Inr. J. Eiochem. Vol. 22, No. 5, pp. 481-487, 1990 Printed in Great Britain. All rights reserved

0020-7 1I X/90 $3.00 + 0.00 Copyright Q 1990 Pergamon Press plc

PURIFICATION AND CHARACTERIZATION OF TWO ACIDIC PHOSPHOLIPASE A, ENZYMES FROM KING COBRA (OPff1QPHAGUS HANNAH) SNAKE VENOM NGET-HONG TAN and M. NOMANBHAYSAIFUDDIN Department of Biochemistry, University of Malaya, Kuala Lumpur, Malaysia

(Received 9 September 1989)

Abstract-l.

The two ma-jot phospholip~e

A, enzymes (OHPLA-DE1 and OHPLA-DE21 of king cobra homogeneity. 2. The isoelectric noints of OHPLA-DE1 and OHPLA-DE2 were 3.8 I and 3.89. resmctivelv and the M,s were 14,000 ad 15,000, respectively, as estimated by Sephadex G-75 gel filtration chromaiography; and 14,000 as estimated by SDS-PAGE. 3. The enzymes were not lethal to mice at a dosage of 10 pg/g body wt by i.v. route. Both phospholipase A, enzymes, however, exhibited moderate edema-inducing and anti-coagulant activities. 4. Bromophenacylation of the enzymes reduced the enzymatic activity drastically but did not affect the edema-inducing activity of the enzymes.

(O~hio~~~gu~ harm&) &nom have be& purified .to el&trophoretic

Determinufion~ o~Froiein and phos~holi~ase A activity

Phospholipas~ A, (EC 3.1.1.4) is widely found in the venoms of snakes, bees and scorpions as well as in mammalian tissues. Many phospholipase A, enzymes have been purified from various sources [for review see Verheij et al. (1980a)]. Besides acting as digesting enzymes, in mammalian tissues the enzyme is also involved in the synthesis of arachidonic acid and its metabolites including prostaglandins and leukotrienes (Hammarstom, 1983). Many phospholipase A2 enzymes from various venom sources, however, also play important roles in the toxic action of the venoms and may exhibit neurotoxic, myotoxic, cardiotoxic, necrotic, hemolytic, anticoagulant, edema-inducing, convulsant or hemorrhagic activities (Rosenberg, 1986, Kini and Iwanaga, 1986, Vishwanath et al., 1987). It is of interest to study the pharmacologically active phospholipase A, as structure-function relation studies may provide information about the molecular features essential for the pharmacological activities of the enzyme. We report in this paper the isolation and characterization of two phospholipase A, enzymes which exhibited moderate edema-inducing and anticoagulant activities from the king cobra (Uphiuphagus hannah) venom. MATERIALS

AND ME’M-IODS

Lyophilized king cobra (0, hannah) venom was obtained from Southeast Asia Venom Institute (Kuala Lumpur, Malaysia). DEAE-Sephacel, Sephadex G-200, Sephadex G-75 and pharmalyte (pH range 2.5-5.5) were obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). SP2300 polyester for gas liquid chromatography and the fatty acid methyl ester standards were purchased from Supelco (Bellefonte, Pa, U.S.A.). All other reagents are of analytical grade and were purchased from Sigma Chemical Company (St Louis, U.S.A.).

Protein concentration was determined by the Bradford method (Bradford, 1976). Phospholipase A activity was determined using the acidimetric method (Tan and Tan, 1988) and the substrate suspension (15 ml) consisted of one part egg yolk, one part 8.1 mM sodium deoxycholate and one part 18 mM calcium chloride. One unit of phospholipase A activity was defined as the release of one /~mol of fatty acid per min. PuriJcation

of

the phosphotipase A enzymes

All of the following operations were carried out at 4°C. A 1 g sample of king cobra venom was dissolved in 15 ml of 20 mM Tris--HCl buffer, pH 7.4 and applied to a Sephadex G-200 gel filtration column (2 x Wcm). The column was eluted with 20 mM Tris-HCI buffer, pH 7.4 at a flow rate of 10 ml/hr and 4.5 ml effluent was collected per tube. Fractions exhibiting strong phospholip~e A activity (tubes 37-46) were pooled and further fractionated by a DEAESephacel ion exchange column (20 x 80 mm) equilibrated with the same Tris-HCl buffer. The column was eluted with a linear, 0.12-0.25 M sodium chloride gradient (800-800 ml) at a constant pH of 7.4 and a flow rate of 18 ml/hr. Per tube 5ml of effluent was collected. The two peaks exhibiting strong phospholipase A activity were separately pooled (tubes 42-51 constituted OHPLA-DE1 fraction whereas tubes 71-82 constituted OHPLA-DE2 fraction) and precipitated by adding ammonium sulfate to 95% saturation. The precipitates were redissolved in 4 ml of 20mM Tris-HCl buffet, pH 7.4 and separately further fractionated by a SeDhadex G-75gel filtration column (2 x 1OOcm) eauilib&ted with the same buffer. The column was elutedwitf; the same Tris-HCl buffer at a tlow rate of 20 ml/hr and 5 ml of effluent was collected per tube. The phospholipase A preparations obtained were designated as-the &if&d 0. h&&h phospholipase A-DE1 (OHPLA-DEI) and phospholipase A-DE2 (OHPLA-DEZ), respectively. Etecrrophoresis and determinarions of molecular weight and isoeteclric point

