Unraveling the Proteome Composition and ... - ACS Publications

Jul 27, 2016 - to 10 mL/h, fraction volume was 2.0 mL, and protein elution was monitored at ..... (Amersham Bioscience, UK) at 20 mA (1.2 mA/cm2) for ...
2 downloads 0 Views 4MB Size
Article pubs.acs.org/jpr

Unraveling the Proteome Composition and Immuno-profiling of Western India Russell’s Viper Venom for In-Depth Understanding of Its Pharmacological Properties, Clinical Manifestations, and Effective Antivenom Treatment Bhargab Kalita, Aparup Patra, and Ashis K. Mukherjee* Microbial Biotechnology and Protein Research Laboratory, Department of Molecular Biology and Biotechnology, Tezpur University, Tezpur 784028, Assam, India S Supporting Information *

ABSTRACT: The proteome composition of western India (WI) Russell’s viper venom (RVV) was correlated with pharmacological properties and pathological manifestations of RV envenomation. Proteins in the 5−19 and 100−110 kDa mass ranges were the most predominate (∼35.1%) and least abundant (∼3.4%) components, respectively, of WI RVV. Non-reduced SDS-PAGE indicated the occurrence of multiple subunits, non-covalent oligomers, self-aggregation, and/or interactions among the RVV proteins. A total of 55 proteins belonging to 13 distinct snake venom families were unambiguously identified by ESI-LC-MS/MS analysis. Phospholipase A2 (32.5%) and Kunitz-type serine protease inhibitors (12.5%) represented the most abundant enzymatic and non-enzymatic proteins, respectively. However, ATPase, ADPase, and hyaluronidase, detected by enzyme assays, were not identified by proteomic analysis owing to limitations in protein database deposition. Several biochemical and pharmacological properties of WI RVV were also investigated. Neurological symptoms exhibited by some RV-bite patients in WI may be correlated to the presence of neurotoxic phospholipase A2 enzymes and Kunitz-type serine protease inhibitor complex in this venom. Monovalent antivenom was found to be better than polyvalent antivenom in immunorecognition and neutralization of the tested pharmacological properties and enzyme activities of WI RVV; nevertheless, both antivenoms demonstrated poor cross-reactivity and neutralization of pharmacological activities shown by low-molecular-mass proteins (50 kDa) in GF-1 and GF-2 fractions, thereby supporting our earlier observations on protein−protein interactions in RVV.18 MALDI-TOF-MS analysis of WI RVV ranges revealed the presence of 92 distinct ions in the m/z range of 5.3−195.9 kDa (Table S1).

Figure 4. Protein family composition of WI RVV proteome. The pie chart represents the relative occurrence of different enzymatic and non-enzymatic proteins families of RVV. The details of each protein are shown in Figure S1 and Table S2.

The venom proteome of WI RVV showed sequence homology with 17 PLA2 isoenzymes reported in Viperidae protein database, and collectively they constitute 32.5% of the RVV proteome (Table 2 and Figure 4). Among the notable identities, a single isoform of the acidic subunit of vaspin (accession no. gi|50874232), a Group IIA PLA2 from Vipera aspis aspis, and two homologues of different isoforms of ammodytin, a myotoxic PLA2 from V. ammodytes montandoni (accession nos. gi|50874310 and gi|50874356), were also identified in WI RVV proteome (Tables 2 and S2, and Figure S1). Remarkably, a single isoform of daboiatoxin, a major PLA2 toxin reported from D. r. siamensis from Myanmar,51 was also identified in the venom proteome of D. russelii from WI and Pakistan.18 Further, LC-MS/MS analysis provided evidence of the existence of three neurotoxic PLA2s, i.e., daboiatoxin (accession no. gi|149241831),52 VRV-PL-VIIIa (accession no. gi|24638087),53 and RV-4 (accession no. gi|400713),54 in WI RVV (Table 2). These data suggest that RVV PLA2s share substantial homology across the Viperidae family. RVV is reported to contain a significant quantity of proteolytic enzymes.2,20,21,33,55,56 Snake venom proteases are broadly classified as serine proteases and metalloproteases.56,57 Proteomic analysis suggested that both of these groups of proteases occur in the RVV under study (Table 2). SVMPs represented the second most abundant group of proteins (24.8%) in WI RVV, thereby reflecting their pathophysiological contribution to clinical symptoms of RV bites in this region.7 A total of five metalloprotease isoforms, including one heavy-chain isoform and two light-chain isoforms of RVV-X,58 a heterodimeric SVMP reported from Vipera russelli, were also identified in WI RVV proteome (Tables 2 and S2, and Figure S1). Remarkably, some of the SVMPs (accession nos. gi|300079900 and gi|162329887) identified in WI RVV (Table 2) also showed sequence similarity with a pro-coagulant, multimeric SVMP possessing fibrinogenolytic activity from eastern India RVV.33 A total of six isoforms of SVSPs (Figure 4), including one homologue of RVV-Vα reported earlier from the venom of V. russelli, were identified, and they contributed 8.0% of WI RVV (Tables 2 and S2, and Figure S1).59 Single isoforms each of βfibrinogenase (accession no. gi|765684342) from D. siamensis and thrombin-like enzyme (accession no. gi|38146946) from Gloydius shedaoensis venom were also observed in WI RVV. Interestingly, tryptic peptides from pro-coagulant serine proteases (VIGGDECNIN)20 and thrombin-like serine pro-

