THROMBOSIS RESEARCH 67; 157-l 65.1992 0049-3646/92 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
ENDOGENOUS HEPARINASE-SENSITIVE ANTICOAGULANT ACTIVITY IN HUMAN PLASMA Simone Cavari, Loredana Stramaccia and Simonetta Vannucchi Institute of General Pathology, University of Firenze, Viale Morgagni 50, 50134 Firenze, Italy (Received 19.2.1992; accepted in revised form 3.6.1992 by Editor S. Coccheri)
ABSTRACT In this paper we show that an anticoagulant activity, which we measure by thrombin time, appears in human plasma after its exhaustive proteolytic digestion. This activity is extremely heat stable, it is resistant to chondroitin ABC lyase (E.C.4.2.2.4) and heparan sulfate lyase (E.C.4.2.2.8), it is sensitive to heparin lyase (E.C. 4.2.2.7) and to nitrous acid treatment : we suggest that it can be identified as authentic heparin. The amount present in 1 ml of plasma of healthy subjects corresponds to 0.1-0.2 I.U. of standard heparin (150 I.U./mg). Proteolytically digested human plasma was submitted to ion-exchange chromatography on DEAE-Sephacel and the anticoagulant activity in the fractions eluted at the different molarities of NaCl was measured by thrombin time. This analysis shows that the anticoagulant activity elutes at very low ionic strength. The possibility that interactions of the endogenous heparin with proteins or protein fragments are responsible for the difficulty in isolating heparin from human plasma is discussed.
INTRODUCTION Heparin is unique among polysaccharides in that it has a variety of important biological activities, the most prominent of which are its anticoagulant activity and its lipemia clearing capacity. The presence of heparin in human plasma is still controversial. Indirect evidence showing the presence of heparin was obtained in several laboratories (1,2), but attempts to isolate chemically Keywords
:
heparin, plasma, heparinase 157
158
HEPARIN IN HUMAN PLASMA
Vol. 67, No. 2
heparin from human plasma gave contradictory results (3,4). Analysis of plasma GAGS isolated by ion-exchange chromatography led to the identification of chondroitin sulfate, keratan sulfate and heparan sulfate (5,6). However, extracts have been obtained from human normal plasma after tryptic digestion of plasma proteins, which inhibited the generation of intrinsic plasma thromboplastin (7). The commercial availability of a wide array of lyase enzymes (8,9), which act specifically on each GAG to produce a unique mixture of acidic oligosaccharide products, suggests a method useful for distinguishing among different GAGS . In this study we describe that the extensive proteolysis of human plasma reveals the presence of an anticoagulant activity, which can be ascribed principally to endogenous heparin, on the basis of its sensitivity to specific enzymatic depolymerizating treatments. However, ion-exchange chromatography on DEAESephacel of digested plasma samples, shows that the anticoagulant activity elutes at very low ionic strength. We suggest that interactions with proteins or protein fragments still occur after proteolytic digestion and are responsible for the difficulty in isolating endogenous heparin from human plasma.
MATERIAL
AND METHODS
Material. Heparin HP 756 from bovine intestinal mucosa (average molecular mass 12.9 kDa; 150 I.U./mg) was provided by Opocrin Research Laboratories (Modena,Italy). Heparan sulfate (from bovine lung) was donated by Professor Cifonelli (Department of Pediatrics, University of Chicago, USA). Chondroitin ABC lyase from Proteus vulgaris (chondroitinase, E.C. 4.2.2.4.), heparinase II (heparin lyase II from Flavobacterium heparinum, E.C.4.2.2.7.), trypsin (E.C. 3.4.21.4.), chymotrypsin (E.C. 3.4.21.1.) and collagenase (E.C. 3.4.24.3.) were provided by Sigma Chemical Company, St.Louis, MO. Papain (E.C. 3.4.22.2.) was from Calbiochem. Heparinase I (from F. heparinum, E.C. 4.2.2.7) and heparitinase (heparan sulfate lyase from F.heparinum, E.C. 4.2.2.8) were obtained from Seikagaku Kogyo (Tokyo, Japan). Pepsin (E.C. 3.4.23.1) and thrombin-reagent were from Boehringer-Mannheim. Gradient PAGE was performed on a standard vertical electrophoresis system supplied by Pharmacia LKB (Uppsala, Sweden). Silver nitrate was from Merck (Darmstadt, Germany) and Azur A was from BDH (Poole,UK). DEAE-Sephacel was obtained by Pharmacia (Uppsala, Sweden). All the other reagents were from Sigma Chemical Company, St.Louis, MO. Plasma sampling. Blood samples were collected from human healthy volunteers, with 0.38% trisodium citrate as anticoagulant. Plasma was obtained by centrifuging the blood at 860 g for 15 min, and the supematant was further centrifuged at 3000 g for 10 min to remove platelets. Pooled plasma was divided in 1 ml aliquots and stored at -80°C until used.