Both PAGE and SDS-PAGE were carried out using 12.5% gels at pH 8.8 (Laemmli, 1970). Molecular weight was determined by SDS-PAGE and Sephadex G-75 gel 481

NGET-HONGTAN and M. NOMANBHAY SAIFUDDIN

482

filtration chromatography using a 12 x 1950 column equilibrated with 20 mM Tris-HCl buffer, pH 7.4. The isoelectric point of the purified enzyme was determined using an 110 ml electrofocusing column (Vesterberg and Svensson, 1966).

dislocation and the edema ratio is defined as the percentage of the weight of the edematous leg to the weight of the normal leg. Minimum edema dose is defined as the amount of sample causing an edema ratio of 130%.

Amino acid composition analysis

Chemical modz$cation by p-bromophenacyi bromide

The purified material was hydrolyzed in 6 M hydrochloric acid in sealed tube at 115°C for 20 hr and analyzed as described earlier (Tan, 1982). Cys/Z was analyzed as cysteic acid. Tryptophan was determined spectrophotometrically (Goodwind and Morton, 1946).

This was carried out essentially as described by Roberts et al. (1977).

Dererminaiion of the substrare positional speczjicify

Synthetic phospholipid L-a-1-stearoyl-2-oleoyl lecithin was hydrolyzed by the purified enzyme as described by Robertson and Lands (I 962) and the free fatty acid reieased was extracted, separated by silica gel H thin-layer plate, methylated and analyzed by gas liquid chromatography using a glass column packed with 10% SF-2300 polyester. Determinations of lethality, recalczfica~ion time and edemainducing activity

Lethality was determined by iv. injection of the sample into the caudal veins of mice (25 + 3 g. 4 mice per dose). Recalcification time was determined using titrated rabbit platelet-poor plasma (Verheij er al., 1980b). The plasma (0.2 ml) was incubated with the sample (0. I ml) at 37°C for I min followed by addition of 25 mM calcium chloride (0.2 ml). Edema-inducing activity was determined according to the method described by Yamakawa et al. (1976). Mice weighing 25 f 3 g in groups of 3 were injected in the foot pads of hind limbs with different doses of sample in 50 ~1 of 0.85% saline. After t hr, the mice were killed by cervical

Statistical analysis

The significance of difference was determined by Student’s f-test. The level of significance chosen was P = 0.01. RESULTS AND

DE33JSSION

~urz~eat~on qf the king cobra L’enorn p~osp~o~jpase enzymes

Sephadex G-200 gel fihration chromatography of 1 g of the crude venom (phosphohpase A 85,000 U) yielded two major peaks, with phospholipase A activity occurred between the two peaks (Fig. 1, insert). Tubes exhibiting strong phospholipase A activity (tubes 37-46, phospholipase A 54,200 U, 60 mg) were pooled and fractionated by DEAE-Sephacel ion exchange chromatography, which yielded nine protein peaks (Fig. 1). Four of the peaks (peaks 4, 5, 6 and 7) exhibited phospholipase A activity. The two major phosphohpase A peaks (OHPLA-DEI fraction, tubes 42-5 I, 12,600 U, 9.0 mg protein; OHPLA-DE2 fraction, tubes 71-82, 9200 U, 2.8 mg protein) were separately pooled and precipitated by ammonia

ix

2b

Tube

;0

Number

$0

f

sb

aI

.E -0.20 ;’ :

I

!., PLA - DE 1

$ 5

E $ -0.15 8

-0.10

Tube

A

Number

Fig. 1. DEAE-Sephacel ion exchange chromatography of the phospholipase A fraction obtained from Sephadex G-200 gel filtration of king cobra venom. Column size is 20 x 80 mm. The ion-exchanger was equilibrated with a 20 mM Tris-HCl buffer, pH 7.4 and elution was carried out at a constant pH using the same buffer and a linear, 0.12-0.25 M sodium chloride gradient. Flow rate was 18 ml/hr and 5 ml of effluent was collected per tube. PLA-DE1 and PLA-DE2 are the two major phospholipase A fractions. The insert near the top right is the Sephadex G-200 gel filtration chromatography of the king cobra venom (1 g). The column (2 x 90 cm) was equilibrated and eluted with a 20 mM Tris-HCl buffer, pH 7.4 and 4.5 ml of effluent was collected per tube. Absorbance at 280 nm (-); phospholipase A activity ( . . ); sodium chloride gradient (0).

t

.