LC-MS/MS Analysis of RVV Ion-Exchange Peaks

The shotgun proteomic approach has emerged as a powerful tool in assembling the toxin arsenal of various snake venoms.15,16,18,42 However, while analyzing complex protein samples, low peptide coverage is a major hurdle. Subsequently, relying on semi-tryptic peptides (one trypsin cleavage site at either one of the terminals) may increase the peptide coverage and lead to enhanced protein identification.47 Therefore, in the current study, tryptic as well as semi-tryptic peptides identified by a MASCOT search were also taken into account for protein identification. The ESI-LC-MS/MS analysis led to identification of 55 enzymes as well as non-enzymatic proteins belonging to 13 distinct snake venom protein families in WI RVV sample (see Table 2, below). The relative abundance of different classes of WI RVV proteins identified by proteomic analysis is depicted in Figure 4. The alignment of MS/MS-derived peptide sequences with the homologous proteins from the Viperidae snake venom database is shown in Figure S1. Enzymatic Proteins Identified in WI RVV

The enzymatic proteins identified in WI RVV include PLA2 (32.5%), SVMP (24.8%), SVSP (8.0%), PDEs (1.4%), NTs (0.4%), LAAOs (0.3%), and PL B (0.1%) (Figure 4). PLA2 is a well-characterized, pharmacologically active snake venom enzyme.39,46,48 The molecular mass of snake venom PLA2 is reported in the range of 10−15 kDa.39 Therefore, elution of most of the PLA2s in the GF-5 fraction was according to the molecular mass of this group of proteins (Table 1a and Figure 3B).35,49 However, similar to our previous observation,18 PLA2 enzyme was detected in almost all the GF peaks of WI RVV (Table 2), advocating interaction of this enzyme with other proteins of RVV to enrich the toxicity of interacting components.18,46,50 F

DOI: 10.1021/acs.jproteome.6b00693 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research

Table 1. (a) Some Enzymatic Activities and Esterolytic Activities of Crude RVV from WI and Its Gel Filtration Fractions and (b) Some Enzyme Activities of Crude RVV from WI and Its Gel Filtration Pools (Values Are Mean ± SD of Triplicate Determinations) (a) Enzymatic and Esterolytic Activities of Crude RVV from WI and Its Gel Filtration Fractions enzyme activity venom

protein content

fraction

(% yield)

crude RVV GF-1 GF-2 GF-3 GF-4 GF-5 GF-6 GF-7 GF-8 GF-9 GF-10

100.0 6.6 ± 0.8 4.3 ± 0.2 8.4 ± 0.9 16.0 ± 0.10 9.6 ± 0.9 14.5 ± 0.11 11.1 ± 0.9 1.8 ± 0.3 4.4 ± 0.3 4.0 ± 0.2

PLA2

a

SVMP

(U/mg) (0.6 (0.2 (0.1 (1.9 (2.7 (4.7 (1.7 (1.2 (0.7 (0.6 (1.2

± ± ± ± ± ± ± ± ± ± ±

0.03) 0.01) 0.01) 0.06) 0.11) 0.20) 0.05) 0.04) 0.02) 0.01) 0.08)

× × × × × × × × × × ×

103 103 103 103 103 103 103 103 103 103 103

b

esterolytic activity

fibrinogenolytic

c

fibrinolytic

c

(U/mg)

(U/mg)

(U/mg)

0.15 ± 0.03 2.33 ± 0.05 0.70 ± 0.01 − − − − − − − −

9.8 ± 0.21 10.7 ± 0.30 7.8 ± 0.25 2.6 ± 0.10 4.5 ± 0.17 5.7 ± 0.25 − 1.4 ± 0.12 − 1.2 ± 0.24 −

0.7 ± 0.04 − − 2.4 ± 0.12 6.0 ± 0.25 0.5 ± 0.02 2.5 ± 0.11 − − 0.7 ± 0.14 −

TAME

d

BAEEe

(U/mg) 1.9 ± 0.08 × 1.9 ± 0.08 × 0.9 ± 0.04 × − 5.7 ± 0.12 × 0.8 ± 0.04 × 2.4 ± 0.10 × − − − −

(U/mg) 103 103 103 103 103 103

2.8 ± 0.08 − 3.3 ± 0.09 2.8 ± 0.05 4.0 ± 0.10 0.5 ± 0.04 0.2 ± 0.01 − − − −

× 102 × × × × ×

102 102 102 102 102

(b) Enzyme Activities of Crude RVV from WI and Its Gel Filtration Pools venom

phosphodiesterasef

LAAOg

ATPaseh

ADPaseh

5′-NTh

hyaluronidasei

fraction

(U/mg)

(U/mg)

(U/mg)

(U/mg)

(U/mg)

