0041-0101/90 53 .00 + .00 © 1990 Pergamon Prtaa pk
Tosieon Vol. 28, No . 3, pp . 319-327, 1990. Printed in Great Britain .
PURIFICATION AND PROPERTIES OF SEVERAL PHOSPHOLIPASES AZ FROM THE VENOM OF AUSTRALIAN KING BROWN SNAKE (PSEUDECHIS AUSTRALIS) CHIICAHISA TAICASAKI, JrN Suzurcr
and
Nosuo TAMIYA
Department of Chemistry, Faculty of Science, Tohoku University, Aobayama, Sentiai 980, Japan (Accepted for publication l5 August 1989)
C. TAI{ASAKr, J. Suzuxr and N. TAMIYA . Purification and properties of several phospholipases A2 from the venom of Australian king brown snake (Pseudechis australis). Toxicon 28, 319-327, 1990 .-Thirteen isoenzymes of phospholipases A2 were purified from the venom of Australian king brown snake, Pseudechis australis. They (except phospholipase A2 Pa-9C) showed normal properties of snake venom phospholipases A2; the apparent mol. wts were about 13,000, the optimum pH values were around 8, calcium ion was indispensable for the enzymatic activity and the optimum calcium ion concentrations were more than 5 mM. Phospholipase A2 Pa-9C had a lag period at the initial stage of the enzymatic reaction. The enzymatic activities determined by the titration method using 1,2-dipalmitoylglycerophosphocholine as a substrate at 37°C were 10,500 units/mg for Pa-1G and 75 units/mg for Pa-13. The lethal activities measured by i.v. injections in mice were 0.09 fig/g body wt for Pa-5 and 6.8 pg/g body wt for Pa-13. The lethal activity correlates with the enzymatic activity (correlation coefficient of 0.92), and both activities showed no relationship to the basicity of the enzyme . Pa-1G is the first acidic phospholipase A2 (pI = 6.4) with high neurotoxicity (0.13 hg/g body wt). INTRODUCTION
A NUMBER of phospholipases A2 [EC 3.1 .1 .4.] have been purified from the venoms of various snakes, and some of them show presynaptic toxicity (for a review, EAKER, 1978). They vary widely on the enzymatic or lethal activities, and there is no direct relationship in the potencies of both activities . JOUBERT (1977) and KARISSON (1979) thought that the basic phospholipases A2 have more lethal and less enzymatic activities than the acidic phospholipases A2 from the comparison studies of three phospholipases A2 from Naja mossambica mossambica or several phospholipases A2 from some snake species respectively . Pseudechis australis (king brown snake), the largest venomous snake in Australia, delivers fairly large amounts of venom, which exhibit lethal activity (LF.ONARDI et al., 1979), strong phospholipases A2 activity and lesser lysophospholipase activity (TAKASAKr and TAMIYA, 1982). From the venom, a toxic and a non-toxic phospholipase A2 (named Pa-11 and Pa-13 respectively) were isolated (IVISHiDA et al., 1985a) and their amino acid sequences were determined (Nrst-itnA et al., 1985b) . Also a post-synaptically acting short319
320
C. TAKASAKI et al.
chain neurotozin, Pseudechis australis a was isolated and sequenced TAMIYA, 1985) .
(TAKASAKI
and
The present paper describes the purification and some properties of 13 iscenzymes (including Pa-11 and Pa-13) of phospholipases AZ from the venom of P. austrahs. MATERIALS AND METHODS Snake venom Desiccated P. australis venom was obtained from Australian Venom Supplies, Turramura, N.S .W ., Australia. Chemicals 1,2-Dipalmitoylglycerophosphocholine (DPPC), 1,2-dilauroylglycerophosphocholine (DLPC), 1,2-dimyristoylglycerophosphcethanolamine (PE) were the products of Calbiochem-Behring, Hoechst Corp . La Jolla, CA, U.S .A. Egg-yolk phosphatidylcholine (PC) was purchased from Wako Pure Chemical Industries, Osaka, Japan, phosphatidyl-r.-serine (PS) from bovine brain was from Tokyo Kasei Kogyo, Tokyo, Japan . CM-cellulose CM-52 and DEAE-cellulose DE-52, SP-Toyopearl, Bio-Rex 70 and Sephadex G-75 were the products of Whatman (Maidstone, Kent, U.K .), Toyo Soda Kogyo (Tokyo, Japan), BIO-RAD Laboratories (Richmond, CA, U.S .A .) and Pharmacia Fine Chemicals (Uppsala, Sweden) respectively . Purity check Purity check was carried out by ion exchange chromatography and reversed phase chromatography using a fast protein liquid chromatography (F'PLC) system (Pharmacia, Uppsala, Sweden). Ion exchange chromalography using a Mono S HR5/5 column was carried out with a linear gradient concentration of NaCI from 0 to 0.5 M in 50 mM sodium acetate buffer (pH 4.7) or in 20 mM Tris-HCl buffer (pH 8.3). Reversed phase chromatography on a PepRPC HRS/S column was carried out with a linear gradient concentration of acetonitrile in 0.1 % trifiuoroacetic acid . About 50 ,ug of each component in 200 pl distilled water was applied and the absorbance at 280 nm was monitored. Assay of phospholipase A2 activity The position of the acyl-ester bond hydrolysis was conûrmed as A~ type by using the radioactive substrate, 1pahnytoyl, 2-[1-"Cloleoylglycerophosphocholine in the same manner described previously (Twluswtcr and Flncusroro, 1989). Phospholipase A2 activity was determined by the titrimetric method. Titration was carried out with 5 mM NaOH by a pH-slat (HSM-10A ; Toa Electronics, Tokyo, Japan) using an emulsion of 4 mM DPPC, 8 mM Triton X-100, 20 mM CaCI, and 0.2 mM EDTA as a substrate at pH 8.0 and 37°C . The reaction velocities were calculated from the slope of the linear part of the curve. A triplicate run was carried out on each experiment . U the standard deviations were greater than 5%,another duplicate run was repeated. One unit of enzymatic activity was defined as the amount of the enzyme that released 1 pmole of fatty acid/min at above conditions . Isaelectric points The iscelectric points were estimated by iscelectric focusing polyacryhunide gel electrophoresis as described by Fnvt.wsox and C~ntAenucH (1971) . Gels were prepared by mixing S% (w/w) acrylamide and 2% (v/v) ampholine (pH gradient 5-8, 7-9 or 9-11). The marker proteins (pI values in parentheses) were betalactoglobulin A from bovine milk (5 .2), myoglobin from sperm whale (8 .2), cytochrome c from horse heart (10 .6), Laticauda semifasciata III (7 .2), erabutoxin c (9.1) and erabutoxin b (9 .7) from a sea krait, Ldticauda semifasciata . Other methods M, determinations by gel-filtration under physiological conditions (Twxeswrct and Tw~ave, 1982), or under denatured conditions (Twr "s x~ and Twsmre, 1985) and amino acid analysis (Twrr " c" er and TAAIIYA, 1985) were carried out as described previously. The assay of lethal activity was carried out by i.v . injection into mice (male, strain ddY, 19-21 g) which were observed for 24 hr . Four mice were used at each dose, with the doses being increased by a factor of 1.2. The r.n values were estimated according to the calculation method described by REeo and Muexcx (1938) .
PLA, in P . australis Venom
Elution
vol .
321
(I)
FIG . 1 . CM~ELLUIA6E COLUMN CHROMATOGRAPHY OF CRUDE VENOM OF P . australis . The desiccated crude venom (2 .5 g) was dissolved in 40 ml of 0.01 M sodium/potassium buffer, pH 6.5, and applied to a column (2 .6 cm x 36 cm) of CM~ellulose CM-52 equilibrated with the same buffer. The elution was carried out with the buffer (130 ml), followed by a linear concentration gradient elution from 0 to 0 .4 M NaCI in the same buffer in a total volume of 4 liter. The
protein fractions indicated by the bars, were collected separately.
