Isolation and genetic characterization of canine adenovirus type 2 variant from raccoon dog (Nyctereutes procynoide koresis) in Republic of Korea
Article information
Abstract
Canine adenovirus type 2 (CAV-2) is a common causative agent of respiratory disease in canines. There have been no reports of CAV-2 variants isolated from raccoon dogs. This study aims to investigate the biological and genetic characteristics of a novel Korean CAV-2 variant. Madin-Darby canine kidney cells were used to isolate the CAV-2 variant from 45 fecal swab samples. Diagnostic tools such as the cytopathic effect (CPE) assay, electron microscopy, polymerase chain reaction, and immunofluorescence and hemagglutination assays were used to confirm the presence of the CAV-2 isolate. A cross-virus neutralization assay was performed to verify the novelty of this CAV variant. Genetic analysis was performed using nucleotide sequences obtained through next-generation sequencing. The isolate was confirmed to be a CAV-2 variant based on the aforementioned methods and designated CAV2232. The number of bases in the fiber and E3 genes of CAV2232 were 1,626 and 414, respectively. Phylogenetic analysis of the fiber and E3 genes confirmed that CAV2232 was classified into a different clade from the known CAV-1 and CAV-2 strains. Mice inoculated with the CAV2232 vaccine developed high virus neutralization antibody titers of 1,024 (210) against CAV2232, while mice inoculated with CAV-1 and CAV-2 vaccines had low virus neutralization antibody titers of 12.9 (23.7) and 6.5 (22.7), respectively, against CAV2232. CAV2232 isolated from wild raccoon dog feces was classified as a novel CAV-2 variant. CAV2232 may therefore be used as an antigen for new vaccine development and serological investigations.
Introduction
Canine adenovirus (CAV) is a DNA virus belonging to the Mastadenovirus genus that infects various animals such as dogs, raccoon dogs, wolves, coyotes, skunks, fennec foxes, bears, and sea lions worldwide [1–3]. CAV types have been categorized into CAV type 1 (CAV-1) and CAV type 2 (CAV-2), based on their biological and molecular characteristics revealed by hemagglutination (HA), virus neutralization (VN), and genetic analyses [4,5]. CAV-1 causes hepatitis, uveitis, and interstitial nephritis in dogs [6], whereas CAV-2 induces infectious laryngotracheitis due to its affinity for the respiratory tract epithelium [7]. Co-infection with canine distemper or a bacterial infection increases the severity of CAV-2 infection resulting in bronchopneumonia or chronic airway disease [8]. The mortality rate of CAV-2 is low, estimated at around 5% to 10%, and the virus is excreted for a long time after infection, posing a high risk of transmission. CAV is secreted by an infected individual and is transmitted through direct or indirect contact with saliva, feces, and urine [9]. The treatment of CAV varies depending on the exhibited symptoms. Frequently used treatments include antibiotics, anti-inflammatory drugs, and antihistamines. CAV-1 and CAV-2 are closely related serologically, and CAV-2 can provide immunity against CAV-1. Therefore, vaccines using the CAV-2 strain are immunogenic against both CAV-1 and CAV-2 [10].
CAV-2 is a common cause of respiratory disease in canines. CAV-2 infections are diagnosed and reported based on viral isolation, histopathological findings, amplification of specific polymerase chain reaction (PCR) genes, and serological assays [5,7]. In the Republic of Korea (ROK), 38 CAV infections were reported in dogs and raccoon dogs between 2000 and 2017 [11]. Two cases of CAV-1 infections were identified in a Eurasian river otter in 2007 and a fennec fox in 2014 [2,12]. The CAV-2, APQA1601 strain, was isolated from a naturally infected dog and its molecular characteristics were reported in 2018 [13]. CAV-2 infection was also discovered in a Korean raccoon dog. The CAV-2/18Ra-54 strain, which was obtained from a raccoon dog between 2017 and 2020, exhibited a high degree of similarity to the Toronto A26/61 strain, which is commonly used as a vaccine [14]. Although CAV vaccination prevents diseases caused by CAV-1 and CAV-2, unvaccinated dogs and wild raccoon dogs continue to be exposed to CAV infections. No previous studies have successfully isolated CAV-2 variant.
CAV genes comprise multiple transcription units, including 5 early (E1A, E1B, E2, E3, and E4), intermediate and late (major late promoter, tripartite leader sequence and protein coding region) transcription units. The late unit produces 5 late mRNA populations, L1–L5. Differences in the number of bases in E3 have been discovered, and these gene differences make it possible to differentiate between CAV-1 and CAV-2 [5,15]. The distinction between CAV-1 and CAV-2 in the E3 gene was confirmed using real-time PCR diagnostic techniques [16]. A few nucleotide differences in the E1B 19 K gene have also been identified among CAV-2 isolates [17]. CAV-1 and CAV-2 have various distinguishing molecular characteristics, and the identification of new CAV in dogs, raccoons, and other animals worldwide provides motivation for further exploration into the genetic diversity of CAV.
