Identification of cobra and viper bites in southern India using enzyme-linked immunosorbent assay

Abstract

Introduction

Snakebite-related deaths are major concerns in populations worldwide that lack access to medical care. Current treatment of snakebites includes administration of polyspecific and polyvalent heterologous equine antiserum. Swift diagnosis of the species of snake responsible is essential to initiate a specific treatment.

Methods

We generated murine monoclonal and rabbit polyclonal antibodies against the venom of the 3 most common venomous snakes in southern India: the Russell viper, the saw-scaled viper, and the Indian cobra. We developed an enzyme-linked immunosorbent assay to distinguish the snake by analyzing the exudates from the snakebite wound.

Results

Monoclonal antibody EC3 reacts specifically with a 45- to 50-kDa venom protein of the saw-scaled viper, antibody NNH6 reacts with 12-kDa venom protein of the Indian cobra, and antibody RVV20 reacts with a 16-kDa venom protein of the Russell viper. We tested the exudates from snakebite wounds in 24 consecutive patients admitted to the Tirunelveli Medical College Hospital with snakebite. Our assay detected 18 patients with Russell viper bite and 1 each for cobra and saw-scaled viper.

Discussion

Our study shows the feasibility of an enzyme-linked immunosorbent assay–based method to identify the 3 major venomous snakes in southern India and holds promise for prompt administration of snake-specific and mechanism-based treatment.

Introduction

Snakebite-associated morbidities and mortalities are a major concern in populations worldwide that lack access to swift medical care.¹ Venomous snakes a devastating environmental and occupational hazard, especially in rural areas of tropical low- and middle-income countries.² Worldwide, it has been estimated that 421 000 to 1 841 000 snakebites occur each year, and 20 000 to 94 000 deaths per year are attributed to those bites.³ According to some estimates, the annual mortality due to snakebite in India may be as high as 58 000 per year.⁴ In the southern Indian state of Tamil Nadu, 3 venomous snakes—namely, the Russell viper (Daboia russelii), the saw-scaled viper (Echis carinatus), and the Indian cobra (Naja naja)—are primarily responsible for most of the snakebites, with the Russell viper being the most common.⁵ The aggressive behavior of these snakes and human encroachment into their natural habitat for irrigation are responsible for an increasing incidence of Russell viper bites.⁶

Viperid snake venoms contain several proteins belonging to 11 different protein families and consist of enzymes (serine proteases, phospholipase A2, and metalloproteases), disintegrins, and snake type C lectins.^7,8 The 2 procoagulant enzymes from D russelii have been characterized molecularly, one activating coagulation factor X⁹ and the other activating coagulation factor V.¹⁰ Together, these enzymes cause severe venom-induced consumption coagulopathy characterized by a fall in plasma coagulation factors leading to fatal hemorrhage, circulatory collapse, and kidney failure.^11,12 In addition, proteins and phospholipases with action similar to vascular endothelial growth factor present in the venom increase vascular permeability, leading to localized swelling, pain, and necrosis at the site of the snakebite.^7,8 Furthermore, abdominal pain, acute kidney failure, neurotoxicity, myotoxicity, cardiac failure, pericardial effusion,¹³ a motor neuron-type of facial palsy,¹⁴ and hypopituitarism have been recorded.¹⁵ The toxins that cause these clinical symptoms are not fully characterized. E carinatus venom has a strong activator prothrombin that induces severe venom-induced consumption coagulopathy similar to D russelii.¹⁶ Furthermore, E carinatus venom induces the activation of neutrophils, prompting the release of their nuclear and granular contents into the extracellular space, thereby creating neutrophil extracellular traps (NETs), a phenomenon known as NETosis. These NETs entrap the venom toxins at the local site, causing occlusion of blood vessels that contributes to severe tissue necrosis at the bite site.¹⁷

In contrast to viperid venoms, venom from elapid snakes such as N naja contains neurotoxins (belonging to the “3-finger toxin” superfamily of proteins) that block the binding of acetylcholine to its receptors in the postsynaptic membranes of muscle cells in the neuromuscular junction and to neuronal nicotinic receptors in the central nervous system.¹⁸ In addition, the Indian cobra venom proteome contain metalloproteases, phospholipases, and cardiotoxins.¹⁹ Death usually occurs from respiratory paralysis.

To administer a rational, mechanism-based, appropriate treatment for a specific snakebite, it is essential to identify the snake species responsible because the pathogenic mechanisms for each snake are different. We have developed a monoclonal/polyclonal sandwich enzyme-linked immunosorbent assay (ELISA)–based method to distinguish the 3 most common snake venoms in southern India quickly so that clinicians can administer snake-specific treatment in a timely manner. We tested the ELISA’s utility in a clinical setting.

