Purification and Characterization of a Novel Heparinase” A Novel Heparinase: Purification and Characterization 1 FIG. 1. Purification profile of crude extract on hydroxylapatite

THE JOURNAL OF BIOLOGICAL CHEMISTRY Cc) 1990 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 265, No. 23, Issue of August 15, pp. 13609-13617, 1990 Printed in U.S. A.

Purification and Characterization of a Novel Heparinase”

(Received for publication, February 6, 1990)

Linde H. Biihmer@, Marthinus J. PitoutS, Pieter L. Steynll, and Leon Visser/l From the *Departments of Physiology, llMicrobiology, and ([Biochemistry, University of Pretoria, 0001 Pretoria, South Africa

A unique heparinase was isolated from a recently discovered Gram-negative soil bacterium. The enzyme (heparinase III) was purified by hydroxylapatite chromatography, chromatofocusing, and gel permeation chromatography. The enrichment was 48X, and the specific activity of catalytically pure heparinase was 127 IU/mg of protein. Similar to the heparinase I from Flauobacterium heparinum, heparinase III also degrades heparin to mainly disaccharide fragments. It is specific for heparin and also breaks down heparan sulfate, but not hyaluronic acid and chondroitin sulfate. Heparinase III, however, differs markedly from heparinase I in several other aspects: it has a higher molecular mass (94 uersus 43 kDa), pI(9.2 uersus 8.5), its K,,, and kcSt are different, and it has a higher energy of activation (15.6 versus 6.3 kcal/mol). Optimal activity was also found at higher pH (7.6 uersus 6.5) and temperature (45 uersus 37 “C). Furthermore, the amino acid composition of heparinase III is quite different from that of heparinase I.

Enzymes which degrade heparin find a wide range of appli- cations. They may be used to characterize the structure (l-5) or endogenous activity (6-9) of heparin and related compounds. There is also great interest in the preparation of heparinase-derived heparin fragments as alternative antico- agulant reagents with less side effects (10-17) and as anti- tumor reagents (18, 19). Heparinase is furthermore used to normalize prothrombin and thromboplastin times of heparin- containing plasma samples (20). Recently, heparinase has also been applied in a heparin filter for the deheparinization of blood after extra corporeal perfusion (21-25). The only com- mercially available heparinase is prepared from Flauobacter- ium heparinum, American Tissue Culture Collection 13125 (26). Production of the enzyme by this organism requires the inclusion of heparin as an inducer (27-29), at fairly high costs in the growth medium. Recently, Joubert and Pitout (30) reported the isolation of a novel Gram-negative soil bacterium which produces heparinase constitutively in high yields (30).

The taxonomic position of this organism is still unclear, and the work reported here was done on Isolate 114. The purification and characterization of heparinase III are described below.

* This work was supported by the National Cancer Association of South Africa. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ To whom correspondence and reprint requests should be ad- dressed.

EXPERIMENTAL PROCEDURES’

Materials

Sigma was the supplier of heparinase I (from F. heparinum), heparin (sodium salt, from porcine intestinal mucosa), heparan sulfate (sodium salt, from bovine kidney), hyaluronic acid (potassium salt, from human umbilical cord), chondroitin sulfate A (sodium salt, from whale cartilage), PVS* (potassium salt), GlcNAc, p-nitrocatechol sulfate, 4 M methanesulfonic acid, 0.2% tryptamine, and Azure A (chloride salt). Casein, soy peptone, and antifoam reagent (p-2000) were obtained from Merck. Yeast extract was purchased from Oxoid. Agar and Freund’s adjuvant were products of Difco. Bio-Rad supplied hydroxylapatite (Bio-Gel HTP) and the protein dye reagent. Phar- malyte 8-10.5 and 3-10, Polybuffer exchanger PBE 118, Sephacryl S- 2OOsf, Sephadex G-25, PD/lO columns, protein markers, lo-15% gels for SDS-PAGE, buffer strips and dye tablets for the Phast System, and Silane A-174 were obtained from Pharmacia LKB Biotechnology Inc. Amberlite MB1 was a product of Serva. Amino acid standards and phenylisothiocyanate were purchased from Pierce. [“H]Iodoacetic acid (188 mCi/mmol) and [14C]Asp (192 mCi/mmol) were supplied by Amersham. Instagel was purchased from Packard Instruments. Ultra filters were products of Amicon Corp. (PM30 and Centricon 30) and Millipore (CX-30). Paragon blue was obtained from Beckman. All other reagents were of analytical grade.

