Molecular Cloning of PhospholipaseA2 from a ThaiRussell's Viper Venom..Gland cDNA Library

ISSARANG NUCHPRA YOON, M.D., Ph.D.*,SUNUTCHA SUNTRARACHUN, M.Sc.**,NARUMOL PAKMANEE, M.Se.**,SURANG NUCHPRAYOON,M.D.,Ph.D.*,

ARKHOM SAI.NGAM, M.Sc.*,JUREEPORN NOIPHROM, D.Se.**,LAWAN CHANHOME,D.V.M.**,VISITH SITPRIJA, M.D., Ph.D. **

Abstract .Snake venom contains several toxins. Russell's viper (D. russellii, RV) is a venomous

snake prevalent in northcm and central Thailand. RV bites can cause disseminated coagulation,hemolysis, and edema of the bitten limbs. To identify protein components of RV venom, we madea cDNA library from RV venom glands, and randomly sequence<!cloned cDNA. We were able toclone a cDNA encoding RV phospholipase A2 (PLA2): PLA2 is.an active enzyme found in severalspecies of snake venom worldwide. PLA2 is thought to be toxic to cell membrane, thereby, cancause local cell and tissue damage. as well as systemic effects in snake bite victims. This PLA2.cDNA clone would facilitate in vivo studies of the pathophysiology of RV bite.

Key word: Russell's Viper Venom, Phospholipase A2, Molecular Genetics

NUCHPRA YOON I, SAI.NGAM A, SUNTRARACHUN S, et alJ Med Assoc Thai 2001; 84 (Suppll): S99.S105

Snake bite is a common medical emer-gency in Thailand and Southeast Asia. Russell'sviper (Daboia russellii, RV, formerly called Viperarusselli) is a venomous snake prevalent in northernand central Thailand as well as in Myanmar, India,Sri Lanka, China, Taiwan, and Indonesia. Russell'sviper bites can cause disseminated intravascular

coagulation (DIC), hemolysis. and severe necrosisand edema of the bitten limb. People bitten by a RVwho show signs of coagulopathy must be promptlytreated with anti-venom(l).

RV is widely distributed if}East and South-east Asia. At least 5 subspecies of RV have beenrecognized based on minor differences in color and

* Department of Pediatrics, Faculty of Medicine, Chulalongom University, Bangkok 10330,** Queen Saovabha Memorial Institute, The Thai Red Cross Society, Bangkok 10330, Thailand.

,.

SlOG . I.NUCHPRAYOONet at

markings; D. russellii russellii in India. D. russelliipulchella in Sri Lanka, D. russellii siamensisinMyanmar, Thailand, and China, D. russellii formo-sensis in Taiwan, and D. russellii limitis in Indo-

nesia(2). Coagulation factor X activator protein hasbeen identified and believed to be the key compo-nent that causes DIC(3). .

Snake venom is a mixture of several pro-teins, many of which are toxins. Although a fewproteins in RV venom have been identified, mostcomponents have not been characterized. Purifiedcomponents may be useful in studying and deve-loping monoclonal antibodies for diagnosis and/ortreatment of RV bites.

To identify protein components of RVvenom, we made a cDNA library from RV venomglands, and ran~omly determined DNA sequencesof the cloned cDNAs. We identified a cDNA clone

encoding a full-length phospholipase A2, and com-pared it with the other RV PLA2.

MATERIAL AND METIlODAnimal specimens

Two adult Russell's vipers, D. russelliisiamensis, weighing 0.70 and 1.16kg, were obtainedfrom Angthong province, Thailand.

Poly (A)+ RNA isolationFour venom glands (0.21, 0.24, 0.38 and

0.40 g) were dissected from the two RVs underether anesthesia and kept immediatelyin liquid nitro-gen until use. Total RNA was isolated by TRlzol LSreagent (Gibco BRL, Grand Island, NY) accordingto manufacturer's recommendation. A yield of 800f.1gof total RNA was then purified for poly (A)+RNA by PolyAT Tract system (Promega. Madison,WI) according to the manufacturer's recommenda-tion. A yield of poly (A)+ RNA was approximately8 f.1g.