Methods

Reagents

Peroxidase-labelled goat antirabbit antibody, polyacrylamide/azlactone matrix (UltraLink Biosupport [Thermo Fisher Scientific]), and 1-Step Ultra TMB-ELISA (Thermo Fisher Scientific) reagent were purchased from Invitrogen Inc. Venoms from N naja (catalog No. V9125), D russelii (catalog No. V2501), and E carinatus (catalog No. V8250) were obtained from Sigma-Aldrich USA. Precast gels for polyacrylamide gel electrophoresis were obtained from Bio-Rad Laboratories. Standard laboratory chemicals and tissue culture reagents were obtained from Thermo Fisher Scientific. HiTrap protein L columns and protein G columns were obtained from VWR Life Sciences. All animals were treated in accordance with the protocol approved by the Institutional Animal Care and Use Committee of the Baylor College of Medicine, Houston, Texas (AN-5063).

Generation and characterization of antibodies

BALB/c mice were immunized with denatured snake venom (boiling in 0.1% sodium dodecyl sulfate for 5 minutes). The mice were immunized in total 3 times, with 50 µg denatured venom in Freund incomplete adjuvant each time. Subcutaneous injections were carried out each time, with an interval of 2 weeks between each immunization. The procedures for fusion, screening of hybridomas, cloning, immunoglobulins (Ig) purification, and class and subclass characterization were carried out following the procedures described previously.²⁰ The antibodies were selected solely on basis of reactivities in ELISA. The clones were grown in RPMI1640 media containing 10% fetal calf serum, and the monoclonal antibodies (mAbs) were isolated from tissue culture supernatant on protein L-agarose. Purified mAbs (5 mg) were immobilized separately in a polyacrylamide/azlactone matrix (5 mg/mL gel), and the cognate antigens were isolated by immunoaffinity chromatography from the crude venoms. These were then used to immunize rabbits to generate the polyclonal antibodies used for the ELISA.

For immunoblot assays, 5 µg crude venom was electrophoresed in sodium dodecyl sulfate–polyacrylamide gel electrophoresis in 5%- to 20%-gradient gels, transferred to polyvinylidene fluoride membranes and incubated with tissue culture supernatants from hybridoma or polyclonal antiserum (1 hour at room temperature), then developed with corresponding species-specific peroxidase-labelled secondary antibodies.

Confirmation of specificity against venom of Indian snake species

D russelii, E carinatus, and N naja venom was procured from the Tamil Nadu Government Irula Co-operative Society under a protocol approved by the Institutional Biosafety Committee (No. RCB/IBSC/20-21/211) of the Regional Centre for Biotechnology in Faridabad, India. The antibodies were tested against these native snake venoms in ELISAs and immunoblot assays.

Clinical sample collection

Samples were collected from snakebite patients admitted to Tirunelveli Medical College Hospital, Tamil Nadu, India. In the emergency department (ED), the snakebite wound was examined. The exudates from the wound were collected after the patient provided informed consent. The wound was cleaned and the gently massaged with sterile cotton swabs. Up to approximately 500 µL of exudate was collected and frozen. The interval between the snakebite and the patient’s arrival at the ED varied from 4 to 24 hours. The protocol for human participants in research was approved by the institutional ethics committee for protection of human participants in research in Tirunelveli Medical College and the Regional Centre for Biotechnology (protocol No. RCB-BBB-IEC-H-36). The frozen samples were sent to the Regional Centre for Biotechnology for further testing. All patients survived the snakebites. The whole blood clotting time was determined at the bedside in the ED. Prothrombin time/international normalized ratio testing and hemoglobin level, white blood cell count, blood urea level, and creatinine level measurements were conducted in the clinical laboratory. Clinical information is presented in Table 1.

Table 1.

Clinical details of the snakebite patients.