Methods

Cultivation of the Organism-Isolate 114 was generously supplied by J. J. Joubert (Dept. of Medical Microbiology, University of Stel- lenbosch, Stellenbosch, South Africa). The organism was stored at room temperature on agar slants which contained the following per liter of water: 28 g of trypticase soy broth, 8 g of glucose, 1 g of yeast extract, 1 g of NaCI, 0.5 g of MgC12, and 12 g of agar. This medium was autoclaved for 15 min at 121 “C. The bacteria were transferred to fresh media every 14 days. Suspension cultures were grown in a lo-liter fermentor (Magnaferm, New Brunswick Scientific Co.) with a medium containing the following per liter: 10 g of casein peptone, 10 g of sucrose, 1 g of yeast extract, 1 g of NaCl, 0.5 g of MgCl,, and 0.2 ml of antifoaming agent. The medium (sucrose separately) was sterilized at 121 “C for 2.5 h. Inocula (10% v/v) of exponential phase cells were used. Cultures were incubated at 30 “C, 180 rpm, for 18 to 30 h with an aeration rate of 100% (80% O,, 20% N2). Growth was measured by turbidity at 540 nm with a Bausch and Lomb Spectronic 20 calorimeter. Cells were harvested by centrifugation in a Beckman J-6 centrifuge with a JS-5.2 rotor at 1,570 x g for 90 min at 4 “C and washed with 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0, and sedimented on a Beckman J-26 centrifuge with rotor 30 at 15,000 X g for 15 min at 4 “C.

Preparation of the Crude Extract-Washed cells were resuspended to 0.2 g of cells/ml of buffer in 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0. Five ml of cell suspension was sonicated on ice for 8 x 15 s with a MSE Soniprep 150 at maximum power (&23 kHz) with a 45-s cooling after each pulse and 4-min cooling after the

’ Portions of this paper (including parts of “Experimental Proce- dures” and “Results” and Figs. I and II) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press.

’ The abbreviations used are: PVS, polyvinyl sulfate; SDS, sodium dodecyl sulfate; GPC, gel permeation chromatography; PAGE, polyacrylamide gel electrophoresis; SC, Sephacryl S-POOsf; SD, Sephadex G-25; CF, chromatofocusing.

13609

13610 A Novel Heparinase: Purification and Characterization

1

FIG. 1. Purification profile of crude extract on hydroxylapatite. Crude extract (4.67 g of protein in 432 ml of 10 mM sodium phosphate, 1 mM 1 mercapt.oethanol, pH 7.0; 60.5 units/mg A280 of protein) was loaded onto a hydroxylapatite column (Pharmacia C26/40) pre- equilibrated with the same buffer. The flow rate was 27 ml/h, and 30.min fractions were collected. Heparinase activity (AA,i2,1/min > 0.2) was found in the shaded area.

5-

0-

5-

4 ELUATE

TABLE I Summary of heparinase III purification

A ratio of 1 IU = 19.9 units was used to compare .lALIL activities with the dye-based activities in the crude extract and the hydroxylapatite-purified fraction. Values in brackets indicate the percentage of the total.

Fractwn Mdhgrams of protein IU IU/mg of protein Enrichment

Crude extract 4.665.6 13,996.g 3.0 1.0 HA2 594.0 (12.7) 9,028.g (64.5) 15.2 5.0 CF4 80.0 (1.7) 5,272.0 (37.7) 65.9 24.0 SC2 29.4 (0.6) 3.722.0 (26.6) 126.6 47.0 X2.2 27.0 iO.Sj 31439.8 (24.6) 127.4 47.5

a b C d e f g

- 67 000

ua & - 43 000

CI - 30 000

I# b - 1.3 20 100

A ti * - 14 400

FIG. 2. SDS-PAGE of active fractions. a, HA2; b, SC2.2 (+dithiothreitol); c, SC2 (-dithiothreitol); d, standards; e, heparinase I; f, CF4; R. crude extract.

fourth pulse. Cell debris were removed by centrifugation in a Beckman J-26 centrifuge with rotor 40 at 60,000 x g for 2 h at 4 “C. The straw- colored supernatant fluid (crude extract) was subjected to further purification. All enzyme preparations and purifications were carried out at 10 ‘C.

Hydroxylapatite Chromatography-Crude extract (25 mg of protein/ml of bed volume) was loaded onto a hydroxylapatite column (Pharmacia 26/40) equilibrated with 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0. Unabsorbed material was washed off with 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0. Absorbed compounds were eluted stepwise with 0.5 M NaCl and 1 M NaCl (both in 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0) and finally with 0.05% Tween 20. Fractions with heparinase activity were pooled, concentrated by ultrafiltration, and dialyzed for 16 h against 100 volumes of 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0.

Chromatofocusing-PBE 118 was equilibrated with 0.025 M trieth-

anolamine HCI, pH 11.0, and packed in a Pharmacia ClO/40 column. Hydroxylapatite-purified enzyme was loaded onto the column and eluted with Pharmalyte g-10.5, pH 8.0, diluted 1:45 with distilled water. Fractions with heparinase activity were pooled, and the pH was adjusted to 7.0 with HCI and concentrated by ultrafiltration.

Gel Permeation Chromatography-Sephacryl S-200sf was resuspended in phosphate-buffered saline, pH 7.0, packed in a Pharmacia C26/100 column, loaded with the chromatofocused heparinase preparation, and eluted with phosphate-buffered saline, pH 7.0, in the presence or absence of 1 mM mercaptoethanol.

Protein Assay-Protein concentrations were measured according to the method of Bradford (31).