cDNA library constructionA cDNA library was constructed using

ZAP express cDNA Synthesis Kit and ZAP expresscDNA Gigapack III Gold Cloning Kit (Stratagene,La Jolla, CA) according to the manufacturer'srecommendation. Briefly,S f.1gof poly (A)+ RNAwas used as substrate to construct the double-stranded cDNA. The cDNA was then modified for

ligating to the vector arms. The ligated DNA was

J Med Assoc Thai June 1001

packed with packaging extract to generate the in-fectious phage particle.

cDNA identification

For excision of the phagemid, the XLI-BLUE MRF' cells were coinfected with an aliquotof the cDNA library and the ExAssist interference-resistant helper phage. The pBK-CMV excisedphagemid was plated in freshly grown XLOLRcells in LB-kanamycin agar plates (50 f.1g1ml)andallowed to grow overnight at 37.C. Plasmid DNA ...was isolated by alkaline miniprep(4) and double-digested with EcoR I and Xho I (New England Bio-labs, Beverly, MA) for 2 hours at 37.C. The pro-duct was analyzed by 1.2 per cent agarose gel elec-trophoresis. Ten randomly chosen plasmids carry-ing cDNA inserts were subjected to DNA sequenc-ing using ABI Prism 310 Genetic Analyser (Perkin-Elmer, Norwalk, CT). The sequencing was per-formed in a 5' direction using T3 sequencing pri-mer. For RVV0l2, the sequencing was performedin both directions using T3 and T7 primers.

Sequence analysisThe cDNA sequences and predicted amino

acid sequences were compared to the sequences inthe GENBANK database using BLAST program(5). The sequence alignments were performed usingCLUSTAL X program.

RESULTS

We successfullymade a cDNA library fromD. russellii siamensis venom gland. Despite ourefforts to use as few snakes as possible, it needed 4venom glands to obtain adequate RNA for libraryconstruction. These two snakes were verified by ataxonomist to be D. russellii siamensis from thesame region. The constructed cDNA library con-tained approximately 1.0 x 106 plaque-formingunits per f.1gof vector arms. Ten cDNA cloneswhich were randomly picked for a preliminaryscreening of the DNA sequence of RV cDNAlibrary contained cDNA inserts varying in lengthfrom 0.6-2.0 kb (average length =1.1 kb). The iden-tification of each clone is shown in Table I. One

clone, RVV0l2, with 610 bp in length showed astrongly significant homology to the PLA2 gene ofthose from several snake venoms. RVVOl2 pos-sesses 60 nucleotides of 5' untranslated region

'1'"

Vol..84 Suppl 1 RUSSELL'S VIPER VENOM PHOSPHOLIPASE A2 S101

Table 1.

Clone Approx. length (kb)

Identification of ten clones picked randomly from RV cDNA library.

Homologous sequences (expected value)a

RVVOO3RVVOO5RVVOO9RVVOIORVVOIIRVV0I2RVVO13RVVOl5.RVVOI6RVV020

0.91.31.11.40.80.61.31.11.02.0

(X78971)b mettalloprotease, Echis pyramidum (2.4)

(AFI55739)b axotrophin, Mus musculus (6e-44)No significant homology found

No significant homology foundNo significant homology found .. ..---

(S29299)b phospholipase A2, D. russeUii (0.0)

No significant homology found(A42972)b coagulation factor X activating enzyme heavy chain, D. russellii (0.0)(AL353819)b related to CMUOAY""' u UiUlSport protein JENI, Neurospora crassa (4:0)

(AK023676)b unnamed protein product, Homo sapiens (ge-13)

Note: a The homologous sequences represent the highest score when performed BLASTX search.b GENBANK accession number.

(UTR), an open reading frame encod 138 aminoacid residues of PLA2 including the first 48 nueleo-tides of signal peptide. and III nucleo,:des of3'-UTR. The nuc!eotide and predicted amino acidsequences of RVV0l2 are shown in Fig. 1.

