Introduction

Rabies is a widespread viral disease of warm-blooded species, including humans, which manifests itself in disorders of the central nervous system, paralyses and encephalomyelitis. The disease is one of the most dangerous and widespread zoonosis [1,2,3]. It is well known that the reservoir and main source of rabies virus infections are bats [4], dogs [5] and wild carnivores [6, 7]. However, the natural reservoir of rabies virus (RV) differs between different geographic regions. Foxes were an important reservoir of rabies in Europe; however, this has largely been controlled by vaccination. In different areas of North America, rabies is endemic in raccoons and skunks, and it was previously endemic in coyotes [8]. In Kazakhstan, foxes (Vulpes vulpes), corsacs (Vulpes corsac) and steppe wolves (Canis lupus campestris) are the most important RV vector species due to their wide geographic distribution, high susceptibility to RV, and tendency to inhabit areas around human dwellings and domestic animals [9]. An effective strategy for preventing rabies comprises the control of rabies in the host reservoir, through vaccination [10].

However, the traditional methods for vaccine administration (subcutaneous, intramuscular or aerosol inoculation) are impracticable when dealing with wildlife. For this reason an oral route for administering the vaccine, via ingestion of various vaccine baits by animals, seems to be the most feasible [1, 11].

A number of reports on oral rabies immunization of carnivores have been published recently. This experimental data confirms the feasibility of this idea of oral immunization [12,13,14,15,16,17,18,19]. Initiation of routine annual preventive vaccination of wild carnivores with existing commercial vaccines (Lysvulpen, SADB19 и SADP5/88, SAG2 и VRG) resulted in a rapid improvement of the epidemiologic situation in some European countries and only single cases of rabies infection are now being reported [20]. To date, Oral rabies vaccination (ORV) has been successfully applied to eliminate rabies among arctic fox (Vulpes lagopus) [21], red foxes (Vulpes vulpes) [22] and skunks (Mephitis mephitis) [23] in Ontario, Canada, with continued expansion of control programs for raccoons (Procyon lotor) into the eastern US and Canada [24, 25].

In the territory of Kazakhstan there are many species of predators, such as steppe wolves (Canis lupus campestris), foxes (Vulpes vulpes) and corsacs (Vulpes corsac). The currently used commercial vaccines are known to be effective only for red foxes and raccoon dogs. There is no information on the effectiveness and safety of these vaccines for wild carnivores such as corsacs and steppe wolves. During the last 4 years (2013-2016) in Kazakhstan the number of rabies cases in humans and wild and domestic animals has been too high [26,27,28]. As a result there is an urgent need for an oral rabies vaccine bait that can effectively immunize corsac foxes and steppe wolves, as these are the main rabies vector species in this region.

Materials and methods

Virus and preparation of the oral vaccine

The cultured strain VRC-RZ2 (Gene Bank accession number: KX009506.1) of rabies fix-virus was used in the study. It was derived from the tissue-and-organ RV isolate AZVI (Almaty Zoo-Veterinary Institute) [29] via alternate and successive passages in suckling mice and continuous BHK-21 cell culture, respectively. This virus strain (rabies virus VRC-RZ2, patent No. 17453 dated December 10, 2004 [30]) was recommended as a base for an oral rabies vaccine [31] since N gene sequencing has shown the strain is vaccine-like and is 99% identical to the Russian strain RV-97 (Gene Bank accession number: AY705430.1).

To make baits a 30% gelatin polymer solution was prepared with distilled water (Sigma–Aldrich, St. Louis, MO, USA) and kept overnight (18 h) for the polymer swelling to occur. Chicken eggs (Allel-Agro Poultry Factory, Kazakhstan) were homogenized, and then meat-and-bone meal was added (Almaty Meat Processing and Packing Factory, Kazakhstan). The polymer solution was heated to 60°C for 20 min, then cooled to 37°C and transferred to the glass with chicken eggs and meat-and-bone meal, before being homogenized at 3000 rpm for 10 min.