Enzyme Assays-Heparinase activity in the crude extract and hydroxylapatite-purified fractions was determined from the disap- pearance of heparin (1 unit = 1 mg of heparin degraded/h). The Azure A assay of Galliher et al. (27) was adapted for microdetermi- nations with the Titertek Multiskan MC (see Miniprint Supplement). Purer preparations (after chromatofocusing and GPC) were assayed by the increase in absorbance of heparin at 232 nm (32). One IU is defined as the amount of enzyme which causes 1 pmol of A4,5-uranic acid to be formed/min, based on a molar extinction coefficient of 5.1 X 10” M-’ cm-’ at 232 nm (33). The procedure of Linhardt et al. (32) was modified for direct determination of initial rates in a temperature- controlled spectrophotometer (see Miniprint Supplement). Hepar- anase, chondroitinase, and hyaluronidase activities were determined similarly, using heparan sulfate, chrondroitin sulfate, and hyaluronic acid, respectively, as substrates. Glycuronidase activity was measured by the decrease of absorbance at 232 nm using heparin fragments (prepared by heparinase digestion as described below) as substrate. Sulfatase activity was assayed by the procedure of Dodgon and Spencer (34) using p-nitrocatechol as substrate.

Polyacrylamide Gel Electrophoresis-SDS-PAGE was carried out on 10-15s gels with a Phast System according to the procedure of Olsson et al. (35, 36).

Isoelectric Focusing-The pI of purified heparinase III was estab- lished by isoelectric focusing using a Pharmacia FBE-3000 apparatus according to the directions of the manufacturer.

Amino Acid Anal,ysis-The amino acid composition of purified

A Novel Heparinase: Purification and Characterization 13611

0.50- - 6.0

A 250-

AA232 v . . . . . . mn

0.25 - - 2.5

100 200 300

mt ELUATE

FIG. 3. Chromatofocusing of hydroxylapatite-purified heparinase. HA2 (118.4 mg of protein in 6.1 ml of 10 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.0, 208.9 units/mg of protein) was chromatofocused on a PBE 118 column (Pharmacia C10/40) equilibrated with 0.025 M triethylamine, pH 11.0. Elution was performed with Pharmalyte B-10.5:H,O (1:45), pH 8.0, at a flow rate of 15 ml/h. Fractions of 10 min were collected.

1

A 230-

0

1 1 “0 “t

.O

AA232 _ . . . . . . mm

1.6

g ELUATE FIG. 4. Purification profile of CF4 on Sephacryl S-2OOsf. CF4 (15.05 mg of protein in 3.5 ml of Pharmalyte

8-10.5:H20::1:45, pH 7.0; 49.7 IU/mg of protein) was chromatographed on a Sephacryl S-200sf column (Pharmacia CZS/lOO) equilibrated with phosphate-buffered saline, pH 7.0. The flow rate was 8 ml/h, and fractions of 15 min were collected.

heparinase III was determined after HCl hydrolysis by reversed phase high performance liquid chromatography with a Waters Pica-Tag System using pre-column derivatization with phenylisothiocyanate according to the method of Bidlingmeyer et al. (37). Tryptophan, which is destroyed by HCl hydrolysis, was determined spectrophotometrically according to the procedure of Edelhoch (38), as well as chromatographically after hydrolysis with methanesulfonic acid by the method of Simpson et al. (39). Cysteine was determined after alkylation with [3H]iodoacetic acid according to the procedure of Anfinsen and Haber (40) (see Miniprint Supplement). Corrections were made for the destruction of Ser, Thr, and Tyr.

N-Terminal Analysis-The N-terminal amino acid was determined by the Edman technique (41, 42) with a spinning cup Beckman apparatus (Model 890) converted to a vapor-liquid-solid phase instrument according to the method of Brandt et al. (43). The phenylthio- hydantoin amino acid was identified on a Hewlett-Packard high

performance liquid chromatography instrument (Model 1084B) and a 3.9- x 300-mm pBondapack Cl8 column, using a gradient of 8-38% methanol in 0.01 M NaAc, pH 4.5, at a flow rate of 1.5 ml/min.

Preparation of Antibodies-A New Zealand white rabbit was im- munized intramuscularly with an emulsion (2:l) of complete Freund’s adjuvant and highly purified heparinase (1.5 mg/ml phosphate-huff- ered saline). Three boosters (containing incomplete instead of complete Freund’s adjuvant) were administered intramuscularly, intra- dermally, or subcutaneously at 14-day intervals thereafter. The total amount of heparinase injected ranged from 70 to 160 pg. The rabbit was bled from the ear on day 69, and the blood was stored overnight at 4 “C to clot. The serum was decanted, centrifuged at 2500 rpm for 15 min, and stored at -30 “C.

Zmmunodiffusion-Immunodiffusion was carried out in agar (1 g/ 100 ml phosphate-buffered saline, 0.05% Tween 20, pH 7.0) according to the procedure of Ouchterlony and Nilsson (44). The plates were


1 1 v, “i

L-Y 600

g ELUATE

AA232 - . . . . . . . . min

FIG. 5. Rechromatography of SC2 on Sephacryl SBOOsf. SC2 (4.8 mg of protein in 1 ml of phosphate- buffered saline, 1 mM mercaptoethanol, pH 7.0; 97.4 IU/mg of protein) was rechromatographed under reducing conditions on a Sephacryl S-2OOsf column (Pharmacia C26/100) equilibrated with phosphate-buffered saline, 1 mM mercaptoethanol, pH 7.0. The flow rate was 8 ml/h, and fractions of 15 min were collected.

a b PI - 9.3

c 8.65 +- a.45 c 8.15

- 7.35

- 6.85

c 6.55

- 5.85

* 5.20 - 4.55 + 3.50

FIG. 6. Isoelectric focusing of heparinase III (a) and standards (b).

incubated overnight at room temperature in a humid chamber. Free protein was removed by five successive washes with phosphate- buffered saline at room temperature over 48 h. Immunoprecipitin lines were stained with 0.5% Paragon blue, 5% acetic acid for 1 h at room temperature. The gel was destained at room temperature with 5% acetic acid, 30% methanol, 20% acetic acid, and 5% acetic acid (each for 1 h).