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
159
Proteolvsis of plasma samples. Aliquots of 1 ml of human pooled plasma were submitted to proteolytic digestion as previously described (10). Briefly papain was added for 24 h at 60°C. Samples were then boiled, and, at intervals of 24 h, the following enzymes, each at final concentration of 1 mg/ml, were added sequentially: trypsin, chymotrypsin, collagenase and pepsin. Samples were boiled at the end of each incubation. Samples were centrifuged at 3000 g x 10 min and the clear supernatants were collected for further analysis, Electronhoresis and enzvrnatic degradations of standard GAGS. 5 pl of standard heparin and heparan sulfate at a concentration of 1 mg/ml were incubated at 37°C for 24 h with 15 mIU of heparinase I or 15 mIU of heparitinase or 0.25 U of heparinase II in 0.1 M sodium acetate, pH 7.0 . Oligosaccharides derived by enzymatic depolymerization of standard heparin and heparan sulfate were analyzed by gradient polyacrylamide electrophoresis as described by Lyon and Gallagher (11). Enzvmatic and chemical degradations of plasma samples. Aliquots of 100 ul of the proteolytically digested plasma samples were incubated at 37°C for 24 h with one of the following enzymes : chondroitinase ABC 0.1 U, heparinase I 15 mIU, heparitinase 15 mIU, heparinase II 0.25 U. Alternatively the sample was treated with HN02 (12), dialyzed and concentrated to the initial volume. Thrombin time assay. Anticoagulant activity was evaluated on the proteolytically digested plasma samples, both before and after enzymatic or nitrous acid treatments, by thrombin time assay. The test was carried out as follows : 100 ul of plasma pool were incubated at 37°C with an equal volume of the treated sample for 2 min, then 100 yl of a thrombin solution (1.5 NII-I/ml) were added and the coagulation time was measured. For the quantitative evaluation of the anticoagulant activity thrombin time assay was performed by incubating undigested plasma in the presence of increasing concentrations of standard heparin. Ion exchanpe chromatoaraphy. 1 ml of the proteolytically digested plasma was applied to a DEAE-Sephacel column (1 cm x 3 cm), equilibrated in 0.05 M NaCl in 0.05 M Tris-HCl pH 7.4, followed by 30 ml of the same buffer. The column was eluted step-wise with increasing concentrations of NaCl in the same buffer. The fractions, each of 10 ml were dialyzed, concentrated to lml and tested for the anticoagulant activity.
RESULTS We previously observed that the proteolytic treatment described in Material and Methods, performed on murine plasma produces a 100% solubility of 35S-labelled molecules (10). In Table 1 we show the development of an anticoagulant activity in human plasma treated with proteases. The treatment of human plasma with papain and trypsin produces a delay of the thrombin time and
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
160
further steps of proteolysis lead to a final subsequent increase. Attempts to increase the release of the anticoagulant activity with the use of further proteases were unsuccessful. TABLE
1
Development of Anticoagulant Activity following Proteolytic Digestion of Pooled Human Plasma . Thrombin time (set) (range, n=9)
Treatments
20-22 38-47 46-55 53-58 57-70
None (Undigested plasma) Plasma treated with papain and trypsin. Plasma treated as above and with chymotrypsin Plasma treated as above and with collagenase Plasma treated as above and with pepsin
Table 2 shows the thrombin times obtained using pooled plasma in the presence of increasing concentrations of standard heparin. On the basis of these values we consider that the anticoagulant activity present in 1 ml plasma at the end of the proteolytic treatment corresponds to 0.1-0.2 I.U. of standard heparin. We decided to characterize this anticoagulant activity by means of enzymatic and chemical treatments (Table 3). The specificity of the heparan sulfate and heparin lyases is shown in Fig.1. TABLE 2 Thrombin Time Assay of Human Pooled Plasma in the Presence of Standard Heparin. Heparin concentration * @g/ml) 0.00
0.25 0.50 0.75 1.00 1.25
Thrombin time (set) 22 32 41 50 61 72
* 100 ul of undigested plasma were incubated with 100 yl of a solution containing the indicated concentrations of heparin.
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
161
Fig. 1: Gradient PAGE analysis of enzymatically depolymerized GAGs.Lanel:heparin; lane 2:hepat-in treated with heparinase II; lane 3 : heparin treated with heparinase I; lane 4: heparin treated with heparan sulfate lyase; lane 5 : heparan sulfate; lane 6: heparan sulfate treated with heparinase II; lane 7: heparan sulfate treated with heparinase I; lane 8: heparan sulfate treated with heparan sulfate lyase.