The predicted 138 amino acid residues ofRVVOl2 were lOOper cent identical to a previouslyreported PLA2 of D. russellii formosellsis (GEN-BANK accession # 829299) which was isolatedfrom a Taiwan RV (Fig. 2A, fI). Comparison ofamino acid sequences of PLA2s among D. russelliisubspecies (ssp. fonnosensis, ssp. russellii and ssp.siamensis) showed a highly significant homology(Fig. 2). In addition to ft, RVVO12 is significantlyhomologous to other reported D. russellii fonno-sens;s PLA2s (96/138 = 69% identity, Fig. 2A, f2)and D. russelli; russellii (8I/121 =67% identity,Fig. 2A, rl). Partial amino acid sequences ofRVVOl2 was identical to one of six reported PLA2sof D. russellii s;amensis obtained from N-terminalsequencing (Fig. 2B, 84),

In addition to RVVO12, we identified aclone (RVVOI5) encoding part of the coagulationfactor X activating enzyme heavy chain gene. Ano-ther cDNA clone (RVVOO3)was remotely homo-logous to a metalloprotease from an African viper,Echis pyramidum. The remaining ~~venclones haveno significant homology to known snake venomcomponents (Table I).

DISCUSSION

Although DIC and renal failure are com-mon findings among all RV bites, certain clinicalfindings vary among geographical regions(2). RV.bites in India and Sri Lanka also have prominent

neurological symptoms, with ptosis and externalophthalmoplegia(l). RV bites in Myanmar oftenpresent with conjunctival edema and other signs ofvascular leakage(8). Some of these findings maybe explained by variations in phospholipase A2(PLA2) in RV venom from different subspecies.

PLA2 is an enzyme that can catalyze phos-pholipid hydrolysis to produce free fatty acid andIysophospholipid. PLA2 has been found in reptilesand mammals of several species, and has beenclassified in several groups based on molecularweight, amino acid sequence and homology, calciumdependence, and cellular localization(9). Low mole-cular weight (-14 kiloda1ton) snake venom PLA2shas been classified into two groups. Calcium-depen-dent group I PLA2s is found in cobras and kraits,while calcium-independent group 11 PLA2s isfound in vipers.

PLA2 is a major constituent of RV venom,up to 70 per cent of dry weight in some specimens(2). RV PLA2 has been isolated bioch~micanyfrom Indian D. russellii russellii and 7 isozymeshave been identified. A purified major component ofD. russelli; russellii venom PLA2, VRV-PL-VIlla,is a basic protein that causes neurotoxic symp-toms, as well as myonecrosis when intramuscularlyinjected into mice(lO). VRV-PL-VI induces onlyedema when injected in mouse foot pad(ll). VRV-PLoY also causes neurotoxic symptoms(l2). VRV-PL-Illb has an anticoagulant effect through inhibi-tion of platelet aggregation(13).Of these D. russelliirussellii PLA2s, partial protein sequence has beenavailable for VRV-PL-VIIIa(l4). Monoclonal anti-bodies against VRV-PL-VIIIa have been shown to

8102 ,. NUCHPRAYOON~ et al. J Med AssoeThai June 2001

1 ggtgcttctgaacccccttcaactctgagaaaaggctgccaactgtctggattcaggagg

M R'~r L . W

61 ATGAGGACTCTCTGGATAGTGGCCGTGTGCCTGATAGGCGTTGAAGGGAACCTTTTCCAG

F Qv A V LI c I G LG E NV

121 TTTGGGGAGATGATCTTGGAAAAGACGGGGAAAGAAGTTGTTCATTCCTACGCCATTTAC

F G E E K T

181 GGATGCTACTGCGGCTGGGGAGGCCAAGGCAGGGCACAGGACGCCACCGACCGCTGCTGC

R A D A D

M I L

G c y G G Qc G w

G K y A I yH sV VE

G Q T R c c

F V T

241 TTTGTGCACGACTGCTGTTACGGGACAGTGAATGACTGCAACCCCAAAACGGCCACCTAT

c p T TGH D c c y V N D N K A y

G I

301 TCCTACAGCTTTGAGAACGGGGATATCGTCTGCGGAGACAACGACCTGTGCCTGAGGACT

cs y s F E DN V G D c R TD N L L

361 GTTTGTGAGTGTGACAGGGCCGCGGCAATCTGCCTTGGACAGAATGTGAATACATACGAC

V c R AE c D A A I c Q T y DN V NL G

y s

421 AAAAACTATGAGTACTACTCAATCTCTCATTGCACGGAGGAGTCAGAGCAATGCTAAgtc

H c s Q *K N y E s Iy E E E cT

481 tctgcaggacgggaaaaagccctccaattacdcaattgtggttgtgctactctattattc

601 aaaaaaaaaa

541 tgaatgcaatactgagcaataaacaggtgccagctctgcactaaatcgaaaaaaaaaaaa

Fig.I. The nucleotide and predicted amino acid sequencesor clone RVVOI2.S' untraoslated region (UTR)'and 3'.UTR are shown in the lowercaseletters, the coding sequences are shown in the uppercaseletters. The uppercase letters (bold) indicate amino acid sequences. Underlined sequence is thesignal peptide,* represents stop codon.

inhibit toxicities of the whole D. russellii russellii

venom to some degree(l4) suggesting that PLA2sare important components of venom from this sub-specie.