Characterization of Heparin Fragments-Heparin (2.5 mg in 0.1 M sodium phosphate, 1 mM mercaptoethanol, pH 7.0) was digested by either heparinase I or III (both at 0.025 IU) for 96 h at 30 “C. The reaction was stopped by acidifying with acetic acid. Heparin fragments were separated on a Sephadex G-25 column (Pharmacia C16/ 100) equilibrated and eluted with 0.5 M NH4Ac.

RESULTS

Purification-Heparinase III was enriched 5~ with respect to the crude extract by hydroxylapatite chromatography, the first purification step (Fig. 1 and Table I). The complex

TABLE II Comparison of the amino acid compositions of heparinase I and III Trp was determined spectrophotometrically as well as after hy-

drolysis with methanesulfonic acid. Cysteine was determined radi- ometrically. All the other amino acids were determined after HCI hydrolysis. The mole percent of the amino acids were calculated assuming the molecular weight of heparinase III as 94,000.

Heparinase III Heparinase I (44)

Amino acid Nearest Mole Nearest Mole integer percent integer percent

Aspartic acid 106 12.1 52 12.8 Glutamic acid 61 6.9 41 10.1 Serine 79 9.0 25 6.2 Glycine 96 10.9 35 8.6 Histidine 12 1.4 11 2.7 Arginine 2 2.7 10 2.5 Threonine 4.2 23 5.6 Alanine 101 11.5 50 12.2 Proline 42 4.8 21 5.2 Tyrosine 26 3.0 16 3.9 Valine 48 5.5 15 3.7 Methionine 16 1.8 3 0.7 Half-cystine 1 <O.l 4 0.9 Isoleucine 40 4.6 16 3.9 Leucine 70 8.0 28 6.9 Phenylalanine 31 3.4 20 5.0 Lysine 70 8.0 37 9.0 Tryptophan 18 2.1 ND” ND Total 878 100 407 100

’ ND, not determined.

mixture of proteins in the crude extract (Fig. Zg) was thereby reduced to about 25 main components (Fig. 2a), with proteins of M, = 94,000 and 72,000 being the most prominent. Further purification was achieved by chromatofocusing with PBE 118 (Fig. 3). Heparinase activity was found in the major compo- nent, CF4, which eluted at a pH of 9.2. This suggests that the p1 of heparinase III may be significantly higher than that of heparinase I (8.5, Ref. 45). Besides heparinase activity, CF4 also contained heparitinase, chondroitinase, and hyaluronidase activity. Sulfatase activity was found in CFl while glycuronidase activity was observed in CF8. The Dresence of

FIG. 7. Lineweaver-Burk plot of heparinase III activity. Initial rates (v) were measured by the AA232 method at 37 “C in 50 mM sodium phosphate, 1 mM mercaptoethanol, pH 7.6. Heparin concentrations varied from 0 to 300 pM. The protein concentration in the reaction mixture was 0.18 rg/ml. An average molecular mass of 11 kDa was used for heparin to calculate its molar concentration.

A Novel Heparinase: Purification and Characterization

K, = 3.4 x lo-’ M

13613

l/[HEPARINl, mM-’

FIG. 8. Inhibition of heparinase III by PVS. Initial rates were measured with the ilAz32 method in the presence of 0 to 9.8 PLM PVS at different heparin concentrations of 15, 25, 35, and 45 pM.

more than one protein in CF4 was also demonstrated by SDS- PAGE (Fig. 2f). Bands were found at M, = 94,000 (main band), 72,000, 49,000, and 14,000. Significantly, no band was seen at 43 kDa, the size of heparinase I.

These proteins were separated on Sephacryl S-2OOsf (Fig. 4). Heparinase activity coincided with the high molecular mass fraction SC2. Chondroitinase and hyaluronidase activity were found in SC3, while heparanase activity was associated mainly with SC3 but to a lesser extent also with SC2. After rechromatography of SC2 under reducing conditions (Fig. 5), more than 90% of the material was found in a symmetrical peak SC2.2, with the same K., value as SC2 in Fig. 4 and corresponding to a molecular mass of 93.9 kDa. SDS-PAGE of SC2 and SC2.2 gave similar results: purified heparinase (SC2.2) showed a single band (Fig. 2c) with an electrophoretic mobility corresponding to a molecular mass of 94 kDa. The reducing agent, dithiothreitol, did not affect the migration position (compare Fig. 2, b and c). These results indicate that heparinase III does not possess interchain disulfides and is most likely a single chain protein with a molecular mass of 94 kDa. This is about twice the size of heparinase I. Heparin-

ase III was purified 48-fold over the crude extract to a specific activity of 127 IU/mg of protein (Table I), which is substan- tially higher than the specific activity of purified heparinase I (26.6 IU/mg of protein, Ref. 45). Recovery of protein (0.6%) and enzyme activity (18.9%) were also higher than the values determined for heparinase from F. heparinum (0.003% and O.B%, respectively, Ref. 45).

