Primary structures and partial toxicological characterization of two phospholipases A2 from Micrurus mipartitus and Micrurus dumerilii coral snake venoms

Accepted Manuscript Primary structures and partial toxicological characterization of two phospholipases A2 from Micrurus mipartitus and Micrurus dumerilii coral snake venoms Paola Rey-Suárez, Vitelbina Núñez, Mónica Saldarriaga-Córdoba, Bruno Lomonte PII:

S0300-9084(17)30061-5

DOI:

10.1016/j.biochi.2017.03.008

Reference:

BIOCHI 5165

To appear in:

Biochimie

Received Date: 7 December 2016 Accepted Date: 13 March 2017

Please cite this article as: P. Rey-Suárez, V. Núñez, M. Saldarriaga-Córdoba, B. Lomonte, Primary structures and partial toxicological characterization of two phospholipases A2 from Micrurus mipartitus and Micrurus dumerilii coral snake venoms, Biochimie (2017), doi: 10.1016/j.biochi.2017.03.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ABSTRACT Snake venom phospholipases A2 (PLA2) share high sequence identities and a conserved structural scaffold, but show important functional differences. Only a few PLA2s

RI PT

have been purified and characterized from coral snake (Micrurus spp.) venoms, and their role in envenomation remains largely unknown. In this report, we describe the isolation, sequencing and partial functional characterization of two Micrurus PLA2s: MmipPLA2

SC

from Micrurus mipartitus and MdumPLA2 from Micrurus dumerilii, two species of clinical importance in Colombia. MmipPLA2 consisted of 119 amino acid residues with a predicted

M AN U

pI of 8.4, whereas MdumPLA2 consisted of 117 residues with a pI of 5.6. Both PLA2s showed the conserved 'group I' cysteine pattern and were enzymatically active, although MdumPLA2 had higher activity. The two enzymes differed notably in their toxicity, with MmipPLA2 being highly lethal to mice and mildly myotoxic, whereas MdumPLA2 was not

TE D

lethal (up to 3 µg/g body weight) but strongly myotoxic. MdumPLA2 displayed higher anticoagulant activity than MmipPLA2 in vitro and caused more sustained edema in the mouse footpad assay. Neither of these enzymes was cytolytic to cultured skeletal muscle

EP

C2C12 myotubes. Based on their structural differences, the two enzymes were placed in separate lineages in a partial phylogeny of Micrurus venom PLA2s and this classification

AC C

agreed with their divergent biological activities. Overall, these findings highlight the structural and functional diversity of Micrurus venom PLA2.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Primary structures and partial toxicological characterization of two

2

phospholipases A2 from Micrurus mipartitus and Micrurus dumerilii coral

3

snake venoms

RI PT

1

4 5

Paola Rey-Suárez1‡, Vitelbina Núñez 1,2, Mónica Saldarriaga-Córdoba3,4, Bruno Lomonte 5

1

7

9 10

Departamento de Ciencias, Universidad Iberoamericana de Ciencias y Tecnología, Chile

4

Centro de Investigación en Recursos Naturales y Sustentabilidad, Universidad Bernardo O'Higgins, Fábrica 1990, segundo piso, Santiago, Chile. 5

12

TE D

14

Keywords: Coral snake, elapid toxins, Micrurus dumerilii, Micrurus mipartitus,

17

20 21

‡

AC C

EP

phospholipase A2, venom.

16

19

Instituto Clodomiro Picado, Facultad de Microbiología, Universidad de Costa Rica, San José, Costa Rica

13

18

Escuela de Microbiología, Universidad de Antioquia, Medellín, Colombia

3

11

15

Programa de Ofidismo y Escorpionismo;

M AN U

2

8

Address correspondence to: Paola Rey Suárez, MSc Programa de Ofidismo/Escorpionismo Universidad de Antioquia

22

Medellín, Colombia

23

E-mail: [email protected]

24

Tel. (+57) 42196649

25

SC

6

ACCEPTED MANUSCRIPT 2

26

ABSTRACT Snake venom phospholipases A2 (PLA2) share high sequence identities and a

28

conserved structural scaffold, but show important functional differences. Only a few PLA2s

29

have been purified and characterized from coral snake (Micrurus spp.) venoms, and their

30

role in envenomation remains largely unknown. In this report, we describe the isolation,

31

sequencing and partial functional characterization of two Micrurus PLA2s: MmipPLA2

32

from Micrurus mipartitus and MdumPLA2 from Micrurus dumerilii, two species of clinical

33

importance in Colombia. MmipPLA2 consisted of 119 amino acid residues with a predicted

34

pI of 8.4, whereas MdumPLA2 consisted of 117 residues with a pI of 5.6. Both PLA2s

35

showed the conserved 'group I' cysteine pattern and were enzymatically active, although

36

MdumPLA2 had higher activity. The two enzymes differed notably in their toxicity, with

37

MmipPLA2 being highly lethal to mice and mildly myotoxic, whereas MdumPLA2 was not

38

lethal (up to 3 µg/g body weight) but strongly myotoxic. MdumPLA2 displayed higher

39

anticoagulant activity than MmipPLA2 in vitro and caused more sustained edema in the

40

mouse footpad assay. Neither of these enzymes was cytolytic to cultured skeletal muscle

41

C2C12 myotubes. Based on their structural differences, the two enzymes were placed in

42

separate lineages in a partial phylogeny of Micrurus venom PLA2s and this classification

43

agreed with their divergent biological activities. Overall, these findings highlight the

44

structural and functional diversity of Micrurus venom PLA2.

46

SC

M AN U

TE D

EP

AC C

45

RI PT

27

ACCEPTED MANUSCRIPT 3

47

1. INTRODUCTION Phospholipases A2 (PLA2) are esterases that hydrolyze glycerophospholipids at the

49

sn-2 ester bond of the glycerol backbone releasing free fatty acids and lysophospholipids

50

[1]. These enzymes are widespread in nature, but have been especially well characterized in

51

mammalian tissues and in arthropod and snake venoms. In the latter, PLA2s often play key

52

roles in the immobilization, killing, and digestion of the prey [2]. These roles result from

53

their bioactivities, which include neurotoxicity (pre- or post-synaptic), myotoxicity (local or

54

systemic), cardiotoxicity, anticoagulant activity, inhibition or stimulation of platelet

55

aggregation, hypotension, and edema-formation, among others [3,4]. Amino acid sequences

56

of many snake venom PLA2s and crystal structures have been reported for some, showing

57

that they share high sequence identities and a conserved structural scaffold [1], but have

58

important differences in their functional properties [4]. On the basis of their primary

59

structures and disulfide bond patterns, PLA2s from snake venoms are classified into group

60

IA, found in the Family Elapidae, and group IIA found in the Family Viperidae [5].

TE D

M AN U

SC

RI PT

48

In the Americas, elapids are represented by coral snakes (genera Micrurus,

62

Micruroides and Leptomicrurus). The genus Micrurus, with 85 species currently accepted

63

by The Reptile Database (http://www.reptile-database.org), is by far the most diverse and

64

abundant [6]. Proteomic and transcriptomic studies of several Micrurus venoms have

65

shown that three-finger toxins (3FTxs) and PLA2s constitute the most abundant

66

components [7]. However, due to the very small amounts of venom produced by these

67

snakes, very few of their components have been isolated and characterized. PLA2s have

68

been isolated from the venoms of only six species: M. fulvius [8,9], M. dumerilii

69

carinicauda [10,11], M. nigrocinctus [12-14], M. frontalis frontalis [15], M. spixi [16], and

70

M. lemniscatus carvalhoi [17-19]. An additional PLA2 was cloned from M. corallinus

AC C

EP

61

ACCEPTED MANUSCRIPT 4

71

venom gland and characterized as a recombinant protein [20]. Thus, PLA2s present in the

72

venoms of Micrurus spp. are still largely uncharacterized. Recent proteomic studies on Micrurus venoms have revealed two divergent

74

compositional patterns [7,21]. A 'PLA2-rich' phenotype has been observed mainly in

75

species that are phylogenetically more derived, whereas a '3FTx-rich' phenotype occurs in

76

species that are phylogenetically more basal, and that probably represents the ancestral