The mixture was poured into a special form with a blister containing the vaccine and then cooled at 4°C for 1 h. The resulting vaccine pellet weighing 25-30 g had a specific scent and a light to dark brown color. The bait contained the RV vaccine virus in a liquid state at a dose of 106.75 TCID50 (the volume of vaccine was 3 mL). The ready bait container looks like a hollow parallel pipe of size 5 × 3×2 cm. The bait pellet contains 1 dose of vaccine.

Animals

For the study, forty-two 3–12 month old female stray mongrel dogs (12-15 kg), 15 corsac foxes (Vulpes corsac) and 15 steppe wolves (Canis lupus campestris) were used. Corsac foxes and wolves (not rabies vaccinated) were obtained from Almaty Zoo. Corsac foxes and wolves were kept alone in special cages. The animals were individually identified with a nylon collar and a numbered tag. Dogs were kept on a leash in a traditional animal facility in four rooms until challenge. Each group was in a separate room (i.e. vaccinated animals were separated from controls). They were fed with dry food and watered twice daily.

Before rabies vaccination and during the previous 4 week quarantine period, they were examined by a certified veterinarian, dewormed (Alben®) and vaccinated against canine distemper, adenovirosis, parvovirosis, and parainfluenza (Nobivac DHPPI®). All the animals were healthy and seronegative for rabies VNA prior the first vaccination.

Vaccination and study design

To study the safety of the oral vaccine (see Table 1) the animals (6 dogs, 3 corsacs and 3 wolves) were fed with 10-times the vaccine dose/animal (10 baits). Daily clinical observation of the animals was conducted for 20 days. At 5 and 10 days after ingestion of the vaccine pellets the saliva of animals was sampled for detection of RV.

Table 1 Safety study groups

To evaluate the duration of the immune response in animals to the RV 8 dogs, 8 corsacs and 8 wolves were used (see Table 2). They were vaccinated through feeding 1 dose (bait) of vaccine per animal. The animals were bled (10 mL/Becton Dickinson Vacutainer) on 14, 30, 60, 90, 120, 150, 180 and 210 dpv. Blood samples were refrigerated at 4 °C for up to 48 h and then centrifuged at 1000 × g for 12 min at 4 °C. Serum (2 mL) was collected and frozen at −20 °C until testing. Prior to testing, serum was thawed and then heat-inactivated at 56 °C for 30 min.

Table 2 Efficacy study groups

To evaluate the duration of the protective immune response against RV-infection 16 dogs were used (see Table 2). The vaccination procedure was the same as used in the study of the duration of immunity. On 14, 90, 180, 180 and 210 dpv the animals were challenged with the virulent RV strain CVS. Four vaccinated and two unvaccinated dogs were used for each time point. Dogs were anesthetized and inoculated bilaterally with 0.5 mL of RV (105.0 MICLD50 (50% mouse intracerebral lethal dose)) into the masseter muscle. All animals challenged with the virulent CVS strain were housed under Animal Biological Safety Level 3 Laboratory conditions. The animals in all three study groups (safety, efficacy and control groups) were observed twice daily for clinical signs of rabies (paralysis, ataxia, acute behavioral change, lack of food consumption, hyper-salivation, vocalization, agitation, tremors, convulsions, unprovoked aggression) and euthanized upon manifestation of two or more clinical signs. All surviving animals were euthanized 90 days post challenge (dpc). The challenge study was conducted only in dogs.

This study was carried out in compliance with national and international laws and guidelines on animal handling. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Research Institute for Biological Safety Problems of the Science Committee of the Ministry of Education and Science of the Republic of Kazakhstan (Permit Number: 0514/97).

Virus neutralization test in mice (VNT)

The test was carried out on mice weighing 8–10 g. The mice were inoculated intracerebrally, using 0.03 mL of inoculum, strictly following the WHO recommended method [32]. They were challenged with a CVS strain at a dose of 50 MICLD50/0.03 mL. The rabies antibody titer was calculated according to Reed and Muench [33].