I-

I

10

I

20

30 Tube

10

20

3b Tube

J

\

I

I

40 Number

40

I

50

60

50

60

Number

Fig. 2. Sephadex G-75 gel filtration of PLA-DE1 (top) and PLA-DE2 (bottom) fractions obtained from DEAE-Sephacel ion exchange chromatography (Fig. I). The column (2 x 100 cm) was equilibrated and eluted with 20 mM Tris-HCI buffer, pH 7.4. A 5 ml sample of effluent was collected per tube. The insert near left of the figures are (A) PAGE of the purified enzymes (B) SDSPAGE of the purified enzymes. The protein migrated from cathode (upper) to anode (lower). Absorbance at 280 nm (-); phospholipase A activity (-- ----).

483

King cobra venom phospholipases A, Table

I, Amino

Amino acid LYS His Arg Asx Thr Ser GlX Pro GUY Ala Cysl2 Val Met Ileu Leu TYr Phe Tryp Total

acid compositions 0. hannah PLA-DE1

3.11 1.00 3.91 17.69 6.83 9.12 8.12 10.07 8.61 1.78 11.70 2.07 0.84 6.18 8.00 7.53 4.19 3.95

3 1 4 18 7 9 9 10 9 8 12 2 I 6 8 8 4 4 123

of phospholipase 0. hannah PLA-DE2 3.15 1.80 6.20 17.85 8.56 7.74 9.64 8.70 7.81 8.01 11.56 5.02 1.01 4.93 4.00 8.10 3.20 1.90

485

A, enzymes from king cobra venom and various

other sources

Naja naja atra PLA*

Naja melanoleuca PLA’

Porcine pancreas PLA’

5 5 20 5 6 8 5 10 II 12 4

4 3 6 20 6 6 6 4 9 II 14 4

9 3 4 23 6 10 7 5 6 8 12 2

5 5 9 4 3

6 3 9 4 3

18 9 8 10 9 8 8 12 5

I

122

The first columns of the data for 0. hannah phospholipase A, enzymes show the normalized molar ratios. The integers show the number of residue of each amino acid. Cys/2 was determined as cysteic acid after performic acid oxidation. Tryptophan was determined spectrophotometrically. *References: N. naia atraPLA (Chane. and Lo, 1972); N. maloneuca PLA (Joubert, 1975); Porcine pancreatic PLA (De Haas et dr.,1970). -

sulfate. The precipitates were redissolved in 20 mM Tri-HCl buffer, pH 7.4, and separately further fractionated by Sephadex G-75 gel filtration chromatography. For OHPLA-DE1 fraction, this yielded three protein peaks of which peak 1 (Fig. 2, TOP), the only peak exhibiting phospholipase A activity (10,900 U, 3.46 mg protein), was designated as the purified 0. hannuh venom phospholipase A-DE1 (purified OHPLA-DEl). Three peaks were also obtained from Sephadex G-75 chromatography of OHPLA-DE2 fraction (Fig. 2, Bottom): only peak 2 exhibited phospholipase A activity (6920 U, 2.02 mg protein) and this was designated as the purified 0. hannah venom phospholipase A-DE2 (purified OHPLADE2). Based on the recoveries of the purification steps, it was estimated that the two phospholipase A enzymes constituted ca 4% of the venom dry weight. Characterization enzymes

of the purljied phospholipase

A

The specific activity of purified OHPLA-DE1 was 3426 pmol/min/mg while that of OHPLA-DE2 was 3150 pmol/min/mg, with egg yolk suspension as substrate and in the presence of sodium deoxycholate. In polyacrylamide and SDS-PAGE, single bands were observed for both purified OHPLA-DE1 and OHPLA-DE2, indicating that the enzymes were homogeneous electrophoretically (Fig. 2, insert). The M,s of OHPLA-DE1 and OHPLA-DE2 were 14,000 and 15,000, respectively, as determined by gel filtration chromatography and 14,000 as determined by SDS-PAGE, indicating that the enzymes exist as monomeric form. The isoelectric points of OHPLA-DE1 and OHPLA-DE2 were determined to be 3.81 and 3.89, respectively. The amino acid compositions of the two purified enzymes are shown in Table 1. Amino acid compositions of phospholipases A, from several other sources are also listed for comparison. It is evident that OHPLA-DE1 and OHPLA-DE2 exhibited very similar amino acid composition and, like other phospholipases A,,

both are rich in cysteine and aspartic acid and its amidated counterpart. Analysis of the hydrolysis products of the action of purified enzymes on the synthetic phospholipid L-U - lstearoyl-2-oleoyl lecithin showed that oleic acid (18: 1) was liberated exclusively, indicating that the two phospholipase A enzymes, like all other snake venom phospholipase A enzymes, hydrolyzed the phospholipid exclusively at the 2 position (Verheij et al., 1980a). Biological properties of the pur$ed phospholipases A