Heparinase (both heparinase II from Sigma and heparinase I from Seigakaku) generates a series of relatively low-molecular weight oligosaccharides from heparin as a result of its known specificity for the abundant hexosaminidic linkage in disaccharides of the structure GlcNS(+/-6s) al-4 IdU(2S). Heparan sulfate lyase has not any apparent depolimeryzing effect on heparin. On the contrary, it generates a highly complex pattern of oligosaccharides from heparan sulfate. Our results indicate that heparinase II as well as heparan sulfate lyase degrades heparan sulfate. The anticoagulant activity present in human digested plasma is heat resistant and it is completely destroyed by nitrous acid treatment; it is resistant to chondroitin ABC lyase (E.C.4.2.2.4.) and sensitive to heparin lyase (E.C.4.2.2.7.). Treatments with heparinase I or heparinase II produce a very similar decrease of the anticoagulant activity . Heparan sulfate lyase (E.C.4.2.2.8.) treatment induces only a little decrease of this actvity (Table 3). Treatments with nitrous acid and heparinase II destroy N-sulfated glycosaminoglycans, i.e. both heparin and heparan sulfate. However, heparinase I is specific for heparin-like structures (see Fig.1). The anticoagulant activity is sensitive to heparinase I and resistant to heparan sulfate-lyase, thus indicating the presence in plasma of polysaccharides having heparin-like activity and structure.
162
HEPARIN IN HUMAN PLASMA
Vol. 67, No. 2
Anticoagulant Activity of Proteolitycally Digested Pooled Human Plasma after Enzymatic and Chemical Treatments. This experiment is representative of three other which gave similar results. Treatment
Thrombin time (set)
None Chondroitinase ABC Heparinase I (Seigakaku) Heparinase II (Sigma. Heparan sulfate lyase Nitrous acid
70 3’8 30 64 25
In order to isolate endogenous heparin from human plasma we submitted the pooled digested plasma to anion-exchange chromatography on DEAE-Sephacel and we evaluated the anticoagulant activity present in the fractions, eluted by increasing molarities of NaCl. As shown in Table 4, the anticoagulant activity present in proteolytically digested human plasma, elutes at a very low ionic strength by a DEAE-Sephacel column. This behaviour suggests that the GAG chains are not completely ‘free’ from interactions (covalent and/or non-covalent) with proteins after the described treatment with proteases. TABLE 4 Anticoagulant Activity of Proteolytically Digested Human Plasma Fractions, eluted by the indicated NaCl molarities from DEAE-Sephacel .This experiment is representative of three other which gave similar results. NaCl molarities Thrombin time (sec.)
0.05
0.1
0.2
0.3
0.4
0.5
0.6
2.0
45
36
22
20
21
22
20
21
DISCUSSION Plasma GAGS isolated by ion-exchange chromatography were identified and characterized by several laboratories ( 5,6 ). The results showed that heparan sulfate constitutes only a small fraction of high-charge GAGS mainly represented by chondroitin sulfate and keratan sulfate. No heparin was isolated by this procedure. Other methods used to extract heparin from human plasma led to the
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
163
isolation of considerable amounts of the compound (10 ug/ml of plasma), but unfortunately they proved difficult to reproduce (1,4). However, previous results obtained in other laboratories (7) and our present investigation demonstrate that an anticoagulant activity occurs after proteolysis of human plasma. In this paper we show that this activity can be ascribed to the presence of polysaccharides with heparin-like structure, because it is sensitive to nitrous acid, heparinase I and heparinase II treatments and it is resistant to heparan sulfate lyase and to chondroitinase ABC. We suggest that all these results can be explained if we suppose that endogenous heparin strongly interacts with proteins. We propose that the anticoagulant activity we measure by thrombin time, results from an ‘unmasking’ of heparin-chains following the action of proteases. The ‘unmasked chains become accessible to heparinase. However, the proteolytic treatment is unable to release completely endogenous heparin from protein interactions. Recently, evidence has indicated that proteins, such as fibroblast growth factor (FGF) bound to heparin are resistant to tryptic digestion (13). This hypothesis is suggested by the results of the anion-exchange chromatography. We have previously shown that the most part of V-labelled GAGS of murine plasma, still after proteolysis, elutes at low ionic strength on anion-exchange chromatography and that both covalent and non-covalent interactions between GAG chains and plasma proteins are responsible of the difficulty in extracting ‘free’ GAGS (10). It is well known that negatively charged heparin interacts with proteins through cooperative unspecific electrostatic binding (14) or through specific interactions depending on structural conformations (15). Interactions between heparin and proteins regulate their biological effects: proteins such as platelet factor 4 and hystidin-rich glycoprotein are known to neutralize the effects of heparin (16); the protease inhibitor AT111 is strongly potentiated following heparin binding (15). We suggest that interaction between heparin and proteins is responsible for the difficulty in isolating endogenous heparin from healthy subjects. Circulating heparin-like molecules have been found in patients affected by a plasma cell disorder (17) or acute monoblastic leukemia (18) and in a patient that died from a fatal bleeding (19). We can actually only hypothesize that heparin-protein(s) interaction has a regulative role of the biological availability of heparin chains, however the recognition that heparin-like inhibitors of coagulation exist in normal subjects has profound implications for a better knowledge of the physiopathology related to the vascular system and haemostasis . Research supported by Minister0 Fubblica Istruzione Italiana Ricerca sul Cancro (AIRC).