, Ate; -----, [NaCI] .
RESULTS Purification of phospholipases A Z
The desiccated P. australis venom (10 g) was divided into four portions and chromatographed on a CM-cellulose CM-52 column at pH 6.5. One of the elution profiles is shown in Fig. 1 . The components in P. australis venom were separated into eighteen fractions (fractions I-XI, XIIA-XIID and XIII-XV in Fig. 1). All fractions showed phospholipase AZ activity and only fraction I showed lysophospholipase activity . Fraction I (2140 total A~ units*) was gel-filtered by passing through a Sephadex G-75 column equilibrated with 50 mM NH4HC0,. The fraction S II ( Ve/ Vt = 0.55-0.62) showed lysophospholipase activity and seems to contain lysophospholipases I and II (TAKASAKI and TA1rmtA, 1982). The fraction S III (Ve/Vt = 0.66-0.85, 1,400 total AZ ~ units) which showed phospholipase AZ activity was rechromatographed on a SP-Toyopearl column (1 .8 cm x 132 cm) equilibrated with 0.02 M sodium acetate buffer, pH 5.2 . Seven fractions (IA-IG) were obtained and the fractions ID (eluted with 0.07 M NaCI) and IG (eluted with 0.12 M NaCI) gave a single peak on the FPLC system respectively . Fraction III gave a single peak on the FPLC system . Fractions IV-VIII were separately rechromatographed on a column of CMcellulose CM-52 equilibrated with 10 mM sodium borate buffer, pH 8.2 . Only the main components from fractions V (component V-3, eluted with 0.02 M NaCI) and VIII (component VIII-2, eluted with 0.02 M NaCI) were obtained in a pure form while the other fractions were complex mixtures. The fractions IX and X were combined, desalted and rechromatographed on a CM-cellulose CM-52 column at pH 6.5. Eleven fractions (fractions IXA-IXE and XA-XF) were obtained (Fig. 2). All the fractions were pooled separately, desalted and freeze-dried . The fraction IXC was rechromatographed on a CMcellulose CM-52 column equilibrated with 10 mM sodium borate buffer, pH 8.2 (eluted with 0.06 M NaCI), followed by a Bio-Rex 70 (minus 400 mesh) column equilibrated with 50 mM ammonium acetate buffer, pH 8.8 and the component IXC-2 was obtained in a *A o unit: the amounts of proteins are expressed as optical absorption of the aqueous solution at 1 cm cell .
280
nm in a
X
0 .2-U ro
Ftc. 2. RECHROMATOGRAPHY OP FRACIION3 IX-X . The combined fractions IX-X (40ml, 3,800 total A ~ units) was applied to a column (3 .3 cm x 47 cm) of CM~ellulose CM-52 equilibrated with 0.01 M sodium phosphate buffer, pH 6.5 . The elution was tamed out with the same buffer containing 0.05 M NaCI (200 ml), followed by a linear wncentration gradient elution from 0.05 to 0.25 M NaCI in the same buffer in a total volume of 4 liter. The protein fractions (IXA-IXE and XA-XF) indicated by the bars were collected separately . , Az,n; -----, [IVaCI] .
pure state. Fractions IXE, XA, B, E and F, XI, XIIA-XIID were separately rechromatographed on a column of CM-cellulose CM-52 equilibrated with 20 mM sodium borate buffer, pH 8.6. The main components from XA (eluted with 0.14 M NaCI), XB (component XB-2, eluted with 0.04 M NaCI), XF (component XF-2, eluted with 0.15 M NaCI), XI (component XI-6, eluted with 0.14 M NaCI), XIIA (component XIIA-4, eluted with 0 .15 M NaCI), XIIB (component XIIB-3, eluted with 0.20 M NaCI) and XIIC (eluted with 0 .20 M NaCI) were pure on the FPLC system. The other fractions (IXE, XE and XIID) were complex mixtures . Fractions XIII and XV were pure on the FPLC system . The component ID (M~ 7,100) did not exhibit phospholipase AZ activity nor lethal potency. The amino acid sequence of component ID is very similar to those of long-chain neurotoxins obtained from elapid snakes, therefore, component ID seems to be a neurotoxin homologue protein which lacks the binding activity to the acetylcholine receptors because of substitutions at three invarient residues (T AKACAK I, 1989). The component XB-2 possesses lethal potency without phospholipase AZ activity . The amino acid composition, M, and co-chromatography on the FPLC system suggested that component XB-2 is similar to the short-chain neurotoxin, Pseudechis australis a (TAxnsAxi and TAMIVA, 1985). The other components which were obtained in a pure state showed phospholipase Az activity and lethal potency. The components XI-6 and XIII were phospholipases AZ Pa-11 and Pa-13 (NISI-)mA et al., 1985a) respectively . The components in fractions IG, III, V-3, VIII-2, IXC-2, XA, XF-2, XIIA-4, XIIB-3, XIIC and XV were named phospholipases AZ Pa-1G, Pa-3, Pa-5, Pa-8, Pa-9C, Pa-10A, Pa-IOF, Pa-12A, Pa-12B, Pa-12C and Pa-15 respectively . The amino acid compositions, yields, M, and the isoelectric points of these isoenzymes are shown in Table 1 . Lethal activities of phospholipases A1 The r n~ values are given in Table 1 . All the phospholipases AZ showed lethal activity in mice with haemoptysis, haemoglobinuria and limb paralysis. Pa-13 was reported as a nontoxic phospholipase AZ previously (NISHII)A et al., 1985a) because the one mouse which
PLA, in P. australir Venom
323
h
OI~l~NO~-~o0°~v~N~OO~OaMNNt~ 10 V1 N v1 V1 0~ t~i N N - ~ ~ ~ '+1 N M V1 M O
M
NooooN~e~OOOM .~O~M~0101000 t~v1N~~01riNfV~~~~fVfVt~i Vi~t~
U N
N00~0:0~01~N01o000~01~000~10 V1[~~nOO~NON~~M-~~~fJO~
pM OO N e~î
vo10 f~1
Ô
00 v1 10
v1 ~~O
~O
O
NM
r`
N
MM ^ 0~ 001~-00 M 001000101 C` MOO .-+I~O M
O
N
lnviv~t~v~ -" ONMON~bt~i-r~t~1Ny
a
N
~
â
d
O O ~ ~ ~+ N ~
10 ~ -
~
M
N
r1 N
~
Ol~ .rp"10~10:01T~-" ...01N10N00 yvivir`~^O~MONVi10e~i ..: c~i' .oo
$8-G V1
MM T01 v10101~°1M00-~°100I~ .~MN-O ~~ri1°10~O~e~i'+i .- :N~10a - ~e+iN~
Ô â
a 01 01 ~-+ .. . .- . 1p r` V -~ 01 00 O; V1 O n °: ~-+ M ~~ V1~ r`O~O(+1MOM~ V1~NMMM~