Patient	Age, y/sex	Location of bite	Pulse, /min	Blood pressure, mm Hg	Hemoglobin,a g/dL	Total white blood cell count,b µL	Clotting time,c min (<8)	Prothrombin timed; prothrombin time/international normalized ratioe; partial thromboplastin time,f s	Urea,g mg/dL	Creatinine,h mg/dL
1	50/F	Left hand	90	160/90	12.5	8600	>20	21.90; 1.75; 39.5	32.0	1.1
2	63/M	Middle finger	78	150/90	13.7	20 800	>20	19.80; 1.58; 42.5	32.7	0.8
3	60/M	Left ankle	NR	NR	NR	NR	NR	NR	NR	NR
4	72/M	Right forearm	NR	NR	NR	NR	NR	NR	NR	NR
5	21/F	Left finger	86	110/60	14.1	8900	<20	15.70; 1.18; 37.4	26.3	1.0
6	67/F	Right leg	86	190/70	7.1	9400	<20	14.80; 1.11; 31.6	17.9	0.6
7	31/M	Right finger	82	120/80	17.4	11 400	<20	13.80; 1.02; 23.0	14.4	0.7
8	53/M	Left foot	89	200/100	NR	NR	<20	14.50; 1.08; 22.9	11.5	8.3
9	40/M	Right hand	90	120/80	14.2	18 300	>20	14.50; 1.08; 21.1	26.6	0.7
10	62/M	Left leg	94	130/80	14.8	9600	<20	NR	29.0	1.1
11	44/M	Right leg	92	130/80	14.6	12 100	<20	17.80; 1.39; 22.6	26.8	1.28
12	55/M	Left foot	106	130/90	14.5	6300	>20	NR	35	0.76
13	31/M	NR	NR	NR	NR	NR	NR	NR	NR	NR
14	50/M	NR	NR	NR	NR	NR	NR	NR	NR	NR
15	42/M	NR	NR	NR	NR	NR	NR	NR	NR	NR
16	37/F	NR	NR	NR	NR	NR	NR	NR	NR	NR
17	42/M	NR	NR	NR	NR	NR	NR	NR	NR	NR
18	55/F	Right hand	90	120/80	10.4	9400	>20	NR	37	1.40
19	35/F	Left foot	82	110/70	10.2	6900	<20	NR	24.6	0.75
20	52/F	Left leg	92	165/100	16.7	9900	<20	NR	22	1
21	32/F	Left foot	87	120/80	11.3	13 000	<20	NR	29.7	0.76
22	40/F	Left foot	88	130/90	11.8	8600	<20	NR	17.5	0.84
23	52/M	Left foot	82	130/80	14.8	9900	<20	NR	34.1	1.29
24	53/M	Left foot	57	140/40	12.6	9400	>20	NR	21.3	1.42

Abbreviations: F, female; M, male; NR, not recorded.

SI conversion factors: To convert hemoglobin to g/L, multiply by 10. To convert white blood cell count to ×10⁹/L, multiply by 0.001. To convert urea to mmol/L, multiply by 0.357. To convert creatinine to µmol/L, multiply by 88.4.

^aReference range: 12-15 g/dL.

^bReference range: 4000-10 000 µL.

^cReference range: <8 minutes.

^dReference range: 11-13.8 seconds.

^eReference range: 0.9-1.5 seconds.

^fReference range: 26-35 seconds.

^gReference range: 15-43 mg/dL.

^hReference range: 0.60-1.20 mg/dL.

ELISA of wound exudate

The wells of 96-well microtiter plates were coated with 1 µg purified mAbs separately in sodium bicarbonate (0.1 M) buffer and stored at 4 °C for 24 to 48 hours before the assay. The wells were blocked for an hour with 5% bovine serum albumin in phosphate-buffered saline. The wound exudates (100 µl per well) were incubated in the ELISA plate well in duplicate for 1 hour. The wells were then washed and incubated with the corresponding rabbit polyclonal antibodies for 1 hour and washed. The bound antigens were detected by peroxidase-labelled goat antirabbit antibody and Ultra TMB-Elisa substrate. All incubations and blocking steps were performed at room temperature without agitation.

Results

Specificity of the antibodies

We generated venom-specific mAbs using hybridoma technology from the splenocytes of immunized BALB/c mice. Monoclonal antibody EC3 reacts with a 45- to 50-kDa antigen in immunoblots in E carinatus (Figure 1A). Monoclonal antibody NNH6 is directed against a 12-kDa venom protein of N naja (Figure 1B), and mAb RVV20 reacts with a 16-kDa protein in the venom of D russelii (Figure 1C). The mAbs EC3 and NNH6 do not cross-react with the other 2 venoms in immunoblots (Figure 1A-1C). RVV20 showed some cross-reactivity with E carinatus venom, but the cross-reactivity is at least 1 order of magnitude less. Furthermore, using specific antibodies for E carinatus and N naja simultaneously, we can easily discriminate between the venoms. We isolated the corresponding antigen by immunoaffinity chromatography of the venoms on immobilized mAb on a polyacrylamide/azlactone matrix from each venom and developed polyclonal antibodies against them by immunizing rabbits with the purified antigens. The specificities of the corresponding polyclonal antibodies are shown in (Figure 1D-1F). The polyclonal antibodies, as expected, showed some species cross-reactivity compared with the corresponding mAbs.