(60%) and Associazione
164
HEPARIN IN HUMAN PLASMA
Vol. 67, No. 2
REFERENCES. 1. DOWES, J. and PEPPER, D.S. A sensitive competitive binding assay for exogenous and endogenous heparins. Thromb. Res., 27.388-396, 1982. 2. ENGELBERG, H. Plasma heparin levels in normal man. Circulation. 23, 578581, 1961. 3. JACOBSON, K.J. and LINDAHL, U. determination Attempted endogenous heparin in blood. Thromb. Haemostas., 42, 84-88, 1979.
of
4. VANNUCCHI, S., RUGGIERO, M. and CHIARUGI, V. Complexing of heparin with phosphatidylcholine : a possible supramolecular assembly of plasm heparin. Biochem. J., 227. 57-65, 1985. 5. CALATRONI, A., DONNELLI, P. V. and DI FERRANTE, N. The glycosaminoglycans of human plasma. ,I!, 48, 332-343, 1969. 6. STAPRANS, I. and FELTS, J.M. Isolation and characterization of glycosarninoglycans in human plasma. J. Clin. Invest., 76, 1984-1991, 1985. 7. FREEMAN, L., ENGELBERG, H., and DUDLEY, B.S. Plasma heparin levels. A method for determination of plasma heparin based on anticoagulant activity. Am.J.Clin.Pathol.. 24, 599-606, 1954. 8. LINHARDT, R.J., COONEY, C.L. and GALLIHER, P.M. lyases. Annl. Biochem. and Biotechnol., 12, 135-137, 1986.
Polysaccharide
9. LINHARDT, R.J., TURNBULL, J.E., WANG, H.M., LOGATHAN, D., GALLAGHER, J.T. Examination of substrate specificity of heparin and heparan sulfate lyases. Biochem, 29, 261 l-26 17, 1990. MAGNELLI, L., RUGGIERO, M., OLDANI, C., 10. PASQUALI, F., Interaction between endogenous CHIARUGI, V. and VANNUCCHI, S. Clin. Chim. circulating sulfated-glycosaminoglycans and plasma proteins. Acta. 192. 19-28, 1990. A general method for the detection 11. LYON, M. and GALLAGHER, J.T. quantities of glycosaminoglycan of submicrogram and mapping oligosaccharides on polyacrylamide gels by sequential staining with AzurA and ammoniacal silver. Anal. Biochem.,m, 63-70, 1990. 12. LAGUNOFF, D. and WARREN, G. Determination of 2- deoxy-2-sulfoami noexose content of mucopolysaccharides. Arch.Biochem., 99, 396-400, 1962.
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
165
13. SOMMER, A. and RIFKIN, D.B. Interaction of heparin with human basic fibroblast growth factor : protection of the angiogenic protein from proteolytic degradation by a glycosaminoglycan. J.Cell Phvsiol., 138, 2 15-220, 1989. 14. LINDAHL, U. and HOOK, M. Glycosaminoglycans and their binding to biological macromolecules. Ann.Rev.Biochem., 47, 385-417, 1978. 15. LINDAHL, U., THUNBERG, L., BACKSTROM, G., RIESENFELD, J., NORDLING, K. , BJORN,I. Extension and structural variability of the antithrombin-binding sequence in heparin. J.Biol. Chem., 259, 12368-12376, 1984. 16. LANE, D.A., PEYLER, G., FLYNN, A.M., THOMPSON, E.A., LINDAHL, Neutralization of heparin-related saccharides by hystidine-rich U. glycoprotein and platelet factor 4. J. Biol. Chem.., 261, 3980-3986,1986. 17. KHOORY, M.S., NESHEIM, M.E., BOWIE, E.J.W., MANN, K.G. Circulating heparan sulfate proteoglycan anticoagulant from a patient with a plasma cell disorder. J. Clin. Invest., 65, 666-674, 1980. 18. BUSSEL, J.B., STEINHERZ, P.G., MILLER, D.R. , HILGARTNER, M.W. A heparin-like anticoagulant in an eigth-month-old boy with acute monoblastic leukemia. Am. J. Pathol.,16. 83-90, 1984. 19. PALMER, R.N., RICK, M.E., RICK, P.D., ZELLER, J.A., GRAALNIK, H.R. Circulating heparan sulfate anticoagulant in a patient with a fatal bleeding disorder. New En@. J. Med.,m, 1696-1699, 1984.
