4 000 fragment respectively. is however equipotent to heparin. After 30 minutes the antithrombotic effect of all the drugs is reduced, but essentially the same relationship between the antithrombotic potency and the molecular weight can be seen. However, the antithrombotic effect of fragment NW 4 000 seemed to have a longer duration than that of heparin. To investigate this matter further, the study was extended to include measurements after 45 and 60 minutes. This comparison of heparin and fragment 11W 4 000 was made at a dose of 30 IU/kg. The relation between the mean antithrombotic score and the dura-tion of the effect is shown in Fig. 4. Each point represents the mean score from 6 determinations.

FIG. 4 Duration of the antithrombotic effect of heparin A and fragment of MW 4 000 ??.

10

20

30

40

minutes

50

60

70

it can be 2stioat22 that the duratizr: of thz antithrom5,tic I!!2 G 003 is 53 and 65 minutes r2effect of heparin and fiagCl?nt This difference is not statistically spectively. established The antithronbotic properties of the LA fraj(Students i-testf. At a dose of 5.2 mg/Xg it was found to ment was also studied. have a moderate effect roughly corresponding to that obtained with 0.06 mg/kg (10 IU/kg) of fragment >!W ii 000.

DISCUSSION Heparin fragments of different molecular weights and with different anti-X,/hPTT activity ratios have been prepared and their anticoagulant and antithrombotic activity has been investigated in a rabbit thrombus model and compared with that of heparin. We found that equal anti-X, doses of heparin and the fragments produced equal initial plasma levels (anti-x,), and further more, that this activity was eliminated from the plasma at the same rate. Each fragment showed a good dose-dependent antithromactivity increased with botic response, i.e. the antithrombotic anti-X, doses increasing anti-X, activity. However - at equal showed markedly different heparin and the different fragments antithrombotic activities. Considerably higher doses of the smallest fragments were required to produce similar protection to that with heparin. It can thus be concluded that there is no direct correlation between the plasma anti-factor X, levels and the antithrombotic activities. This was also supported by the results obtained with the LA fragment that produced an antithrombotic effect without any detectable ex vivo anti-Xa activity in the plasma. properties of the As was to be expexted from the in vitro fragments, they produced a very short prolongation OL APTT in the rabbits. Very small, if any, differences in the APTT rebetween the three active fragments and also between difsponse This might be due to ferent doses of the fragments were found. doses (APTT activity) too the fact that the rabbits were given low (l-10 ID/kg) to be accurately assayed by the method used. and dose dependent proHeparin, however, produced a significant longation of the APTT. All fragments were found to be much more potent antithrombotic drugs than heparin when compared on the basis of administeFor example, doses of 2, 6 and 12 IU/kg of fragred APTT units. scores comparable to 10, ment !YW 4 000 produced antithrombotic less prolongation of the APTT 30 and 60 IU/kg of heparin. Also, in rabbit plasma was required with the fragment, as compared to This is to obtain a comparable antithrombotic response. heparin, We can conclude particularly pronounced with fragment MW 4 000. that neither the dose injected (APTT activity) nor the prolongation of the clotting time obtained reflects the antithrombotic activity. the Of the parameters studied when using active fragments, A very low to be the most important one. molecular weight seems for the MW 2 100 fragment, where antithrombotic potency was found A minimum moleas fragment MW 4 000 was equipotent to heparin. - below which the antithrombotic activity dramaticular weight on comparing the Furthermore, cally decreases - seems to exist. active fragment MW 4 000 with the LA fragment, it becomes obvious

Vs1.35,

So.6

EFFECTS OF HFPARIN

483

FRAGXZNTS

that

t1he active fragment is much more potent. Although the anticoagulant activity seems to be important to the results show that the in obtain a good antithrombotic effect, vitro snd ex vivo anticoagulant activity parameters studied alone poorly reflect the antithrombotic properties of heparin fragments dependent on the molecular weight -. Additional factors - probably One such factor in the in vivo situation. appear to be involved may be the interaction with the vascular endothelium that has This may also explain the been demonstrated with heparin (18). antithrombotic effect of the LA heparin. Some of the drawbacks of heparin therapy today are the increased bleeding tendency and the relatively short duration of A high APTT value is effect, requiring frequent administration. generally considered to reflect an increased risk of bleeding. This study shows that a heparin fragment of a certain molecular only a weight has a strong antithrombotic potencv while exerting slight effect on the clotting time, thus indicating a reduced The results also suggest a longer duration of bleeding tendency. the antithrombotlc effect of a fragment as compared to heparin. These properties, together with the previously reported reduced effect on platelets (8) and lipoprotein lipase (19), make a low molecular weight heparin a promising candidate for a new antithrombotic drug.

ACKNOWLEDGEMENT We are grateful to Dr. L.-O. Andersson for his revision of this manuscript and to Ms. G. Sdderstrsm, Ms. K. Irvall, Ms. T. Uthne and Ms. K. Neiger for skilful technical assistance. We also wish to thank Ms. M. Stendahl for typing this manuscript.