77

venom condition. Moreover, it has been observed that 'PLA2-rich' venoms generally present

78

more complex protein composition than their '3FTx-rich' counterparts. This complexity is

79

mainly determined by an expansion in the number of PLA2 isoforms, especially those

80

associated with myotoxicity [7]. As a contribution to knowledge on the diversity of coral

81

snake PLA2s, in this report we describe the isolation, sequencing, and partial functional

82

characterization of two these enzymes: MmipPLA2 from Micrurus mipartitus, and

83

MdumPLA2 from Micrurus dumerilii, respectively. These two species are responsible for

84

the majority of coral snake bites in Colombia, and envenomings inflicted by them induce

85

mainly neurotoxic effects leading to progressive paralysis of skeletal muscle, including

86

respiratory muscles [22]. M. mipartitus and M. dumerilii produce venoms that are '3FTx-

87

rich' and 'PLA2-rich', respectively: in the venom of M. mipartitus, 3FTxs (60% of the total

88

protein content) predominate over PLA2s (30%), whereas in the venom of M. dumerilii

89

PLA2s (52%) predominate over 3FTxs (28%) [23,24]. As the most abundant PLA2s in their

90

respective venoms, MmipPLA2 and MdumPLA2 exemplify the great diversification of

91

structural and functional characteristics of toxic PLA2s within the genus Micrurus.

AC C

EP

TE D

M AN U

SC

RI PT

73

92 93

2. MATERIALS AND METHODS

94

2.1. Venoms and isolation of toxins

ACCEPTED MANUSCRIPT 5

Venoms of M. mipartitus and M. dumerilii were donated by the serpentarium of the

96

University of Antioquia, and were pools obtained from 15 adult specimens of both sexes of

97

each species, from the region of Antioquia, Colombia. PLA2s were isolated by reverse-

98

phase-HPLC as previously described [23,24]. In brief, 2 mg venom aliquots were dissolved

99

in 200 µL of 0.1% trifluoroacetic acid (TFA; solution A) and separated on a C18 column

100

(Pinnacle, 5 µm particle diameter; 250 x 4.6 mm), using a Shimadzu Prominence-20A

101

chromatograph monitored at 215 nm. Elution was performed at a flow rate of 1 mL/min by

102

applying the following gradient of solution B (acetonitrile, containing 0.1% TFA): 5% B

103

for 5 min, 5–15% B over 10 min, 15–45% B over 60 min, and 45–70% B over 12 min.

104

Major fractions previously identified as PLA2s in proteomic analyses of M. mipartitus and

105

M. dumerilii venoms were manually collected, dried by vacuum centrifugation, and stored

106

at -20°C. These two PLA2s correspond to peaks 20 (MmipPLA2) and 26 (MdumPLA2),

107

respectively, in the numbering used in proteomic studies [23,24]. Electrophoretic

108

homogeneity of MmipPLA2 and MdumPLA2 was evaluated by SDS-PAGE. After

109

reduction with 5% 2-mercaptoethanol at 100 °C for 5 min, 20 µg of each protein was

110

loaded onto a 15% gel, and run in a Mini-Protean Tetra® electrophoresis system (Bio-Rad)

111

at 150 v. Proteins were visualized by Coomassie blue R-250 staining.

113 114

SC

M AN U

TE D

EP

AC C

112

RI PT

95

2.2 ESI mass spectrometry (MS) To determine the isotope-averaged mass of MmipPLA2 and MdumPLA2, each

115

protein was dissolved in 10 µL of 0.1% formic acid in 50% acetonitrile, loaded into a metal-

116

coated capillary tip (Proxeon), and directly infused into a nano-ESI source coupled to a Q-

117

Trap® 3200 (Applied Biosystems) mass spectrometer. Ionization was performed at 1300 V

118

and spectra were acquired in positive Enhanced Multi-Charge mode, in the m/z range 500–

ACCEPTED MANUSCRIPT 6

119

1700. Charge-state and deconvolution of the ion series were analyzed with the aid of the

120

Bayesian protein reconstruction tool of BioAnalyst® v.1.5 software (ABSciex), and

121

confirmed by manual calculation.

123

RI PT

122

2.3 Amino acid sequencing

Each toxin was dissolved in 50 mM ammonium bicarbonate and subjected to

125

reduction with dithiothreitol (10 mM) and alkylation with iodoacetamide (50 mM). An

126

aliquot of each reduced-alkylated protein was subjected to N-terminal sequencing on a

127

PPSQ-33A Protein Sequencer (Shimadzu). The rest of this material was digested with

128

sequencing grade bovine trypsin (overnight) or chymotrypsin (4 h) at 37 °C, using a

129

protein: enzyme ratio of 100:1 [25]. The resulting peptides were separated by RP-HPLC on

130

a C18 column (2.1×150 mm; Phenomenex), eluted at 0.3 ml/min with a 0–70% acetonitrile

131

gradient over 40 min, and manually collected. Each peak was subjected to MS/MS

132

fragmentation for amino acid sequencing. For MALDI-TOF-TOF analyses, fragmentation

133

spectra were acquired on a Proteomics Analyzer 4800-Plus instrument (Applied

134

Biosystems) using α-cyano-hydroxycinnamic acid as matrix, at 2 kV in positive reflectron

135

mode, 500 shots/spectrum, and a laser intensity of 3000. Spectra were initially searched

136

using the Paragon® algorithm of ProteinPilot 4.0 (ABSciex) against the UniProt/SwissProt

137

database, and further interpreted manually. Peptides that did not produce well-resolved

138

fragmentation spectra using MALDI were further subjected to nESI-MS/MS by direct

139

infusion in a QTrap® 3200. Selected doubly- or triply-charged peptide ions were analyzed

140

in Enhanced Resolution mode (250 amu/s), and fragmented using the Enhanced Product

141

Ion tool, with Q0 trapping. Settings were: Q1, unit resolution; collision energy, 25–45 eV;

AC C

EP

TE D

M AN U

SC

124

ACCEPTED MANUSCRIPT 7

142

linear ion trap Q3 fill time, 250 ms; and Q3 scan rate, 1000 amu/s [26]. All resulting spectra

143

were manually interpreted with the aid of the BioAnalyst® 1.5 Manual Sequencing tool.

145 146

2.4 Phylogenetic relationships

RI PT

144

The search for homologous proteins to MmipPLA2 and MdumPLA2 was done using BLAST

[27]

(BLASTp:

search

protein

databases

using

a

protein

query,

148

http://blast.ncbi.nlm.nih.gov/Blast.cgi) and UniProt (http://www.uniprot.org/blast/). For

149

each query sequence, the sequences with the lowest E-value (value close to 0) which

150

presented the highest percentage of identity were selected. A total of eight related amino

151

acid sequences (some of them predicted from transcriptomes) of venom proteins from

152

Micrurus (M. altirostris, M. corallinus, M. dumerilii, M. fulvius, M. nigrocinctus and M.

153

tener) were selected. A Naja naja sequence (CAA45372-P15445) was used as an out-

154

group. Amino acid sequences were aligned in BioEdit version 7.0 [28] using default

155

parameters. After including gaps to maximize alignments, the final number of amino acid

156

positions was 122.

TE D

M AN U

SC

147

Phylogenetic relationships among the new PLA2s and related enzymes were

158

analyzed using the Bayesian inference implemented in MrBayes v3.0B4 [29], which is well

159

known for its ability to deal with divergent datasets. A search for the amino acid

160

substitution model that best described the evolutive process of the multiple sequence

161

alignment of the PLA2s was performed with the ProtTest software [30], using the Akaike

162

selection criterium. The command block lset rates = gamma with prset aamodelpr = fixed

163

(WAG) was used. The analysis was performed by running a minimum of 1×107 generations

164

in four chains. All post-burn-in estimates (sampled every 1000 generations) were

165

combined, and phylogeny and program parameter estimates were summarized from this

AC C

EP

157

ACCEPTED MANUSCRIPT 8

166

combined posterior distribution [31]. Nodes were considered supported if the posterior

167

probabilities were >0.95.