ENDOGENOUS HEPARINASE-SENSITIVE ANTICOAGULANT ACTIVITY IN HUMAN PLASMA Simone Cavari, Loredana Stramaccia and Simonetta Vannucchi Institute of General Pathology, University of Firenze, Viale Morgagni 50, 50134 Firenze, Italy (Received 19.2.1992; accepted in revised form 3.6.1992 by Editor S. Coccheri)
ABSTRACT In this paper we show that an anticoagulant activity, which we measure by thrombin time, appears in human plasma after its exhaustive proteolytic digestion. This activity is extremely heat stable, it is resistant to chondroitin ABC lyase (E.C.4.2.2.4) and heparan sulfate lyase (E.C.4.2.2.8), it is sensitive to heparin lyase (E.C. 4.2.2.7) and to nitrous acid treatment : we suggest that it can be identified as authentic heparin. The amount present in 1 ml of plasma of healthy subjects corresponds to 0.1-0.2 I.U. of standard heparin (150 I.U./mg). Proteolytically digested human plasma was submitted to ion-exchange chromatography on DEAE-Sephacel and the anticoagulant activity in the fractions eluted at the different molarities of NaCl was measured by thrombin time. This analysis shows that the anticoagulant activity elutes at very low ionic strength. The possibility that interactions of the endogenous heparin with proteins or protein fragments are responsible for the difficulty in isolating heparin from human plasma is discussed.
INTRODUCTION Heparin is unique among polysaccharides in that it has a variety of important biological activities, the most prominent of which are its anticoagulant activity and its lipemia clearing capacity. The presence of heparin in human plasma is still controversial. Indirect evidence showing the presence of heparin was obtained in several laboratories (1,2), but attempts to isolate chemically Keywords
:
heparin, plasma, heparinase 157
158
HEPARIN IN HUMAN PLASMA
Vol. 67, No. 2
heparin from human plasma gave contradictory results (3,4). Analysis of plasma GAGS isolated by ion-exchange chromatography led to the identification of chondroitin sulfate, keratan sulfate and heparan sulfate (5,6). However, extracts have been obtained from human normal plasma after tryptic digestion of plasma proteins, which inhibited the generation of intrinsic plasma thromboplastin (7). The commercial availability of a wide array of lyase enzymes (8,9), which act specifically on each GAG to produce a unique mixture of acidic oligosaccharide products, suggests a method useful for distinguishing among different GAGS . In this study we describe that the extensive proteolysis of human plasma reveals the presence of an anticoagulant activity, which can be ascribed principally to endogenous heparin, on the basis of its sensitivity to specific enzymatic depolymerizating treatments. However, ion-exchange chromatography on DEAESephacel of digested plasma samples, shows that the anticoagulant activity elutes at very low ionic strength. We suggest that interactions with proteins or protein fragments still occur after proteolytic digestion and are responsible for the difficulty in isolating endogenous heparin from human plasma.
MATERIAL
AND METHODS
Material. Heparin HP 756 from bovine intestinal mucosa (average molecular mass 12.9 kDa; 150 I.U./mg) was provided by Opocrin Research Laboratories (Modena,Italy). Heparan sulfate (from bovine lung) was donated by Professor Cifonelli (Department of Pediatrics, University of Chicago, USA). Chondroitin ABC lyase from Proteus vulgaris (chondroitinase, E.C. 4.2.2.4.), heparinase II (heparin lyase II from Flavobacterium heparinum, E.C.4.2.2.7.), trypsin (E.C. 3.4.21.4.), chymotrypsin (E.C. 3.4.21.1.) and collagenase (E.C. 3.4.24.3.) were provided by Sigma Chemical Company, St.Louis, MO. Papain (E.C. 3.4.22.2.) was from Calbiochem. Heparinase I (from F. heparinum, E.C. 4.2.2.7) and heparitinase (heparan sulfate lyase from F.heparinum, E.C. 4.2.2.8) were obtained from Seikagaku Kogyo (Tokyo, Japan). Pepsin (E.C. 3.4.23.1) and thrombin-reagent were from Boehringer-Mannheim. Gradient PAGE was performed on a standard vertical electrophoresis system supplied by Pharmacia LKB (Uppsala, Sweden). Silver nitrate was from Merck (Darmstadt, Germany) and Azur A was from BDH (Poole,UK). DEAE-Sephacel was obtained by Pharmacia (Uppsala, Sweden). All the other reagents were from Sigma Chemical Company, St.Louis, MO. Plasma sampling. Blood samples were collected from human healthy volunteers, with 0.38% trisodium citrate as anticoagulant. Plasma was obtained by centrifuging the blood at 860 g for 15 min, and the supematant was further centrifuged at 3000 g for 10 min to remove platelets. Pooled plasma was divided in 1 ml aliquots and stored at -80°C until used.