Ô
~O
â
M N
ô â âa
N
QO N
M
~
a
a ô
-10 O~V1 Mt~ -Vl p1 'r V1 0 ~-+ 00 1~ O O N1 01 0 N
1/1
~~O
Vl
~~O f~1 N
OG N
4 ô ~a
N
~O f+1
ô
N ~~O N '+i
O
â
â m F
~01[~ -+M10NNOONOONM VINN 1°~vioô~o00ri1°0~~10a - -1DMN ... ,.. --
o0 01
N ~r.NI~O1~ O-" 0101~-~NNt~r+00 oÔl~t+ilOVjO~~MO~M10e+iN~~NN
N $g01 00 MN
00
~+î N
~O M [~
h
N01oOf+110~~ON01000~'" 01t+1M-+01 [~~ef l~vi0lpMriOt~i~ ~t~i .-:t+i~if`l~
M
00-[~ M O; V1M O~ 01100100NN NM T 101~M [~ V1 OO~+iM .-;t~i'+i~ t~iN ~~ N n
a
M ~O; ~ 00 ~-+10N 0-~00 .+[~ O+M 10000 l~ 1~MOO V10ÔMMNt1~ 1ÔfV N01etN00
VI pl~
NM
N OÔ
Ô
V 3
O
00
~ NO 00
NN
[~ 1.,
0
4.rO N CO
M Ô
3 0
â m~ ~ ,, °
a
>~
~Q~rKC7â0aU>~~~FOx.~~ .~1QyF~~zG7
a
âha° .~
E
8
°a
324
C. TAKASAKI et al .
10
ô~ E à
O
m
O 9
0
E 1
"U "X
O m Ol
e
0 .1 1 10 Phospholipase A2 activity (kunitslmg)
F1G . 3. CORRELATION OF TOXICrrY AND ENZYMATIC ACTIVITY OF P. sUStratlS PH06PHOLIPASFS Az. Enzymatic activity was assayed on an emulsion of 4mM DPPC, 8 mM Triton X-100, 20 mM CaCI, and 0.2 mM EDTA at pH 8.0 and 37°C . Toxicity was defined as the reciprocal of LDm . The correlation coefficient was calculated as 0.92. a, Pa-1G; b, Pa-3 ; c, Pa-5 ; d, Pa-8; e, Pa-9C; f, Pa10A; g, Pa-IOF ; h, Pa-1l; i, Pa-12A ; j, Pa-12B ; k, Pa-12C; 1, Pa-13; m, Pa-15.
was injected with Pa-13 (7.4 Ftg/g body wt) had survived . In this study, the Pa-13 injected mice showed the same symptoms with those which were injected the other phospholipases AZ, and the Ln~ value of Pa-13 was estimated to be 6.8 ~g/g body wt. Some of the isozymes (Pa-5, Pa-l0A and Pa-11) were tested for the effects on neuromuscular transmission and muscle contractility on chick biventer cervicis and mouse diaphragm preparations. They reduced responses of both preparations to indirect stimulation in a concentration-dependent manner (RownN et al., 1989). The other phospholipases AZ obtained in a pure state caused the same symptoms on mice as those caused by Pa-1l . Enzymatic properties The position of the acyl~ster bond hydrolysis was confirmed as AZ type by using the radioactive substrate (data are not shown) . All the phospholipases AZ from P. australis showed very similar enzymatic properties although their specific activities varied by more than 100 fold. All of them had their optimum pH values around 8.0 and showed activities of more than 80% relative to the maximum velocity of each enzyme in the pH range of 7.3-8.8 . Each enzyme was assayed at various calcium ion concentrations (0-50 mM) in the presence of 0.2 mM EDTA . All of the enzymes were inactive in the absence of calcium, therefore, calcium ion was indispensable for the reaction. The apparent optimum calcium ion concentrations for the enzymatic activities except Pa-9C were around 20 mM in the presence of 0.2 mIvl EDTA, and over the range of 2 to 50 mM CaC12 concentration, they showed more than 90% of relative activities . Phospholipase AZ Pa-9C showed a lag period in its enzymatic reaction, and the lag period was elongated by higher concentration of calcium ion, therefore, the optimum calcium ion concentration for Pa-9C could not be determined . The enzymatic activities under the standard conditions are given in Table 1 .
PLA, in P. australis Venom Tea~e 2. SuasrxezE sPecuTCrrgs Substrate
P. australis
oP
A,
PHOSPHOUP~s
PH08PHOLIPII)
EYPC
325 ON 1T~ FATTY ACII) I:STER PARTS OP
DPPC
DLPC
Phospholipase A,
Ym '
Apparent f~,(mM)
V"
Apparent I~,(mM)
V ,"
Pa-5 Pa-l0A Pa-11 Pa-13
9,100 5,240 3,710 79
1 .6 2.4 4.9 1 .6
9,330 4,740 3,560 89
0.75 0.53 0.40 0.35
16,000 6,240 6,200 171
Apparent
K(mM)
0.51 0.25 0.20 0.27
Abbreviations used : EYPC, egg-yolk phosphatidylcholine; DPPC, l,2-dipalmitoyl-glycerophosphocholine; DLPC, 1,2-dilauroylglyarophosphocholine . " =Eanoles/min/mg .
Correlation of enzymatic activity and toxicity The specific enzymatic activities of P. australis phospholipases Az were plotted against the toxicity defined as the reciprocal number of LD P (Fig. 3). There is a close relationship between the two activities (correlation coeflïcient of 0.92). Substrate specificity Further investigations on the enzymatic properties were carved out for phospholipases AZ Pa-5, Pa-10A, Pa-11 and Pa-13 (Tables 2 and 3). As shown in Table 2, P. australis phospholipases A2 hydrolyzed both of saturated- (DPPC) and unsaturated- (egg yolk PC) fatty acid ester at the 2-position of phosphatidylcholine with almost the same V~ values . However, the 1~ values for the saturated-fatty acid ester substrate were smaller than those for the unsaturated fatty acid ester substrate. As compared with the longer chain saturated-fatty acid ester substrate, the shorter one is hydrolyzed with a larger V~ and almost the same ~ values. All of the P. australis phospholipases AZ hydrolyzed phosphatidylcholine (PC) better than phosphatidyl-l,-serine (PS) or phosphatidylethanolamine (PE). They did not hydrolyze 1,2-dipalmitoylphosphatidic acid.
T~Le
3.
StmsrxeTE
Substrates
sPacrnictTTes PHOSPHOLIPA .4ES A,
EYPC
Phospholipase A, Pa-5 Pa-l0A Pa-11 Pa-13
DPPC
P.
australis
DMPE
PS
oF
(units/mg) 6,820 3,380 1,890 60
7,380 4,160 2,820 75
335 363 182 11
490 425 234 4
The values are the specific activities at a substrate concentration oC 4 mM. Abbreviations used : EYPC, egg-yolk phosphatidylcholine; DMPE, 1,2-dimyristoylglycerophosphoethanolamine; PS, phosphatidyl-L-serine .