Our clone of PLA2, 610 bp in length isthe most complete report of PLA2 cDNAs to

date. The 5' -UTR (60 bp) and 3' UTR (lll bp)of RVVO12 is longer than the report by Tsai(7)(15 bp 5'-UTR and 12 bp 3'-UTR, GENBANKaccession 4#:X68386). PLA2's cDNA-predictedamino acid sequences of RVVOl2 is homologousto all reported PLA2 protein sequences from ano-

-=,RUSSELL'S VIPER VENOM PHOSPHOLIPASE A2 S103'Vol.84 Suppl 1

(A)

1< 20RVVO12 :f1f2["1

,1ETL',nVAVCLII:;VEI}j'ILF';'f EHl,<. ' ",~,' ' '

4RTL\;7IVAVr:LIGVEGlJ1F,;.:r K;I T

"mTLWIVA VCLIGt-EGHL?' ;'?~1 Il'IL?';':f~E:ll'v

rnctIwivavcIigvegNLFQF eMI

60 1<

RWO12 :f1 :!if2 : TP["1 : i~E

qDATDRCCFVHDCCYgV

KI

1<

'

I' 1< E

liO 1<

E IFS IS

I' S SDIktGk 6 sY YGCYCGWGGqG

-l-,~~,:,-:,>:~:,,;'::;'~~QG

-;I~~i:'-"":'i:~;,":~i~~':!':~

-;1::;':'\'i:,,~;,~'::;':~Q':;

-;;:;i:"l"-:',:~~.,;'::;I:;r;:;

50505034

80 1< 100

'ISGIRt

: 100: 100: 100

84

. i:'JH'~ -;~:;':.D:Er\jl;D_~~~~~~[~~I'i:'NI='~ '::::':::::?~NGD:::'/':\:;miD

'-'NI=';' -:'::"'::F..,i~"'~NT",--',-;m!T1',~.~E'; ;>:-~:F':~I=;~~':"~"=~D;;I:CnPK a YSYSF2nG IVCGdN

RW012f1f2["1

120

:-- :138:138: 138: 121

6CECDR AA C qN NTY1K Y

(B)

82

*

8 JCOO9 :5381-251-154RVVO12

NL5QF

Fig. 2.

1<

ys S ct 3EQC

* 40, *IS"eflWAl'l!la~IKDATDRCCFVHD.GC : SO~ ~ : 22IISiGC'lCG'PGG",r. '. DATDRCCFVHDC}: : SOS' iGCYCG1>rGr.'G~"':~DATDRCCFVHr;:~~C : 49SN r'GC'L~O;'T,TGr..'dCi,&.DAT DRCCF\TH[;C;C' : 49

: SO: SOfGCYCG~GG~G~DATDRCCF\TH~~F

f GCYCG.ioiTGG e;G;-:~":DA T DRCCE' \.'H DCC

ygcycgwgg 9 datdrccfvhdcc

Comparison of D. russeUii'sPLA2s. (A) AUgmentof mature PLA2's cDNA-predicted amino acidsequences of RVVOl2,fl; D. russeUiiformosensis GENBANKaccession# 829299, U; D. russelliiformosensisGENBANKaccession# 829299,d; D.russelliirusselliiGENBANKaccession# 2392602.