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
159
Proteolvsis of plasma samples. Aliquots of 1 ml of human pooled plasma were submitted to proteolytic digestion as previously described (10). Briefly papain was added for 24 h at 60°C. Samples were then boiled, and, at intervals of 24 h, the following enzymes, each at final concentration of 1 mg/ml, were added sequentially: trypsin, chymotrypsin, collagenase and pepsin. Samples were boiled at the end of each incubation. Samples were centrifuged at 3000 g x 10 min and the clear supernatants were collected for further analysis, Electronhoresis and enzvrnatic degradations of standard GAGS. 5 pl of standard heparin and heparan sulfate at a concentration of 1 mg/ml were incubated at 37°C for 24 h with 15 mIU of heparinase I or 15 mIU of heparitinase or 0.25 U of heparinase II in 0.1 M sodium acetate, pH 7.0 . Oligosaccharides derived by enzymatic depolymerization of standard heparin and heparan sulfate were analyzed by gradient polyacrylamide electrophoresis as described by Lyon and Gallagher (11). Enzvmatic and chemical degradations of plasma samples. Aliquots of 100 ul of the proteolytically digested plasma samples were incubated at 37°C for 24 h with one of the following enzymes : chondroitinase ABC 0.1 U, heparinase I 15 mIU, heparitinase 15 mIU, heparinase II 0.25 U. Alternatively the sample was treated with HN02 (12), dialyzed and concentrated to the initial volume. Thrombin time assay. Anticoagulant activity was evaluated on the proteolytically digested plasma samples, both before and after enzymatic or nitrous acid treatments, by thrombin time assay. The test was carried out as follows : 100 ul of plasma pool were incubated at 37°C with an equal volume of the treated sample for 2 min, then 100 yl of a thrombin solution (1.5 NII-I/ml) were added and the coagulation time was measured. For the quantitative evaluation of the anticoagulant activity thrombin time assay was performed by incubating undigested plasma in the presence of increasing concentrations of standard heparin. Ion exchanpe chromatoaraphy. 1 ml of the proteolytically digested plasma was applied to a DEAE-Sephacel column (1 cm x 3 cm), equilibrated in 0.05 M NaCl in 0.05 M Tris-HCl pH 7.4, followed by 30 ml of the same buffer. The column was eluted step-wise with increasing concentrations of NaCl in the same buffer. The fractions, each of 10 ml were dialyzed, concentrated to lml and tested for the anticoagulant activity.
RESULTS We previously observed that the proteolytic treatment described in Material and Methods, performed on murine plasma produces a 100% solubility of 35S-labelled molecules (10). In Table 1 we show the development of an anticoagulant activity in human plasma treated with proteases. The treatment of human plasma with papain and trypsin produces a delay of the thrombin time and
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
160
further steps of proteolysis lead to a final subsequent increase. Attempts to increase the release of the anticoagulant activity with the use of further proteases were unsuccessful. TABLE
1
Development of Anticoagulant Activity following Proteolytic Digestion of Pooled Human Plasma . Thrombin time (set) (range, n=9)
Treatments
20-22 38-47 46-55 53-58 57-70
None (Undigested plasma) Plasma treated with papain and trypsin. Plasma treated as above and with chymotrypsin Plasma treated as above and with collagenase Plasma treated as above and with pepsin
Table 2 shows the thrombin times obtained using pooled plasma in the presence of increasing concentrations of standard heparin. On the basis of these values we consider that the anticoagulant activity present in 1 ml plasma at the end of the proteolytic treatment corresponds to 0.1-0.2 I.U. of standard heparin. We decided to characterize this anticoagulant activity by means of enzymatic and chemical treatments (Table 3). The specificity of the heparan sulfate and heparin lyases is shown in Fig.1. TABLE 2 Thrombin Time Assay of Human Pooled Plasma in the Presence of Standard Heparin. Heparin concentration * @g/ml) 0.00
0.25 0.50 0.75 1.00 1.25
Thrombin time (set) 22 32 41 50 61 72
* 100 ul of undigested plasma were incubated with 100 yl of a solution containing the indicated concentrations of heparin.
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
161
Fig. 1: Gradient PAGE analysis of enzymatically depolymerized GAGs.Lanel:heparin; lane 2:hepat-in treated with heparinase II; lane 3 : heparin treated with heparinase I; lane 4: heparin treated with heparan sulfate lyase; lane 5 : heparan sulfate; lane 6: heparan sulfate treated with heparinase II; lane 7: heparan sulfate treated with heparinase I; lane 8: heparan sulfate treated with heparan sulfate lyase.