(U/mg)

crude RVV GF-1 GF-2

11.8 ± 0.08 22.6 ± 0.11 1.0 ± 0.02

19.8 ± 0.92 23.7 ± 0.96 19.7 ± 0.86

(4.5 ± 0.15) × 103 (9.5 ± 0.26) × 103 (0.3 ± 0.13) × 103

(6.4 ± 0.25) × 103 (8.5 ± 0.28) × 103 (0.2 ± 0.19) × 103

(1.7 ± 0.05) × 104 (4.1 ± 0.11) × 104 −

63.4 ± 2.11 71.4 ± 2.54 −

a

One unit is defined as a decrease by 0.01 in absorbance at 740 nm after 10 min of incubation. bOne unit is defined as change in absorbance at 450 nm per min at 37 °C. cOne unit is defined as 1.0 μg of tyrosine equivalent liberated per min per mL of enzyme. dOne unit is defined as an increase by 0.01 in absorbance at 244 nm during the first 10 min of the reaction at 37 °C. eOne unit is defined as an increase by 0.01 in absorbance at 254 nm during the first 5 min of the reaction at 37 °C. fOne unit is defined as micromoles of p-nitrophenol released per min. gOne unit is defined as 1 nmol of kynurenic acid produced per min. hOne unit is defined as micromoles of Pi released per min. iOne unit is defined as a 1% decrease in turbidity at 405 nm in comparison to control (100% turbidity). In part (b), no data were recorded for GF-3 through GF-10.

tease (Russelobin) (TSTYIAPLSLPSSPPR)21 from Pakistan RVV were also detected in WI RVV (accession nos. gi| 38146946 and gi|765684342, Table 2), indicating some evolutionary relationship between venom of RVs from WI and Pakistan. The highest esterolytic (TAME and BAEE) activity was observed in GF-4 (Table 1a); however, the highest fibrinogenolytic and fibrinolytic specific activity was demonstrated by GF-1 and GF-4, respectively (Table 1a). This result indicates fibrinolytic and fibrinogenolytic enzymes may presumably be distinct proteases in RVV, and therefore they were predominately eluted in different GF fractions (Table 1a). Remarkably, a few SVSPs and SVMPs purified from the venom of RVs from Pakistan and eastern India, respectively, demonstrated fibrinogenolytic activity without showing fibrinolytic property, supporting our assumption.20,33 Proteomic as well as biochemical analyses suggested that Rusviprotease-like RVV-SVMPs33 predominated in GF-1 fraction of WI RVV (Tables 1a and 2), which is also responsible for fibrinogenolytic activity displayed by this venom (Figure S2A,B). Further, WI RVV displayed better fibrinogenolytic activity as compared to fibrinolytic activity (Table 1a and Figure S2A−D). The GF-4 fraction displayed the highest BAEE-esterase as well as appreciable fibrinogenolytic activity (Table 1a), which may be correlated to the elution of 32−34 kDa FV-activating fibrinogenolytic serine protease isoenzymes20 in this fraction (Figure 3B and Table 2). LC-MS/MS analysis showed the elution of a SVMP (accession no. gi|300079900), showing sequence similarity with a SVMP possessing fibrinogenolytic

activity from eastern India RVV33 in GF-9 (Table 2). This may presumably be due to specific/non-specific interaction of a low quantity of SVMP with low-molecular-mass proteins of WI RVV to augment their toxicity.45 Interestingly, the protease(s) accountable for fibrinogenolytic activity in GF-7 was not identified by LC-MS/MS probably analysis owing to limitation in venom protein database deposition. Snake venom PDEs are single subunit enzymes with molecular weight ranging from 100 to 140 kDa.60,61 A highmolecular-weight protein (∼103 kDa) that eluted in the GF-1 fraction (Figure 3A,B) with a relative contribution of only 1.4% of WI RVV proteome (Figure 4) was identified as PDE by proteomic analysis (Table 2). The PDE from WI RVV shared 17.4% sequence similarity with PDE from Macrovipera lebetina (Tables 2 and S2, and Figure S1). Nucleotidases (AMPase, ADPase and ATPase) are yet another group of important but poorly characterized highmolecular-weight snake venom enzymes43,62 eluted in the GF-1 fraction (Tables 1b and 2, and Figure 3B). Fascinatingly, ATPase and ADPase enzymes of WI RVV (Table 1b) were not identified by proteomic analysis, which may be due to limitation in protein database depository.18 LC-MS/MS analysis identified two isoenzymes of 5′-NT/AMPase that comprised 0.4% of the WI RVV (Table 2 and Figure 4, Table S2 and Figure S1). LAAO is a class of enzyme that catalyzes the oxidative deamination of an L-amino acid to an α-keto acid, thereby liberating ammonia and hydrogen peroxide; the latter is particularly detrimental to cells.63 Due to their high molecular weight (60−150 kDa),27 LAAOs were eluted in GF-1 and GF-2 G

DOI: 10.1021/acs.jproteome.6b00693 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research

Table 2. Summary of Different Venom Proteins Identified in Western Indian Russell’s Viper Venom by ESI-LC-MS/MS Analysis of Ion-Exchange Peaks sl. no.

protein

NCBI accession no.