Post mortem examination

Following euthanasia or death of the dogs, the brain and submaxillary salivary glands were collected from all the animals. Impression smears from brains and salivary glands were tested for the presence of RV antigen by a direct fluorescent antibody test (FAT) using a fluorescent conjugate specific for RV nucleocapsid (Biorad, France, 72112 conjugate) [34].

Saliva collection

The presence of VRC-RZ2 virus in salivary excretions and blood samples was studied in the vaccine safety animals (n=12) and controls groups (n=6). Saliva samples were collected from the 18 animals before and on 5 and 10 dpv. Saliva samples were tested for the presence of nucleoprotein rabies ribonucleic acid (RNA) by real time –Polymerase Chain Reaction (RT-PCR). Briefly, viral RNA was extracted from 150 µl of saliva samples using a QiAmp viral RNA mini kit (Quiagen, France), according to the manufacturer’s instructions.

Statistical analyses

Geometric mean serum VNA titers were determined for each experimental group, and ANOVA was used to measure the significance of the observed differences. Differences in protection results among experimental groups were determined by use of the Fisher exact test. Values of P ≤ 0.05 were considered to be significant.

Results

Safety assessment of the vaccines in stray dogs and wild carnivores

Clinical observations: This study showed that immunization of animals with an oral vaccine formulation did not have any negative impact on the overall clinical status (behavior, appetite, etc.) of the animals throughout the observation period. In the safety group vaccinated with 10-times field dose/animal, no animals showed any signs of disease or changes in behavior or appetite during the period of clinical observation, similar to the animals in the negative control group.

Saliva excretion: RV antigen was not detected in saliva sampled from animals prior to vaccination and at 5 and 10 dpv.

Efficacy of the vaccines in stray dogs and wild carnivores

Rabies virus neutralizing antibodies

Dogs immunized with the dose 106.75 TCID50 had VNA levels of 0.59 ± 0.12 IU/mL at 14 dpv, 1.37 ± 0.48 IU/mL at 30 dpv, 1.02 ± 0.40 IU/mL at 90 dpv, and 0.48 ± 0.48 IU/mL at 210 dpv (Fig.1a). On 14 dpv, all the wild carnivores had detectable levels of neutralizing antibodies with mean titers ranging from 0.50 ± 0.07 IU/mL (for wolves) to 0.59 ± 0.10 IU/mL (for corsacs) (Fig. 1c, 1b). In the weeks after vaccination, all the vaccinated wolves and corsacs had even higher levels of neutralizing antibodies: 0.70 ± 0.10 – 0.71 ± 0.08 IU/mL at 30 dpv, 1.06 ± 0.08 – 1.28 ± 0.21 IU/mL at 60 dpv and 0.41 ± 0.09 – 047 ± 0.06 at 180 dpv (Fig. 1c, 1b). The highest level of VNA (˃1.0 IU/ml) was detected at 60 dpv in all vaccinated animals. The VNA level in the blood sera of vaccinated wildlife (corsacs and wolves) was by ~0.08-1.15 IU/mL lower than in the blood sera of dogs while at 180 dpv the VNA level in wild carnivores was c.0.5 IU/mL lower (Fig.1b, 1c). Analysis of the results of this study show that on 60 and 210 dpv the VNA levels in the blood sera of immunized animals differed significantly (P ≤ 0.001) between the 3 groups.

Fig. 1
figure 1

Dynamics of the accumulation of RV neutralizing antibodies. Sera were collected on days 0, 14, 30, 60, 90, 120, 150, 180 and 210 after vaccination and titrated by the virus neutralisation test (VNT). a - RV neutralizing antibody response in stray dogs after OV; b – RV neutralizing antibody response in corsac foxes after OV; c – RV neutralizing antibody response in steppe wolves after OV. Titers are expressed in international units per milliliters (IU/ml). Geometric means and standard deviation are calculated for vaccinated groups

Challenge study

The vaccinated and control animals were challenged with the virulent virus. All vaccinated dogs were healthy for 180 dpv while control animals got ill with specific clinical signs of rabies. One of the four vaccinated dogs that was challenged on 210 dpv manifested the first clinical signs of rabies on 14 day post challenge (dpc) and died on 24 dpc with the same signs as in control animals (Table 3). However, the control animals (unvaccinated dogs) showed the first clinical signs on day 6 post administration of the virulent virus and died of rabies on 10-13 dpc (Fig.2a).