In this study, we isolated a new CAV-2 strain from the feces of raccoon dogs using Madin-Darby canine kidney (MDCK) cells. The CAV-2 strain, designated CAV2232, was passaged through the same cells 5 times. We performed biological and molecular analyses and serotype classification of CAV2232 and confirmed its novelty as a new CAV-2 variant. These results are expected to contribute to the prevention of CAV-2 infection and the development of new diagnostic methods.
Materials and Methods
Ethical approval
We carried out the animal experiments in accordance with the protocols approved by the Institutional Animal Care and Use Committees (approval number: 2024–849).
Samples and reference viruses
In total, 45 fecal swab samples were obtained from 20 water deer (Hydropotes inermis) and 25 raccoon dogs (Nyctereutes procyonoides koreensis) in 2022. All fecal samples were centrifuged at 3,500 × g for 15 minutes at 4°C to remove fecal debris. The supernatants were filtered through a 0.45 μm syringe filter (Millipore, USA) to prevent bacterial infection and used for virus isolation. The Utrecht (Korea Veterinary culture collection [KVCC] no: VR2400007) and APQA1701 (KVCC no: VR200005) strains, belonging to CAV-1 and CAV-2, respectively, were used as reference viruses for PCR, HA, and VN testing.
Virus isolation and titration
MDCK cells (CCL-34; ATCC, USA) were cultured in Dulbecco's Modified Eagles Medium (DMEM) containing penicillin, streptomycin, amphotericin B, and 10% heat-inactivated fetal bovine serum (FBS; Gibco BRL, USA). The MDCK cells were plated into 24-well plates, the fecal samples (100 μL) were added, and the mixture was incubated in a 5% CO2 incubator at 37°C for 1 hour. After removing the fecal samples, 1 mL DMEM containing 5% FBS was added to each well. The plates were maintained under the same conditions as described above for 5 days. When cytopathic effects (CPEs) were observed in the MDCK cells, the cell supernatants were collected for further passage and re-inoculated into freshly prepared MDCK cells. If CPEs were not observed after the second passage, the samples were excluded from virus isolation. The growth kinetics of the isolate were determined, and the viral titers were calculated in accordance with the Reed and Muench method.
Immunofluorescence assay
The immunofluorescence assay (IFA) was performed in the same manner as previously described [13]. Briefly, the MDCK cells infected with the CAV2232 strain were fixed in cold acetone and exposed to mouse monoclonal antibodies against CAV-1 or CAV-2 (VMRD Inc., USA) at 37°C for 1 hour. The cells were then stained with FITC-conjugate (KPL Laboratories, USA) and observed under a fluorescence microscope. Samples showing intranuclear fluorescence were positive for CAV-1 and CAV-2.
Electron microscopy
Electron microscopy (EM) was performed using a previously reported method [13]. Briefly, the MDCK cells infected with CAV2232 strain were collected and fixed with glutaraldehyde (2.5%) and osmium tetroxide (1%) solutions in phosphate buffered saline (PBS; pH, 7.2) at 4°C for 1 hour. After embedding in resin, cell sections were stained with uranyl acetate and lead citrate. CAV2232 was purified with cesium chloride and negatively stained with 1% uranyl acetate. Viral particles in cell sections and on the grid were examined using an electron microscope (H-7100FA; Hitachi, Japan).
PCR analysis
For the diagnosis and differentiation of CAV, PCR was performed by modifying and supplementing a method described previously [13]. Viral DNA was extracted from the CAV2232 infected cells according to the manufacturer’s instructions. In the first PCR, 2 specific primer sets were used to detect CAV. The second PCR amplified the E3 gene to differentiate between the CAV types (Table 1). Each PCR was performed using a PCR premix (Bioneer, ROK) that included 10 μL denatured DNA, 1 μL of each primer (50 pmol), and 38 μL distilled water in a 50 μL total volume. After the first denaturation step (95°C for 5 minutes), the profile was subjected to 35 cycles of denaturation, annealing, and extension. The temperature and time steps were 95°C for 30 seconds, 55°C for 30 seconds, 72°C for 1 minute, and a final extension step of 72°C for 5 minutes. The expected sizes of the PCR products were determined by electrophoresis.