168

2.5 Homology modeling

RI PT

169

Sequences with highest sequence identity to MmipPLA2 and MdumPLA2 toxins

171

were searched using the Swiss-Model automated protein structure homology-modeling

172

server (http://swissmodel.expasy.org; [32]). This program was used to generate

173

corresponding three-dimensional structure models, which were superimposed and

174

examined using Swiss-PDB Viewer v.4.0 (http://www.expasy.org/spdbv; [33]). The best

175

models were obtained when using the crystal structure coordinates of Naja naja 1PSH (2.3

176

Å resolution) as a template for MmipPLA2 (63.6% amino acid sequence identity), and Naja

177

naja 1A3D (1.8 Å resolution) for MdumPLA2 (70.1% identity). Resulting models of

178

MmipPLA2 and MdumPLA2 were evaluated with ProCheck [34] and represented with

179

UCSF Chimera (https://www.cgl.ucsf.edu/chimera/).

M AN U

TE D

181

2.6 Phospholipase A2 activity

EP

180

SC

170

Enzymatic activities of MmipPLA2 and MdumPLA2 venom were assayed on the

183

synthetic monodisperse substrate 4-nitro-3-octanoyloxy-benzoic acid (4-NOBA). Various

184

amounts of each toxin dissolved in 25 µL of 10 mM Tris, 0.1 M NaCl, 10 mM CaCl2, pH

185

8.0 buffer, were mixed with 200 µL of this buffer and 25 µL of NOBA (1 mg/mL in

186

acetonitrile), and incubated at 37 °C for 60 min [35]. The reaction product was quantified at

187

405 nm in a microplate reader (Nunc). Wells containing all reagents, except the enzyme,

188

were used as a blank. All samples were assayed in triplicate wells.

189

AC C

182

ACCEPTED MANUSCRIPT 9

190

2.7 Cytotoxicity upon C2C12 myotubes Cytolytic activities of MmipPLA2 and MdumPLA2 were evaluated on the murine

192

C2C12 myogenic cell line (ATCC CRL-1772), differentiated to the myotube stage, as

193

previously described [36]. Forty micrograms of each toxin in 150 µL of assay medium

194

(Dulbecco's modified Eagle Medium supplemented with 1% fetal bovine serum) was added

195

to the myotubes in 96-well plates. After 3 h of incubation at 37 °C, culture supernatants

196

were collected and the activity of lactate dehydrogenase (LDH) released from damaged

197

cells was determined using a kinetic UV assay (LDH, Biocon). Controls for 0% and 100%

198

cytotoxicity consisted of culture medium alone, and medium containing 0.1% Triton X-

199

100, respectively. Experiments were done in triplicate.

200 201

2.8 Anticoagulant activity

M AN U

SC

RI PT

191

Various amounts of MmipPLA2 and MdumPLA2, dissolved in 50 µL PBS (0.12 M

203

NaCl, 0.04 M sodium phosphate, pH 7.2) were preincubated with 500 µL of citrated human

204

plasma for 10 min at 37 °C. Then, 100 µL of 0.25 M CaCl2 was added and clotting times

205

were recorded. All samples were assayed in triplicate, and PBS alone was used as a control.

208

EP

207

2.9 Lethality in mice

AC C

206

TE D

202

All animal experiments were done in Swiss-Webster mice (18-20 g body weight) in

209

accordance with guidelines of the Ethics Committee of Universidad de Antioquia

210

(Resolución Rectoral No.18, 084). Lethality induced by MdumPLA2 was evaluated by

211

injecting different doses (3 µg to 50 µg) in 250 µL of PBS by the intraperitoneal (i.p.)

212

route, in groups of four mice. A control group received PBS alone. Deaths were recorded

ACCEPTED MANUSCRIPT 10

213

during 48 h and the Spearman-Karmer method [37] was used to estimate the median lethal

214

dose. The lethal activity of MmipPLA2 had been reported previously [23].

215

2.10 Edematogenic activity

RI PT

216

Induction of edema was assessed by injecting 8 µg of MmipPLA2 or MdumPLA2 in

218

50 µL of PBS in the footpad of the posterior right paw of three mice. The amount of toxin

219

used in this assay was chosen based on previous experience with other venom PLA2s. The

220

same volume of sterile PBS was injected in a control group. The thickness of the footpad

221

was then measured at different time points (0.5, 1, 2, 3 and 24 h) with a low-pressure spring

222

caliper (Ultra-Cal III Sylvac) as described [38].

M AN U

SC

217

223 224

2.11 Myotoxic activity

Myotoxicity was evaluated by quantification of plasma levels of creatine kinase

226

(CK) in mice receiving an intramuscular (i.m.) injection of various doses of MmipPLA2

227

and MdumPLA2 (2, 4, or 8 µg/100 µL PBS) in the gastrocnemius muscle. A control group

228

received an identical injection of PBS alone. After 3 h (an interval chosen based on

229

previous experience with myotoxic PLA2s), a blood sample was collected from the tip of

230

the tail into heparinized capillaries, centrifuged, and the plasma was assayed for CK

231

activity using an UV kinetic assay (CK-Nac, Biocon; Brea, CA, USA) [39]. Mice were

232

killed by carbon dioxide inhalation in a gas chamber 6 h after injection, and samples of

233

injected muscles were obtained and processed for histological assessment of muscle

234

damage using formalin-fixed and hematoxylin-eosin stained sections [40].

AC C

EP

TE D

225

235 236

2.12 Statistical analyses

ACCEPTED MANUSCRIPT 11

237

Numerical results were expressed as the mean ± SD, when appropriate. The

238

significance of differences between means of two groups was assessed by Student's t-test,

239

implemented in the InStat v.3 software. Differences were considered significant for p<0.05.

RI PT

240 241

3. RESULTS

243

3.1 Isolation and determination of molecular masses of MmipPLA2 and MdumPLA2.

SC

242

Two major PLA2 fractions, obtained from RP-HPLC separation of the venoms of M.

245

mipartitus (Fig.1A) and M. dumerilii (Fig.2A) were named MmipPLA2 and MdumPLA2,

246

respectively. Their retention times under the described chromatographic conditions were

247

46.7 min (MmipPLA2) and 49.6 min (MdumPLA2). The relative abundances of MmipPLA2

248

and MdumPLA2 were approximately 10% and 14% of total venom protein, respectively, as

249

estimated by integration of their absorbance peak areas at 215 nm. The homogeneity of

250

these proteins was evaluated by SDS-PAGE, and in both cases single bands with a

251

migration at nearly 13 kDa was observed (insets in Figs.1A and 2A). By nESI-MS,

252

deconvolution of the multi-charged state m/z values of the isolated proteins yielded isotope-

253

average molecular masses (Mav) of 13,210 ± 2 Da for MmipPLA2 (Fig1B/C), and 13,288 ±

254

2 Da for MdumPLA2 (Fig.2B/C).

256 257

TE D

EP

AC C

255

M AN U

244

3.2 Amino acid sequencing of MmipPLA2 and MdumPLA2. Complete primary structures of MmipPLA2 and MdumPLA2 were determined by a

258

combination of N-terminal Edman degradation and MS/MS sequencing. MmipPLA2

259

consisted of 119 amino acid residues (Fig.3A), 31 of which were obtained by Edman

260

degradation, and the remaining 88 determined by MS/MS sequencing of peptides obtained

ACCEPTED MANUSCRIPT 12

after digestion with trypsin (m/z 1,650.0+1, 3,091.3+1, 1,329.6+1, 1,915.8+1 and 1,723.9+1) or

262

chymotrypsin (m/z 974.5+1, 949.5+1, 716.3+2, 952.5+1, 856.4+1 and 1,032.5+1). On the other

263

hand, MdumPLA2 consisted of 117 amino acid residues (Fig.3B), 31 obtained by Edman

264

degradation, and the rest determined by MS/MS sequencing of tryptic (m/z 1,882.1+1,

265

2,795.7+1, 2,236.0+1 and 1,968.9+1) and chymotryptic (m/z 2,730.4+1, 2,677.4+1, 1,977.1+1

266

and 2,557.3+1) peptides. As expected for group I PLA2s from elapid venoms, both

267

MmipPLA2 and MdumPLA2 present 14 Cys residues, all of which adhere to the strictly

268

conserved pattern that defines this enzyme group.