protein score

coverage (%)

mol wt (kDa)

homology with protein from

GF peaka

Enzymatic Proteins Phospholipase A2 (PLA2) 1 basic phospholipase A2 2 basic phospholipase A2 3 3 basic phospholipase A2 RVV-VD 4 ammodytin I1(D) isoform 5 6 7 8 9 10 11 12

basic phospholipase A2 homologue basic phospholipase A2 VRV-PL-VIIIa acidic phospholipase A2 DsM-A2/DsM-A2′ acidic phospholipase A2 Cvv-E6b acidic phospholipase A2 RV-7 chain H, structure of daboiatoxin acidic phospholipase A2 Ts-A3 acidic phospholipase A2 homologue vipoxin A chain 13 ammodytin I1(B′) variant 14 ammodytin I1(F) isoform 15 acidic phospholipase A2 ammodytin I1 16 17 Snake 1 2

vaspin acidic subunit (1) variant basic phospholipase A2 RV-4 Venom Metalloprotease (SVMP) chain A, venom metalloproteinase coagulation factor X activating enzyme light chain 3 factor X activator light chain 2 4 coagulation factor X-activating enzyme beta-chain 5 factor X activator heavy chain Snake Venom Serine Protease (SVSP) 1 venom serine proteinase-like protein 2 2 serine protease, partial 3 beta-fibrinogenase-like 4 factor V activator RVV-V alpha 5 serine protease VLSP-3 6 thrombin-like enzyme Phosphodiesterase (PDE) 1 phosphodiesterase Nucleotidase (NT) 1 snake venom 5′-nucleotidase 2 5′-nucleotidase, partial L-Amino Acid Oxidase (LAAO) 1 L-amino acid oxidase 2 L-amino acid oxidase Lm29 Phospholipase B (PLB) 1 phospholipase B

15.5 15.5

Daboia russelii Daboia russellii russellii Daboia russellii russellii Vipera ammodytes montandoni Gloydius halys Daboia russellii russellii Daboia siamensis Crotalus viridis viridis Daboia siamensis Daboia siamensis Trimeresurus stejnegeri Vipera ammodytes meridionalis Vipera berus berus Vipera aspis atra Vipera ammodytes ammodytes Vipera aspis aspis Daboia siamensis

3,5,9 1,3

26.2 33.3

47.6 14.5

Daboia siamensis Daboia russelii

2,5,6 2,3

404.7 375.4

22.2 23.4

18.3 18.3

Daboia russellii russellii Daboia siamensis

1,2 1

gi|300079900

329.2

13.3

69.5

Daboia russellii russellii

1,2,3,5,6,9

gi|13959655 gi|297593758 gi|765684342 gi|134129 gi|380875417 gi|38146946

484.9 222.9 181.3 161.4 103.1 59.2

8.9 12.7 13.3 20.3 12.0 5.9

28.9 28.8 28.0 26.2 28.3 26.5

Macrovipera lebetina Echis coloratus Daboia siamensis Daboia siamensis Macrovipera lebetina Gloydius shedaoensis

2 2 2,3 1,2,3 2 3

gi|586829527

964.4

17.4

96.1

Macrovipera lebetina

1, 2

gi|395455152 gi|586829529

151.3 121.4

6.8 11.5

64.4 45.0

Gloydius brevicaudus Macrovipera lebetina

1 1

gi|395406796 gi|704043548

366.3 54.4

9.9 3.7

56.9 58.5

Daboia russellii russellii Lachesis muta

1,2 1

gi|727360709

161.9

6.5

64.5

Echis coloratus

1

4 4 1,2,3,6,7,8,9 4,5,6,7,8 6,7 5,6,7,8,9 6 3,5,6,7,8,9

gi|71912223 gi|298351762 gi|3914259 gi|50874384

11115.2 7409.7 5514.4 1203.5

66.4 76.0 57.0 43.5

16.3 13.7 13.6 15.4

gi|27151648 gi|24638087 gi|408407661 gi|82209451 gi|400714 gi|149241831 gi|82201337 gi|2851544

704.1 688.5 557.0 544.0 530.6 460.3 443.2 312.1

20.5 75.2 74.6 10.1 69.6 34.4 10.1 45.1

13.9 13.6 15.6 15.4 15.4 14.0 15.5 13.6

gi|50874310 gi|50874356 gi|25453141

307.8 278.8 231.0

48.6 47.8 36.2

15.4 15.4 15.4

gi|50874232 gi|400713

150.4 97.6

39.9 23.9

gi|162329887 gi|251205

942.0 441.9

gi|300079896 gi|73621140

4 1,2,3,4,5,6,7,9 1,2,3,4,6,8,9 3,5,6,7,9,10 5 1,6,9 3,5,6,7,8,9,10 5 1,2,3,5,6,7,8,9,10 3,5,7,9 5 3,5,6 8,10 9,10 1,2,10

Nonenzymatic Proteins Kunitz-type Serine Protease Inhibitor (KSPI) 1 Kunitz-type protease inhibitor 2 Kunitz protease inhibitor-II 3 Kunitz-type serine protease inhibitor B2 4 Kunitz-type serine protease inhibitor 4 5 Kunitz-type serine protease inhibitor B1 6 Kunitz-type serine protease inhibitor B4 7 Kunitz-type serine protease inhibitor C6 8 Kunitz-type serine protease inhibitor C3 Cysteine-Rich Secretory Protein (CRISP) 1 cysteine-rich seceretory protein Dr-CRPK

gi|379647506 gi|87130864 gi|239977248 gi|123913154 gi|239977245 gi|239977254 gi|239977259 gi|239977252