326
C. TAKASAKI et at.
DISCUSSION
Thirteen isoenzymes of phospholipase AZ were purified from the venom of the king brown snake, P. australis. When the venom was milked from five specimens separately and subjected to CM-cellulose column chromatography at pH 6.5, at least twelve protein fractions which showed phospholipase AZ activity were obtained, although the yields varied and some fractions were absent in some cases (data not shown). This observation suggested that many isoenzymes of phospholipase AZ are produced in one specimen. The presence of various isoenzymes may have protected the species from extinction by the loss of lethality of its venom by mutation . The mol . wt determinations under the physiological or denatured and reduced conditions for all the 13 isoenzymes gave similar results (Mr were about 13,000), suggesting a monomeric nature of the enzyme proteins. The mol . wts of phospholipases AZ are either 13,500 [for example, Notechis scutatus scutatus notexin (KAitLSSON et al., 1972), L . semifasciata PLA I, III and IV (YOSxmn et al., 1979), porcine pancreas phospholipase AZ (hIIEUVVENHLTIZEN et al., 1973)] or 28,000 [for example, Crotalus adamanteus a and ß (WELt.s and HAxAxAN, 1969), Trimeresurus flavoviridis PLA (Isi-nhtAxu et al., 1980)] suggesting dimerization . From the amino acid analysis, P. australis phospholipases AZ are characterized by a high content of aspartic acid, glycine, alanine, half-cystine and lysine residues, as commonly observed in phospholipase AZ molecules from other sources . Cystine residues tend to give lower values owing to destruction during hydrolysis, therefore, the numbers of half-cystine residues seem to be 14, the same as the other phospholipases AZ. In fact, Pa-11 and Pa-13 contain 14 half-cystine residues at the same positions as those of other phospholipases AZ from the venoms of elapid snakes (NISHIDA et al., 1985b). Phospholipase AZ Pa-9C showed a lag period at the initial stage of the enzymatic reaction. Similar observations were reported with L . semifasciata phospholipases AZ, LsPLA III and IV (Y06HIDA et al., 1979) and L . colubrina phospholipase A2, LcPLA-II (TAKASAIü et al., 1988). In every case, the length of the lag period was shortened, without changing the highest velocity, by the addition of the reaction product (long-chain-fatty acid). In considering the toxicity of the phospholipases AZ, KARL.SSON (1979) concluded that the basic phospholipases AZ in general appear to be toxic. JousntT (1977) also reported that in three N. mossambica mossambica venom phospholipases AZ, the most basic one (CM-III) had the most lethal and least enzymatic activity . Some theories that the toxicity domain might be the basic region have been proposed (RrroN~A and GusEN~Ex, 1985; KTHI and IWANAGA, 1986; Tsat, L-H . et al., 1987) . In P. australis phospholipases AZ, toxicity correlates with enzymic activity but not with the basicity of the protein molecules . Pa-1G is the first example of an acidic phospholipase AZ with rather high toxicity. The enzymatic activities of P. australis phospholipases AZ varied more than 100-fold from each other, but showed very similar substrate specificities . All of the P. australis enzymes hydrolyze the phospholipid substrates in the order of DPPC > egg yolk PC » PE ~ PS (Table 3). The ~ values of P. australis enzymes are very similar to each other with each substrate, and the potencies of the enzymatic activities depend on the V~ values . L . semifasciata LsPLA I hydrolyzed egg yolk PC » DPPC PE and did not hydrolyze PS (TAKASAKI, unpublished work). The dü%rence of the substrate preference of these enzymes may come from the diti"'erence of the substrate recognition site structure of
PLA, in P. australis Venom
327
the enzyme molecules. The amino acid sequences of phospholipases A2 Pa-1G, Pa-3, Pa5, Pa-9C, Pa-10A, Pa-12A, Pa-12C and Pa15 will be reported in the next paper ~TAKASAKI et al., 1990) . Acknowledgements-We are grateful
to Professor K. Ooux.+ of Tohoku University, Sendai, Japan for the use of the Pharmacia fast protein liquid chromatography system. We thank Mr H. AsE for the amino acid analysis . REFERENCES E~tc©t, D. (1978) Studies of presynaptically newotoxic phospholipases A,. In: Versatility ojProteins pp. 41331 (L~, C. H., Ed.) New York: Academic Press. Fnvr.nYSON, G. R. and Ctnw~ecx, A. (1971) Isoelectric focusing in polyacrylamide gel and its preparative application. Analyt . Biochem. 40, 292-311. Lsfn~+xu, K., KntNU, H. and OHNO, M. (1980) Purification and properties of phospholipase A from venom of Trimeresurus flavoviridis (habu snake) . J. Biochem., Tokyo 88, 443-451. Jouesar, F. J. (1977) Naja mossambica mossambica venom. Purification, some properties and the amino acid sequences of three phospholipases A (CM-I, CM-II and CM-III). Biochem. biophys. Acra 493, 216-227. K~ttrssox, E. (1979) Chemistry of protein toxin in snake venom. In : Handbook ojExperimental Pharmacology Vol. 52, pp . 159-212 (La, C. Y., Ed .) Berlin: Springer. Knxtssorr, E., EaxEx, D. and RYni nr, L. (1972) Purification of a presynaptic newotoxin from the venom of the Australian tiger snake Notechis scutatus scutatus . Toxicon 10, 405-413. K~rr~, R. M. and IWANAGA, S. (1986) Structure-function relationships of phospholipases. Toxicon 24, 527-541 . Leoxexnt, T. M., Hownn~, M. E. H. and SPENCE, I. (1979) A lethal myotoxin isolated from the venom of the Australian king brown snake (Pseudechis australis) . Toxicon 17, 549-555. N~uvvErrHUtzex, W., S~OecH, P. and DE HMS, G. H. (1973) The isolation and properties of two prephospholipases A, from porcine pancreas . Eur. J. Biochem. 40, 1-7. Ntstnne, S., TIIUSr~fe, M ., S~zu, T., Tetusnrct, C. and T~tve, N. (19ß5a) Isolation and properties of two phospholipases A, from the venom of an Australian elapid snake (Pseudechis australis). Toxicon 23, 73-85. NL~e, S., TEnasFmw+, M. and TMaYe, N. (19ß5b) Amino acid sequences of phospholipases A, from the venom of an Australian elapid snake (Pseudechis australis). Toxicon 23, 87-104. Ran, L. J. and MuENCx, H. (1938) A simple method of estimating fifty per cent end points. Am . J. Hygiene 27, 493-497. ROWAN, E. G., Hnnv~, A. L., Tetc~xt, C. and T~~A, N. (1989) Neuromuscular effects of three phospholipases A, from the venom of the Australian king brown snake Pseudechis australis. Toxicon 27, 551-560. Rrrown, A. and GuaEx~K, F. (1985) Ammodytoxin A, a highly lethal phospholipase A, from Yipera ammodytes ammodytes venom. Biochem. biophys. Acta 828, 306-312. T,+tusera, C. (1989) Amino acid sequence of a long~hain ewotoxin homologue, Pa ID, from the venom of an Australian elapid snake, Pseudechis australis. J. Biochem., Tokyo 106, 11-16. TAY " GYt , C . and FutcuAro~ro, M. (1989) Phospholipases B from the Japanese yellow hornet (I'espa xanthoptera) venom. Toxicon 27, 449-458. T~rus~¢t, C. and Tesrt3r~, N. (1982) Isolation and properties of lysophospholipases from the venom of an Australian elapid snake, Pseudechis australis. Biochem. J. 203, 269-276. TARASAKI, C. and T~tnre, N. (1985) Isolation and amino acid sequence of a short~hain neurotoxin from an Australian elapid snake, Pseudechis australis. Biochem. J. 232, 367-371. T~rusern, C., Kmrvxe, S., Koxusrnv, Y. and T~Ye, N. (1988) Isolation, properties and amino acid sequences of a phospholipase A, and its homologue without activity from the venom of a sea snake, Lrtticauda colubrina, from the Solomon Islands. Biochem. J. 253, 869-875. Tex.~sero, C., Yurtw~, F. and Ke.ttvesxtrct, T. (1990) Amino acid sequences of eight phospholipases A, from the venom of Australian king brown snake, Pseudechis australis. Toxicon 28, 329-339. Tsar, L-H., Ltu, H.-C. and Ct~tANG, T. (1987) Toxicity domain in presynaptically toxic phospholipase A, of snake venom. Biochem . biophys. Acta 916, 94-99. WEr,ts, M. A. and HeNexaN, D. J. (1969) Studies on phospholipase A. I. Isolation and characterization of two enzymes from Crotalus adamanteus venom. Biochemistry 8, 41424. Yosrnne, H., Kuno, T., Stmvrcet, W. and T~v~, N. (1979) Phospholipase A of sea snake Laticauda semijasciata venom. J. Biochem., Tokyo 85, 379-388.