Zsoelectric Focusing-Purified heparinase III focused at pH 9.2 (Fig. 6), which is in agreement with the p1 determined by chromatofocusing (Fig. 3). This value differs from the p1 of 8.5 of heparinase I (45).

Amino Acid Analysis-The amino acid compositions of heparinase I and III are compared in Table II and differ significantly, confirming that heparinase III is not a dimer of heparinase I. The lower content of the basic amino acids, lysine and arginine of heparinase III (10.7 mol%), in comparison to heparinase I (11.5 mol%, Ref. 45) was noticeable. On the other hand, the former contains less acid and acid amide amino acids: 19.0 uersus 22.9 mol% (45), which may contribute to the higher p1 of heparinase III versu.s I (see above).

The low content of sulfur-containing amino acids in hepa-


.G

FIN;. 9. Effect of ionic strength, pH, and temperature on heparinase III activity. Initial rates were measured with the 1A?:,>method at a protein concentration of 0.18 &ml. 0, NaCl concentration varied between 0 and 0.223 M in 50 mM sodium phosphate, 1 !IIM mercapt.oethanol, pH 7.6. , citrate-phosphate-borate buffer, pH 5.3 to 9.7 was used. A, Temperatures ranged from 5- 65 “C.

30 -

c*

20 -

10 -

b d

c .. I - - ,

I T,“C -

I 0.2 0.3

IONIC STRENGTH+-+3 I I

FIG. 10. Immunodiffusion plate. Well a contains rabbit antiserum against heparinase III (20 ~1 undiluted); well b, heparinase III (108 ~g); well c, heparinase I (4.4 pg); and u&l d, heparinase III (8.6 !4).

rinase III was conspicuous. Only one cysH/enzyme could be determined after alkylation with [“Hliodoacetic acid (see Min- iprint Supplement). Heparinase I, on the other hand, contains 4 residues of cysH per molecule (45). Good agreement was found between the Trp determinations obtained by methanesulfonic acid hydrolysis (17.8 residues/m01 of protein) and spectroscopy (19 residues/m01 of protein).

The molar (t) and specific (E 0.1%) extinction coefficients of heparinase III at 280 nm were calculated from the observed Trp and Tyr contents and were 1.36 X lo” M-’ cm-’ and 1.44, respectively.

N-Terminal Analysis-Just one type of N-terminal amino acid, uiz. alanine, was found after Edman degradation. Results from SDS-PAGE, GPC, and amino acid analysis had already shown that heparinase III is not a disulfide-bonded dimer of heparinase I. Covalent linkage of two subunits by, e.g. Lys and Glu, as is found in cross-linked fibrin (46), has not been ruled out. However, if this were the case, heparinase III most likely would have shown two different N-terminal amino acids. The results therefore strongly indicate that heparinase III is a single chain protein.

e e 10

pH 04

Kinetics and Substrate Specificity-The kinetic constants of heparinase III were determined by Lineweaver-Burk analysis (Fig. 7). The K,,, for heparin was found to be 3.4 pM and the V,,,,, as 36.8 Fmol/min at a total enzyme concentration of 0.18 pg/ml. The corresponding values reported for heparinase I activity were 8.0 pM and 98.5 pmol/min, respectively, at an enzyme concentration of 0.5 pg/mI. Heparinase III thus shows a greater affinity toward heparin than heparinase I. The k,,, value of heparinase I (4.23 X 10” min-‘, Ref. 45) is also lower than that of heparinase III (1.93 x lo4 min-‘). This is in agreement with the higher specific activity of heparinase III (127.4 uersus 26.6 IU/mg of protein).

Heparinase III was very specific, acting only on heparin and to a lesser extent on heparan sulfate (11% of the activity on heparin), but not on chondroitin sulfate nor on hyaluronic acid. The latter two, as well as GlcNAc, also did not inhibit heparin degradation when present at 22 mg/lOO ml. PVS, on the other hand, was found to be a competitive inhibitor of heparinase III activity. The Ki was determined to be 1.7 X lo-’ M by the new Dixon method (Fig. 8). Heparinase I was found to bind with a higher affinity to PVS (K; of 3.0 X lo-’ M, Ref. 23), suggesting that PVS is a tight binding inhibitor (47) for heparinase I.