SC

RI PT

261

The amino acid sequences of MmipPLA2 and MdumPLA2 are aligned and

270

compared in Fig.3C, showing an identity of 65%. Sequence differences between these two

271

enzymes are also reflected by their theoretical pI values, calculated using the Compute

272

Mw/pI tool (http://web.expasy.org/compute_pi/). MmipPLA2 was predicted to be a basic

273

protein with a pI of 8.4, in contrast to MdumPLA2, which had a predicted acidic pI of 5.6.

274

The acidic and basic residues of both proteins are shown by color codes in Fig. 3C. The

275

sequences reported here for MmipPLA2 and MdumPLA2 have been deposited in the

276

UniProt Knowledgebase under the accession numbers C0HKB9 and C0HKB8,

277

respectively.

279 280

TE D

EP

AC C

278

M AN U

269

3.3 Phylogenetic analysis The amino acid sequence alignment of MmipPLA2 with related enzymes from

281

elapid snake venoms showed highest identities with PLA2s isolated, or cDNA-predicted,

282

from M. altirostris (75%, F5CPF1; 69% AED89576), M. tener (68%, AET85561) and M.

283

fulvius (67%, AAZ29512) (Fig.4). Highest identities of MdumPLA2 were observed

284

compared to PLA2s from M. fulvius (88%, A0A0F7YZM3), M. tener (87%, AET85561),

ACCEPTED MANUSCRIPT 13

285

M. corallinus (70%, Q8AXW7), M. nigrocinctus (70%, P81167) and M. dumerilii (69%,

286

MIDCA1) (Fig.4). A phylogeny of PLA2 sequences aligned in Fig. 4 was reconstructed by Bayesian

288

inference, using the Akaike selection criterion and the WAG+G amino acid substitution

289

model. These analyses recovered two lineages, the first of which (red circle in Fig.5) is well

290

supported, with a posterior probability (PP) = 1.0, and includes MmipPLA2 and two PLA2s

291

of M. altirostris. Although support for the second lineage was not strong (PP = 0.73) it was

292

clearly separated from the first and included the PLA2s of M. corallinus, M.dumPLA2, M.

293

dumerilii carinicauda (MIDCA1), and PLA2s from M. nigrocinctus, M. fulvius, and M.

294

tener, respectively (green circle in Fig.5)

M AN U

SC

RI PT

287

295 296

3.4 Homology modeling of MmPLA2 and MdPLA2

Three-dimensional structural models predicted for MmipPLA2 and MdumPLA2 are

298

shown in Fig.6. Evaluation of their stereochemical quality using Ramachandran plots in

299

ProChek showed that the MmipPLA2 model (Fig.6A) had 99% of its amino acid residues in

300

allowed regions (88.1% in most favored regions, 9.9% in additional allowed regions, and

301

1.0% in generously allowed regions). In turn, the model for MdumPLA2 (Fig.6B) had

302

100% of residues in allowed regions (87.4% in most favored regions, 11.7% in additional

303

allowed regions, and 1.0% in generously allowed regions). Superposition of the alpha-

304

carbon backbone of the models predicted for the two enzymes (Fig.6C) showed the

305

conservation of their general scaffold, with most significant differences corresponding to

306

amino acid positions 17 and 18, where MmipPLA2 has two additional amino acids (Lys17

307

and

308

between residues 28 and 36, possibly related to the change of Lys31 in MmipPLA2 to Ser31

AC C

EP

TE D

297

Pro18) compared to MdumPLA2. Other differences were observed in the region

ACCEPTED MANUSCRIPT 14

309

in MdumPLA2, and in the C-terminal region, where only a few residues were conserved

310

between both protein sequences.

311

3.5 Functional characterization of MmipPLA2 and MdumPLA2

RI PT

312

Both MmipPLA2 and MdumPLA2 are enzymatically active, as shown by their

314

ability to hydrolyze 4-NOBA, with MdumPLA2 being considerably more active than

315

MmipPLA2 (Fig.7A). This difference appears to correlate with their myotoxic potencies in

316

vivo (Fig.7B). MdumPLA2 induced a potent, dose-dependent increase of plasma CK levels

317

after i.m. injection into the gastrocnemius muscle, whereas the effect of MmipPLA2 was

318

minimal under identical conditions. In addition, MdumPLA2 induced a significantly more

319

sustained edema in the mouse footpad, compared to MmipPLA2, at 2 and 3 h after injection

320

(Fig.7C). By 24 h edema induced by both enzymes had disappeared, as footpad thickness

321

measurements returned to normal (not shown). Histological examination of the injected

322

muscle tissue corroborated the difference in myotoxicity between the two enzymes.

323

Injection of MdumPLA2 induced widespread damage to muscle fibers, as well as a profuse

324

leukocyte infiltration (Fig.8B), while injection of the same dose of MmipPLA2 resulted in

325

only occasional foci of necrotic muscle fibers and markedly less inflammatory cell

326

infiltration (Fig.8A). No signs of reddish urine were observed in mice after the injection of

327

either toxin.

M AN U

TE D

EP

AC C

328

SC

313

Despite the ability of both toxins to induce necrosis of mature skeletal muscle in

329

vivo, albeit at markedly different potencies, neither was cytolytic to the myogenic C2C12

330

myotubes in vitro (data not shown). In vitro, both enzymes showed an anticoagulant effect

331

upon human plasma, although with different potencies. The anticoagulant effect of

332

MmipPLA2 was only mild, (increasing two-fold the coagulation time (13 min) in

ACCEPTED MANUSCRIPT 15

333

comparison to that of the control (6 min), whereas MdumPLA2 was highly anticoagulant,

334

prolonging clot formation time ten-fold longer than the control (60 min). A major functional difference between MmipPLA2 and MdumPLA2 relates to their

336

lethality in mice. The former enzyme was previously reported [23] to have an i.p. LD50 of

337

0.1 µg/g body weight. In contrast, the i.p. injection of up to 50 µg/mouse (3 µg/g body

338

weight) of MdumPLA2 did not result in any evident alterations or death.

RI PT

335

340

SC

339

4. DISCUSSION

The most abundant PLA2 of M. mipartitus venom, MmipPLA2, representing nearly

342

10% of the total proteins, was isolated and partially characterized in this study. Described

343

as fraction 20 in a previous proteomic analysis of M. mipartitus venom [23], MmipPLA2 is

344

highly lethal to mice (i.p. LD50 of 0.10 µg/g), which suggests the possibility of a neurotoxic

345

action. Although this possibility has not been tested in the present study, a number of

346

PLA2s in elapid snake venoms display presynaptic activity, thus potentially playing a role

347

in the overall neurotoxicity induced by the venom. In a previous study by Renjifo et al.

348

[41], M. mipartitus whole venom showed a triphasic effect during the inhibition of indirect

349

twitches in neuromuscular preparations, characterized by a small decrease, followed by a

350

brief transient increase, and a final inhibition of twitches. This effect has previously been

351

shown to occur with other presynaptic PLA2 neurotoxins, and has been suggested to

352

involve PLA2 activity [42-45]. In the case of M. mipartitus, this was further corroborated

353

when an attenuation of the triphasic effect was observed after the PLA2 activity of venom

354

was inhibited by replacing Ca2+ with Sr2+ in the physiological salt solution [41].

AC C

EP

TE D

M AN U

341

355

In addition to its potent lethal activity in mice, MmipPLA2 displays weak

356

anticoagulant activity in vitro, and induces edema and mild myotoxicity after i.m. injection.

ACCEPTED MANUSCRIPT 16

The weak myotoxicity of MmipPLA2 is in agreement with previous studies showing the

358

negligible myotoxic effect of M. mipartitus crude venom [23]. On the other hand, the

359

relative role of MmipPLA2 in the overall neurotoxic effect of M. mipartitus venom remains

360

to be established. It has been reported that this venom induces a more relevant postsynaptic

361

blockade effect [43], in agreement with its large proportion of 3FTxs [23]. Indeed, the most

362

abundant (25%) venom protein, Mipartoxin-I, is a potently lethal 3FTx which was shown to

363

induce a clear neuromuscular blockade in both avian and mouse nerve-muscle preparations

364

through its action on the cholinergic nicotinic receptor [46].