Tosieon Vol. 28, No . 3, pp . 319-327, 1990. Printed in Great Britain .
PURIFICATION AND PROPERTIES OF SEVERAL PHOSPHOLIPASES AZ FROM THE VENOM OF AUSTRALIAN KING BROWN SNAKE (PSEUDECHIS AUSTRALIS) CHIICAHISA TAICASAKI, JrN Suzurcr
and
Nosuo TAMIYA
Department of Chemistry, Faculty of Science, Tohoku University, Aobayama, Sentiai 980, Japan (Accepted for publication l5 August 1989)
C. TAI{ASAKr, J. Suzuxr and N. TAMIYA . Purification and properties of several phospholipases A2 from the venom of Australian king brown snake (Pseudechis australis). Toxicon 28, 319-327, 1990 .-Thirteen isoenzymes of phospholipases A2 were purified from the venom of Australian king brown snake, Pseudechis australis. They (except phospholipase A2 Pa-9C) showed normal properties of snake venom phospholipases A2; the apparent mol. wts were about 13,000, the optimum pH values were around 8, calcium ion was indispensable for the enzymatic activity and the optimum calcium ion concentrations were more than 5 mM. Phospholipase A2 Pa-9C had a lag period at the initial stage of the enzymatic reaction. The enzymatic activities determined by the titration method using 1,2-dipalmitoylglycerophosphocholine as a substrate at 37°C were 10,500 units/mg for Pa-1G and 75 units/mg for Pa-13. The lethal activities measured by i.v. injections in mice were 0.09 fig/g body wt for Pa-5 and 6.8 pg/g body wt for Pa-13. The lethal activity correlates with the enzymatic activity (correlation coefficient of 0.92), and both activities showed no relationship to the basicity of the enzyme . Pa-1G is the first acidic phospholipase A2 (pI = 6.4) with high neurotoxicity (0.13 hg/g body wt). INTRODUCTION
A NUMBER of phospholipases A2 [EC 3.1 .1 .4.] have been purified from the venoms of various snakes, and some of them show presynaptic toxicity (for a review, EAKER, 1978). They vary widely on the enzymatic or lethal activities, and there is no direct relationship in the potencies of both activities . JOUBERT (1977) and KARISSON (1979) thought that the basic phospholipases A2 have more lethal and less enzymatic activities than the acidic phospholipases A2 from the comparison studies of three phospholipases A2 from Naja mossambica mossambica or several phospholipases A2 from some snake species respectively . Pseudechis australis (king brown snake), the largest venomous snake in Australia, delivers fairly large amounts of venom, which exhibit lethal activity (LF.ONARDI et al., 1979), strong phospholipases A2 activity and lesser lysophospholipase activity (TAKASAKr and TAMIYA, 1982). From the venom, a toxic and a non-toxic phospholipase A2 (named Pa-11 and Pa-13 respectively) were isolated (IVISHiDA et al., 1985a) and their amino acid sequences were determined (Nrst-itnA et al., 1985b) . Also a post-synaptically acting short319
320
C. TAKASAKI et al.
chain neurotozin, Pseudechis australis a was isolated and sequenced TAMIYA, 1985) .
(TAKASAKI
and
The present paper describes the purification and some properties of 13 iscenzymes (including Pa-11 and Pa-13) of phospholipases AZ from the venom of P. austrahs. MATERIALS AND METHODS Snake venom Desiccated P. australis venom was obtained from Australian Venom Supplies, Turramura, N.S .W ., Australia. Chemicals 1,2-Dipalmitoylglycerophosphocholine (DPPC), 1,2-dilauroylglycerophosphocholine (DLPC), 1,2-dimyristoylglycerophosphcethanolamine (PE) were the products of Calbiochem-Behring, Hoechst Corp . La Jolla, CA, U.S .A. Egg-yolk phosphatidylcholine (PC) was purchased from Wako Pure Chemical Industries, Osaka, Japan, phosphatidyl-r.-serine (PS) from bovine brain was from Tokyo Kasei Kogyo, Tokyo, Japan . CM-cellulose CM-52 and DEAE-cellulose DE-52, SP-Toyopearl, Bio-Rex 70 and Sephadex G-75 were the products of Whatman (Maidstone, Kent, U.K .), Toyo Soda Kogyo (Tokyo, Japan), BIO-RAD Laboratories (Richmond, CA, U.S .A .) and Pharmacia Fine Chemicals (Uppsala, Sweden) respectively . Purity check Purity check was carried out by ion exchange chromatography and reversed phase chromatography using a fast protein liquid chromatography (F'PLC) system (Pharmacia, Uppsala, Sweden). Ion exchange chromalography using a Mono S HR5/5 column was carried out with a linear gradient concentration of NaCI from 0 to 0.5 M in 50 mM sodium acetate buffer (pH 4.7) or in 20 mM Tris-HCl buffer (pH 8.3). Reversed phase chromatography on a PepRPC HRS/S column was carried out with a linear gradient concentration of acetonitrile in 0.1 % trifiuoroacetic acid . About 50 ,ug of each component in 200 pl distilled water was applied and the absorbance at 280 nm was monitored. Assay of phospholipase A2 activity The position of the acyl-ester bond hydrolysis was conûrmed as A~ type by using the radioactive substrate, 1pahnytoyl, 2-[1-"Cloleoylglycerophosphocholine in the same manner described previously (Twluswtcr and Flncusroro, 1989). Phospholipase A2 activity was determined by the titrimetric method. Titration was carried out with 5 mM NaOH by a pH-slat (HSM-10A ; Toa Electronics, Tokyo, Japan) using an emulsion of 4 mM DPPC, 8 mM Triton X-100, 20 mM CaCI, and 0.2 mM EDTA as a substrate at pH 8.0 and 37°C . The reaction velocities were calculated from the slope of the linear part of the curve. A triplicate run was carried out on each experiment . U the standard deviations were greater than 5%,another duplicate run was repeated. One unit of enzymatic activity was defined as the amount of the enzyme that released 1 pmole of fatty acid/min at above conditions . Isaelectric points The iscelectric points were estimated by iscelectric focusing polyacryhunide gel electrophoresis as described by Fnvt.wsox and C~ntAenucH (1971) . Gels were prepared by mixing S% (w/w) acrylamide and 2% (v/v) ampholine (pH gradient 5-8, 7-9 or 9-11). The marker proteins (pI values in parentheses) were betalactoglobulin A from bovine milk (5 .2), myoglobin from sperm whale (8 .2), cytochrome c from horse heart (10 .6), Laticauda semifasciata III (7 .2), erabutoxin c (9.1) and erabutoxin b (9 .7) from a sea krait, Ldticauda semifasciata . Other methods M, determinations by gel-filtration under physiological conditions (Twxeswrct and Tw~ave, 1982), or under denatured conditions (Twr "s x~ and Twsmre, 1985) and amino acid analysis (Twrr " c" er and TAAIIYA, 1985) were carried out as described previously. The assay of lethal activity was carried out by i.v . injection into mice (male, strain ddY, 19-21 g) which were observed for 24 hr . Four mice were used at each dose, with the doses being increased by a factor of 1.2. The r.n values were estimated according to the calculation method described by REeo and Muexcx (1938) .
PLA, in P . australis Venom
Elution
vol .
321
(I)
FIG . 1 . CM~ELLUIA6E COLUMN CHROMATOGRAPHY OF CRUDE VENOM OF P . australis . The desiccated crude venom (2 .5 g) was dissolved in 40 ml of 0.01 M sodium/potassium buffer, pH 6.5, and applied to a column (2 .6 cm x 36 cm) of CM~ellulose CM-52 equilibrated with the same buffer. The elution was carried out with the buffer (130 ml), followed by a linear concentration gradient elution from 0 to 0 .4 M NaCI in the same buffer in a total volume of 4 liter. The
protein fractions indicated by the bars, were collected separately.