Conditions for Optimal Actiuity-The effects of ionic strength, pH, and temperature on heparinase III are shown in Fig. 9. Optimal activity was found at 0.03 M NaCl, 50 mM sodium phosphate, 1 mM mercaptoethanol (i.e. r/, = 0.15), pH 7.6 and 45 “C. The activation energy was calculated to be 15.63 kcal/mol and the temperature coefficient (&) as 2.48. Heparinase I activity was optimal under quite different conditions: 0.1 M NaCl, 0.25 M sodium acetate, 2.5 mM calcium acetate, 10 mM sodium phosphate (i.e. I’/, = 0.4), pH 6.5 and 37 “C! (45). The activation energy as well as the temperature coefficient were also lower: 6.3 kcal/mol and 1.45, respectively (45). The presence of calcium (0.2 mM) in the reaction mixture did not affect the activity, neither did EDTA (0.1 mM) or

mercaptoethanol (1 mM). On the other hand, the enzyme was

13615 A Novel Heparinase: Purification and Characterization ,

0.50.

A232

0.25~

g ELUATE FIG. 11. Elution profile of heparin fragments on Sephadex G-25. Heparin fragments prepared by

extensive digestion of heparin with heparinase I or III were separated on a Sephadex G-25 column (Pharmacia C16/100) equilibrated with 0.5 M NH4Ac. The flow rate was 4 ml/h, and 30-min fractions were collected.

completely inhibited by 0.1 mM HgC12. Heparinase I was also inhibited by HgC& (26), not affected by mercaptoethanol(45), but stimulated by calcium (26).

Conditions for Optimal Stability-Heparinase III was found to be most stable at pH 7.0, just as heparinase I (45). The presence of glycerol (0.8 M) or NaCl (0.5 M) stabilized the enzyme when stored as a diluted solution. A higher residual activity was also found if 1 mM mercaptoethanol was included in the buffer as a reducing agent. Preliminary results suggest that heparinase III is more stable than heparinase I. The former enzyme could be exposed to a higher temperature (37 versuS 30 “C) with the retention of a comparable amount of enzyme activity (87 uersus 93%) after incubation for 96 h under optimal conditions (45).

Immunodiffusion-No immunoprecipitation of heparinase I was observed after Ouchterlony double diffusion with rabbit antiserum against heparinase III. The latter antigen gave a strong precipitin band (Fig. 10).

Degradation Products-Heparin fragments produced by heparinase III were compared by GPC to those prepared with heparinase I (Fig. 11). According to the known profile of heparinase I-derived fragments (48), SD1 can be identified as the disaccharide-, SD2 the tetrasaccharide-, and SD3 as the hexasaccharide-containing fraction. The chromatogram clearly shows that heparinase III also produced mainly disaccharide fragments from heparin. The results, however, also reveal that the reaction with heparinase III proceeded further than with heparinase I, since more disaccharide and less higher molecular mass fragments were observed. This was also found to be the case with heparinase II (49). Subtle differences in the specificities of the enzymes thus seem to exist.

DISCUSSION

A heparin lyase was purified to homogeneity according to SDS-PAGE and GPC. In comparison with the properties of the known heparinase I from F. heparinarum (45), this enzyme proved to be unique. It has a molecular mass of about twice that of heparinase I: 94 versus 43 kDa, but is not a dimer of heparinase I. Molecular mass determinations were

done under reducing conditions and just one type of N- terminal amino acid, alanine, could be determined.

The unique structure of heparinase III was also evident from its different amino acid composition compared to heparinase I (45). Most conspicuous was the occurrence of only 1 cysteine in heparinase III compared to the 4 in heparinase I (45). Retention of the reduced form of Cys proved to be of great importance for the reactivity of the enzyme, since Hg2 was shown to completely inhibit heparinase III, while mercaptoethanol preserved activity during prolonged storage. The low content of sulfur-containing amino acids of heparinase I (1.6 mol%) and III (1.8 mol%) might be related to their functions, uiz. the degradation of heparin in order to replenish the sulfur pool of the bacteria when sulfur becomes limiting. Taking the synthesis of the enzymes under limiting sulfur concentrations into account, natural selection most probably put heparinases with high contents of methionine or cysteine at a disadvantage.

Interestingly, the two enzymes also differed immunologi- cally as was evident from Ouchterlony double diffusion analysis which showed no cross-reactivity between heparinase I and anti-heparinase III serum. Heparinase I and III not only differed structurally but also with regard to kinetic parameters: heparinase III showed a lower K,,, (3.4 uersus 8.0 pM)

and a higher kc,,, (1.9 x lo4 uersus 4.2 X lo3 min-‘) than heparinase I (45). This means that not only does heparinase III bind more strongly to heparin, but it also degrades it more effectively. The higher efficacy of heparinase III is clearly demonstrated in its specificity constant, k,,,/K,,,, which is an order of magnitude higher than that of heparinase I (5.7 x

lo7 uersus 5.3 X 10’ M-’ mix--‘, Ref. 45). In addition to the K,,, and k,,,, other parameters also differed

between heparinase I and III. The energy of activation for example was higher, 15.6 uersw 6.3 kcal/mol (45). Ca*+ was found not to influence heparinase III activity, whereas it was shown to activate heparinase I (26). Furthermore, optimal activity was found under different conditions: pH 7.6, r/z = 0.16 and 45 “C for heparinase III versus pH 6.5, I’/2 = 0.4 and 37 “C for heparinase I (45). Since conditions for optimal heparinase 111 activity are nearer to plasma values (0.154 M

13616 A Novel Heparinme: Purification and Characterization

NaCl, pH 7.4, 37 “C, Ref. 50) and heparinase III is more effective and heat-stable than heparinase I, the former enzyme should be superior for the deheparinization of blood.

