Near-complete genome of SARS-CoV-2 Delta variant of concern identified in a symptomatic dog (Canis lupus familiaris) in Botswana

SARS-CoV-2, a betacoronavirus causing COVID-19, likely originated from bats or pangolins. Previous coronaviruses have crossed species barriers, and prior to SARS-CoV-2, dogs were known to host alphacoronaviruses (e.g., canine enteric coronavirus) with typically mild disease. Emerging evidence supports anthroponotic transmission of SARS-CoV-2 between humans and animals, with infection biologically plausible in dogs and cats via shared ACE2 receptor. Dogs generally show low susceptibility and often asymptomatic infections with low RNA levels, while cats and ferrets are more susceptible. Infections have also been reported in wildlife (minks, tigers, lions, ferrets). Potential reservoirs, including immunocompromised humans and animals, may contribute to viral evolution and new variants. In Botswana, COVID-19 has not been designated a notifiable animal disease, and there are limited guidelines for managing infected animals. Using a One Health perspective, the study aimed to document and genetically characterize SARS-CoV-2 infection in a symptomatic dog whose owners had recent COVID-19 diagnoses, and to contextualize the canine viral genome within local human SARS-CoV-2 diversity.

The paper references studies documenting SARS-CoV-2 infections in companion animals (dogs and cats), experimental and natural susceptibility across species (ferrets, minks, lions), and the role of animals in viral evolution. It notes most pet infection reports originate outside Africa, with few from the continent. One Health frameworks are discussed for zoonoses management. Prior case reports identified Delta variant infections in dogs in the USA and Spain, and infections in Namibian dogs, supporting the plausibility of human-to-pet transmission and generally mild or asymptomatic clinical courses in pets.

- Case and sampling: A 5-year-old crossbreed male dog in Gaborone, Botswana, living with three adults under 10-day quarantine after SARS-CoV-2 diagnosis (rapid antigen tests on 2 August 2021), presented with a 4-day dry hacking cough, mild depression, appetite loss. Clinical evaluation on 5 August 2021 ruled kennel cough as less likely based on history, vaccination records, and absence of symptoms in in-contact dogs. The dog had prior vaccinations (Vanguard Plus 5 on 27/04/2018; Rabisin on 17/08/2019). - Sample collection and handling: Under restraint and sedation (0.3 mL dexmedetomidine 0.5 mg/mL; 1.0 mL ketamine 50 mg/mL; 0.2 mL butorphanol 10 mg/mL), buccal and nasopharyngeal swabs and bronchoalveolar lavage (BAL) were collected on 5–6 August 2021. Specimens were transported at 2–8°C to the Botswana Harvard HIV Reference Laboratory (BHHRL), ISO 15189 recognized by SADCAS. - Nucleic acid extraction and RT-qPCR: Extraction used a Nucleic Acid Extraction Kit (Wuhan MGI Tech) as per manufacturer. Diagnostic RT-qPCR employed the 2019-nCoV RNA (PCR-Fluorescence Probing) Assay (Sun Yat-sen University, Da An Gene). Controls included extraction control, master-mix-only control, positive control, and no-template control. To increase RNA yield, specimens were extracted twice; after centrifugation at 22,000 × g for 15 min, pellets were resuspended in lysis buffer before bead-based automated extraction. RT-qPCR targeted the N gene; initial nasopharyngeal/buccal swabs showed weak positivity with qCt ~36; BAL and deep nasal swabs on 6 August had Ct 38 and 36, respectively. - Tiling PCR and next-generation sequencing: Residual RT-qPCR RNA underwent ARTIC network tiling PCR using Superscript IV and ARTIC V3 primers (Inqaba Biotech). Amplicons were sequenced on Oxford Nanopore MinION Mk1B with MinKNOW 21.05.12. - Bioinformatics and genome submission: FASTQ reads were processed with Genome Detective. Sequence quality and coverage were assessed with Nextclade v1.7.4 (accessed 10–11–2021). Lineage assignment used pangolin (v3.1.20, accessed 2022-02-28) and later Pango v4.0.6 (PLEARN v1.3, 2022-04-21). The consensus genome was deposited to GISAID (EPI_ISL_8900015). - Phylogenetic and mutation analysis: Mutation profiling highlighted Spike and non-Spike mutations. For phylogeny, all near-full-length Delta variant sequences from GISAID available at the time (n=31, including from Botswana) and a larger set from Botswana (as of 21 April 2022; n≈1303, largely human-derived) were included. Sequences were aligned with NextAlign; maximum likelihood trees were inferred using FastTree after model selection with ModelTest v3.7; bootstrap set to 1000. Reference NC_044512 was used. Trees were inspected with SeaView and visualized in FigTree v1.4.3; posterior probabilities >0.80 considered significant.