, Ate; -----, [NaCI] .
RESULTS Purification of phospholipases A Z
The desiccated P. australis venom (10 g) was divided into four portions and chromatographed on a CM-cellulose CM-52 column at pH 6.5. One of the elution profiles is shown in Fig. 1 . The components in P. australis venom were separated into eighteen fractions (fractions I-XI, XIIA-XIID and XIII-XV in Fig. 1). All fractions showed phospholipase AZ activity and only fraction I showed lysophospholipase activity . Fraction I (2140 total A~ units*) was gel-filtered by passing through a Sephadex G-75 column equilibrated with 50 mM NH4HC0,. The fraction S II ( Ve/ Vt = 0.55-0.62) showed lysophospholipase activity and seems to contain lysophospholipases I and II (TAKASAKI and TA1rmtA, 1982). The fraction S III (Ve/Vt = 0.66-0.85, 1,400 total AZ ~ units) which showed phospholipase AZ activity was rechromatographed on a SP-Toyopearl column (1 .8 cm x 132 cm) equilibrated with 0.02 M sodium acetate buffer, pH 5.2 . Seven fractions (IA-IG) were obtained and the fractions ID (eluted with 0.07 M NaCI) and IG (eluted with 0.12 M NaCI) gave a single peak on the FPLC system respectively . Fraction III gave a single peak on the FPLC system . Fractions IV-VIII were separately rechromatographed on a column of CMcellulose CM-52 equilibrated with 10 mM sodium borate buffer, pH 8.2 . Only the main components from fractions V (component V-3, eluted with 0.02 M NaCI) and VIII (component VIII-2, eluted with 0.02 M NaCI) were obtained in a pure form while the other fractions were complex mixtures. The fractions IX and X were combined, desalted and rechromatographed on a CM-cellulose CM-52 column at pH 6.5. Eleven fractions (fractions IXA-IXE and XA-XF) were obtained (Fig. 2). All the fractions were pooled separately, desalted and freeze-dried . The fraction IXC was rechromatographed on a CMcellulose CM-52 column equilibrated with 10 mM sodium borate buffer, pH 8.2 (eluted with 0.06 M NaCI), followed by a Bio-Rex 70 (minus 400 mesh) column equilibrated with 50 mM ammonium acetate buffer, pH 8.8 and the component IXC-2 was obtained in a *A o unit: the amounts of proteins are expressed as optical absorption of the aqueous solution at 1 cm cell .
280
nm in a
X
0 .2-U ro
Ftc. 2. RECHROMATOGRAPHY OP FRACIION3 IX-X . The combined fractions IX-X (40ml, 3,800 total A ~ units) was applied to a column (3 .3 cm x 47 cm) of CM~ellulose CM-52 equilibrated with 0.01 M sodium phosphate buffer, pH 6.5 . The elution was tamed out with the same buffer containing 0.05 M NaCI (200 ml), followed by a linear wncentration gradient elution from 0.05 to 0.25 M NaCI in the same buffer in a total volume of 4 liter. The protein fractions (IXA-IXE and XA-XF) indicated by the bars were collected separately . , Az,n; -----, [IVaCI] .
pure state. Fractions IXE, XA, B, E and F, XI, XIIA-XIID were separately rechromatographed on a column of CM-cellulose CM-52 equilibrated with 20 mM sodium borate buffer, pH 8.6. The main components from XA (eluted with 0.14 M NaCI), XB (component XB-2, eluted with 0.04 M NaCI), XF (component XF-2, eluted with 0.15 M NaCI), XI (component XI-6, eluted with 0.14 M NaCI), XIIA (component XIIA-4, eluted with 0 .15 M NaCI), XIIB (component XIIB-3, eluted with 0.20 M NaCI) and XIIC (eluted with 0 .20 M NaCI) were pure on the FPLC system. The other fractions (IXE, XE and XIID) were complex mixtures . Fractions XIII and XV were pure on the FPLC system . The component ID (M~ 7,100) did not exhibit phospholipase AZ activity nor lethal potency. The amino acid sequence of component ID is very similar to those of long-chain neurotoxins obtained from elapid snakes, therefore, component ID seems to be a neurotoxin homologue protein which lacks the binding activity to the acetylcholine receptors because of substitutions at three invarient residues (T AKACAK I, 1989). The component XB-2 possesses lethal potency without phospholipase AZ activity . The amino acid composition, M, and co-chromatography on the FPLC system suggested that component XB-2 is similar to the short-chain neurotoxin, Pseudechis australis a (TAxnsAxi and TAMIVA, 1985). The other components which were obtained in a pure state showed phospholipase Az activity and lethal potency. The components XI-6 and XIII were phospholipases AZ Pa-11 and Pa-13 (NISI-)mA et al., 1985a) respectively . The components in fractions IG, III, V-3, VIII-2, IXC-2, XA, XF-2, XIIA-4, XIIB-3, XIIC and XV were named phospholipases AZ Pa-1G, Pa-3, Pa-5, Pa-8, Pa-9C, Pa-10A, Pa-IOF, Pa-12A, Pa-12B, Pa-12C and Pa-15 respectively . The amino acid compositions, yields, M, and the isoelectric points of these isoenzymes are shown in Table 1 . Lethal activities of phospholipases A1 The r n~ values are given in Table 1 . All the phospholipases AZ showed lethal activity in mice with haemoptysis, haemoglobinuria and limb paralysis. Pa-13 was reported as a nontoxic phospholipase AZ previously (NISHII)A et al., 1985a) because the one mouse which
PLA, in P. australir Venom
323
h
OI~l~NO~-~o0°~v~N~OO~OaMNNt~ 10 V1 N v1 V1 0~ t~i N N - ~ ~ ~ '+1 N M V1 M O
M
NooooN~e~OOOM .~O~M~0101000 t~v1N~~01riNfV~~~~fVfVt~i Vi~t~
U N
N00~0:0~01~N01o000~01~000~10 V1[~~nOO~NON~~M-~~~fJO~
pM OO N e~î
vo10 f~1
Ô
00 v1 10
v1 ~~O
~O
O
NM
r`
N
MM ^ 0~ 001~-00 M 001000101 C` MOO .-+I~O M
O
N
lnviv~t~v~ -" ONMON~bt~i-r~t~1Ny
a
N
~
â
d
O O ~ ~ ~+ N ~
10 ~ -
~
M
N
r1 N
~
Ol~ .rp"10~10:01T~-" ...01N10N00 yvivir`~^O~MONVi10e~i ..: c~i' .oo
$8-G V1
MM T01 v10101~°1M00-~°100I~ .~MN-O ~~ri1°10~O~e~i'+i .- :N~10a - ~e+iN~
Ô â
a 01 01 ~-+ .. . .- . 1p r` V -~ 01 00 O; V1 O n °: ~-+ M ~~ V1~ r`O~O(+1MOM~ V1~NMMM~