2751.7 1635.1 1311.9 719.1 423.7 351.9 223.0 161.0

51.1 25.0 34.5 52.4 48.8 39.3 43.8 41.7

10.4 9.9 9.3 9.5 9.3 9.4 10.3 9.4

Daboia Daboia Daboia Daboia Daboia Daboia Daboia Daboia

gi|190195321

160.1

30.1

26.7

Daboia russelii

H

russelii russellii russellii siamensis russellii russellii siamensis siamensis siamensis siamensis

1,2,3,4,5,9 DOI: 10.1021/acs.jproteome.6b00693 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research Table 2. continued sl. no.

protein

2

cysteine-rich seceretory protein Ch-CRPKa, partial Disintegrin 1 jerdostatin Nerve Growth Factor (NGF) 1 venom nerve growth factor Vascular Endothelial Growth Factor (VEGF) 1 snake venom vascular endothelial growth factor toxin VR-1 2 vascular endothelial growth factor A Snaclec 1 2 3 4 5 6 7 a

dabocetin alpha subunit P68 alpha subunit Snaclec 3 dabocetin beta subunit Snaclec 4 P31 beta subunit P31 alpha subunit

NCBI accession no.

protein score

coverage (%)

Nonenzymatic Proteins gi|190195307 147.5 12.2

mol wt (kDa) 24.7

homology with protein from

GF peaka

Crotalus horridus

2,3,5,6

gi|292659514

183.4

45.7

4.9

Protobothrops jerdonii

6,7,8,9

gi|400499

128.5

31.6

13.3

Daboia russellii russellii

1,2,3,4,5,6

gi|327478537

650.9

34.0

16.3

Daboia russellii russellii

1,2,3,4,5

gi|327488518

55.7

9.4

22.4

Vipera ammodytes ammodytes

6

gi|300490462 gi|300490470 gi|73620111 gi|300490464 gi|73620112 gi|300490488 gi|300490478

1659.6 1090.2 900.4 756.2 443.1 113.2 111.8

36.4 61.4 46.0 26.7 49.3 16.0 34.5

17.9 18.0 16.9 18.0 16.8 17.4 18.2

Daboia Daboia Daboia Daboia Daboia Daboia Daboia

4,5 3,5,6,8 1,2,3,5 3,4,5 1,2,3,9 3 3

russellii russellii siamensis siamensis russellii russellii siamensis russellii russellii russellii limitis

The distributions of the proteins in different GF peaks are shown in the last column.

Rusvikunin-II) were purified and characterized from Pakistan RVV.22,44 Cystein-rich secretory proteins (CRISPs) are a class of nonenzymatic venom proteins which consist of a single polypeptide chain of 20−30 kDa mass with 16 conserved cysteine residues.40 By proteomic analysis, RVV sample from Pakistan was reported to contain 2.6% of CRISPs.18 A total of two CRISPs comprising of 6.8% of venom proteome and sharing sequence homology with CRISPs from viperid venom, such as Ch-CRPKa and Dr-CRPK, were identified in WI RVV proteome (Tables 2 and S2, and Figure S1). Disintegrins are cysteine-rich, low-molecular-weight (4−15 kDa) viperid venom components generated by the proteolytic cleavage of Class P II metalloproteases.42 To date, there is no report on purification and characterization of disintegrins from RVV, albeit their presence in RVVs was confirmed by proteomic analyses.16,18 A single disintegrin with a relative abundance of 4.9% was identified in WI RVV by proteomic analysis, and it shared sequence similarity with jerdostatin, isolated from Trimeresurus (Protobothrops) jerdonii snake venom (Tables 2 and S2, and Figure S1). Venom NGFs belong to a specific family of venom proteins called “neurotrophic factors”, and their molecular weights range from 25 to 54 kDa.41 NGFs belong to a class of relatively less characterized venom protein, and their pharmacological effect in victims or rationale of existence in snake venoms is controversial. To date, no NGF has been purified from RVV, notwithstanding the occurrence of NGF in RVVs from southern India, Sri Lanka, and Pakistan was demonstrated by proteomics analysis.16−18 A single NGF constituting 4.8% of WI RVV proteome (Figure 4) was identified by tandem mass spectrometry analysis (Tables 2 and S2, and Figure S1). Snake venom VEGFs are 23−33 kDa proteins.67 Although VEGF has not been purified and characterized from RVV; nevertheless, its presence in venoms of D. russelii and D. siamensis was demonstrated by proteomic analyses.16,18,51 By LC-MS/MS analysis, two VEGF isoforms (Table 2) constituting 1.8% of the WI RVV proteome were identified (Figure 4).

fractions of RVV (Figure 3B), which was also confirmed by enzymatic assay and LC-MS/MS analysis (Tables 1b and 2). Proteomic analysis of WI RVV fractions identified two LAAO isoenzymes (Tables 2 and S2, and Figure S1), which comprised approximately 0.3% of the RVV proteome (Figure 4). Therefore, LAAO is a minor component of WI RVV. Phospholipase B (PLB) or lysophospholipases catalyzes the hydrolysis of monoacyl phosphatides with the release of free fatty acid and the formation of glycerophosphoryl derivatives.64 To date, very little information is available on this class of enzymes, and their role in pathophysiology of snake bites remains to be elucidated. By proteomic analysis, the presence of a single PLB enzyme (0.1% of WI RVV) showing sequence homology with PLB enzyme from Echis coloratus venom was detected in GF-1 of RVV under study (Tables 2 and S2, and Figure S1). Non-enzymatic Proteins of RVV