Heparinase (both heparinase II from Sigma and heparinase I from Seigakaku) generates a series of relatively low-molecular weight oligosaccharides from heparin as a result of its known specificity for the abundant hexosaminidic linkage in disaccharides of the structure GlcNS(+/-6s) al-4 IdU(2S). Heparan sulfate lyase has not any apparent depolimeryzing effect on heparin. On the contrary, it generates a highly complex pattern of oligosaccharides from heparan sulfate. Our results indicate that heparinase II as well as heparan sulfate lyase degrades heparan sulfate. The anticoagulant activity present in human digested plasma is heat resistant and it is completely destroyed by nitrous acid treatment; it is resistant to chondroitin ABC lyase (E.C.4.2.2.4.) and sensitive to heparin lyase (E.C.4.2.2.7.). Treatments with heparinase I or heparinase II produce a very similar decrease of the anticoagulant activity . Heparan sulfate lyase (E.C.4.2.2.8.) treatment induces only a little decrease of this actvity (Table 3). Treatments with nitrous acid and heparinase II destroy N-sulfated glycosaminoglycans, i.e. both heparin and heparan sulfate. However, heparinase I is specific for heparin-like structures (see Fig.1). The anticoagulant activity is sensitive to heparinase I and resistant to heparan sulfate-lyase, thus indicating the presence in plasma of polysaccharides having heparin-like activity and structure.
162
HEPARIN IN HUMAN PLASMA
Vol. 67, No. 2
Anticoagulant Activity of Proteolitycally Digested Pooled Human Plasma after Enzymatic and Chemical Treatments. This experiment is representative of three other which gave similar results. Treatment
Thrombin time (set)
None Chondroitinase ABC Heparinase I (Seigakaku) Heparinase II (Sigma. Heparan sulfate lyase Nitrous acid
70 3’8 30 64 25
In order to isolate endogenous heparin from human plasma we submitted the pooled digested plasma to anion-exchange chromatography on DEAE-Sephacel and we evaluated the anticoagulant activity present in the fractions, eluted by increasing molarities of NaCl. As shown in Table 4, the anticoagulant activity present in proteolytically digested human plasma, elutes at a very low ionic strength by a DEAE-Sephacel column. This behaviour suggests that the GAG chains are not completely ‘free’ from interactions (covalent and/or non-covalent) with proteins after the described treatment with proteases. TABLE 4 Anticoagulant Activity of Proteolytically Digested Human Plasma Fractions, eluted by the indicated NaCl molarities from DEAE-Sephacel .This experiment is representative of three other which gave similar results. NaCl molarities Thrombin time (sec.)
0.05
0.1
0.2
0.3
0.4
0.5
0.6
2.0
45
36
22
20
21
22
20
21
DISCUSSION Plasma GAGS isolated by ion-exchange chromatography were identified and characterized by several laboratories ( 5,6 ). The results showed that heparan sulfate constitutes only a small fraction of high-charge GAGS mainly represented by chondroitin sulfate and keratan sulfate. No heparin was isolated by this procedure. Other methods used to extract heparin from human plasma led to the
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
163
isolation of considerable amounts of the compound (10 ug/ml of plasma), but unfortunately they proved difficult to reproduce (1,4). However, previous results obtained in other laboratories (7) and our present investigation demonstrate that an anticoagulant activity occurs after proteolysis of human plasma. In this paper we show that this activity can be ascribed to the presence of polysaccharides with heparin-like structure, because it is sensitive to nitrous acid, heparinase I and heparinase II treatments and it is resistant to heparan sulfate lyase and to chondroitinase ABC. We suggest that all these results can be explained if we suppose that endogenous heparin strongly interacts with proteins. We propose that the anticoagulant activity we measure by thrombin time, results from an ‘unmasking’ of heparin-chains following the action of proteases. The ‘unmasked chains become accessible to heparinase. However, the proteolytic treatment is unable to release completely endogenous heparin from protein interactions. Recently, evidence has indicated that proteins, such as fibroblast growth factor (FGF) bound to heparin are resistant to tryptic digestion (13). This hypothesis is suggested by the results of the anion-exchange chromatography. We have previously shown that the most part of V-labelled GAGS of murine plasma, still after proteolysis, elutes at low ionic strength on anion-exchange chromatography and that both covalent and non-covalent interactions between GAG chains and plasma proteins are responsible of the difficulty in extracting ‘free’ GAGS (10). It is well known that negatively charged heparin interacts with proteins through cooperative unspecific electrostatic binding (14) or through specific interactions depending on structural conformations (15). Interactions between heparin and proteins regulate their biological effects: proteins such as platelet factor 4 and hystidin-rich glycoprotein are known to neutralize the effects of heparin (16); the protease inhibitor AT111 is strongly potentiated following heparin binding (15). We suggest that interaction between heparin and proteins is responsible for the difficulty in isolating endogenous heparin from healthy subjects. Circulating heparin-like molecules have been found in patients affected by a plasma cell disorder (17) or acute monoblastic leukemia (18) and in a patient that died from a fatal bleeding (19). We can actually only hypothesize that heparin-protein(s) interaction has a regulative role of the biological availability of heparin chains, however the recognition that heparin-like inhibitors of coagulation exist in normal subjects has profound implications for a better knowledge of the physiopathology related to the vascular system and haemostasis . Research supported by Minister0 Fubblica Istruzione Italiana Ricerca sul Cancro (AIRC).