Ô
~O
â
M N
ô â âa
N
QO N
M
~
a
a ô
-10 O~V1 Mt~ -Vl p1 'r V1 0 ~-+ 00 1~ O O N1 01 0 N
1/1
~~O
Vl
~~O f~1 N
OG N
4 ô ~a
N
~O f+1
ô
N ~~O N '+i
O
â
â m F
~01[~ -+M10NNOONOONM VINN 1°~vioô~o00ri1°0~~10a - -1DMN ... ,.. --
o0 01
N ~r.NI~O1~ O-" 0101~-~NNt~r+00 oÔl~t+ilOVjO~~MO~M10e+iN~~NN
N $g01 00 MN
00
~+î N
~O M [~
h
N01oOf+110~~ON01000~'" 01t+1M-+01 [~~ef l~vi0lpMriOt~i~ ~t~i .-:t+i~if`l~
M
00-[~ M O; V1M O~ 01100100NN NM T 101~M [~ V1 OO~+iM .-;t~i'+i~ t~iN ~~ N n
a
M ~O; ~ 00 ~-+10N 0-~00 .+[~ O+M 10000 l~ 1~MOO V10ÔMMNt1~ 1ÔfV N01etN00
VI pl~
NM
N OÔ
Ô
V 3
O
00
~ NO 00
NN
[~ 1.,
0
4.rO N CO
M Ô
3 0
â m~ ~ ,, °
a
>~
~Q~rKC7â0aU>~~~FOx.~~ .~1QyF~~zG7
a
âha° .~
E
8
°a
324
C. TAKASAKI et al .
10
ô~ E à
O
m
O 9
0
E 1
"U "X
O m Ol
e
0 .1 1 10 Phospholipase A2 activity (kunitslmg)
F1G . 3. CORRELATION OF TOXICrrY AND ENZYMATIC ACTIVITY OF P. sUStratlS PH06PHOLIPASFS Az. Enzymatic activity was assayed on an emulsion of 4mM DPPC, 8 mM Triton X-100, 20 mM CaCI, and 0.2 mM EDTA at pH 8.0 and 37°C . Toxicity was defined as the reciprocal of LDm . The correlation coefficient was calculated as 0.92. a, Pa-1G; b, Pa-3 ; c, Pa-5 ; d, Pa-8; e, Pa-9C; f, Pa10A; g, Pa-IOF ; h, Pa-1l; i, Pa-12A ; j, Pa-12B ; k, Pa-12C; 1, Pa-13; m, Pa-15.
was injected with Pa-13 (7.4 Ftg/g body wt) had survived . In this study, the Pa-13 injected mice showed the same symptoms with those which were injected the other phospholipases AZ, and the Ln~ value of Pa-13 was estimated to be 6.8 ~g/g body wt. Some of the isozymes (Pa-5, Pa-l0A and Pa-11) were tested for the effects on neuromuscular transmission and muscle contractility on chick biventer cervicis and mouse diaphragm preparations. They reduced responses of both preparations to indirect stimulation in a concentration-dependent manner (RownN et al., 1989). The other phospholipases AZ obtained in a pure state caused the same symptoms on mice as those caused by Pa-1l . Enzymatic properties The position of the acyl~ster bond hydrolysis was confirmed as AZ type by using the radioactive substrate (data are not shown) . All the phospholipases AZ from P. australis showed very similar enzymatic properties although their specific activities varied by more than 100 fold. All of them had their optimum pH values around 8.0 and showed activities of more than 80% relative to the maximum velocity of each enzyme in the pH range of 7.3-8.8 . Each enzyme was assayed at various calcium ion concentrations (0-50 mM) in the presence of 0.2 mM EDTA . All of the enzymes were inactive in the absence of calcium, therefore, calcium ion was indispensable for the reaction. The apparent optimum calcium ion concentrations for the enzymatic activities except Pa-9C were around 20 mM in the presence of 0.2 mIvl EDTA, and over the range of 2 to 50 mM CaC12 concentration, they showed more than 90% of relative activities . Phospholipase AZ Pa-9C showed a lag period in its enzymatic reaction, and the lag period was elongated by higher concentration of calcium ion, therefore, the optimum calcium ion concentration for Pa-9C could not be determined . The enzymatic activities under the standard conditions are given in Table 1 .
PLA, in P. australis Venom Tea~e 2. SuasrxezE sPecuTCrrgs Substrate
P. australis
oP
A,
PHOSPHOUP~s
PH08PHOLIPII)
EYPC
325 ON 1T~ FATTY ACII) I:STER PARTS OP
DPPC
DLPC
Phospholipase A,
Ym '
Apparent f~,(mM)
V"
Apparent I~,(mM)
V ,"
Pa-5 Pa-l0A Pa-11 Pa-13
9,100 5,240 3,710 79
1 .6 2.4 4.9 1 .6
9,330 4,740 3,560 89
0.75 0.53 0.40 0.35
16,000 6,240 6,200 171
Apparent
K(mM)
0.51 0.25 0.20 0.27
Abbreviations used : EYPC, egg-yolk phosphatidylcholine; DPPC, l,2-dipalmitoyl-glycerophosphocholine; DLPC, 1,2-dilauroylglyarophosphocholine . " =Eanoles/min/mg .
Correlation of enzymatic activity and toxicity The specific enzymatic activities of P. australis phospholipases Az were plotted against the toxicity defined as the reciprocal number of LD P (Fig. 3). There is a close relationship between the two activities (correlation coeflïcient of 0.92). Substrate specificity Further investigations on the enzymatic properties were carved out for phospholipases AZ Pa-5, Pa-10A, Pa-11 and Pa-13 (Tables 2 and 3). As shown in Table 2, P. australis phospholipases A2 hydrolyzed both of saturated- (DPPC) and unsaturated- (egg yolk PC) fatty acid ester at the 2-position of phosphatidylcholine with almost the same V~ values . However, the 1~ values for the saturated-fatty acid ester substrate were smaller than those for the unsaturated fatty acid ester substrate. As compared with the longer chain saturated-fatty acid ester substrate, the shorter one is hydrolyzed with a larger V~ and almost the same ~ values. All of the P. australis phospholipases AZ hydrolyzed phosphatidylcholine (PC) better than phosphatidyl-l,-serine (PS) or phosphatidylethanolamine (PE). They did not hydrolyze 1,2-dipalmitoylphosphatidic acid.
T~Le
3.
StmsrxeTE
Substrates
sPacrnictTTes PHOSPHOLIPA .4ES A,
EYPC
Phospholipase A, Pa-5 Pa-l0A Pa-11 Pa-13
DPPC
P.
australis
DMPE
PS
oF
(units/mg) 6,820 3,380 1,890 60
7,380 4,160 2,820 75
335 363 182 11
490 425 234 4
The values are the specific activities at a substrate concentration oC 4 mM. Abbreviations used : EYPC, egg-yolk phosphatidylcholine; DMPE, 1,2-dimyristoylglycerophosphoethanolamine; PS, phosphatidyl-L-serine .