Several non-enzymatic proteins are reported in RVV that contribute significantly to the clinical manifestations of RV envenomation.22,44,65 Although reports on non-enzymatic proteins from WI RVV are limited, characterization of such proteins from RVV from other regions of the Indian subcontinent was well documented.22,23,28,44,65,66 KSPIs are low-molecular-weight proteins (6−10 kDa) and reported to occur in RVV.22 LC-MS/MS analysis demonstrated KSPIs were the most abundant non-enzymatic proteins (12.5%) of WI RVV. KSPIs, like PLA2s, were also identified in almost all the GF peaks of RVV (Table 2) indicating their interactions with other components of venom. The same observation was also put forwarded for Pakistan RVV.18 Proteomic analysis revealed the presence of eight KSPI isoforms in WI RVV proteome (Tables 2 and Table S2, and Figure S1). Interestingly, two tryptic fragments, FCNLAPESGR and QTCGAPR, of Rusvikunin, and one tryptic peptide, RIYYNLESNK, of Rusvikunin II were also detected in WI RVV proteome (accession nos. gi|379647506 and gi| 239977248, Table 2). The above two KSPIs (Rusvikunin and I

DOI: 10.1021/acs.jproteome.6b00693 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research

Table 3. In Vitro Blood Clotting, PT, APTT, Hemolytic Activity, and Thrombocytopenia of Crude RVV and Its GF Fractions (Values Are Mean ± SD of Triplicate Determinations)f pharmacological properties venom

plasma clotting activity

a

platelet counte

hemolysis (%)

coagulant/ b

fraction

(U/mg)

anti-coagulant

PT

APTT

Crude RVV GF-1 GF-2 GF-3 GF-4 GF-5 GF-6 GF-7 GF-8 GF-9 GF-10

1.7 ± 0.08 × 104 7.4 ± 0.10 × 104 11.5 ± 0.21 × 104 10.0 ± 0.15 × 104 8.7 ± 0.11 × 104 1.5 ± 0.04 × 104 11.9 ± 0.14 × 104 1.0 ± 0.03 × 104 2.6 ± 0.09 × 104 ND ND

C C C C C AC AC AC AC − −

8.3 ± 0.2f 6.3 ± 0.2f 8.6 ± 0.3f 9.9 ± 0.4f 11.2 ± 0.3f 15.0 ± 0.4f 16.5 ± 0.7f 14.2 ± 0.6 14.7 ± 0.4f 13.0 ± 0.5 13.8 ± 0.6

c

16.8 ± 0.7f 19.1 ± 0.8f 16.2 ± 0.4f 21.6 ± 0.3f 23.9 ± 0.1f 45.6 ± 0.4f 48.7 ± 0.6f 38.0 ± 0.3f 35.7 ± 0.8 36.8 ± 0.5 36.9 ± 2.3

direct

d

0.2 ± 0.01 1.0 ± 0.03 − − 1.3 ± 0.04 0.9 ± 0.02 − − − 0.8 ± 0.02 −

indirect

d

18 ± 0.5 26.1 ± 1.1 6.7 ± 0.4 2.5 ± 0.1 4.8 ± 0.3 16.4 ± 0.5 27.3 ± 1.1 0.3 ± 0.1 1.1 ± 0.1 8.1 ± 0.2 5.6 ± 0.1

(×106 cells/mL) 3.3 ± 0.12f 2.4 ± 0.10f 2.8 ± 0.11f 2.6 ± 0.09f 2.8 ± 0.03f 4.5 ± 0.15f 5.2 ± 0.14 5.2 ± 0.16 5.4 ± 0.23 5.3 ± 0.21 5.2 ± 0.19

Plasma clotting activity was assayed with crude RVV or GF fractions at a concentration of 5 μg/mL against 300 μL of goat platelet poor plasma (PPP). ND indicates not detected. C and AC represent coagulant and anti-coagulant activities, respectively. bAverage prothrombin time (PT) of control PPP was 13.5 ± 0.6 s. The assay was performed with 1.0 μg protein. cAverage thromboplastin time (APTT) of control PPP was 35.0 ± 0.9 s. The assay was performed with 1.0 μg protein. dHemolytic activity was assayed against 5% (v/v) mammalian erythrocytes with RVV/fraction at a final concentration of 5 μg/mL. ePlatelet count of control was (5.2 ± 0.21) × 106 cells/mL. fSignificance of difference with respect to control, *p < 0.05. a

Correlation of WI RVV Proteome with Its in Vitro Pharmacological Properties and Clinical Manifestations Post RV Bite in WI