SC

RI PT

357

The most abundant protein of M. dumerilii venom (14%) corresponds to fraction 26

366

of the chromatographic profile [24], and is here named MdumPLA2. This is the second

367

PLA2 isolated from M. dumerilii venom. Dal Belo et al. [10,11] first isolated MIDCA1, a

368

PLA2 of 120 aminoacid residues. With a theoretical molecular mass of 15,552 Da, MIDCA

369

differs from MdumPLA2 by 2.5 kDa. These two PLA2s share a high percentage of

370

conserved amino acids (69%). MIDCA1 produces a presynaptic neuromuscular blockade in

371

vertebrate nerve-muscle preparations, showing triphasic effects in mammalian preparations

372

probably caused by activation of sodium channels and complemented by the blockade of

373

nerve terminal potassium channels [10]. However, MIDCA1 has low enzymatic activity

374

and does not induce myotoxicity [10]. In contrast, MdumPLA2 is a highly myotoxic

375

enzyme, inducing skeletal muscle damage with a creatine kinase release of similar

376

magnitude to that reported for PLA2 myotoxins isolated from M. nigrocinctus [12, 14].

377

Studies on the myonecrotic effect induced by the latter PLA2 toxins indicated that the

378

sarcolemma was the first structure to be affected, with focal disruptions followed by

379

hypercontraction of myofilaments and loss of sarcoplasmic reticulum integrity as evidenced

380

by numerous small vesicles in the cellular space [12]. The venom of M. dumerilii induces a

AC C

EP

TE D

M AN U

365

ACCEPTED MANUSCRIPT 17

381

pattern of myonecrosis [24] similar to that described for the PLA2 isolated from M.

382

nigrocinctus. In spite of the high toxicity of MdumPLA2 for skeletal muscle tissue, this protein

384

did not induce cytolysis in C2C12 myotubes in vitro. In this regard, MdumPLA2 resembles

385

some myotoxic PLA2s from elapid snakes such as Notexin, from the venom of Notechis

386

scutatus, which is not cytotoxic in cultured myotubes [36]. Nevertheless, another myotoxic

387

PLA2 from the venom of M. lemniscatus carvalhoi, Lemnitoxin, has been reported to be

388

cytolytic to C2C12 myotubes [19]. Therefore, it is likely that differences in the structural

389

motifs and molecular mechanisms leading to myotoxicity exist among PLA2s found in

390

elapid venoms, a subject that warrants further study.

M AN U

SC

RI PT

383

MdumPLA2 was not lethal to mice when administered i.p. at doses as high as 50

392

µg/animal. Similar findings were reported for four PLA2s isolated from M. lemniscatus

393

carvalhoi venom (Mlx-08, Mlx-09, Mlx-11 or Mlx-12) which all lacked toxicity by

394

intravenous injection at doses comparable to those of the present study [17]. Thus,

395

MdumPLA2 displays a functional profile mainly characterized by strong myotoxicity and

396

lack of lethal effect in mice, in contrast to that of MmipPLA2, which displays minimal

397

myotoxicity but high lethality. Interestingly, MdumPLA2 presents the amino acid residues

398

Arg15, Ala100, Asn108, and a hydrophobic residue at position 109, which have been

399

proposed as a characteristic feature of group I PLA2s that display myotoxic activity [47].

400

This array of amino acids has also been found in another strongly myotoxic PLA2,

401

Lemnitoxin, isolated from the venom of M. lemniscatus [19]. Also of note, Arg15 is

402

replaced by Gly15 in MmipPLA2, a susbstitution that might be related to its very low

403

myotoxic activity.

AC C

EP

TE D

391

ACCEPTED MANUSCRIPT 18

PLA2s isolated from the venom of M. fulvius have been found to cause a rapid and

405

potent hemolysis in mice, leading to a drastic drop in hematocrit and excretion of intense

406

reddish urine, in which hemoglobinuria was demonstrated [9]. Neither of the two enzymes

407

studied here induced the excretion of reddish urine, indicating the absence of extensive

408

intravascular hemolysis and/or myoglobinuria after their injection.

RI PT

404

In conclusion, two PLA2s from Micrurus venoms were isolated and characterized,

410

showing significant differences in their primary structures and functional properties. In

411

mice, MmipPLA2 mainly displays lethal activity, probably as a consequence of presynaptic

412

toxicity, whereas MdumPLA2 mainly exerts muscle damage, in the absence of a lethal

413

effect. In agreement with their functional divergence, the two enzymes partition into

414

separate lineages in the partial phylogeny of PLA2s from Micrurus venoms reported here.

415

As discussed elsewhere [7], Micrurus venoms expressing the PLA2-predominant phenotype

416

are associated with a conspicuous myotoxic activity, while this toxic effect is absent or only

417

mild in 3FTx-predominant venoms. Phylogenetic and functional characteristics of the two

418

PLA2s described here follow this trend, since MmipPLA2 clusters with enzymes from M.

419

altirostris, a 3FTx-rich species [48], whereas MdumPLA2 groups with enzymes from

420

species with PLA2-predominant venoms known to exert myotoxicity [26,49].

422 423 424 425

M AN U

TE D

EP

AC C

421

SC

409

Conflicts of interest statement The authors declare no potential conflicts of interest regarding this manuscript.

Acknowledgements

426

The authors thank Colciencias (111556933661 and 617), University of Antioquia

427

UdeA, and Vicerrectoría de Investigación, University of Costa Rica (741-B3-760) for

ACCEPTED MANUSCRIPT 19

partial financial support of this study. The valuable advice of Prof. Emeritus Mark O.

429

Lively (Wake Forest University School of Medicine) is also gratefully acknowledged. This

430

work was performed in partial fulfilment of the Ph.D. degree of Paola Rey-Suárez at the

431

University of Antioquia, Colombia.

RI PT

428

AC C

EP

TE D

M AN U

SC

432

ACCEPTED MANUSCRIPT 20

433

REFERENCES

434

[1] R.M. Kini, Excitement ahead: structure, function and mechanism of snake venom

435

phospholipase A2 enzymes. Toxicon 42 (2003) 827-840. [2] J.B. Harris, T. Scott-Davey, Secreted phospholipases A2 of snake venoms: effects on the

437

peripheral neuromuscular system with comments on the role of phospholipases A2 in

438

disorders of the CNS and their uses in industry. Toxins 17 (2013) 2533-2571.

RI PT

436

[3] R.M. Kini, Phospholipase A2, a complex multifunctional protein puzzle, in: R.M. Kini,

440

(Ed.), Venom Phospholipase A2 Enzymes: Structure, Function and Mechanism. New

441

York, Wiley, Chichester, 1997, pp. 1-28.

443 444 445

[4] J.M. Gutiérrez, B. Lomonte, Phospholipases A2: unveiling the secrets of a functionally versatile group of snake venom toxins. Toxicon 62 (2013) 27-39.

M AN U

442

SC

439

[5] R.H. Schaloske, E.A. Dennis, The phospholipase A2 superfamily and its group numbering system. Biochim. Biophys. Acta 1761 (2006) 1246-1259.

446

[6] N.J. da Silva Jr, M.A. Buononato, D.T. Feitosa, As cobras-corais do Novo Mundo. In:

447

da Silva Jr., N. J. (Org.). As Cobras-Corais do Brasil: Biologia, Taxonomia, Venenos

448

e Envenenamentos, Editora PUC Goiás. Goiânia, GO. Brasil, 2016, pp. 46-78. [7] B. Lomonte, P. Rey-Suárez, J. Fernández, M. Sasa, D. Pla, N. Vargas, M. Bénard-Valle,

450

L. Sanz, C. Corrêa-Netto, V. Núñez, A. Alape-Girón, A. Alagón, J.M. Gutiérrez, J.J.

451

Calvete, Venoms of Micrurus coral snakes: evolutionary trends in compositional

452

patters emerging from proteomic analyses. Toxicon 122 (2016) 7-25..

