Introduction
As a highly urbanized yet ecologically diverse megacity, the Seoul Metropolitan Area combines densely populated districts with surrounding mountains and green spaces. Urbanization and ecological heterogeneity in Seoul shape local environmental conditions and influence wildlife communities inhabiting these settings (Kim et al., Reference Kim, Kim, Song and Lee2016; Janicke et al., Reference Janicke, Kim and Cho2020). This urban–ecological interface allows raccoon dogs (Nyctereutes procyonoides) to thrive in green spaces while remaining close to domestic animals and humans. In such environments, interactions among wildlife, domestic animals, and mosquito vectors may facilitate the circulation of vector-borne parasites. Given the veterinary and zoonotic relevance of Dirofilaria immitis, documenting its occurrence in urban wildlife is useful for understanding local transmission contexts. These ecological interactions highlight the value of examining parasitic infections in wildlife living at the urban–ecological interface.
Dirofilaria immitis, commonly known as the canine heartworm, is a mosquito-borne filarioid nematode that inhabits the pulmonary arteries and right ventricle of its hosts, causing dirofilariasis (McCall et al., Reference McCall, Genchi, Kramer, Guerrero and Venco2008; Simon et al., Reference Simon, Siles-Lucas, Morchon, Gonzalez-Miguel, Mellado, Carreton and Montoya-Alonso2012). Clinical manifestations range from coughing and dyspnoea to weight loss and, in severe cases, heart failure. Transmission occurs via mosquitoes that ingest microfilariae during a blood meal and subsequently transmit infective larvae to new hosts (Gomes-de-Sa et al., Reference Gomes-de-Sa, Santos-Silva, Moreira, Barradas, Amorim, Cardoso and Mesquita2023). Although dogs are the primary host, infections have also been reported in cats, wild carnivores, and occasionally humans (Dantas-Torres and Otranto, Reference Dantas-Torres and Otranto2013; Saha et al., Reference Saha, Bonnier, Chong, Chieng, Austin, Hu and Shkolnik2022).
Recent studies have documented natural infections in various wildlife species, indicating that wild carnivores can be exposed to mosquito-borne filarioid parasites (Upton et al., Reference Upton, Budke and Verocai2023; Ramos et al., Reference Ramos, Hakimi, Salomon, Busselman, Curtis-Robles, Hodo, Hamer and Verocai2024). The raccoon dog, native to East Asia and widely distributed across the Korean Peninsula, is increasingly observed in urban and peri-urban areas, where it adapts readily to anthropogenic environments (Hong et al., Reference Hong, Kim, Lee and Min2013, Reference Hong, Kim, Min and Lee2018). Reports of raccoon dog rescues in Seoul suggest that domestic dogs are at greater risk of parasite transmission than humans (Kim, Reference Kim2023; 2025). This has increased interest in the potential role of raccoon dogs in local transmission cycles. Surveys in Japan and Russia have already documented notable infection rates in raccoon dogs (Kido et al., Reference Kido, Wada, Takahashi, Kamegaya, Omiya and Yamamoto2011; Kravchenko et al., Reference Kravchenko, Itin, Kartashev, Ermakov, Kartashov, Diosdado, Gonzalez-Miguel and Simon2016), but comparable data from Korea remain limited.
Beyond the filarioid nematode itself, attention has also focused on its intracellular endosymbiont Wolbachia, which is an obligate endosymbiont of D. immitis and is expected to be present in all D. immitis worms (Bandi et al., Reference Bandi, Trees and Brattig2001). This bacterium plays a critical role in the biology of many filarioid worms, influencing reproduction, development and survival, and has also been implicated in host inflammatory responses associated with filarial pathology (Taylor et al., Reference Taylor, Bandi and Hoerauf2005). Characterizing Wolbachia in wildlife hosts may therefore provide additional insight into the transmission dynamics and pathological consequences of heartworm infection. To our knowledge, no published molecular study has simultaneously examined Dirofilaria immitis and its endosymbiont Wolbachia in raccoon dogs. In this study, D. immitis infection in raccoon dogs rescued in Seoul was investigated, and the genetic sequences of both the parasite and its endosymbiont Wolbachia were analysed. To our knowledge, this represents the first molecular record of D. immitis and Wolbachia in raccoon dogs in Korea. These findings provide baseline molecular data and contribute to understanding the occurrence and genetic characteristics of these agents in raccoon dogs within an urban environment.