REFERENCES 1.

JOHNSON, mucosal 1976.

E.A. and MULLOY, B. 'neparin preparations.

2.

HijQK, M., BJijRK, I., HOPWOOD, J. and LINDAHL, U. Anticoagulant activity of heparin: Separation of high-activity and low-activity heparin species by affinity chromatography on immobilized antithrombin. FEBS Lett, 66, 90-93, 1976.

3.

LAM, L.H., SILBERT, J.E. and of active and inactive forms Comm, 69, 570-577, 1976.

4.

ANDERSSON, L.-O., BARROWCLIFFE, T.W., HOLMER, E., JOHNSON, E.A. and SIMS, G.E.C. Anticoagulant properties of heparin fractionated by affinity chromatography on matrix-bound antithrombin III and by gel filtration. Thromb Res, 9, 575-583, 1976.

3.

LINDAHL, U., BACKSTRijM, G., THUNBERG, L. Evidence for a 3-O-sulfated D-glucosamine antithrombin-binding sequence of heparin. US.-\, 77, 6551-6555, 1980.

The molecular weight Carbohydr Res, 51,

ROSENBERG, of heparin.

R.D. The Biochem

range of 119-127,

separation Biophys Res

and LEDER, 1-G. residue in the Proc Nat1 Acad

Sci

0.

u'i:-iy RIESESFELD, J., TH'L'SBERG, L., iiuii_. ) >!. a " I. LI\;D.-\L, 'L. _+.Z antithronbin-'binding sequence of heparin; locaticn ,>r esser.groups. tial N-sulfate J Biol Chem, 256, 2339-2394, i981.

7.

LINDAHL, U., BxCKSTRii>l, G., Hiii;iK,?!., THUSBERG, L., FRASSSO?; A. L. -4. and LINKER, Structure of the antithroabin-binding site in heparin. Proc Nat1 Acad Sci USA, 76, 3193-3202, 1979.

8.

HOLMER, E., LINDAHL, U., BgCKSTRB:l, G., THCSBERG, L., SANDG. and ANDERSSON, L.-O. BERG, H., SBDERSTRGM, Anticoagulant activities and effects on platelets of a heparin fragment with high affinity for antithrombin. Thromb Res, 18, 861869, 1980.

9.

CHOAY, J., LORMEAU, ORESTE, P., TORRI, oligosaccharides.

10.

J.C., G. and Thromb

PETITOU, M., SINAY, P., CASU, B., GATTI, G. Anti-Z, active heparin Res, 18, 573-578, 1980.

THUNBERG, L., BACKSTRijM, G., GRUFDBERG, H., RIESENFELD, and LINDAHL, U. The molecular size of the antithrombinbinding sequence in heparin. FEBS Lett, 117, 203-206,

J, 1980.

11.

ANDERSSOS, L.-O., BARROWCLIFFE, T.W., HOLMER, E., JOHNSON, G. Molecular weight dependency of the E.A. and SBDERSTRGM, heparin potentiated inhibition of thrombin and activated Effect of heparin neutralization in plasma. factor X. Thromb Res, 15, 531-541, 1979.

12.

THUNBERG, L., JACKSON, C.R. anticoagulant 243, 1979.

13.

HOLMER, E., KURACHI, weight-dependence of on the inhibition of XII, tor XI,, factor them J, 193, 395-400,

14.

YIN, E.T., WESSLER, S. and STOLL, P.J. Biological properties of the naturally occurring plasma inhibitor to activaJ Biol Chem, 246, 3703-3711, 1971. ted factor X.

15.

CARTER, C.J., KELTON, J.G., HIRSH, J. and GENT, M. tionship between the antithrombotic and anticoagulant Thromb Res, fects of low molecular weight heparin. 174, 1981.

16.

BITTER, T. reaction.

17.

WESSLER, S., REIMER, S.M. and SHEPS, inducing activity in of a thrombosis Physiol, 14, 943-946, 1959.

LINDAHL, On the activity

U., TENGBLAD, A., LAURENT, T.C. and molecular weight-dependence of the of heparin. Biochem J, 181, 241-

The molecular K. and SBDERSTRGM, G. the rate-enhancing effect of heparin ~ factor X,, factor IX,, facthrombin Bioand kallikrein by antithrombin. 1981.

and MUIR, H.M. A modified Anal Biochem, 4, 330, 334,

uranic 1962. M.C. human

acid

Relaef21, 169-

carbazole

Biological serum. J

assay APP~

?

2.

CLI>!ELIIYS, B., t’ne il,

9.

surface 773-T82,

of

BCSCZ,

cultored r97a.

c.

and xijijx;, II. ‘,:UClai:

endozielial

3indir.g cells.

of

:leparir. on T :? r 0 z ‘2 8.e s ,

BESGTSSOZ, G., OLIVECROSA, T., Hij?%X,>!., RIESESFEL3, LIXDAHL, U. Interaction of Lipoprstein lipsse with ar.d inodified heparin-like 625-633, 1350.

polysaccharides.

Biochezi

J. and native C, 189,