Ferret badger rabies origin and its revisited importance as potential source of rabies transmission in Southeast China

Epidemiological survey on FB-associated human rabies cases

A retrospective survey on rabies epidemics was carried out in Zhejiang, Jiangxi and Anhui provinces in southeast China. The data were collected and summarized from the provincial CDC surveillance system and epidemiological records. Some information was obtained verbally after interviewing animal traders or hunters in the endemic areas. In human rabies cases and its potential association with FB transmission, we conducted a preliminary investigation of FB population density, exposure frequency of sick FB to humans, and management of rabies PEP after a FB bite.

FB and dog brain specimens and Fluorescent Antibody Test (FAT)

Dead or live FB collected in fields and houses, and dogs that had bitten people were sent by selected farmers to the designated laboratories for diagnoses. The surveyed mountainous areas were Huzhou, Hangzhou, Jinhua, Quzhou, Lishui in Zhejiang province; Nanchang, Jingdezhen, Shangrao, Wuyuan in Jiangxi province; and Huangshan, Xuancheng, Anqing, Jingxian in Anhui province. The animal heads were packed appropriately and shipped to our laboratory under cold conditions. The brains were removed by opening the skull under sterile conditions.

Animal experiments related to this study were approved by the Committee of Animal Welfare and Ethics of Veterinary Research Institute, Academy of Military Medical Sciences. Humane endpoints were used throughout this study in accordance with the ethical principles for in vivo studies. All animals including ferret badgers, dogs and mice that showed clinical signs of infection were killed humanely. This study did not refer to any issue of human ethics as only the epidemiological data were collected and analyzed, and no sample was collected from any healthy and contracted humans.

The FAT on brain specimens was performed according to methods described elsewhere [14]. Briefly, the brain tissue impressions were made on slides. After air drying, the slides were fixed with 80% acetone for 20 min. The FITC-conjugated anti-rabies virus nucleoprotein monoclonal antibodies (made in Laboratory of Epidemiology, Veterinary Institute, Changchun, for routine rabies diagnosis) was added and incubated for 40 min at 37ºC. After three-times' wash using PBS buffer (0.01 M, pH 7.4) -Tween-20 (0.1%), the slides were left to dry in the air at ambient temperature, and observed under a UV fluorescent microscope (Zeiss Corporation, Germany) for typical rabies virus staining.

Mouse inoculation test (MIT)

For FAT positive specimens, the suspension of the sample was injected intracerebrally to 1-day-old suckling mice (Kunming mice, Animal Core Laboratory, Changchun Institute of Biological Products) according to protocols described elsewhere [15]. The mice were observed for ~28 days. The brain smear was made to re-check for the presence of rabies virus antigen using the FAT if the inoculated mice died from day 6 to 28.

Sequences of the nucleoprotein and glycoprotein genes and phylogenetic analysis

Total RNA of the infected FB brains were extracted using Trizol reagent (Invitrogen, Carlsbad, CA). The nucleoprotein (N) and glycoprotein (G) genes were amplified by RT-PCR following the protocol described by Nadin-Davis et al [16] using the total RNA with the following primers. The positions of the primers were referred against strain FJ009 (GenBank Acc. No. FJ866836.1).

RV-N-F (1-25): 5'-ACGCTTAACAACAAAACCATAGAAG-3';

RV-N-R (1515-1538): 5'-CGGATTGACGAAGATCTTGCTCAT-3';

RV-G-F (3291-3315): 5'-CATCCCTCAAAAGACTTAAGGAAAG-3';

RV-G-R (4918-4941): 5'-CCGAGGAGATGAGGTCTTCGGGAC-3',

The TaKaRa (TaKaRa Corp Ltd, Dalian, China) was the contracted company responsible for sequencing the amplicons. We generated a phylogenetic tree using the Neighbor-joining (NJ) method in MEGA 4 (MegAlign, DNASTAR Software Suite, Version 7.1.0, Copyright 1989-2006, DNASTAR Inc.). The Bootstrap values were calculated from 1,000 repeats using 70% as the cut-off.