326
C. TAKASAKI et at.
DISCUSSION
Thirteen isoenzymes of phospholipase AZ were purified from the venom of the king brown snake, P. australis. When the venom was milked from five specimens separately and subjected to CM-cellulose column chromatography at pH 6.5, at least twelve protein fractions which showed phospholipase AZ activity were obtained, although the yields varied and some fractions were absent in some cases (data not shown). This observation suggested that many isoenzymes of phospholipase AZ are produced in one specimen. The presence of various isoenzymes may have protected the species from extinction by the loss of lethality of its venom by mutation . The mol . wt determinations under the physiological or denatured and reduced conditions for all the 13 isoenzymes gave similar results (Mr were about 13,000), suggesting a monomeric nature of the enzyme proteins. The mol . wts of phospholipases AZ are either 13,500 [for example, Notechis scutatus scutatus notexin (KAitLSSON et al., 1972), L . semifasciata PLA I, III and IV (YOSxmn et al., 1979), porcine pancreas phospholipase AZ (hIIEUVVENHLTIZEN et al., 1973)] or 28,000 [for example, Crotalus adamanteus a and ß (WELt.s and HAxAxAN, 1969), Trimeresurus flavoviridis PLA (Isi-nhtAxu et al., 1980)] suggesting dimerization . From the amino acid analysis, P. australis phospholipases AZ are characterized by a high content of aspartic acid, glycine, alanine, half-cystine and lysine residues, as commonly observed in phospholipase AZ molecules from other sources . Cystine residues tend to give lower values owing to destruction during hydrolysis, therefore, the numbers of half-cystine residues seem to be 14, the same as the other phospholipases AZ. In fact, Pa-11 and Pa-13 contain 14 half-cystine residues at the same positions as those of other phospholipases AZ from the venoms of elapid snakes (NISHIDA et al., 1985b). Phospholipase AZ Pa-9C showed a lag period at the initial stage of the enzymatic reaction. Similar observations were reported with L . semifasciata phospholipases AZ, LsPLA III and IV (Y06HIDA et al., 1979) and L . colubrina phospholipase A2, LcPLA-II (TAKASAIü et al., 1988). In every case, the length of the lag period was shortened, without changing the highest velocity, by the addition of the reaction product (long-chain-fatty acid). In considering the toxicity of the phospholipases AZ, KARL.SSON (1979) concluded that the basic phospholipases AZ in general appear to be toxic. JousntT (1977) also reported that in three N. mossambica mossambica venom phospholipases AZ, the most basic one (CM-III) had the most lethal and least enzymatic activity . Some theories that the toxicity domain might be the basic region have been proposed (RrroN~A and GusEN~Ex, 1985; KTHI and IWANAGA, 1986; Tsat, L-H . et al., 1987) . In P. australis phospholipases AZ, toxicity correlates with enzymic activity but not with the basicity of the protein molecules . Pa-1G is the first example of an acidic phospholipase AZ with rather high toxicity. The enzymatic activities of P. australis phospholipases AZ varied more than 100-fold from each other, but showed very similar substrate specificities . All of the P. australis enzymes hydrolyze the phospholipid substrates in the order of DPPC > egg yolk PC » PE ~ PS (Table 3). The ~ values of P. australis enzymes are very similar to each other with each substrate, and the potencies of the enzymatic activities depend on the V~ values . L . semifasciata LsPLA I hydrolyzed egg yolk PC » DPPC PE and did not hydrolyze PS (TAKASAKI, unpublished work). The dü%rence of the substrate preference of these enzymes may come from the diti"'erence of the substrate recognition site structure of
PLA, in P. australis Venom
327
the enzyme molecules. The amino acid sequences of phospholipases A2 Pa-1G, Pa-3, Pa5, Pa-9C, Pa-10A, Pa-12A, Pa-12C and Pa15 will be reported in the next paper ~TAKASAKI et al., 1990) . Acknowledgements-We are grateful
to Professor K. Ooux.+ of Tohoku University, Sendai, Japan for the use of the Pharmacia fast protein liquid chromatography system. We thank Mr H. AsE for the amino acid analysis . REFERENCES E~tc©t, D. (1978) Studies of presynaptically newotoxic phospholipases A,. In: Versatility ojProteins pp. 41331 (L~, C. H., Ed.) New York: Academic Press. Fnvr.nYSON, G. R. and Ctnw~ecx, A. (1971) Isoelectric focusing in polyacrylamide gel and its preparative application. Analyt . Biochem. 40, 292-311. Lsfn~+xu, K., KntNU, H. and OHNO, M. (1980) Purification and properties of phospholipase A from venom of Trimeresurus flavoviridis (habu snake) . J. Biochem., Tokyo 88, 443-451. Jouesar, F. J. (1977) Naja mossambica mossambica venom. Purification, some properties and the amino acid sequences of three phospholipases A (CM-I, CM-II and CM-III). Biochem. biophys. Acra 493, 216-227. K~ttrssox, E. (1979) Chemistry of protein toxin in snake venom. In : Handbook ojExperimental Pharmacology Vol. 52, pp . 159-212 (La, C. Y., Ed .) Berlin: Springer. Knxtssorr, E., EaxEx, D. and RYni nr, L. (1972) Purification of a presynaptic newotoxin from the venom of the Australian tiger snake Notechis scutatus scutatus . Toxicon 10, 405-413. K~rr~, R. M. and IWANAGA, S. (1986) Structure-function relationships of phospholipases. Toxicon 24, 527-541 . Leoxexnt, T. M., Hownn~, M. E. H. and SPENCE, I. (1979) A lethal myotoxin isolated from the venom of the Australian king brown snake (Pseudechis australis) . Toxicon 17, 549-555. N~uvvErrHUtzex, W., S~OecH, P. and DE HMS, G. H. (1973) The isolation and properties of two prephospholipases A, from porcine pancreas . Eur. J. Biochem. 40, 1-7. Ntstnne, S., TIIUSr~fe, M ., S~zu, T., Tetusnrct, C. and T~tve, N. (19ß5a) Isolation and properties of two phospholipases A, from the venom of an Australian elapid snake (Pseudechis australis). Toxicon 23, 73-85. NL~e, S., TEnasFmw+, M. and TMaYe, N. (19ß5b) Amino acid sequences of phospholipases A, from the venom of an Australian elapid snake (Pseudechis australis). Toxicon 23, 87-104. Ran, L. J. and MuENCx, H. (1938) A simple method of estimating fifty per cent end points. Am . J. Hygiene 27, 493-497. ROWAN, E. G., Hnnv~, A. L., Tetc~xt, C. and T~~A, N. (1989) Neuromuscular effects of three phospholipases A, from the venom of the Australian king brown snake Pseudechis australis. Toxicon 27, 551-560. Rrrown, A. and GuaEx~K, F. (1985) Ammodytoxin A, a highly lethal phospholipase A, from Yipera ammodytes ammodytes venom. Biochem. biophys. Acta 828, 306-312. T,+tusera, C. (1989) Amino acid sequence of a long~hain ewotoxin homologue, Pa ID, from the venom of an Australian elapid snake, Pseudechis australis. J. Biochem., Tokyo 106, 11-16. TAY " GYt , C . and FutcuAro~ro, M. (1989) Phospholipases B from the Japanese yellow hornet (I'espa xanthoptera) venom. Toxicon 27, 449-458. T~rus~¢t, C. and Tesrt3r~, N. (1982) Isolation and properties of lysophospholipases from the venom of an Australian elapid snake, Pseudechis australis. Biochem. J. 203, 269-276. TARASAKI, C. and T~tnre, N. (1985) Isolation and amino acid sequence of a short~hain neurotoxin from an Australian elapid snake, Pseudechis australis. Biochem. J. 232, 367-371. T~rusern, C., Kmrvxe, S., Koxusrnv, Y. and T~Ye, N. (1988) Isolation, properties and amino acid sequences of a phospholipase A, and its homologue without activity from the venom of a sea snake, Lrtticauda colubrina, from the Solomon Islands. Biochem. J. 253, 869-875. Tex.~sero, C., Yurtw~, F. and Ke.ttvesxtrct, T. (1990) Amino acid sequences of eight phospholipases A, from the venom of Australian king brown snake, Pseudechis australis. Toxicon 28, 329-339. Tsar, L-H., Ltu, H.-C. and Ct~tANG, T. (1987) Toxicity domain in presynaptically toxic phospholipase A, of snake venom. Biochem . biophys. Acta 916, 94-99. WEr,ts, M. A. and HeNexaN, D. J. (1969) Studies on phospholipase A. I. Isolation and characterization of two enzymes from Crotalus adamanteus venom. Biochemistry 8, 41424. Yosrnne, H., Kuno, T., Stmvrcet, W. and T~v~, N. (1979) Phospholipase A of sea snake Laticauda semijasciata venom. J. Biochem., Tokyo 85, 379-388.