A list of the polymerase chain reaction primers utilized
Hemagglutination
HA was performed in the same manner as previously described [13]. Briefly, HA was performed by incubating serial 2-fold dilutions of 3 types of CAVs propagated in MDCK cells in 50 µL of PBS at 4°C with 50 µL of 0.6% erythrocytes from a guinea pig, goose, fowl, mouse, and pig. The HA titer was calculated as the reciprocal of the highest dilution of CAV showing an HA reaction.
Sequence and phylogenetic analyses
Viral DNA extracted from CAV2232 was sent to an external agency (Sanigen Inc., ROK) for confirmation of the complete genome sequence by next-generation sequencing (NGS). The whole-genome sequence was compared with those of the CAV reference strains. In addition, phylogenetic analyses were performed based on the fiber (F) and E3 genes of the CAV2232 and CAV strains obtained from the GenBank database. Nucleotide alignments among the CAVs were examined using Clone Manager Basic ver. 9 software (Sci-Ed Software, USA). Two phylogenic trees were constructed by applying the neighbor-joining method using MEGA ver. 10.0 software (http://www.megasoftware.net).
Preparation of trial vaccines
To verify the new CAV serotypes, the Utrecht strain for CAV-1, APQA1701 strain for CAV-2, and CAV2232 strain were propagated in MDCK cells. The titers of the 3 types of CAVs were confirmed to be > 106.5 TCID50/mL. The 3 CAV types were inactivated with 0.1% formalin at 37°C for 24 h and blended with aluminum hydroxide adjuvant (Rehydragel LV; Chemtrade Logistics Inc., Canada) at the ratio of 9:1 to prepare inactivated CAV antigens for testing.
Animal experiments
Three types of inactivated CAV antigens were prepared, and six 4-week-old mice used at the Institute for Cancer Research were divided into 3 groups. Six mice per group were immunized twice intramuscularly with 0.1 mL of the vaccine at 2-week intervals. Blood samples were collected from the mice, and each serum sample was subjected to a VN test.
Cross-virus neutralization test
Cross-VN tests for CAVs were performed in 96-well microplates using MDCK cells. Heat-inactivated serum (50 µL) was diluted 2-fold with Eagle’s medium and mixed with an equal volume of 200 TCID50/0.1 mL of CAVs. The serum was incubated at 37℃ for 1 hour. Subsequently, samples containing 20,000 MDCK cells were added to each well. The plates were placed in a CO2 incubator for 4 days. Each well was observed under a microscope to confirm the CPE caused by the virus. VN antibody (VNA) titers against CAVs were expressed as the reciprocal of the highest serum dilution that completely inhibited viral propagation.
Results
Biological characterization of CAV2232
One virus was isolated from MDCK cells inoculated with raccoon dog fecal samples. As shown in Fig. 1A, CPEs resembling a bunch of grapes were observed. To ensure that a sufficient amount was produced for accurate virus identification, the isolate was continuously passaged through the same cell. MDCK cells inoculated with the isolate after passage 5 showed clear and similar CPEs. Based on the unique CPE, it was assumed to be a CAV. MDCK cells inoculated with the isolate were fixed and incubated with specific antibodies against CAV-1 and CAV-2. Specific fluorescence was observed in the nuclei of MDCK cells stained only with the CAV-2 monoclonal antibody (Fig. 1B). The isolate was designated CAV2232 because it was isolated from the 32nd sample in 2022. After propagation in the same cells grown in 25 cm2 culture flasks, the viral titers of CAV2232 peaked at 2 days post-inoculation, reaching 105.5 TCID50/mL (Fig. 2). HA titers of CAV2232 were measured in 5 types of erythrocytes. As shown in Table 2, the isolate agglutinated only guinea pig blood cells with 8 HA units. Many CAV particles within MDCK cells infected with CAV2232 were observed by EM (Fig. 3A), and the purified CAV2232 particles showed a typical adenovirus form (Fig. 3B). The isolate was confirmed to be CAV-2 based on the results from the CPE assay, IFA, HA, and EM tests.
(A) Cytopathic effects of the canine adenovirus (CAV) isolate in infected Madin-Darby canine kidney (MDCK) cells (200×). Indirect immunofluorescence assay using monoclonal antibodies against (C) CAV-1 and (B) CAV-2 and (D) normal MDCK cells. Scale bars: (A–D) 100 µm.
Growth kinetics of CAV2232 at passage 5 according to the day of harvest from Madin-Darby canine kidney cells. CAV2232 showed the highest viral titer of 105.5 TCID50/mL at 2 days post-inoculation.