TE D

449

[8] L.D. Possani, A. Alagón, P.L. Fletcher Jr., M.J. Varela, J.Z. Juliá, Purification and

454

characterization of a phospholipase A2 from the venom of the coral snake, Micrurus

455

fulvius microgalbineus (Brown and Smith). Biochem. J. 179 (1979) 603-606.

AC C

EP

453

456

[9] R. Arce-Bejarano, B. Lomonte, J.M. Gutiérrez, Intravascular hemolysis induced by the

457

venom of the Eastern coral snake, Micrurus fulvius, in a mouse model: identification

458

of directly hemolytic phospholipases A2. Toxicon 90 (2014) 26-35.

459

[10] C.A. Dal Belo, G.B. Leite, M.H. Toyama, S. Marangoni, A.P. Corrado, M.D. Fontana,

460

A. Southan, E.G. Rowan, S. Hyslop, L. Rodrigues-Simioni, Pharmacological and

461

structural characterization of a novel phospholipase A2 from Micrurus dumerilii

462

carinicauda venom. Toxicon 46 (2005) 736-750.

ACCEPTED MANUSCRIPT 21

463

[11] C.A. Dal Belo, M.H. Toyama, D.O. Toyama, S. Marangoni, F.B. Moreno, B.S.

464

Cavada, M.D. Fontana, S. Hyslop, E.M. Carneiro, A. Boschero, Determination of the

465

amino acid sequence of a new phospholipase A2 (MIDCA1) isolated from Micrurus

466

dumerilii carinicauda venom. Protein J. 24 (2005) 147-153. [12] O. Arroyo, J.P. Rosso, O. Vargas, J.M. Gutiérrez, L. Cerdas, Skeletal muscle necrosis

468

induced by a phospholipase A2 isolated from the venom of the coral snake Micrurus

469

nigrocinctus nigrocinctus. Comp. Biochem. Physiol. 87B (1987) 949-952.

RI PT

467

[13] J.P. Rosso, O. Vargas-Rosso, J.M. Gutiérrez, H. Rochat, P.E. Bougis, Characterization

471

of α-neurotoxin and phospholipase A2 activities from Micrurus venoms. Eur. J.

472

Biochem. 238 (1996) 231-239.

SC

470

[14] A. Alape-Girón, B. Stiles, J. Schmidt, M. Girón-Cortés, M. Thelestam, H. Jörnvall, T.

474

Bergman, Characterization of multiple nicotinic acetylcholine receptor-binding

475

proteins and phospholipases A2 from the venom of the coral snake Micrurus

476

nigrocinctus nigrocinctus. FEBS Lett. 380 (1996) 29-32.

M AN U

473

[15] Francis BR, da Silva Júnior NJ, Seebart C, Casais e Silva LL, Schmidt JJ, Kaiser II,

478

Toxins isolated from the venom of the Brazilian coral snake (Micrurus frontalis

479

frontalis) include hemorrhagic type phospholipases A2 and postsynaptic neurotoxins.

480

Toxicon 35 (1997), 1193-1203.

TE D

477

[16] A.L. Terra, L.S. Moreira-Dill, R. Simões-Silva, J.R. Monteiro, W.L. Cavalcante, M.

482

Gallacci, N.B, Barros, R. Nicolete, C.B. Teles, P.S. Medeiros, F.B. Zanchi, J.P.

483

Zuliani, L.A. Calderon, R.G. Stábeli, A.M. Soares, Biological characterization of the

484

Amazon coral Micrurus spixii snake venom: isolation of a new neurotoxic

485

phospholipase A2. Toxicon 103 (2015) 1-11.

EP

481

[17] D.A. Oliveira, C. Harasawa, C.S. Seibert, L.L. Casais e Silva, D.C. Pimenta, I. Lebrun,

487

M.R.L. Sandoval, Phospholipases A2 isolated from Micrurus lemniscatus coral snake

488 489

AC C

486

venom: behavioral, electroencephalographic, and neuropathological aspects. Brain Res. Bull. 75 (2008) 629-639.

490

[18] N.D. de Carvalho, R.C. Garcia, A.K. Ferreira, D.R. Batista, A.C. Cassola, D. Maria, I.

491

Lebrun, S.M. Carneiro, S.C. Afeche, T. Marcourakis, M.R.L. Sandoval, Neurotoxicity

492

of coral snake phospholipases A2 in cultured rat hippocampal neurons. Brain Res. 1552

493

(2014) 1-16.

ACCEPTED MANUSCRIPT 22

494

[19] L.L. Casais-e-Silva, C.F.P. Texeira, I. Lebrun, B. Lomonte, A. Alape-Girón, J.M.

495

Gutiérrez, Lemnitoxin, the major component of Micrurus lemniscatus coral snake

496

venom, is a myotoxic and proinflammatory phospholipase A2. Toxicol. Lett. 257

497

(2016) 60-71. [20] U.C. Oliveira, A. Assui, A.R. da Silva, J.S. de Oliveira, P.L. Ho, Cloning and

499

characterization of a basic phospholipase A2 homologue from Micrurus corallinus

500

(coral snake) venom gland. Toxicon 42 (2003) 249-255.

RI PT

498

[21] J. Fernández, N. Vargas, D. Pla, M.Sasa, P. Rey-Suárez, L. Sanz, J.M. Gutiérrez, J.J.

502

Calvete, B. Lomonte, Snake venomics of Micrurus alleni and Micrurus mosquitensis

503

from the Caribbean region of Costa Rica reveals two divergent compositional patterns

504

in New World elapids. Toxicon 107 (2015) 217-233.

M AN U

SC

501

505

[22] R. Otero-Patiño, Snake bites in Colombia. In: Gopalakrishnakone P, Faiz SMA,

506

Gnanathasan CA (eds.) Clinical Toxinology. Springer Science Business Media,

507

Dordrecht, 2014, 1-44. DOI 10.1007/978-94-007-6288-6_41-2. [23] P. Rey-Suárez, V. Núñez, J.M. Gutiérrez, B. Lomonte, Proteomic and biological

509

characterization of the venom of the redtail coral snake, Micrurus mipartitus

510

(Elapidae), from Colombia and Costa Rica. J. Proteomics 75 (2011) 655-667.

TE D

508

[24] P. Rey-Suárez, V. Núñez, J. Fernández, B. Lomonte. Integrative characterization of the

512

venom of the coral snake Micrurus dumerilii (Elapidae) from Colombia: Proteome,

513

toxicity, and cross-neutralization by antivenom. J Proteomics 136 (2016) 262-273.

514

[25] M. Van der Laat, J. Fernández, J. Durban, E. Villalobos, E. Camacho, J.J. Calvete, B.

515

Lomonte, Amino acid sequence and biological characterization of BlatPLA2, a non-

516

toxic acidic phospholipase A2 from the venom of the arboreal snake Bothriechis

517

lateralis from Costa Rica. Toxicon 73 (2013) 71-80.

519 520 521 522 523 524

AC C

518

EP

511

[26] J.J. Calvete, P. Juárez, L. Sanz, Snake venomics, strategy and applications. J. Mass Spectrom. 42 (2007) 1405-1414.

[27] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignment search tool. J. Mol. Biol. 215 (1990) 403-410. [28] T.A. Hall, BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Series 41 (1999) 95-98. [29] F. Ronquist, J.P. Huelsenbeck, MrBayes 3: Bayesian phylogenetic inference under

ACCEPTED MANUSCRIPT 23

525 526 527 528

mixed models. Bioinformatics 19 (2003) 1572-1574. [30] F. Abascal, R. Zardoya, D. Posada, ProtTest: selection of best-fit models of protein evolution. Bioinformatics 21 (2005) 2104-2105. [31] T.A. Castoe, J.M. Daza,

E.N. Smith, M. Sasa, U. Kuch, J.A. Campbell, P.T.

Chippindale, C.L. Parkinson, Comparative phylogeography of pitvipers suggests a

530

consensus of ancient Middle American highland biogeography. J. Biogeogr. 36 (2009)

531

88-103.

534 535

Repository and associated resources. Nucleic Acids Res. 37 (2009) 387-392.

SC

533

[32] F. Kiefer, K. Arnold, M. Künzli, L. Bordoli, T. Schwede, The SWISS-MODEL

[33] N. Guex, M.C. Peitsch, SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18 (1997) 2714-2723.