Figure 1.

Specificity of the antibodies used. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis of 5 µg unfractionated venom was performed in 5%- to 20%-gradient gels; the results were transferred to polyvinylidene fluoride membranes and incubated with tissue culture supernatants from hybridomas: (A) mAb EC3, (B) mAb NNH6, (C) mAb RVV, or (D, E) polyclonal rabbit antiserum for 1 hour and developed with peroxidase-labelled goat antimouse (A-C) or goat antirabbit (D-F) antibodies. Abbreviation: mAb, monoclonal antibody.

Reactivity of snakebite wound exudates

We developed a sandwich ELISA using the purified mAbs as capturing antibodies and the polyclonal antibodies for the detection of the cognate antigen. We tested the exudates from snakebite wounds from 24 consecutive snakebit patients admitted to the Tirunelveli Medical College Hospital ED. Patients’ clinical information is presented in Table 1. All the patients survived the bite. In 18 patients, the wound exudate showed clinically significant reactivity to anti–Russell viper venom antibody (Figure 2) but not to the antibodies to the venoms of E carinatus or N naja. These results are consistent with the fact that the Russell viper is the most common snake in this region. Only 1 bite was clearly identified as coming from a cobra (Figure 2B)²¹ and another one from E carinatus (Figure 2D).¹⁷ The identity of the snake could not be identified in 4 patients possibly because little venom was present in the swab collected from the snakebite site, delay in seeking medical care, or the detection limits of the method used.

Figure 2.

Sandwich ELISA of the wound exudate. ELISA plate wells containing immobilized mAbs to Naja naja, Daboia russelii, or Echis carinatus were incubated separately with exudates from the snakebite wound (100 µL) for 1 hour at room temperature, and the bound venoms were detected in polyclonal rabbit antibodies followed by peroxidase-labelled goat antirabbit antibodies. (A) The color development and (B-D) graphs for optical densities (at λ 450 nm) are shown for each patient for 3 venoms: (B) Naja naja, (C) Daboia russelii, and (D) Echis carinatus. For positive controls (first 3 rows in A), 100 to 500 ng unfractionated venom was added to the corresponding wells instead of wound exudate. Abbreviations: –, negative control; +, positive control; ELISA, enzyme-linked immunosorbent assay; mAb, monoclonal antibody; OD, optical density.

Discussion

When a patient with snakebite is brought to medical attention, the identity of the snake is uncertain in most cases. The patients are anxious, and they may misidentify a snake or attribute the symptoms to other, unrelated injuries, such as a thorn prick or an insect bite. Furthermore, snakebites often occur at night, which produces additional strain on limited hospital staff. There is a substantial overlap in the symptoms of different snake species, and identifying the responsible snake based solely on symptoms may be misleading.²¹ To correlate the pathologic symptoms with a specific snake species and to provide a snake-specific mechanism-based treatment, quick and precise identification of the snake species is crucial.

Current treatment of snakebites includes supportive care and administration of polyvalent heterologous equine antiserum,²² even though the pathophysiologic mechanisms vary. This approach results in waste of resources and exposes patients to unnecessary heterologous serum and risk of severe anaphylaxis on subsequent exposure. If the identity of the snake is known, one can pursue a more targeted, mechanism-based therapy with specific inhibitors. Early anticoagulation may be beneficial in viperid snakebites because they cause severe venom-induced coagulopathy and bleeding from depletion of coagulation factors by activating the coagulation cascade. Recently developed coagulation factor–specific inhibitors could be a rational treatment for viperid venom–induced coagulopathy. For elapid snakebites, mechanical ventilatory support will be more appropriate. Rapid identification of the snake species is essential for these strategies because different pathophysiologic mechanisms are in play. The assay described in this article is simple and straightforward, and it can be performed effectively in 3 to 4 hours without the use of sophisticated instruments or technologies. The reagents can be modified for lateral flow assays used in rapid antigen testing.

In summary, we have developed a monoclonal/polyclonal antibody–based sandwich ELISA to identify the species of the snake from exudates collected from snakebite wounds to facilitate more specific individualized treatment.

Conflicts of interest

The authors declare no commercial or financial conflict of interest.

Funding

The authors acknowledge the funding by grant-in-aid from the Regional Centre for Biotechnology, Department of Biotechnology, Government of India, to P.G. and from US National Institutes of Health grant No. HL13950 to P.T.