Hemagglutinating ability of 3 types of canine adenoviruses on erythrocytes from 5 animal species
(A) Viral particles in the nucleus and cytoplasm of Madin-Darby canine kidney cells infected with CAV2232. (B) The virus was purified with cesium chloride and confirmed to be a typical adenovirus with a diameter of 60–80 nm.
Molecular characterization of CAV2232
Molecular genetic methods were used to accurately identify CAV2232. Two partial F genes of CAV2232 were amplified by PCR using 2 primer sets for CAV. As shown in Fig. 4A, 2 expected PCR products from CAV2232 were detected at 408 and 549 bp on a 2.0% agarose gel. DNA from CAV2232 was extracted along with previously known CAV-1 and CAV-2 genes, and their E3 genes were amplified by PCR using a primer set capable of differentiating between CAV types. The sizes of the PCR products of CAV-1, CAV-2, and CAV2232 were 499, 1,032, and 369 bp, respectively (Fig. 4B). In other words, CAV2232 was confirmed to have an E3 gene of a different size than those of CAV-1 and CAV-2. The entire gene sequence of CAV2232 was determined using NGS. CAV2232 consists of 31,093 nucleotides and encodes 29 open reading frames. To explore their genetic relationship, we compared the entire CAV2232 nucleotide sequence to those of the 4 CAV-2 strains including APQA1601 and APQA1701. As shown in Fig. 5A, CAV2232 had a defective E3 gene. The difference in the size of the E3 gene of CAV2232 was so large that homology with CAV-2 could not be clearly arranged (Fig. 5B). The base numbers of the hexon, F, and E3 genes among the 3 CAVs were compared, and the hexon gene was found to be the same size (2,670 bp) across all 3 CAVs. CAV-1 and CAV-2 were confirmed to have 1,629 bases in the F gene, and CAV2232 was found to have 1,626 bases. Based on the fiber gene of CAVs, the CAV showing the highest homology to CAV2232 strain was APQA1701 strain (86.7%) isolated from a naturally infected Korean dog. However, the base numbers of the E3 genes among CAV-1, CAV-2, and CAV2232 were 585, 1,095, and 414, respectively (Table 3). Differences in base number among the 3 CAVs lead to low homology. Two phylogenetic trees constructed using CAV F and E3 genes provided a better understanding of the genetic relatedness between CAV-1, CAV-2, and CAV2232. As shown in Fig. 6, CAVs are largely divided into 2 types. In addition, CAV2232 represents a lineage distinct from previously reported CAVs. Therefore, CAV2232 was designated as a CAV-2 variant.
Polymerase chain reaction (PCR) using 2 types of diagnostic primer sets confirmed that the isolate was canine adenovirus. M, 1 kb ladder; lanes 1 and 2, CAV2232 (A). PCR was performed using primers that separated the E3 gene-amplified PCR products into different sizes depending on the canine adenovirus (CAV) type (B). M, 1 kb ladder; lane 1, CAV-1; lane 2, CAV-2; lane 3, CAV2232.
(A) Comparison of the full nucleotide sequences of 3 canine adenoviruses (CAVs) revealed that part of the E3 gene of CAV2232 was deleted. (B) The E3 gene of CAV2232 has low similarity with those of CAV-2.
Comparison of base number of hexon, fiber and E3 genes among 3 canine adenoviruses
Phylogenetic trees were created based on the nucleotide sequences of the (A) fiber and (B) E3 genes from known canine adenoviruses (CAVs). In both trees, the CAV2232 clade is distinct from CAV-2.
Virus neutralization analysis
To verify the novel CAV types, we immunized mice with 3 types of experimental CAV-inactivated antigens. The VN test results of immunized mouse sera showed a clear difference in VNA titers against homologous and heterologous CAVs (Table 4). Mice inoculated with the CAV-1 vaccine showed a mean VNA titer of 512 (29) against the homologous antigen, whereas mice inoculated with the CAV-2 and CAV2232 vaccines showed mean VNA titers of 4 (22) and 51.9 (25.7), respectively. Mice inoculated with the CAV-2 vaccine had a mean VNA titer of 630.3 (29.3) against the homologous antigen; however, mice inoculated with the CAV-1 and CAV2232 vaccines had mean VNA titers of 51.9 (25.7) and 128 (27), respectively. Mice vaccinated with CAV2232 showed a mean VNA titer of 1,024 (210) against CAV2232. In contrast, mice inoculated with CAV-1 and CAV-2 vaccines had mean VNA titers of 12.9 (23.7) and 6.5 (22.7), respectively, against CAV2232. VNA titers for homologous CAV antigens were high, whereas those for heterologous CAV antigens were low.