Although heparinase III differs structurally and kinetically from heparinase I, both react with the same glycosidic bond in heparin, cleaving it nonhydrolytically by elimination (51). In both cases, disaccharide units are the main end products (49). On the basis of these findings, heparinase III should also be classified as a heparin lyase, class 4.2.2.7, according to the system of the Enzyme Commission. In order to distinguish it from the known flavobacterial enzymes (I and II), the desig- nation heparinase III for the enzyme from Isolate 114 is recommended.

Recently, a fourth heparin lyase, also with unique features, was isolated from the anaerobe mouth bacterium Bucteriodes heparinolyticus (52). The molecular mass was determined to be 63 kDa, and the p1 was 9.5. Degradation products and enzyme specificity also resembled those of heparinase I, II, and III. Furthermore, this enzyme was also inhibited by Hg2+. On the grounds of these unique characteristics (molecular mass and PI), it is suggested that this enzyme should be identified as heparinase IV.

Without further information about the primary, secondary, and tertiary structures of these heparinases or the nucleotide sequences of their genes, it is difficult to speculate on their phylogenetic relationship. It might be that these enzymes originated from the same ancestral gene by divergent evolu- tion as is the case with the serine proteases trypsin, chymo- trypsin, elastase, thrombin, and plasmin (53). In these proteins, a large proportion of amino acids were conserved, and the catalytic centers are similar. On the other hand, the heparinases might have originated from separate genes and developed the same functional activity by convergent evolu- tion. The bacterial serine protease, subtilisin, for example, only resembles the trypsin family in so far as the catalytic center is concerned. Its overall architecture is entirely different (54).

REFERENCES

1. Linker, A., and Hovingh, P. (1977) Fed. Proc. 36, 43-46 2. Linker, A., and Hovingh, P. (1979) &p&z: Structure Cellular Functions

and Clinical Ao~lications (McDuffie. N. M.. ed) nn. 3-24. Academic Press. Orlando, FL . .

_.

3. Casu, B., Oreste, P., Torri, G., Zoppetti, G., Choay, J., Lormeau, J. C., Petitou, M., and Sinay, P. (1981) Biochem. J. 197,599-609

4. Linker. A.. and Hovineh. P. (1984) Carbohvdr. Res. 127.75-94 5. Merchant,’ Z. M., Kim,‘Y. S., Rice, K. G., and Linhardt, R. J. (1985)

Biochem. J. 229,369-377 6. Reeves, W. H., Kanwar, Y. S., and Farquhar, M. G. (1980) J. Cell Biol. 86,

735-753 7. Shimada. K.. Gill, P. J., Silbert. J. E., and Douelas. W. H. J. (1981) J. Clin.

Inuest:68; 995-1002’ -

8. Lander, A. D., Fujii, D. K., Gospodarowicz, D., and Reichardt, L. F. (1982) J. Cell Bid. 94, 574-585

9. Castellot, J. J. Jr.! Addonizio, M. L., Rosenberg, R., and Karnovsky, M. J. (1981) J Cell B~ol. 90,372-379

10. Linhardt, R. J., Grant, A., Cooney, C. L., and Langer, R. (1982) J. Biol. Chem. 267.7310-7313

11. Beraavist. D.. Nilsson. B.. Hedner. U.. Pedersen. P. C.. and Ostereaard. P. By {19&j) Yi’hromb. &s.‘38,589-601

I

12. Diness, V., Nielsen, J. I., Pedersen, P. C., Wolffbrand, K. H., and Oster- gaard, P. B. (1986) Thronb. Res. 65,410-414

Neerstand, H., Ostergaard, P., Bergqwst, D., Matzsch, T., and Hedner, U. (1987) Fibrinolysti 1,39-43

Ostergaard, P. B., Nilsson, B., Bergqvist, D., Hedner, U., and Pedersen, P. C. (1987) Thromb. Res. 46.739-749

Dine&, V.,‘and Ostergaard, P: B. (1986) Thromb. Hoemostask S&318-322 Matsch, T., Bergqvist, D., Hedner, U., and Ostergaard, P. (1987) Thromb.

Haemostasis 67,97-101 Fareed, J., Walenga,, J. M., Hoppenstead, D., Huan, X., and Racanelli, A.

(1988) Haemostasls l&3-15

13

14

15. 16

17.

18

19.

i::

22.

23.

24.

25.

26. 27.

28.

29.

30.

i::

33. 34. 35.

36.

37.

it:

40. 41. 42.

43.

44.

Folkman, J., Langer, R., Linhardt, R. J., Haudenschild, C., and Taylor, S. (1983) Science 221,719-725

Cmm, R., Szabo, S., and Folkman, J. (1985) Nature 230,1375-1378 Hutt, E. D., and Kingdon, S. (1972) J. Lab. Clin. Med. 79,1027-1034 Linhardt, E. J., Coonev. C. L.. Tanuer. D.. Zannetos. C. A.. Larsen. A. K..

and Langer, R. (19&j A& B&hem. Biktechnol. @,41-55 Lancer. R.. Linhardt. R. J.. Coonev. C. L.. TaDDer. D.. and Klein. M. D.