Materials and methods
Sample collection and preparation
Between October 2024 and August 2025, the Seoul Wildlife Centre provided carcasses of 51 raccoon dogs rescued from urban areas of Seoul (Figure 1). Among them, 26 were female and 25 were male. Carcasses were refrigerated at approximately 4°C until necropsy. Necropsies were performed to determine mortality causes and to monitor potential zoonotic pathogens. For the purpose of pathogen screening, approximately 150–200 mg of lung tissue sample was collected specifically from grossly abnormal regions exhibiting visible alterations, such as inflammation, discoloration, or focal lesions, which are routinely targeted for molecular analysis. Because this work was initiated retrospectively after the need to investigate heartworms became apparent, opportunities for specimen collection were limited. When possible, efforts were made to collect more than one worm per raccoon dog to allow for assessment of greater genetic diversity.
Figure 1. Study area and geographic location of the sampling sites in Seoul, South Korea. The left panel shows a national context map with Seoul highlighted and the arrow indicates the area enlarged in the right panel. Blue stars and orange circles represent PCR-positive and PCR-negative outcomes, respectively. Base map: ESRI National Geographic; coordinate reference system: WGS 84 (EPSG:4326). Map created with QGIS v3.40.10 (https://qgis.org/). A scale bar and a north arrow are included.
During postmortem examinations, the presence of adult heartworms was examined by gross inspection of the heart and pulmonary arteries. The raccoon dogs generally showed a suboptimal nutritional state (mean body condition score 2.6 on a 1–9 scale). Frequent co-infections, including scabies (38/51) and roundworm infections (28/51), were also observed. Among the animals infected with D. immitis, the lungs exhibited characteristic vascular pathology, including severe villous intimal proliferation of the pulmonary arteries, marked arterial dilation, and partial luminal obliteration or recanalization.
DNA extraction and PCR amplification
Adult heartworms and lung tissue samples obtained during necropsy were homogenised with 10 × volume of phosphate-buffered saline (PBS) using 2.8-mm stainless steel beads (Precellys™ Hard Tissue Homogenising CK28-R, Bertin Technologies SAS, France) in a bead beater (Precellys Evolution, Bertin Technologies SAS, France) at 6500 rpm for 30 s, repeated twice with a 30-s interval. The homogenates were centrifuged at 4000 rpm for 1 min, and 200 µL of the supernatant was used for DNA extraction. For whole-blood samples, 100 µL was diluted with an equal volume of PBS prior to extraction. DNA was extracted using the NucleoSpin® VET kit (Macherey-Nagel, Düren, Germany) according to the manufacturer’s instructions.
PCR amplification was performed using primers previously described (Casiraghi et al., Reference Casiraghi, Anderson, Bandi, Bazzocchi and Genchi2001), with cycling conditions modified for this study (Table 1). The target genes were the cytochrome c oxidase subunit 1 (cox1) of Dirofilaria spp. and the 16S ribosomal RNA (16S rRNA) of Wolbachia spp. Each reaction contained 5 µL of template DNA, 1 µL of each primer (10 pmol µL−1), and AccuPower PyroHotStart Taq PCR Premix (Bioneer, Daejeon, Korea), with distilled water added to a final volume of 20 µL. PCRs were performed using a ProFlex PCR System (Applied Biosystems, Foster City, CA, USA). For the Dirofilaria cox1 gene, the thermal cycling conditions consisted of an initial denaturation at 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 20 s, annealing at 52°C for 20 s and extension at 72°C for 1 min, with a final extension at 72°C for 5 min. For Wolbachia 16S rRNA amplification, the cycling profile included an initial denaturation at 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 30 s and extension at 72°C for 1 min, with a final extension at 72°C for 5 min. Amplicons were separated on 2% agarose gels at 135 V for 22 min and visualised under UV illumination.