Table 3 Protective response of dogs to rabies virus challenge after a single immunization
Fig. 2
figure 2

(a) Dog survival after challenge (single immunization): on days 14, 90, 180 and 210 post vaccination the vaccinated (n = 4 per time points) and control (n = 2 per time points) dogs were challenged with 0.5 mL (105.0 MICLD50/ml) of the virulent CVS RV strain. On 10-13 dpc a 100% mortality in the unvaccinated group was recorded. In the group of vaccinated dogs the survival rate was 75% on 210 dpv. (b) Effectiveness of the vaccine: the diagram shows mean VNA titers in vaccinated dogs on 14, 30, 60, 90, 120, 150, 180 and 210 dpv, 75% of them (upper 95% CI of mean) having VNA titers as high as 1.33 IU/mL and 25% of them having VNA titers as low as 0.59 IU/mL (lower 95% CI of mean). (c) VNA titers prior to and after challenge: on days 14, 90, 180 and 210 post vaccination the VNA titer was determined prior to challenge as well as (*) on day 30 after it. Before challenge the animals had VNA titers that were lower (P>0.05) than 30 days after. The only exception was a slight decrease in the VNA titers in vaccinated dogs on day 90 post challenge (P < 0,05)

The specific cause of death was confirmed with the help of an immunofluorescence test. All samples from control (unvaccinated but challenged) animals were rabies virus positive. The immunofluorescence test failed to detect RV in brain and salivary gland samples from vaccinated dogs.

Vaccine efficacy

On the 14th dpv 85% of the single dose immunized animals demonstrated sufficient levels of VNA. Starting from the 30th day and until the completion of the experiment all vaccinated animals had VNA levels required for their protection against virulent RV (Fig.2b, 2c). The titers were stable through 180 dpv, but by 210 dpv they had decreased to < 0.50 IU/mL. However, at the same time on 210 dpv the survival rate among vaccinated but challenged dogs was 75%. The tested vaccine is an effective and safe preparation for rabies prophylaxis.

Discussion

Rabies in Kazakhstan is observed mainly in naturally occurring epizooties. The main regional foci for the infection is in Southern and Eastern Kazakhstan, the Almaty and Zhambyl oblasts. Various species of wild, farm and domestic animals are affected by the disease. Many species of preying animals inhabit this territory of Kazakhstan: wolves, feral dogs, raccoon dogs, foxes, corsacs, etc. The most cost-effective strategy for rabies prophylaxis in wildlife is immunization with an oral vaccine.

Conceptually the use of oral vaccines for rabies prophylaxis among wild carnivores was approved over 30 years ago [35]. Many researchers have shown that live rabies vaccines (generated on the basis of fixed rabies viruses SAD and ERA) are effective for oral vaccination (OV). These strains are able to protect foxes against experimental infection with virulent strains of RV [12]. The SAD (State Alabama Dufftring) strain was originally isolated from a rabid dog in the USA. Subsequently it became a source of fixed strains ERA, SAD-Bern, SAG2 and so on [12]. At present these fixed strains are successfully used for OV of wildlife in Europe [36].

In our study we used the fixed RV strain VRC-RZ2 derived from tissue-and-organ RV isolate AZVI [29, 31]. According to the published data AZVI originated from the fixed RV strain Sheep-VGNKI [37]. Sheep-VGNKI is a variant of the Pasteur strain that is adapted to the brain tissue of sheep, rabbits, ferrets and mice [1]. When comparing the N gene of strain VRC-RZ2 it appears 99% identical to the Russian RV strains RV-97 and Moscow (GenBank), with nucleotide substitutions at positions 235 and 637. Equivalent substitutions occurred in the strain SAD Bern/2006/POL in the N gene at position 198 [38]. This is probably through the increased number of passages of the RV vaccine strain in biological systems.