(1982) Eruyme En;. 6,433-441 “’ _. Langer, R., Linhardt, R. J., Galliher, P. M., Flanagan, M. M., Cooney, C.

L., and Klein, M. D. (1982) Ado. Chem. Ser. 199,493-509 Langer, R., Linhardt, R. J., Hoftberg, S., Larsen, A. K., Cooney, C. L.,

Tapper, D., and Klein, M. (1982) Science 217,261-263 Linhardt, R. J., Cooney, C. L., Tap

and Langer, R. (1984) Arch. Bioc er D Zannetos, C. A., Larsen, A. K., & Y .P_

m. Bzophys. 9,41-55 Linker, A., and Hovingh, P. (1972) Methods Enzymol. 28,902-911 Galliher, P. M:, Cooney, C. L., Langer, R., and Linhardt, R. J. (1981) Appl.

Enuiron. Mzcrobiol. 4 1, 360-365 Galliher, P. M., Linhardt, R. J., Conway, L. J., Langer, R., and Cooney, C.

L. (1982) Eur. J. Appl. Microbial. Biotechnol. 16,252-257 Cerbelaud, E. C., Conway, L. J., Galliher, P. M., Langer, R. S., and Cooney,

C. L. (1986) Ap Joubert, J. J., an a

1. Emwon. Microbial. 61,640-646 Pitout, M. J. (1985) Experientin 41, 1541

Bradford, M. M. (1976) Anal. Biochem. 72,248-254 Linhardt. R. J.. Fitzeerald. G. L.. Coonev. C. L.. and Lancer. R. (1982)

Biochin. Bio’hys. Acta iO2,19?-203 ” ’ ” Linker, A., an B Hovmgh, P. (1972) Biochemistry 11,563-568 Dodgon, K. S., and Spencer, B. (1957) Methods Biochem. Anal. 4,211-255 Olsson. I.. Fredrikson. U. B.. Deaerman. M.. and Olsson. B. (1988) Electro-

phoies&9,16-22 ’ - OLssarQ Whe_eler,.R., Johansson, c.~, Ekstrom, B., Stafstrom, N., Bikhab-

Bidlingmi and Jakobson, G. (1988) Ekctrophoresk 9,22-27

:yer, B. A., Cohen, S. A., and Tarvin, T. L. (1984) J. Chromotogr. 336,9: 3-104

Edelhoch. H. (1967) Biochemist 6,1948-1954 Simpson, ‘R. J., Neuberger, M.%., and Liu, T.-Y. (1976) J. Biol. C/tern.

25 -A, 1936-1940 Antinsen, C., and Haber, E. (1961) J. Biol. C&m. 236,1361-1363 Edman, P., and Begg, G. (1967) Eur. J. Biochem. 1, SO-91 Hunkapiller, M. W., Hewick, R. M., Dreyer, W. J., and Hood, L. E. (1983)

Methods Enzymol. 9 1,399-413 Brandt, W. F., Alk, H., Chauhan, M., and von Holdt, C. (1984) FEBS L&t.

174.228-232 - -7 - - - - - -

45.

46.

47.

48. 49.

Ouchterlony, O., and Nilsson, L. A. (1978) Handbook of Experimental Immunology (Weir, D. M., ed) pp. 19.1-19.39, Blackwell Scientific Pub- lications Ltd., Oxford

Yang, V. C., Lmhardt, R. J., Bernstein, H., Cooney, C. L., and Langer, R. (1985) J. Bi+ Chem. 260,1849-1857

Pig30, J. J., Fmleyson, J. S., and Peyton, M. P. (1968) Science 160,892

Dixon, M., and Webb, E. C. (1979) Enzymes, 3rd Ed, pp. 47-206, Longman Green, London

Silva, M. E., and Dietrich, C. P. (1975) J. Biol. Chem. 260.6841-6846 McLean, M. W., Long, W. F., and Williamson, F. B. (1985) Proceedings of

the 8th International Symposium on Glycoconjugates Vol. 1, pp. 73-74, Paeger Publishers, New York

50.

51. 52.

53.

Diem, K., and Lentner, C. (1971) Scientific Tables. 7th Ed, pp. 517-557, CIBA Limited, Base1

Linker, A., and Hovingh, P. (1965) J. Biol. Chem. 240,3724-3728 Nakamura. T.. Shibata. Y.. and Fuiimura. S. (1988) J. Clin. Microbial. 26.

54.

1070-1071 * ’ Dayhoff? M. O., Hunt, L. T., and Hurst-Calderone, S. (1978) Atlas of

Protern Sequence and Structure (Dayhoff, M. O., ed), Vol. 5, Sup lement 3, p

P .363-369, National Biomedical Research Foundations, Was mgton Ti

Mark and, F. S., and Smith, E. L. (1967) J. Biol. &em. 242,5198-5211

A Novel Heparinase: Purification and Characterization 13617

D C

-

5 I

r

‘1 i

1 I

l-

Documents

Purification and Characterization of a Novel Heparinase” A Novel Heparinase: Purification and Characterization 1 FIG. 1. Purification profile of crude extract on hydroxylapatite