(60%) and Associazione
164
HEPARIN IN HUMAN PLASMA
Vol. 67, No. 2
REFERENCES. 1. DOWES, J. and PEPPER, D.S. A sensitive competitive binding assay for exogenous and endogenous heparins. Thromb. Res., 27.388-396, 1982. 2. ENGELBERG, H. Plasma heparin levels in normal man. Circulation. 23, 578581, 1961. 3. JACOBSON, K.J. and LINDAHL, U. determination Attempted endogenous heparin in blood. Thromb. Haemostas., 42, 84-88, 1979.
of
4. VANNUCCHI, S., RUGGIERO, M. and CHIARUGI, V. Complexing of heparin with phosphatidylcholine : a possible supramolecular assembly of plasm heparin. Biochem. J., 227. 57-65, 1985. 5. CALATRONI, A., DONNELLI, P. V. and DI FERRANTE, N. The glycosaminoglycans of human plasma. ,I!, 48, 332-343, 1969. 6. STAPRANS, I. and FELTS, J.M. Isolation and characterization of glycosarninoglycans in human plasma. J. Clin. Invest., 76, 1984-1991, 1985. 7. FREEMAN, L., ENGELBERG, H., and DUDLEY, B.S. Plasma heparin levels. A method for determination of plasma heparin based on anticoagulant activity. Am.J.Clin.Pathol.. 24, 599-606, 1954. 8. LINHARDT, R.J., COONEY, C.L. and GALLIHER, P.M. lyases. Annl. Biochem. and Biotechnol., 12, 135-137, 1986.
Polysaccharide
9. LINHARDT, R.J., TURNBULL, J.E., WANG, H.M., LOGATHAN, D., GALLAGHER, J.T. Examination of substrate specificity of heparin and heparan sulfate lyases. Biochem, 29, 261 l-26 17, 1990. MAGNELLI, L., RUGGIERO, M., OLDANI, C., 10. PASQUALI, F., Interaction between endogenous CHIARUGI, V. and VANNUCCHI, S. Clin. Chim. circulating sulfated-glycosaminoglycans and plasma proteins. Acta. 192. 19-28, 1990. A general method for the detection 11. LYON, M. and GALLAGHER, J.T. quantities of glycosaminoglycan of submicrogram and mapping oligosaccharides on polyacrylamide gels by sequential staining with AzurA and ammoniacal silver. Anal. Biochem.,m, 63-70, 1990. 12. LAGUNOFF, D. and WARREN, G. Determination of 2- deoxy-2-sulfoami noexose content of mucopolysaccharides. Arch.Biochem., 99, 396-400, 1962.
Vol. 67, No. 2
HEPARIN IN HUMAN PLASMA
165
13. SOMMER, A. and RIFKIN, D.B. Interaction of heparin with human basic fibroblast growth factor : protection of the angiogenic protein from proteolytic degradation by a glycosaminoglycan. J.Cell Phvsiol., 138, 2 15-220, 1989. 14. LINDAHL, U. and HOOK, M. Glycosaminoglycans and their binding to biological macromolecules. Ann.Rev.Biochem., 47, 385-417, 1978. 15. LINDAHL, U., THUNBERG, L., BACKSTROM, G., RIESENFELD, J., NORDLING, K. , BJORN,I. Extension and structural variability of the antithrombin-binding sequence in heparin. J.Biol. Chem., 259, 12368-12376, 1984. 16. LANE, D.A., PEYLER, G., FLYNN, A.M., THOMPSON, E.A., LINDAHL, Neutralization of heparin-related saccharides by hystidine-rich U. glycoprotein and platelet factor 4. J. Biol. Chem.., 261, 3980-3986,1986. 17. KHOORY, M.S., NESHEIM, M.E., BOWIE, E.J.W., MANN, K.G. Circulating heparan sulfate proteoglycan anticoagulant from a patient with a plasma cell disorder. J. Clin. Invest., 65, 666-674, 1980. 18. BUSSEL, J.B., STEINHERZ, P.G., MILLER, D.R. , HILGARTNER, M.W. A heparin-like anticoagulant in an eigth-month-old boy with acute monoblastic leukemia. Am. J. Pathol.,16. 83-90, 1984. 19. PALMER, R.N., RICK, M.E., RICK, P.D., ZELLER, J.A., GRAALNIK, H.R. Circulating heparan sulfate anticoagulant in a patient with a fatal bleeding disorder. New En@. J. Med.,m, 1696-1699, 1984.