(B) Alignment PLA2's cDNA-predicted amino acid sequences of RVVO12 and N-terminal sequenc-ing sequences of various D. russellii siamensis mature PLA2s superscript (» and SJCOOO9;GENBANK accession # JCOOO9.

ther RV subspecies (Fig. 2) suggesting similarfunctions. Geographical variation of clinical effects

, could be explained by some of these differencesbetween PLA2 of Thai D. russellii siamensis and

Indian D. russellii russellii. Complementary DNAsequence of D. russellii siamensis PLA2 in thisstudy has been found to be identical to a neurotoxiccDNA of Taiwan viper (D. russellii formosensis)PLA2(6). However, since D. russellii formosensisPLA2 has been reported to be neurotoxic(6), it does

not explain the lack of neurotoxicity in Thai D.russellii siamensis RV bites. It is possible thatneurotoxicity of PLA2 in mice does not predictneurotoxicity in man or D. russellii formosensisvenom neurotoxicity could be due to other proteincomponents than PLA2.

There are several PLA2 sequences obtainedfrom each RV subspecies N"terminal amino acidsequencing(7). These findings suggest that tt;ere areseveral PLA2 gene-sthat arise from gene duplication

8104 I. NUCHPRAfOON et al.

among snakes in this subspecies. Based on their par-tial primary sequences, the D. russellii siamensisPLA2 could be categorized into three groups (Fig.2a); (1)5I-I and51-2(83%identity),(II)52,5_JC0009an 53 (89-90% identity), and (m) RVV0l2 and 54(100% identity). Full-length cDNAs of group I and11 PLA2 D. russellii siamensis have not yet beenreported. Conceivably, .several isofofms of PLA2scould be derived froro.homo- and hetero-dimeriza-

tion of these PLA2 proteins, resulting in variousclinical effects. It is not yet known whether u. rus-sellii russellii from various regions will have iden-tical PLA2s sequence.

In addition to PLA2, at least 3 clones(RVVOO3,RVVOl2 and RVVOIS) are involved in

J Med AssocThai June 2001

the toxicity of the venom. Therefore, cDNA libraryconstruction is an effective way to identify andclone genes of interest from venom glands. Cloningof D. russellii siamensis PLA2 will allow expres-sion of this enzyme by recombinant technology.The recombinant protein can be potentially helpfulto u~md-ythe pathophysiologyof RV venom,andto produce an antibody against it for diagnostic ortherapeutic purposes.

n.i:~OWLEDGEMENTThe authors wish to thank the Molecular

Biology Program Research Affairs, the Faculty ofMedicine, Chulalongkom University and ThailandResearch Fund (TRF) for supporting this study.

(Received (or publication on March 21. 2001)

REFERENCES

I. Warren DA ed. WHOISEARO guideline for theclinical management of snake bites in the South-east Asian regions. Southeast Asian J Trop MedPublic Health 1999; 30 (Suppll): 5-35.Warren DA. Snake venorns in science and clini-cal medicine-I. Russell's viper. biology, venomand treatment of bites. Trans R Sac Trop MedHyg 1989; 83: 732-40.Takeya H, Nishida, S. MiYaIaT, et al. Coagula- lOotion factor X activating enzyme from RusseU'sviper venom (RVV-X)-a novel metalloproteinase.with disintegrin (platelet aggregation inhibitor)-like and c-type lectin-like domains. J BioI Chem 11.1992;267: 141~17.Maniatis T, Fritsch EF, Sambrook J. Molecularcloning: A Laboratory Manual. Cold Spring HaborLaboratory, Cold Spring Habor, NY 1982-Altschul SF, Madden TL. Schaffer AA, et al.. 12.Gapped BLAST and PSI-BLAST: a new genera-tion of protein d3t3M~ search programs. NucleicAcids Res 1997;25: 3389-02.Wang YM, Lu PI, Ho CL, et al. Characterization 13.and molecular cloning of neurotoxic phospho-lipase A2 from Taiwan viper (Vipera rwsellifonnosensis). Eur I Biochem 1992; 209: 635-41.Tsai IH, Lu PI, Su JC. Two types of Russell'sviper revealed by variation in phospholipase A2 14.from venom of the subspecies. Toxicon 1996; 34:99-09.

Myint-Lwin. Philllips RE. Tun-Pe, Warren DA,

2.

3.

4.

5.

6.

7.

8.

9.

Tin-Nu-Swe, Maung-Maung-Lay. Bites byRussell's viper (Vipera russelli siamensis) inBunna: Haemostatic, vascular, and renal distur-bances and response to treatment. Lancet 1985:1259-64. .