M AN U

532

RI PT

529

536

[34] R.A. Laskowski, M.W. MacArthur, D. Moss, J.M. Thornton, PROCHECK: a program

537

to check the stereochemical quality of protein structures J. Appl. Cryst., 26 (1993) 283-

538

291.

[35] D. Mora-Obando, J. Fernández, C. Montecucco, J.M. Gutiérrez, B. Lomonte,

540

Synergism between basic Asp49 and Lys49 phospholipase A2 myotoxins of viperid

541

snake venom in vitro and in vivo. PLoS One. 9 (2014) e109846. doi:

542

10.1371/journal.pone.010984.

TE D

539

[36] B. Lomonte, Y. Angulo, S. Rufini, W. Cho, J.R. Giglio, M. Ohno, J.J. Daniele, P.

544

Geoghegan, J.M. Gutiérrez, Comparative study of the cytolytic activity of myotoxic

545

phospholipases A2 on mouse endothelial (tEnd) and skeletal muscle (C2C12) cells in

546

vitro. Toxicon 37 (1999) 145-158.

548

[37] World Health Organization, Progress in the Characterization of Venoms and

AC C

547

EP

543

Standardization of Antivenoms WHO Offset Publication No. 58, Geneva, 1981.

549

[38] B. Lomonte, A. Tarkowski, L.Å. Hanson, Host response to Bothrops asper snake

550

venom: analysis of edema formation, inflammatory cells, and cytokine release in a

551

mouse model. Inflammation 17 (1993) 93-105.

552

[39] J.M. Gutiérrez, B. Lomonte, L. Cerdas, Isolation and partial characterization of a

553

myotoxin from the venom of the snake Bothrops nummifer. Toxicon. 24 (1986) 885-

554

894.

ACCEPTED MANUSCRIPT 24

555

[40] C. Herrera, A. Rucavado, D.A. Warrell, J.M. Gutiérrez, Systemic effects induced by

556

the venom of the snake Bothrops caribbaeus in a murine model. Toxicon 63 (2013) 19-

557

31. [41] C. Renjifo, E.N. Smith, W.C. Hodgson, J.M. Renjifo, A. Sanchez, R. Acosta, J.H.

559

Maldonado, A. Riveros, Neuromuscular activity of the venoms of the Colombian coral

560

snakes Micrurus dissoleucus and Micrurus mipartitus: an evolutionary perspective.

561

Toxicon 59 (2012) 132-142.

563

[42] C. Montecucco, O. Rossetto, How do presynaptic PLA2 neurotoxins block nerve terminals? Trends Biochem. Sci. 25 (2000) 266-270

SC

562

RI PT

558

[43] M.J. Su, C.C. Chang, Presynaptic effects of snake venom toxins which have

565

phospholipase A2 activity (β-bungarotoxin, taipoxin, crotoxin), Toxicon 22 (1984)

566

631-640.

M AN U

564

567

[44] A.L. Harvey, Presynaptic effects of toxins Int. Rev. Neurobiol. 32 (1990) 201–239

568

[45] J. Pungercar, I. Krizaj, Understanding the molecular mechanism underlying the

569

presynaptic toxicity of secreted phospholipases A2. Toxicon, 50 (2007) 871-892 [46] P. Rey-Suárez, R. Stuani-Floriano, S. Rostelato-Ferreira, M. Saldarriaga, V. Núñez, L.

571

Rodrigues-Simioni, B. Lomonte, Mipartoxin-I, a novel three-finger toxin, is the major

572

neurotoxic component in the venom of the redtail coral snake Micrurus mipartitus

573

(Elapidae). Toxicon 60 (2012) 851-863.

TE D

570

[47] A. Alape-Girón, B. Persson, E. Cederlund, M. Flores-Díaz, J.M. Gutiérrez, M.

575

Thelestam, T. Bergman, H. Jörnvall, Elapid venom toxins: multiple recruitments of

576

ancient scaffolds, Eur. J. Biochem. 259 (1999) 225-234.

578 579 580

[48] C. Corrêa-Netto, I.L. Junqueira-de-Azevedo, D.A. Silva, P.L. Ho, M. Leitão-de-

AC C

577

EP

574

Araújo, M.L. Alves, L. Sanz, D. Foguel, R.B. Zingali, J.J. Calvete, J.J., Snake venomics and venom gland transcriptomic analysis of Brazilian coral snakes, Micrurus altirostris and M. corallinus. J. Proteomics 74 (2011) 1795-1809.

581

[49] J. Fernández, A. Alape-Girón, Y. Angulo, L. Sanz, J.M. Gutiérrez, J.J. Calvete, B.

582

Lomonte, Venomic and antivenomic analyses of the Central American coral snake,

583

Micrurus nigrocinctus (Elapidae), J. Proteome Res. 10 (2011) 1816-1827.

584

ACCEPTED MANUSCRIPT 25

585

LEGENDS FOR FIGURES

586

Figure 1: (A) RP-HPLC separation of Micrurus mipartitus venom (2 mg) on a C18 column

588

(250 x 4.6 mm) eluted at 1 mL/min with an acetonitrile gradient (dotted line). MmipPLA2

589

was collected in the peak eluting at 46.7 min, indicated with an arrow. The protein was

590

analyzed by SDS-PAGE (15% gel) under reducing conditions (insert in A). MM: molecular

591

mass standards, in kDa. The insert (B) shows the deconvolution of the multi-charged ion

592

series obtained by nESI-MS analysis (C).

M AN U

593

SC

RI PT

587

Figure 2: (A) RP-HPLC separation of Micrurus dumerilii venom (2 mg) on a C18 column

595

(250 x 4.6 mm) eluted at 1 mL/min with an acetonitrile gradient (dotted line). MdumPLA2

596

was collected in the peak that eluted at 49.6 min, indicated with an arrow. The protein was

597

analyzed by SDS-PAGE (15% gel) under reducing conditions (insert in A). MM: molecular

598

mass standards, in kDa. The insert (B) shows the deconvolution of the multi-charged ion

599

series obtained by nESI-MS analysis (C).

TE D

594

EP

600

Figure 3: Amino acid sequence of (A) MmipPLA2 and (B) MdumPLA2. Overlapping

602

peptides obtained after reduction of disulfide bonds and alkylation of cysteines followed by

603

digestion with proteases (T: trypsin, blue; or C: chymotrypsin, green) were de novo

604

sequenced by MALDI-TOF/TOF MS and Edman degradation (yellow). (C) Alignment of

605

amino acid sequences of MmipPLA2 and MdumPLA2 using BLASTp. Sequence matches

606

are shaded in gray, cysteines are indicated in bold, basic residues in blue, and acidic

607

residues in red. For reference to colors in this figure the reader is referred to the web

608

version of the article.

AC C

601

ACCEPTED MANUSCRIPT 26

609 610

Figure 4: Multiple sequence alignment of MmipPLA2 and MdumPLA2 with related

611

proteins isolated or predicted from M. altirostris (F5CPF1- AED89576), M. corallinus

612

(Q8AXW7),

613

A0A0F7YZM3), M. nigrocinctus (P81166) and M. tener (AET85561) venoms. Protein

614

access codes are indicated on the left. Cysteine residues are highlighted with a gray

615

background.

carinicauda

(MIDCA1),

M.

fulvius

(AAZ29512-

RI PT

dumerilii

SC

M.

616

Figure 5: Phylogenetic relationships of MmipPLA2 and MdumPLA2 with related sequences

618

included in the alignment shown in Fig.4. The tree was built using the Bayesian inference

619

implemented in version 3.0B4. Nodes with posterior probabilities >0.95 are indicated with

620

a black dot. The rooted tree was drawn to scale (see scale bars) by use of the FigTree

621

software (http://tree.bio.ed.ac.uk/software/figtree/). Scale bar indicates number of amino

622

acid changes per site. Green circle - PLA2 from PLA2-rich venoms; red circle - PLA2 from

623

3FTx-rich venoms. This color code definition also applies to lines with the same colors.

624

The black line indicates the outgroup.