Both OHPLA-DE1 and OHPLA-DE2 were not lethal to mice at a dose of 10 /*g/g by i.v. injection. This is not surprising as most other acidic snake venom phospholipases A, are either not lethal or exhibited low lethality in mice (Rosenberg, 1986). The phospholipase A, enzymes, however, exhibited moderate anticoagulant activity (Table 2). Samples of 20 mg of the enzymes prolonged recalcification time of titrated rabbit platelet poor plasma to > 1 hr. This is consistent with the current view that generally acidic phospholipase A, enzymes exhibit only weak to moderate anticoagulant activity (Verheij et al., 1980b). Table 2. Effect of king cobra venom phospholipase A, enzymes on the recalcification times of titrated rabbit olatelet-Door olasma’ Recalcification

DoseW 30 20 10 5 2.5

time

OHPLA-DE1

OHPLA-DE2

>I hr > 1 hr 36k4min 8*2min 7 + 1 min

> 1 hr >I hr 18k3min 8k2min 6+ lmin

*Value expressed as mean + SD (n = 3). The clotting mixture contained 0.2 ml titrated rabbit platelet poor plasma, 0. I ml sample in saline and 0.2ml 25mM calcium chloride. Control for recalcification time was 4 f I min.

486

NGET-HONG TAN and Table 3. Edema-inducing

M. NOMANBHAYSAIFUDDIN

activity of king cobra A, enznnes

and venom phospholipase Edema ratio (%)

Venom 10.0 7.5 5.0 2.0 I .o 0.5 0.25 Minimum

edema dose’ (jog)

‘Mean + SD (n = 3). bND, not determined. ‘Values in parentheses dose.

157*3 ND ND 152+4 140*2 136+ I 126k I 116f I 0.7 (0.5-1.0)

OHPLA-DE1 142 + 3 129+ I 122*2 III *2 ND ND ND ND 10.0 (9.2-10.8)

OHPLA-DE2 141 f I 133f3 123f I III f I ND ND ND ND 9.7 (8.7-l 1.1)

represent the 95% fiducial limits to the minimum

Both enzymes also produced an increase in wet weight of the mouse foot pad when injected intradermally (Table 3). The minimum edema doses for OHPLA-DE1 and OHPLA-DE2 were 10.0 pg (95% fiducial limits 9.2-10.8 pg) and 9.7 pg (95% fiducial limits 8.7-l 1.1 pg), respectively, while the amounts of enzyme causing an edema ratio of 120% were 7.0 and 6.8 pg, respectively for OHPLA-DE1 and OHPLA-DE2. The edema-inducing activity of the king cobra venom phospholipase A, enzymes is thus weaker than that of T. jlavoviridis and V. russelli venom phospholipase A, enzymes: the amounts of enzyme causing an edema ratio of 120% for these enzymes ranged from 0.85 to 3 pg (Vishwanath et nl., 1987, 1988). On the other hand, the minimum edema dose for king cobra venom was 0.7pg (Table 3). Thus, king cobra venom exhibited a much greater edema inducing activity than the two purified phospholipase A, enzymes, and therefore it can be concluded that phospholipase A, is not likely to be the major edema-inducing factor of king cobra venom. Chiu et al. (1989) also demonstrated that the edemainducing activity of T. mucrosquamatus venom was much greater than that of the purified phospholipase A,. However, phospholipase A, may act synergistically to potentiate the edema-inducing activity of other venom component(s). Damerau et al. (1975), e.g. have demonstrated that cobra venom phospholipase A and direct lytic factor were synergistic in histamine-releasing action, which may cause edema. Bonta et al. (1972) also reported the potentiation of direct lytic factor by phospholipase A, in the pulmonary vessel lesion phenomenon. It is well established that p-bromophenacyl bromide reacts specifically with the active-site histidine residue of phospholipase AZ and thereby inactivates the enzyme (Verheij et al., 1980a). Thus, treatment of the purified phospholipase A, enzymes with pbromophenacyl bromide for 24 hr reduced the enzymatic activity of OHPLA-DE1 and OHPLA-DE2 to 10% and