Dennis EA. Diversity of group types. regulation.and function of phospholipase A2. J Bioi Chem1994;269: 13057-60.Gowda TV, Schmidt I, Middlebrook JL. Primarysequence determination of the most basic myo-necrotic phospholipase A2 from the venom ofVipera russeUi.Toxicon 1994; 32: 665-73.Vishwanath BS, Kini RM, Gowda TV. Purifica-tion and panial biochemical characterization ofan cdema inducing phospholipase A2 from Viperarusselli (RusseU's viper) snake venom. Toxicon1988;26: 713-20.Kasturi S, Gowda TV. Analysis of Vipera russelli .

venom using polyclonal antibodies preparedagainst its purified toxic phospholipase A2 VRVPL-V. Biochem Int 1992; 27: 155-64.Prasad BN, Kemparaju K, Bhan KG, et al. A.platelet aggregation inhibitor phospholipase A2from RusseUi's viper (Vipera rwselli) venom:isolation and characterization. Toxicon 1996; 34:1173-85.Gowda TV, Middlebrook JL. Monoclonal anti-bodies to VRV-PL-Vllla, a basic multitoxic phos-pholipase A2 from Vipera russelli venom. Toxi-con 1994;32: 955-64.

,7Vol. 84 Suppl 1. RUSSELL'S VIPER VENOM PHOSPHOLIPASE A2 SlUS

n1.,1ftft'U~'UvJ~«1vJa!JLtI«LtJ-a~1n';D~S~f\~U~B~tiiD1-J'W~LL~1LIft"1

ijfoTn~f'1 U'lfth::tl1. W.U., th.fI.., iJ1RU 7lf~1U, 1Y1.U..,. ..=< =< .. .. ""'..':'1/1111',:UYln'(u, 1Y1.J.J., ,1W1 UDtlW1»U, 1Y1.U. , Ut]Un wnuCIJ.1Y1.U. ,

(UYl1111U,ff.W.U..', lfn~f'1 U'lfth::tl1. W.U., thfl.", ?~~1f ii~m"', W.1I., t.l1.fI.". . .. ':N

., ..a11f/Jtl

~LI).I1L'II1LthH~1:Iilw'U 1Mn'f).lYll~mf'U~;j DLlA::mfln Al~'tiD~ 1 YlO r.rilCln~LI).I1L 'II1nflDl~~D1n1'n~Dfl1~LI ;j~fi1.. .. . . .. .. ..Li1f1LaDflu.fI~u.fInu.~::~D1n1'1'U1).1'U1L1n1i1C1mnfl u1 D~~ln \ \I~~~Nl1¥i1:l).l1n).l1!1~~10Wfl Ln~~ 1fiNn~~D~~).Ifliiu.. .~lnfiD).I~~uJ.m '111 L'I:ID~::inm1')Lfln::~"( 1hPi\lilLij\lD~fi\.h::nD'U\ um'1in \ ~infl~1:I \\lm'1ftn1:l1~fi\l LnNl).l11f1

If1Auii\l~DAh~~tmJA LD~ (PLA2) Iflo PLA2 LihILD\li!lwilw1J1fil\l~~~A10Wfl~1t~n Llri::fllf1'hin\~infl~1:IfiD.I" '..1 \ ".1'.1. t; - . ".I ~ <!-~..

LDmp.fL'!JA~ ~~r.JA ~L\lmtJD\!nYll~ltJ '11).1Y1~m1mflD1m1f11~ "I 'lJD~~YI\!nju.~1L'!J1nfl m'1l1n1:l11~DO\l PLA2 D1~

';1!1 hiNl).l11rtftn1:l1W mnnlL Uf',1iiND1n11fi~ "1 ilLnfl~lnjLl).l1L 'II1nfl1fi

il1infrt)/ : ¥i~LI).I1L'II1.~DAI~a!luJN LD~. Dqjvr\lqfllAflf

D«Tl4t\ 'f'f\h=a'2. D1~ 'l«~ If\rIf1 Cl'rnTl1U. UL'I::flN::

"MUn!lLMflm4U'WYlI"I2544; 84 (QW~Lfht 1): 599-5105

. tnAi'lnI')~1'L111f'11i{f1{ F1tU::uWY1tJf'{li{"f ~mA~n'rU~'I11iY1tncitJ.

.. ~m\lLi{11111. i{11101'1nfl1Y1tJ. nl~'Y1W 'I 10330