Table 1. PCR primers used for the molecular detection of Dirofilaria spp. and Wolbachia
Sequencing and phylogenetic analysis
In total, 6 D. immitis cox1 amplicons and 4 Wolbachia 16S rRNA amplicons obtained from raccoon dog samples were sequenced by Macrogen (Seoul, Korea). The resulting sequences were edited and aligned using BioEdit v7.7.1 (https://thalljiscience.github.io/) to generate consensus sequences, which were compared with reference sequences in the GenBank database using the Basic Local Alignment Search Tool (BLAST). Phylogenetic analyses were performed in MEGA X (Molecular Evolutionary Genetics Analysis software) using the Tamura–Nei model with 1000 bootstrap replications (Kumar et al., Reference Kumar, Stecher, Li, Knyaz and Tamura2018). To contextualize the genetic diversity of the Korean isolates, 15 published cox1 sequences from dogs, cats, and different regions were incorporated, producing a dataset of 21 partial cox1 sequences for haplotype and phylogenetic analysis. Haplotype analysis was conducted using DnaSP v6 (Rozas et al., Reference Rozas, Ferrer-Mata, Sanchez-delbarrio, Guirao-Rico, Librado, Ramos-Onsins and Sanchez-Gracia2017) to calculate haplotype diversity (Hd), nucleotide diversity (π) and the number of polymorphic sites (S) based on aligned cox1 sequences. In this study, a haplotype was defined as a unique cox1 sequence variant based on nucleotide differences within the partial cox1 region – a standard operational definition in filarioid nematode studies using mitochondrial markers.
Statistical analysis
Differences in detection frequency between sexes were compared using Fisher’s exact test. Paired outcomes between lung tissue PCR and whole-blood PCR were assessed using McNemar’s test, applying both continuity-corrected and exact methods as appropriate. Statistical analyses were performed with SPSS Statistics v24 (IBM, Armonk, NY, USA), and a P-value of < 0.05 was considered statistically significant.
Results
Detection of D. immitis and Wolbachia endosymbionts in raccoon dogs
Among the 51 raccoon dogs examined, adult D. immitis were detected in 13 animals within the pulmonary arteries and right ventricle (Figure 2), with worm burdens ranging from 2 to 9. PCR of lung tissue confirmed infection in 17 animals (33.3%), including all grossly positive cases and 4 additional subclinical infections. Of these, 11 individuals (21.6%) were also positive by whole-blood PCR, all of which corresponded to animals already confirmed by lung tissue PCR and gross inspection. Wolbachia DNA was successfully amplified from the lung tissue of 9 raccoon dogs (17.6%), all of which were positive for D. immitis. The overall results of D. immitis detection by different diagnostic methods, together with Wolbachia detection in lung tissue, are summarized in Table 2.
Figure 2. Gross pathology of a raccoon dog heart, demonstrating adult D. immitis worms within the right ventricle and pulmonary artery. A 1-cm scale bar is included for reference.
Table 2. Comparative detection of D. immitis and Wolbachia in raccoon dogs using multiple diagnostic methods and sample types
No statistically significant differences in D. immitis infection rates were observed between male and female raccoon dogs. The odds ratios were 0.51 (P = 0.334) for adult worm detection, 0.55 (P = 0.415) for lung tissue PCR, and 0.47 (P = 0.496) for whole-blood PCR, all indicating non-significant associations.
A paired comparison of lung tissue PCR and whole blood PCR showed a significant difference in detection outcomes (lung +/blood − = 6 vs. lung −/blood + = 0). The continuity-corrected chi-square test yielded χ2 (1, n = 51) = 4.17, P = 0.041, and the exact test also confirmed statistical significance (P = 0.031). These results indicate that lung tissue PCR identified more positive cases than whole-blood PCR
Phylogenetic analysis of D. immitis
Six partial cox1 sequences of D. immitis were obtained from raccoon dogs (GenBank accession No. PV628747, PV628786, PV628865, PX517297, PX517298 and PX517305), and all sequences were deposited in the GenBank. A maximum likelihood tree was reconstructed using partial cox1 sequences of D. immitis and related filarioid nematodes, with Ascaris lumbricoides included as the outgroup (Figure 3). The dataset included Korean isolates from dogs, cats and raccoon dogs, together with reference sequences from diverse geographical origins. The tree clearly separated D. immitis from other filarioid species such as D. repens, Brugia malayi, B. pahangi, Loa loa, and Onchocerca lupi. Within the D. immitis clade, Korean dog and cat isolates, together with raccoon dog-derived sequences, clustered with reference sequences from Asia, Europe, and the United States, forming a highly conserved group with strong bootstrap support (> 90%).