- Detection and sequencing: SARS-CoV-2 RNA was detected in a symptomatic dog’s nasopharyngeal/buccal swabs (qCt ~36) and confirmed in BAL (Ct 38) and deep nasal swabs (Ct 36). - Genome recovery: A near-full-length genome (~90% coverage) was generated and deposited (GISAID EPI_ISL_8900015). - Lineage/clade: The genome was classified in clade 20O and assigned to a Delta sublineage. Using pangolin v3.1.20 (2022-02-28), it was AY.94; using Pango v4.0.6 (2022-04-21), it was AY.43, reflecting dynamic lineage updates. - Mutation profile: Spike mutations included T19R, T478K, D614G, P681R, D950N; other mutations included M:I82T, N:D63G, N:R203M, N:G215C, ORF7a:T120I, ORF7a:V82A, ORF7b:T40I, NSP4:T492I, NSP5:S113A, NSP6:Q257T and T77A, NSP12:P323L, NSP13:P77L. - Phylogenetics: The canine sequence clustered among contemporaneous Delta sequences circulating in Botswana (AY.43, AY.116, B.1.617.2), with no novel mutations distinguishing it from human sequences. In an exploratory analysis of Botswana Delta sequences up to April 2022 (n≈1303; predominantly human-derived), clades included G (n=23), GH (n=5), GK (n=1219), GR (n=37), GRA (n=13), GV (n=1), O (n=5); Delta sublineages included AY.1 (n=9), AY.112 (n=40), AY.116 (n=326), AY.45 (n=168), AY.46 (n=664), AY.43 (n=16), B.1.617.2 (n=24), among others. - Epidemiologic context: This represents the first documented canine SARS-CoV-2 genome from Botswana and one of the first near-full-length nonhuman SARS-CoV-2 sequences from Africa. The dog lived with three recently infected adults, supporting likely human-to-dog transmission.

Findings support that dogs can acquire SARS-CoV-2, likely from infected owners during periods of close contact. The dog exhibited milder clinical signs than those typically observed in symptomatic humans during the same wave, consistent with reports of mostly asymptomatic or self-limiting infections in pets. The canine genome’s clustering within circulating human Delta sublineages and the absence of unique mutations reinforce the plausibility of anthroponotic transmission rather than sustained pet-to-pet or pet-to-human spread. The case underscores the importance of One Health approaches, and suggests individuals with COVID-19 should minimize contact with pets during illness to reduce spillover risks. Ongoing genomic surveillance in domestic animals can help detect potential variant spillovers and monitor interspecies transmission dynamics.

This case report documents the first near-complete SARS-CoV-2 genome from a dog in Botswana (Delta variant sublineage) and one of the earliest nonhuman sequences from Africa. The infection likely reflects human-to-dog transmission in a household with active COVID-19. The study highlights the need to limit human–animal contact during infection and to implement surveillance of SARS-CoV-2 in domestic animals to better understand and mitigate interspecies transmission risks. Future work should include concurrent sampling and sequencing of human household contacts, longitudinal sampling of pets, and broader animal surveillance across epidemic waves.

- No human SARS-CoV-2 genomes from the infected owners were available for direct comparison, as they were diagnosed by rapid antigen tests. - Phylogenetic analysis relied on the canine genome and available human sequences from public datasets, limiting precise source attribution. - Environmental acquisition of infection (e.g., contaminated surfaces) cannot be excluded. - High Ct values indicate low viral load, which may limit sequencing completeness and variant detection.