The following parameters can be used to assess the safety of fixed viruses intended for the preparation of oral vaccines: RV neuropathogenicity for laboratory animals following extraneural inoculation; invasiveness index; ability of RV to excrete within the saliva of a vaccinated animal; ability to increase its virulence while being passaged; probability of spontaneous virus transmission from a sick animal to a healthy one [39,40,41,42]. The vaccines selected for OV should also not induce sickness in young animals when oral administered at a 10-times field dose, per animal. Similar studies have been conducted by other authors using fixed RVs such as RV-71, ERA, SAD-Bern, SAD-B19 and Vnukovo-32 in oral preparations because they manifest residual pathogenicity for the majority of rodent species following oral administration and are completely apathogenic for target animals [42, 43]. Our experiments have demonstrated that in the saliva and blood of animals sampled after being fed 10-times the field dose of the vaccine the RV antigen was not detected. According to the results of the previous study [44] VRC-RZ2 strain has low virulence, is completely apathogenic for various animal species after extraneural inoculation and is irreversible (to neuropathogenic virulence). These findings allowed us to conclude that this strain is safe for animal vaccination.

Aside from the characteristics of the RV strain, the quality and availability of the baits play an important role in OV efficacy. In the course of our study on this vaccine a bait pellet containing a blister with liquid RV (strain VRC-RZ2) vaccine was developed. Meat-and-bone meal and chicken eggs were used as the major components. Gelatin served as a binding agent. In spite of the experimental confirmation of OV efficacy among wild carnivores the information about preferable routes for vaccine administration into a target body is contradictory. Some researchers consider the mucosa of the oral cavity and throat to be an effective route for vaccine virus application, while others believe that for effective immunization the vaccine should be transported into the intestine. To deliver the virus antigen to immunocompetent cells in the mouth cavity and the rings of the oropharynx liquid forms of the preparation are used [45,46,47].

Numerous laboratory and field tests have confirmed the efficacy of oral rabies vaccines (based on fixed virus) for mammals such as red foxes (Vulpes vulpes), gray foxes (Urocyon cinereoargenteus), raccoon dogs (Nyctereutes procyonoides), arctic foxes (Alopex lagopus), jackals, Ethiopian wolves (Canis simensis), steppe wolves (Canis latrans), raccoons (Procyon lotor), skunks (Mephitis mephitis), as well as dogs [1, 12]. There is also sufficient information on the immunogenicity and protective ability of many modified strains such as SAD, ERA, SAD Bern, SADB19, SAG1, SAG2 in carnivorous animals [12].

In our experiments immunization of stray dogs, corsacs and wolves with a dose of 6.75 TCID50 and higher induced the production of virus-neutralizing antibodies. Published data shows that for the formation of strong immune responses the virus dose in the vaccine preparation should be not less than 6.00 log LD50/cm3 for foxes and raccoon dogs, and not less than 7.00 log LD50/mL for dogs [48, 49].

Our observations on the immune responses of vaccinated animals during 180 days demonstrated that the level of VNA accumulation in the blood sera of vaccinated dogs ranged from 0.59-1.37 IU/mL. These titers were sufficient to protect these dogs from virulent RV. The VNA accumulation in the blood sera of vaccinated wildlife (corsacs and wolves) was 0.08-1.15 IU/mL lower than in the blood sera of vaccinated dogs. In our opinion this may be explained by physiological characteristics of the wildlife in question, or through changes to the natural life mode, e.g. the artificial housing conditions that results in an immunodepressive state. Our data is evidence that the oral form of the vaccine preparation induces VNA production in corsacs and wolves to titers sufficient for prevention of the disease for 180 dpv.

This study has shown that the tested oral bait vaccine (based on the RV strain VRC-RZ2) meets the immunogenicity requirements for preparations of this type [50] and may be tested on wildlife in the field.