Snaclecs are characterized as larger quaternary structures of disulfide linked homo- or heterodimers; the molecular weight of the monomers range from 8 to 16 kDa.28,42 The first snaclec from RVV (Pakistan) was reported by Mukherjee et al.28 The proteome analysis of WI RVV resulted in identification of seven snaclec isoforms eluted in GF-1 to GF-5 fractions, and they constituted only 1.8% of the WI RVV proteome (Figure 4, Tables 2 and S2, and Figure S1). Isoforms of both α and β subunits of Dabocetin (accession nos. gi|300490462 and gi| 300490464), a heterodimeric snaclec from D. r. siamensis venom,68 were identified in WI RVV proteome (Tables 2 and S2, and Figure S1). Further, peptide sequences sharing homology to P68 α subunit, P31 α and β subunits of snaclec from D. siamensis, and CTL-3 and CTL-4A from other viperid venoms were also identified in WI RVV (Tables 2 and S2, and Figure S1). In addition, one of the peptide fragments, GSHLLSLHNIAEADFVLK identified in RV snaclec from Pakistan RVV28 was also detected (accession no. gi|73620111, Table 2) in WI RVV. Furthermore, six isoforms of structural proteins including actin, keratin, calmodulin, 60s ribosomal proteins, and myosin were also identified in WI RVV proteome. However, they may not be the real components of RVV and were probably contaminated from the venom gland tissues during the process of venom extraction. Therefore, these structural proteins representing only a very minor component of RVV (0.7%) were not considered for calculation of relative abundances of proteins in WI RVV. Notably, Rusvitoxin-like cytotoxic peptide (isolated from eastern India RVV), demonstrating close identity with snake venom three-finger toxins, cytotoxins, and cardiotoxins,65 could not be identified by proteomic analysis of WI RVV, advocating that geographical differences in RVV composition may show different pharmacological properties and associated clinical manifestations of RV bites.

A sound correlation between venom enzyme function and pathophysiological symptoms was demonstrated.2,10,13,18 Interference in blood coagulation is the major clinical manifestation following RV envenomation,1,2,4,6,7 suggesting that hemostatically active proteins from RVV play a significant role in the pathophysiology of RV envenomation.2,56 A fully grown Russell’s viper contains 225−250 mg of venom. A fully lethal RV bite results in injection of 40−45 μg/mL of RVV in an adult human’s blood.34 WI RVV at all the tested doses exhibited procoagulant activity in in vitro conditions (Table 3 and Figure 5), attributed to the predominance of pro-coagulant SVMPs and SVSPs (Table 2 and Figure 4) in its venom.18,20,21,33,56 Notably, high-molecular-mass pro-coagulant enzymes (>30 kDa) were separated in GF-1 to GF-4 fractions (Figure 3A and Table 3), which corroborates well with the previous reports.20,21,56,69,70

Figure 5. Effect of different concentrations of crude RVV on Ca2+ clotting time of goat platelet-poor plasma. Values are mean ± SD of triplicate determinations. Significance of difference with respect to control, *p < 0.05. J

DOI: 10.1021/acs.jproteome.6b00693 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research Nevertheless, WI RVV proteins in the mass range of 27−6 kDa (GF-5 to GF-8 gel filtration fractions) (Figure 3B) showed anticoagulant activity (Table 3). The above proteins are mostly PLA2s, KSPIs, and disintegrins (Table 2), and their anticoagulant action via enzymatic and/or non-enzymatic mechanism(s) has already been demonstrated.22,28,34,44,48,71,72 Despite containing a higher proportion of anti-coagulant proteins (GF5 to GF-8) compared to pro-coagulant proteins (GF-1 to GF4) in WI RVV (Table 1a), in in vitro conditions crude RVV demonstrated pro-coagulant activity because pro-coagulant potency of the high-molecular-weight proteins surpasses the anti-coagulant activity of low-molecular-weight proteins (Table 3).2,18 There are several ways by which RVV interferes with the hemostatic system of victims.56 FXa activator pro-coagulant proteases such as RVV-X activators (accession nos. gi|251205, gi|73621140, gi|300079896, and gi|300079900) present in WI RVV displayed remarkable prothrombin activation property (Figure S3A,B). Further, activation of FV to FVa by RVV-V activator (accession no. gi|134129) accelerated factor Xacatalyzed prothrombin conversion by several folds.73,74 Therefore, it is rational to assume that FV activating, pro-coagulant serine protease isoenzymes as well as FVa like components of WI RVV (accession no. gi|134129) activates circulating factor V which in turn interferes with coagulation process.20 The LCMS/MS analysis unequivocally demonstrated the elution of most of the pro-coagulant SVMPs as well as SVSPs in GF-1 to GF-4 fractions (Tables 2 and 3), thus showing a good correlation of proteomic data with in vitro pharmacological activity of WI RVV. Furthermore, neutralization of both enzymatic and pro-coagulant activity of SVMPs and SVSPs of WI RVV by PAV or MAV to a similar extent (see below) advocated involvement of SVMPs and SVSPs in inducing procoagulant activity. Thrombin-like SVSPs of RVV also coagulate the blood.21,56 Crude WI RVV in in vitro conditions did not show fibrinogen clotting activity, albeit GF-2 and GF-3 fractions converted human fibrinogen to tenuous fibrin clots (data not shown). This pharmacological property may be correlated with the predominance of Russelobin-like21 fibrinogen clotting SVSPs (accession nos. gi|765684342 and gi|38146946) in these fractions (Table 2). Several of the RV envenomed patients from Maharashtra, WI, showed clinical features of enhanced vascular permeability, hypotension, and thrombocytopenia (decrease in circulating platelets).3−5 Snake venom VEGFs, also identified in WI RVV (Table 2), exhibit potent hypotensive and enhancement of vascular permeability effects.67 To date, the precise mechanism(s) of RV-induced thrombocytopenia is unknown. Crude WI RVV and gel filtration fractions GF-1 to GF-5 significantly reduced the platelet count (thrombocytopenia); nevertheless, the highest activity was displayed by GF-1 (Table 3). Proteomic analysis demonstrated that the predominance of snaclec and/or SVMPs in GF-1 to GF-5 fractions of WI RVV (Table 2) may be accountable for thrombocytopenia (Table 3).75,76 However, the contribution of other protein(s) in RVV-induced thrombocytopenia needs to be explored. WI RVV exhibited aggregation of platelets in a dose-dependent manner (Figure 6). The occurrence of considerable amounts of platelet-modulating proteins, such as PDE, NT, LAAO, SVSP, snaclec, disintegrins, and Rusvikunin complex (KSPI) (Table 2 and Figure 4), explains the in vitro potent platelet aggregation property of WI RVV (Figure 6)28,33,43,44,63,71,77