TE D

EP

625

M AN U

617

Figure 6: Homology modeling of the three-dimensional structure of (A) MmipPLA2 and

627

(B) MdumPLA2 was performed using Swiss-Model. The best models were obtained when

628

using Naja naja naja 1PSH.1 as a template for MmipPLA2 (63.6% identity), and Naja naja

629

naja 1A3D.1 for MdumPLA2 (70.1% identity). The resulting models were evaluated with

630

ProCheck. Structure superposition shown in (C) was achieved using UCSF Chimera.

631

Arrows point to regions having significant differences in the predicted backbones (see

632

section 3.4 for details).

AC C

626

ACCEPTED MANUSCRIPT 27

633

Figure 7: Functional characterization of MmipPLA2 and MdumPLA2. (A) Phospholipase

635

A2 activity of MmipPLA2 and MdumPLA2 was assayed using the synthetic monodisperse

636

substrate 4-nitro-3-octanoyloxy-benzoic acid (4-NOBA). Points represent means of three

637

replicas, and SD is smaller than symbol size; (B) Myotoxic activity was evaluated by

638

quantification of plasma creatine kinase (CK) activity 3 h after injection of each toxin (2, 4,

639

or 8 µg/100 µL PBS) i.m. in the gastrocnemius muscle of mice. Points represent means ±

640

SD of four mice per group. (C) Edematogenic effects were evaluated by the increase in

641

footpad thickness after injecting each toxin (8 µg/50 µL PBS) subcutaneously. Symbols

642

represent means ± SD of three mice per group. The responses to MmipPLA2 and

643

MdumPLA2 are significantly different (p<0.05) at 2 and 3 h. PBS: phosphate-buffered

644

saline control.

M AN U

SC

RI PT

634

TE D

645

Figure 8: Hematoxylin-eosin stained sections of mouse gastrocnemius muscle 6 h after the

647

i.m. injection of 8 µg of MmipPLA2 (A) or MdumPLA2 (B), obtained after 6 h. Note the

648

intense leukocyte infiltrate in the necrotic areas of (B), but not in (A). (C) Muscle from a

649

mouse injected with phosphate-buffered saline, showing normal histological appearance.

650

The scale bar in (C) also applies to the other panels.

AC C

651

EP

646

ACCEPTED MANUSCRIPT 28

AC C

EP

TE D

M AN U

SC

RI PT

Figure 1

ACCEPTED MANUSCRIPT 29

AC C

EP

TE D

M AN U

SC

RI PT

Figure 2

ACCEPTED MANUSCRIPT 30

AC C

EP

TE D

M AN U

SC

RI PT

Figure 3

ACCEPTED MANUSCRIPT 31

Figure 4

MdumPLA2 A0A0F7YZM3 AET85561 MIDCA1 Q8AXW7 P81166

NLIDFKNMIK NLIHFKNMIE NLIHFKNMIE NLIQFLNMIQ NLINFQRMIQ NLYQLKNMIK

MmipPLA2 F5CPF1 AED89576 AET85561 AAZ29512

....|....| ....|....| ....|....| ....|....| ....|....| 60 70 80 90 100 CYGDAESIYG CTPFLTYYSY ECSE--RQDL CRGNGTKCKA FVCNCDRLAA CYTDAYRFYR CWPFLTLYSH TCSN--RKVI CRGNTTKCKA FVCNCDRVAA CYGEAETVHK CNPFWTFYSY ECSE--GQLT CRDNDTNCKE FVCNCDLEAA CYGEAEKVHG CWPKWTLYSY DCSN--GQLT CKDNNTKCKD FVCNCDRTAA CYDTAEKVHG CWPKWTLYSY DCSN--GQLT CKDNNTKCKD FVCNCDRTAA

MdumPLA2 A0A0F7YZM3 AET85561 MIDCA1 Q8AXW7 P81166

CYGEAEKVHG CYGEAEKVHG CYGEAEKVHG CYDTAKKVFG CYGAAEKYHR CYDTAKHVCK

MmipPLA2 F5CPF1 AED89576 AET85561 AAZ29512

....|....| ....|....| .. 110 120 LCFAKAPYNK KNYNINLNR- CK NCFAKAPYNK RNYNN----- CK NCFAKAPYIE ENYNINLNR- CT LCFAKAPYNN KNYNIDLKR- CQ LCFAKAPYND KNYNIDLKR- CQ

E-value 3.0e-90 2.0e-64 2.0e-62 5.0e-60 4.0e-58

MdumPLA2 A0A0F7YZM3 AET85561 MIDCA1 Q8AXW7 P81166

ICFAKAPYDD LCFAKAPYND LCFAKAPYNN LCFAKAPYND LCFGRAPYNK LCFAKAPYNN

2.0e-89 3.0e-68 2.0e-67 1.0e-60 8.0e-59 2.0e-48

DCSN--GQLT DCSN--GQLT DCSN--GQLT DCSE--GKLT TCSSQTGSVT DCSE--GKLT

AC C

EP

TE D

CWPKWTLYSY CWPKWTLYSY CWPKWTLYSY CSPYFTMYSY CSPKLTLYTS CSPSMTMYSY

NNFMINNPRKNYNIDPSRKNYNIDLKRKNYKIDLTKR NNENINPNRKNFKIDPTKG

SGGSGTPVDD SGGSGTPVDD SGGSGTPVDD RGGSGTPVDE AGGSGTPVDE YGGSGTPVDE

LDRCCKVHDD LDKCCQVHDD LDKCCQVHDD LDRCCQVHDN LDRCCKVHDD LDKCCQVHDK

SC

DFADYGCYCG DFADYGCYCG DFADYGCYCG AFANYGCYCG DFTNYGCYCG SFTNYGCYCG

M AN U

CT--TKRSVL CT--TKRSWW CT--TKRSWW CTTPGREPLV CTT--RRSAW CT--NTRHWV

RI PT

MmipPLA2 F5CPF1 AED89576 AET85561 AAZ29512

....|....| ....|....| ....|....| ....|....| ....|....| 10 20 30 40 50 NLIHFSSMIK CTIPGSKPVP DYSDYGCYCG KGGSGTPVDA LDRCCQVHDK NLIQFGNMIQ CTIPGSSPLL DYADYGCYCG RGGSGTPVDK LDRCCQAHDK NLYQFGNMIN CTMPGGSPLL DYADYGCYCG SGGGGTPVDD LDRCCQAHDN NLIHFKNMIE CT--TKRSWW DFADYGCYCG SGGSGTPVDD LDKCCQVHDD NLIHFKNMIE CTTK--RSWW HFADYGCYCG SGGSGTPVDD LDKCCQVHDK

CQ CQ CQ CQ CR CQ

CKDNNTKCKD CKDNNTKCKD CKDNNTKCKD CKDNNTKCKA CKDNGTKCKA CKDNNTKCKD

FVCNCDRTAA FVCNCDRTAA FVCNCDRTAA AVCNCDRTAA FVCNCDRTAA FVCNCDRTAA

Score 621 451 438 422 412

%Id 100 75 69 68 67

615 476 471 426 414 346

100 88 87 69 70 70

ACCEPTED MANUSCRIPT 32

AC C

EP

TE D

M AN U

SC

RI PT

Figure 5

ACCEPTED MANUSCRIPT 33

AC C

EP

TE D

M AN U

SC

RI PT

Figure 6

ACCEPTED MANUSCRIPT 34

AC C

EP

TE D

M AN U

SC

RI PT

Figure 7

ACCEPTED MANUSCRIPT 35

AC C

EP

TE D

M AN U

SC

RI PT

Figure 8

ACCEPTED MANUSCRIPT

Coral snake venoms are rich in three-finger toxins (3FTx) and phospholipases A2 (PLA2).

•

MmipPLA2 (M. mipartitus) and MdumPLA2 (M. dumerilii) were isolated and partially characterized.

•

The two PLA2s were located in separate lineages in a partial phylogeny of Micrurus PLA2s.

•

The enzymes differed in enzymatic, anticoagulant, myotoxic and edematogenic activities.

•

MmipPLA2 was at least 30-fold more lethal to mice than MdumPLA2 when injected i.p.

AC C

EP

TE D

M AN U

SC

RI PT

•