Figure 3. Phylogenetic tree of D. immitis based on partial cox1 sequences. Raccoon dog-derived sequences (accession No. PV628747, PV628786, PV628865, PX517297, PX517298 and PX517305) obtained in this study are shown in bold. Reference sequences from dogs, cats, mosquitoes and humans originating from different countries were retrieved from GenBank. Each accession number is listed alongside the corresponding species names. The scale bar represents 0.2 nucleotide substitutions per site.
Haplotype analysis was conducted using 21 partial cox1 sequences belonging to the same D. immitis phylogenetic cluster shown in the reconstructed tree (Figure 3). All raccoon dog-derived sequences, together with the Korean dog isolate, were assigned to Hap_1 alongside reference sequences from other geographic regions. In contrast, a single Korean cat-derived sequence (accession No. KF918397.1) was placed in Hap_2. These results indicate low genetic variation within this cluster. Detailed summary statistics are presented in Table 3.
Table 3. Genetic variation in D. immitis sequences belonging to a single phylogenetic cluster identified in Figure 3
Hd: Haplotype diversity.
π: Nucleotide diversity.
S: Number of segregating (polymorphic) sites.
Aligned region: Total length of the aligned sequences.
Effective length: Length excluding gaps and missing data, used for diversity analysis.
Phylogenetic analysis of Wolbachia endosymbionts
Four partial 16S rRNA sequences of Wolbachia endosymbionts were obtained from D. immitis isolates derived from raccoon dogs. These sequences were deposited in the GenBank database (accession No. PV635601, PV637187, PX517299 and PX517300). A phylogenetic tree was reconstructed using partial 16S rRNA sequences of Wolbachia, together with reference sequences representing different Wolbachia supergroups (Figure 4). Rickettsia rickettsii was used as the outgroup. The tree showed that the 4 Wolbachia sequences from raccoon dog-derived D. immitis clustered with other D. immitis-associated Wolbachia, while being clearly separated from Wolbachia sequences of other filarioid species.
Figure 4. Phylogenetic tree of Wolbachia endosymbionts inferred from partial 16S rRNA sequences. Wolbachia 16S rRNA sequences (accession No. PV635601, PV637187, PX517299 and PX517300) obtained from raccoon dog-derived D. immitis in this study are shown in bold. Additional Wolbachia sequences from filarioid nematodes and arthropods from diverse geographic origins were retrieved from GenBank. Each accession number is listed alongside the corresponding species names. The bar denotes 0.02 nucleotide substitutions per site.
Discussion
The present study provides the first molecular evidence of D. immitis and associated endosymbiont Wolbachia in wild raccoon dogs inhabiting urban Seoul, Korea. The prevalence (33.3%) of D. immitis infection observed in this study is markedly higher than that previously reported in Korea. Nam et al. documented a prevalence of 17.8% (17/95) in raccoon dogs based on both necropsy and heartworm antigen testing nationwide, indicating that the infection rate in the capital region (Seoul) is higher than the nationwide prevalence (Nam et al., Reference Nam, Kim, Yang and Hyun2013). When compared to studies in neighbouring Japan, our finding is substantially higher than the rate reported from the Nishi-Tama District of Tokyo, Japan (6.0%, 2/29, postmortem examination) and Kanagawa Prefecture (7.4% by antigen test to 16.0% by microscopic identification) (Nakagaki et al., Reference Nakagaki, Suzuki, Hayama and Kanda2000; Kido et al., Reference Kido, Wada, Takahashi, Kamegaya, Omiya and Yamamoto2011). Such regional differences may reflect not only variation in diagnostic methods and sample types, but also potential influences of host age and geographical factors on the reported occurrence of D. immitis.