Mean cross-virus neutralization antibody titers of the mice immunized with 3 types of inactivated canine adenovirus vaccines
Discussion
Many rescued wild animals are sent to and receive treatment at wildlife rescue centers in the ROK. In 2022, attempts were made to identify the pathogens responsible for disease using the feces of rescued wild animals. We isolated one causative organism from the feces of raccoon dogs. We propagated the isolate in MDCK cells and designated it CAV2232. It was confirmed to be a CAV-2 variant based on the CPE assay, IFA, HA, EM, PCR, NGS, and VN test results.
In this study, we inoculated the isolate into MDCK cells and found that CAV2232 grew well in these cells. However, it did not grow in the Vero cells. CAV2232 is presumed to be circulating within wildlife. This is because the CAV1601 and CAV-2/18Ra-542020 strains, which proliferate well in Vero cells, are presumed to be the origin of vaccine strains such as the Toronto A26/61 strain [13,14]. CAV2232-5P showed a viral titer of 105.5 TCID50/mL in MDCK cells at 2 days post-inoculation. This was 10 times lower than those of the previously reported APQA1601 and APQA1701 strains [13,18], indicating that CAV2232 had not yet properly adapted to MDCK cells or showed inherent differences in proliferative capacity. CAV2232 only agglutinated with the red blood cells of guinea pigs. The HA activity of CAV2232 was similar to those of the OD-N and APQA1701 strains isolated from dogs with respiratory diseases in Japan and the ROK [18,19].
When the full nucleotide sequence of CAV2232 was compared with those of CAV-2 including APQA1701 strain, some parts of the E3 gene of CAV2232 were found to be missing in CAV-2 strains. The number of bases in the E3 gene of CAV2232 was 414, 171 fewer bases than the 585 bases of CAV-1. Although double-stranded DNA viruses, such as CAV, have lower mutation rates than single-stranded DNA or RNA viruses [20], CAV2232 was found to have a shortened E3 gene. The CAV E3 gene encodes several proteins that help the virus evade host immune responses and promote cell survival [21]. The shortened E3 gene in CAV2232 strain could result from a mutation that leads to the loss of nucleotides without affecting the reading frame. This type of mutation may lead to a truncated protein that could retain some function or potentially gain a new function. Changes in the number of bases in the CAV E3 gene can lead to the emergence of new CAV genotypes and serotypes. Therefore, it is believed that additional research is needed on the function of the shortened E3 gene. Phylogenetic analyses based on the F and E3 genes of 19 CAV strains showed that CAV2232, unlike the CAV-1 or CAV-2 groups, formed a new clade. Although CAV2232 showed the highest homology with the CAV-2 strain, it remains unclear how CAV2232 was introduced into raccoon dogs. Further studies are required to investigate the epidemiology and pathogenicity of CAV2232 in canine.
Three types of inactivated antigens containing CAV-1, CAV-2, and CAV2232 were inoculated into mice to identify the new CAV2232 serotypes. Mice developed high VNA titers against homologous antigens but a significant difference of < 4-fold was observed against the heterologous antigens. These findings indicate that CAV2232 has different antigenic properties and may escape cross-protection conferred by existing vaccines containing CAV-1 or CAV-2. Mutations in the partial E3 gene region of CAV-2 isolated from Indian dogs have been confirmed and classified into a genetically different clade [22]. Therefore, it is essential to monitor the genetic diversity and evolution of CAV in various canines and develop vaccines that can elicit broad-spectrum immunity against the CAV-2 variant.
In conclusion, we isolated a CAV from a raccoon dog, designated CAV2232, and confirmed it to be CAV-2 based on CPE, PCR, EM, and IFA analyses. After determining the complete nucleotide sequence using NGS, we compared the viral genes with those of previously reported CAVs. Phylogenetic analyses of the F and E3 genes classified CAV2232 into a clade different from CAV-1 and CAV-2. Therefore, CAV2232 may be a candidate for the development of a novel vaccine and as an antigen for serological monitoring in dogs and raccoon dogs.
Notes
The authors declare no conflict of interest.
Author’s Contributions
Conceptualization: Yang DK: Data curation: Yang DK, Ahn S; Formal analysis: Yang DK, Funding acquisition: Cho YS, Yang DK; Investigation: Yang DK; Methodology: Kim M, Ahn S, Kim JT, Park J, Oh S; Project administration: Yang DK, Software: Yang DK, Lee HJ; Validation: Lee HJ; Writing–original draft: Yang DK; Writing–review & editing: all authors.
Funding
This study was supported financially by a grant (B-1543083-2023-24-01) from the Animal and Plant Quarantine Agency, Ministry of Agriculture, Food and Rural Affairs (MAFRA), Republic of Korea.