Figure 6. Concentration-dependent platelet modulating activity of crude RVV on PRP obtained from goat blood. Values are mean ± SD of triplicate determinations. Significance of difference with respect to control, *p < 0.05.

Other major clinical features of RV envenomation are systemic bleeding from vital organs, bleeding from gums, hematemesis (blood vomiting), hematuria (blood in urine), subcutaneous ecchymosis (bruises), and hemorrhage.2,4,5,7 The in vivo hemolysis observed in RV envenomed victims2 may be associated with the indirect hemolytic activity exhibited by RVV PLA2s (Table 3).28,34,48 The abundant pro-coagulant SVMPs and SVSPs of WI RVV (Tables 1a and 2, and Figure 4) are responsible for consumption coagulopathy (uncoagulable blood) in RV bite patients.2,7,33,56 Subsequently, abundant anti-coagulant proteins such as PLA2, KSPI, and snaclec of WI RVV (Tables 2 and 3, and Figure 4) also exert anti-coagulation in victims.22,28,35,44,48,78 Further, hyaluronidase of WI RVV (Table 1b) acts as a spreading factor to augment the bleeding disorder and lethal potency of RVV.79 RV envenomation also resulted in acute renal failure in victims from western Maharashtra (WI),5,6 and about 7% of the patients demonstrated neurotoxic symptoms like severe ptosis with flaccid paralysis,3 which are uncommon clinical manifestations in RV bite in eastern India.2 The proteomic analysis has identified three neurotoxic PLA2s contributing 3.6% of WI RVV proteome. Notably, neurotoxic symptoms were also observed in NSA mice model injected with Rusvikunin complex isolated from Pakistan RVV.22,44 Occurrence of Rusvikunin-like complex (∼15−27 kDa) in WI RVV was demonstrated in GF-7 fraction (Figure 3C). Moreover, the MS/MS derived tryptic peptides detected in ion-exchange fractions of GF-7 demonstrated sequence similarity with KSPIs (the major component of Rusvikunin complex) from Pakistan RVV.22 Taken together, neurotoxic PLA2s and Rusvikunin-like protein complex may collectively be responsible for inducing neurological symptoms in RV bite patients in WI. Probably cytotoxic PLA2s, PLA2 homologues, and RVV-X are the primary causes of acute renal failure in RV bite victims.70,76 Therefore, the above clinical features may be well correlated with presence of high proportion of PLA2 as well as RVV-X enzymes in WI RVV (Table 2 and Figure 4). The ATPase enzyme may induce aggressive shock in victim/prey by depletion of ATP.3,5 Taken together, proteomic analysis has provided substantial evidence to address the clinical manifestations shown by RV bite patients in WI. K

DOI: 10.1021/acs.jproteome.6b00693 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research Neutralization Studies and Immuno-Profiling of WI RVV against Commercial Antivenom

Intravenous administration of equine antivenom is the only choice of treatment for snakebite. However, in addition to variations in venom composition, several other factors, including poor immunogenicity of low-molecular-weight venom components,22,44,66 are severe impediments in the development of an effective antivenom. 12,80 Therefore, improvement in efficacy and availability of antivenoms should be the primary target for snakebite management. PAV and MAV exhibited differential neutralization of tested enzymatic and pharmacological properties of WI RVV (Figure 7A,B); nevertheless, the neutralization potency of MAV surpassed that of PAV, in agreement with our previous observation.18 Both PAV and MAV neutralized the enzymatic activity of high-molecular-weight proteins such as ATPase, ADPase, AMPase, hyaluronidase, PDE, LAAO, and SVMP to a significantly higher extent (p < 0.01) as compared to neutralization of relatively low-molecular-weight RVV enzymes such as TAME, BAEE, PLA2, and SVSPs (Figure 7A), suggesting high-molecular-weight proteins served as better antigens. An interesting observation made in this study was poor neutralization of PLA2 activity of crude WI RVV (Figure 7A), although immunoblot analysis indicated better recognition of GF-5 containing most of the PLA2s (Figure 8C,D). This result suggests that the catalytic site of RVV PLA2 enzyme is a poor immunogen. Immuno-profiling of crude RVV and its GF fractions by ELISA (Figure 8A) and immunoblot analysis (Figure 8B−D) also suggested that WI RVV proteins were better recognized by MAV compared to PAV. Further, immunological cross-reactivity of PAV and MAV with highand mid-molecular-mass range RVV proteins (GF-1 to GF-6) was found significantly higher (p < 0.05) compared to recognition of abundant of low-molecular-mass components (