In Europe, there have also been reports indicating that the prevalence of D. immitis infection varies among different wild carnivores. For example, prevalence in raccoon dogs was reported as 31.1% (28/90) in Krasnodar Krai, Russia (Kravchenko et al., Reference Kravchenko, Itin, Kartashev, Ermakov, Kartashov, Diosdado, Gonzalez-Miguel and Simon2016), and a single infected individual was also documented in Romania (Ionica et al., Reference Ionica, Deak, Boncea, Gherman and Mihalca2022), while jackals and foxes showed infection rates of 23.3% and 20.4% in Russia, respectively, and also golden jackal showed an infection rate of 19.05% in Romania. Notably, some regions in Europe have reported high prevalence not only in raccoon dogs but also in other wild carnivores. Taken together, these studies suggest that large-scale investigations of D. immitis infection should include both diverse geographic regions and multiple wild carnivore hosts, including raccoon dogs, since D. immitis is a mosquito-borne disease and different carnivore species may serve as potential reservoirs for one another.
In Korea, D. immitis infection has been widely reported in various animal hosts, including dogs and cats. A meta-analysis estimated a prevalence ranging from approximately 11.7% to 16.5% in pet dogs (Pak, Reference Pak2009). And a separate survey of stray dogs in Seoul animal shelters reported a prevalence of 9.8% (Kim et al., Reference Kim, Kwak, Kim, Park, Kim and Lee2014). Feline infection has been also documented, with an overall prevalence of 6.0% reported in stray cats (Park et al., Reference Park, Lee, Lee, Oh, Maheswaran, Seo and Song2014). Xenomonitoring studies highlight the importance of mosquito-mediated transmission in endemic regions. For example, a 2021 study conducted in urban parks in Ulsan detected D. immitis DNA in mosquito pools and reported a high minimum infection rate of 6.4 per 1000 mosquitoes, indicating a significant risk of active transmission in urban green spaces (Cha et al., Reference Cha, Yoon, Lee and Jang2022). This broad epidemiological evidence underscores the zoonotic potential of D. immitis in Korea, further supported by documented human pulmonary dirofilariasis cases, the first of which was reported in 2000 with additional cases subsequently confirmed (Lee et al., Reference Lee, Park, Yong, Im, Jung, Jeong, Lee, Yong and Shin2000).
In order to examine how different sample types affect PCR-based detection, we compared results obtained from lung tissue and whole-blood samples. Lung tissue PCR consistently revealed additional infections compared to gross examination and whole-blood PCR. This indicates that lung tissue PCR is more sensitive than blood-based assays, a difference that can be explained by the occurrence of occult or asymptomatic infections (Venco et al., Reference Venco, Genchi, Genchi, Grandi and Kramer2008) and by the limitations of postmortem blood sampling, which is prone to coagulation, DNA degradation, and PCR inhibition.
With respect to Wolbachia, it was detected in a smaller proportion of cases relative to D. immitis. This discrepancy may reflect several biological factors, as the density of Wolbachia varies by host and developmental stages (Ferri et al., Reference Ferri, Bain, Barbuto, Martin, Lo, Uni, Landmann, Baccei, Guerrero, de Souza Lima, Bandi, Wanji, Diagne and Casiraghi2011). Moreover, the mitochondrial cox1 gene used for D. immitis detection is present in multiple copies per cell, which enhances sensitivity, whereas Wolbachia 16S rRNA is typically present at a single or low number and is difficult to amplify efficiently, thereby reducing detectability (Sawasdichai et al., Reference Sawasdichai, Chaumeau, Dah, Kulabkeeree, Kajeechiwa, Phanaphadungtham, Trakoolchengkaew, Kittiphanakun, Akararungrot, Oo, Delmas, White and Nosten2019). The reduced detection rate of Wolbachia in the current study is therefore likely attributable to a combination of these biological and molecular factors.
