Detection of Nipah virus in Pteropus medius in 2019 outbreak from Ernakulam district, Kerala, India

The study was approved by the Institutional Animal Ethics Committee (IAEC) of ICMR-NIV, Pune (IAEC/2019/MEZ/04). The permission from Principal Chief Conservator of Forest, Kerala state was obtained for trapping of bats and performing this study.

Bats were captured from the five-kilometer radius of the index case residence (Ernakulum district) and the college area (Idukki district) between 5th and 10th day after detection of index case. The five sites from where the bats were trapped were: Vavakkad, Aluva, Thuruthipuram in Ernakulam district and Thodupuzha, Muttam in Idukki district as per the proximity to the residence and college (Fig. 2).

The selected sites except for Thuruthipuram had bat roosts with numbers ranging from a hundred to several thousand. Thuruthipuram was chosen due to the presence of guava and almond tree plantations, frequently visited by bats, which were just 100–200 m away from the residence of the index case. Vavakkad site was within 5 km from the index case house and had a permanent roost with approximately 200 bats. Thodupuzha site was within a 5 km radius from the college, where the index case was studying and had a massive roost (nearly 10–20 thousand bats), spread over about half an acre of land. Muttam was towards the Idukki dam area and was one of the oldest roosts in the area spread over an acre of land with approximately 50,000 bats.

Bats [Pteropus medius (109) and Rousettus leschenaultii (32)] were trapped randomly using mist nests hoisted on aluminum inter-connectible poles at the height of 48 to 56 ft. The nets were set up to trap the bats between 6 and 7 PM before the foraging activities of the bats were initiated. The trapped bats were collected in the early morning (4 to 6 am) and were anesthetized with isoflurane inhalation. Throat and rectal swabs specimens were collected from 141 bats in virus transport medium (HiViral™ Transport Medium, HIMEDIA) and stored immediately in dry ice. The blood samples (around 2–3 ml) were collected from the wing (cephalic) vein of 78 bats [Pteropus spp (n = 58) and Rousettus spp (n = 32)] and serum was separated. After recovery from anesthesia, the bats were released back [7, 8]. All the samples (serum, throat, and rectal swabs) were transported to ICMR-NIV, Pune in dry ice (Table 1). Ninety-two bats from five sites were euthanized with Isoflurane inhalation overdose. They were transported to ICMR-NIV, Pune, in liquid nitrogen to check for the presence of the NiV RNA in visceral organs. Necropsy was performed on 92 captured bats taking all biosafety precautions in the Biosafety Level (BSL)-4 laboratory at ICMR-NIV, Pune. The organs (brain, heart, lung, liver, spleen, intestine, kidney, and reproductive organs) were collected aseptically. These organs were then triturated in minimum essential medium (MEM) (Life Technologies Corporation, USA), and tissue homogenate was stored at − 80⁰ C for further use.

RNA was extracted from the throat, rectal swabs and tissue homogenates using the Magmax RNA extraction kit (Applied Biosystems, USA) as per manufacturer’s instructions. NiV-specificqRT-PCR was performed as described earlier [9].

Efforts were made for the amplification of bat specimens using the PCR multiplexing method. The multiplexing reaction was performed with the primer pools designed using the Primal Schema tool http://​primal.​zibraproject.​org/​ using Kerala NiV sequences MH523640 [11]. The primers designed for the complete NiV sequencing were of 400 base pairs (bps) with an overlap of 75 bps. The steps involved are as follows: 5 μl of the RNA (liver, spleen, kidney, lung, and rectal swab) was amplified using 35 cycles of PCR. The amplified PCR product was loaded on 1.5% agarose gel and electrophoresed. The PCR products were purified using the Qiagen gel extraction kit (Qiagen, Germany). The amplicon generated from this extracted product was used from the second strand synthesis step, as described in the TruSeq Stranded mRNA LT Library preparation kit (Illumina, USA). This purified amplicon was then used further for library preparation as described in the TruSeq Stranded mRNA LT Library preparation kit (Illumina, USA), beginning at the second strand synthesis step as described earlier [12]. The generated reads were analyzed using CLC Genomics workbench version 11.0. Reference-based mapping was used to retrieve the genomic sequence of the NiV. Reference NiV sequences were downloaded from the GenBank database and were aligned using MEGA software version 7 [13]. An evolutionary tree was generated using the best-predicted model for the N gene of the Nipah virus. A bootstrap replication of 1000 cycles were performed to assess statistical robustness.

One (MCL-19-bat-572 Pteropus bat; collected from site-1 Thodupuzha, Idukki district) out of the total 141 rectal swabs collected was positive for NiV RNA by qRT-PCR while none of the throat swabs [n = 141] samples were positive. The visceral organs, including liver/spleen, kidney, lung, heart, reproductive organ, and intestine of MCL-19-bat-572 showed NiV RNA positivity [Cycle threshold (Ct) values ranged from 26.47 to 36.51]. MCL-19-bat-574 also collected from site-1 showed NiV positivity in liver/spleen, kidney, brain, lung, heart [Ct values ranged from 28.96 to 34.14]. MCL-19-bat-618 Pteropus bat trapped from site-3 (Aluva, Ernakulam district) showed positivity in Liver/spleen [Ct value 33.62] (Table 1).

Anti NiV-IgG bat antibodies were detected in 12/58 (20.68% positivity) Pteropus bat sera from sites 1,3,4 and 5, whereas none of the Rousettus sera were found positive (Table 1).

The complete genome of the NiV (18,172 nucleotide bps) could be retrieved from the kidney sample of one Pteropus bat, using the reference-based mapping approach [9]. NiV sequences could also be retrieved from the liver/spleen sample of the two Pteropus bats with a length of approximately 17,200 and 15,100 bps. The details of the NiV genomic regions retrieved from the bat samples are given in Supplementary Figure 1. The evolutionary tree demonstrated a distinct branch for the Indian NiV sequences (Fig. 3). NiV human sequences from Kozhikode, Kerala 2018 outbreak, and the NiV bat sequences from the current 2019 study area branched out from the Bangladesh sequences. The NiV strains circulating in the southern region of India are distinct from the B genotype and formed a separate cluster. The net-evolutionary nucleotide divergence found between the NiV strains from South India (Kerala) to Bangladesh and Malaysia strains is 1.96 and 8.24%, respectively. Gene-wise nucleotide and amino acid net-evolutionary divergence for Malaysia and Bangladesh strains concerning Indian sequences are given in Table 2. Bangladesh and West Bengal sequences showed a divergence of 0.75%, while Kerala and Bangladesh sequences showed the divergence of 2.2%. Hence we hypothesize the existence of two types of genotypic sequences in India and propose a separate genotype for Kerala, Indian strains as ‘I-India’ genotype. The percent nucleotide divergence for the sequences used in the study and the amino acids changes for the representative sequences are given as Supplementary files 1, and 2 respectively. The pairwise comparison for each NiV gene with respect to its reference Bangladesh strain is given in supplementary Table 3.