Rabies virus (RV) neutralizing antibodies in dog and FB serum samples

Serum samples were collected from the captured FBs and dogs in our expanded surveyed areas. All the FBs were alive and appeared healthy when the serum was collected. Dog sera were from watchdogs belonging to residents of different villages in Zhejiang, Jiangxi and Anhui provinces. The method of titration of virus neutralizing antibodies (VNA) by Fluorescent Antibody Virus Neutralization (FAVN) test was described elsewhere [17]. Briefly, 3-fold serial dilutions of standard serum (0.5 International Units, IU/ml) and test serum samples were prepared in microplates in quadruplicate. Fifty _μl of challenge rabies virus (CVS-11) containing 100 TCID₅₀ was also added to each well. After 60 min incubation at 37°C in a humidified 5% CO₂ incubator, 50 μl cell suspension containing 2 × 10⁴ cells was added to each well and the plates were incubated for 48 h at 37°C. After fixation at room temperature for 30 min in 80% acetone, the cell monolayers were stained by addition of FITC-conjugated anti-rabies nucleoprotein monoclonal antibodies (Laboratory of Epidemiology, Veterinary Institute, Changchun, China) to each well. Staining was carried out in an incubator at 37°C for 30 min and fluorescence was observed by UV microscope (Zeiss, Germany). Wells exhibiting no specific fluorescence were considered antibody positive. Neutralizing antibody titers were calculated using the Spearman-Kärber formula and expressed in IU/ml by comparison with a reference serum (13.5 IU/ml, product of AFSSA, France), 0.1 IU/ml is considered as the cut-off of seroconversion after rabies virus infection in both ferret badgers and dogs.

Epidemiological survey on FB-associated human rabies cases and preliminary ecological investigation on FB population

The population density of FBs is unknown. Empirically, hunters lay their traps according to the number of FB tracks. On average, 2-5 tracks could be found within 80 Chinese acres (1 Chinese acre = 660 m²). The number of FB is about 38-95 per km². However, this rough estimation needs ecological support. In our investigation, there was at least one hunter in almost every village. Seasonally, more than 1 FB was caught by a hunter per day. Despite frequent contact with rabies- susceptible animals, no hunters are aware of the potential danger of rabies exposure. No PEP occurred after FB bites. Furthermore, the hunters and farmers live in remote rural areas and mountainous regions, and their income (1000 RMB or less per month) cannot afford the high cost of PEP (1300-2000 RMB) in China. Therefore, the recommendation of PEP is not only ignored, but also intentionally neglected due to the associated costs.

FAT and MIT

In total, we collected 209 dead FB from our surveyed regions, and 76 of 209 were partially decomposed when the brains were removed for diagnosis. Eight of 209 brain samples were confirmed rabies positive using FAT (3.8%), and RV was successfully isolated by MIT from the samples. The rabies incubation period in 1-day-old suckling mice was 7-12 days.

From the 56 brain samples we collected from live FBs in Jiangxi, Anhui and Zhejiang provinces, only 1 sample was rabies positive by FAT (1.8%, 1/56). Since captured and injured by the hunter, this positive animal did not present obvious clinical signs of rabies before euthanized for diagnosis. From the 15 dog brain samples collected in the same regions, 2 were rabies positive by FAT (13%, 2/15). All RVs from FAT positive samples were successfully isolated by MIT.

Rabies VNA in FB and dog serum samples

Sequence and phylogenetic analysis

We compared RV N sequences from rabid FB isolates with those in dogs from China, foxes from Europe/Middle East, dogs in Africa, and rabies vaccine strains from Asia, Europe and the US (Figure 1). Dog RVs in China were categorized into two groups. Group 1 compromises RVs all across south and southeast China, and group 2 is formed by Chinese isolates distributed in Guizhou, Jiangsu, Henan, Jilin provinces, and is closely related to Asian and occidental vaccine strains, and cosmopolitan dog RVs (Figure 1). In general, Chinese dog-associated lineages shared nucleotide identity from 87 to 98% (Figure 1). Same phylogenetic pattern was reconstructed using the G gene sequences (data not shown).

The FB RVs were segregated in an independent lineage (Figures 1). The nucleotide identity of FB isolates with current south and southeast Chinese dog-related lineages (from Zhejiang, Jiangxi, Fujian, Jiangsu, Hunan, Guangxi, Guizhou, Hubei, Yunnan, Henan) ranged from 87 to 89% (94.5-98.8% homology by amino acid sequence comparison). However, the intra-group identity in the 9 FB isolates only varied from 98.7 to 99.1% (99.0-100% homology by amino acid sequences). A notable difference was detected on the G protein motif at residual position 331-333 with Serine-Valine-Arginine (S-V-R) in FB isolates. This motif has not been found in any isolates from dogs in China and the corresponding motif in dog isolates was Serine-Isoleucine-Arginine (S-I-R), which does not exist in the FB isolates (using PV as the reference sequence, GenBank Acc. No. M13215).