Phylogenetic analysis based on partial cox1 sequences clearly separated D. immitis from related filarioid nematodes. Korean isolates, including those from raccoon dogs, clustered tightly with global reference sequences, and haplotype analysis likewise indicated low genetic diversity within D. immitis. This clustering pattern, combined with the minimal haplotype variation detected, demonstrates the strong conservation of the cox1 region. It is highly effective for species-level identification, though not optimized for resolving intra-species genetic structure. More recently, efforts have shifted toward primer strategies that accommodate sequence variability to better resolve haplotype diversity, such as two separate cox1 fragment primers (Alsarraf et al., Reference Alsarraf, Carretón, Ciuca, Diakou, Dwużnik-Szarek, Fuehrer, Genchi, Ionică, Kloch, Kramer, Mihalca, Miterpáková, Morchón, Papadopoulos, Pękacz, Rinaldi, Alsarraf, Topolnytska, Vismarra, Zawistowska-Deniziak and Bajer2023) and a degenerate cox1 primer set (Huggins et al., Reference Huggins, Atapattu, Khieu, Traub and Colella2025). Similarly, Wolbachia sequences from D. immitis grouped within supergroup C, which belongs to one of the 6 major Wolbachia supergroups (A–F). Among these, supergroups A and B are typically associated with arthropods, C and D with filarioid nematodes, E with arthropods and F occurs in both arthropods and filarioid nematodes (Werren et al., Reference Werren, Baldo and Clark2008; Lefoulon et al., Reference Lefoulon, Bain, Makepeace, d’Haese, Uni, Martin and Gavotte2016). This result supports the phylogenetic placement of Wolbachia from Korean raccoon dogs within the well-established association of D. immitis and its endosymbiont, further reflecting the host-specific nature of this symbiosis.
Although this study provides novel insights, some limitations should be acknowledged. First, our sampling was restricted to the Seoul area, which may not represent nationwide prevalence or genetic variability. Second, a total of 21 sequences, including the 6 newly generated in this study, were used for phylogenetic analysis and haplotype assessment. As described in the ‘Materials and methods’ section, the work relied on retrospective wildlife necropsy materials, meaning that only 6 samples retained DNA of sufficient quality for sequencing. Although these sequences were valuable for species-level confirmation and comparative analyses, the limited number constrains fine-scale interpretation of population-level variation. Third, although cox1 remains a robust and widely used marker for species identification, integrating other loci (e.g., 12S rRNA, internal transcribed spacer regions, and nuclear markers) could broaden insights into genetic diversity and enhance resolution for population studies.
The present study establishes molecular detection of D. immitis and its endosymbiont Wolbachia in wild raccoon dogs in urban Seoul, offering new insight into their occurrence in urban wildlife. Our findings underscore the role of raccoon dogs as potential wildlife reservoirs of heartworms in urban and peri-urban environments and highlight the need for continuous monitoring and expanded molecular characterization across hosts and regions. In addition, this study demonstrates that lung tissue represents a valuable specimen type for improving the detection of subclinical infections and enhancing the accuracy of heartworm surveillance.
Data availability statement
Sequence data supporting this study are publicly available in GenBank under accession numbers PV628747, PV628786, PV628865, PX517297, PX517298, PX517305 (D. immitis cox1) and PV635601, PV637187, PX517299, PX517300 (Wolbachia 16S rRNA).
Acknowledgements
The authors would like to thank the staff of the Seoul Wildlife Centre for their assistance in providing raccoon dog specimens and for their commitment to wildlife rescue, treatment and rehabilitation, which play a vital role in fostering coexistence between humans and wildlife. This project was supported by the National Institute of Wildlife Disease Control and Prevention as ‘Specialized Graduate School Support Project for Wildlife Disease Specialists’.
Author contributions
J.R. conceived and designed the study, conducted all laboratory experiments, molecular analyses, statistical analyses, and drafted the manuscript. J.R., H.L. and H.K.L. carried out necropsies, and HL and HKL were responsible for sampling and data recording. K.K. and C.S.R. supervised the project and provided research support. W.G.Y. reviewed the manuscript. W.G.Y. and S.C.Y. served as the corresponding author and contributed to overall guidance and manuscript revision.
Financial support
Sample collection and laboratory analyses were fully supported by the institutional budget of the Seoul Metropolitan Government Research Institute of Public Health and Environment. This study was partially supported by the Research Institute for Veterinary Science, Seoul National University.
Competing interests
The authors declare there are no conflicts of interest.
Ethical standards
This study did not involve live animals. All raccoon dog carcasses were obtained through the Seoul Wildlife Centre, following standard procedures for wildlife rescue and necropsy submission. Therefore, approval from the Institutional Animal Care and Use Committee (IACUC) was not required.
Open access
