A 60 year-old woman who developed symptoms on the 12th February 2020 was diagnosed with COVID–19 disease on 24 February 2020. She kept a 17 year-old neutered male Pomeranian which had a number of pre-existing diseases, including a Grade II heart murmur, systemic and pulmonary hypertension, chronic renal disease, hypothyroidism and a previous history of hyperadrenocorticism (Dr Florence Chan pers comm). One female domestic helper (secondary case A) in the household developed a fever on 16th February and was subsequently confirmed to be infected. The remaining three members of the household, who were not known to be infected were sent to a quarantine centre on 26th February. That evening, the dog was transferred to a holding facility managed by AFCD and nasal, oral, and rectal swabs as well as a faecal sample were collected. Additional specimens for virus detection were collected from the dog on 28th February, 2nd March, 5rd March and 9th March. A blood sample was collected on 3rd March for serological testing.
Throughout the period in quarantine the dog remained bright and alert with no obvious change in clinical condition. On 7th March a saliva sample was collected from the second domestic helper (secondary case B) who was asymptomatic. Specimens from the three patients were confirmed to be SARS-CoV–2 positive by RT-PCR at the Public Health Laboratory Centre and viruses genetically sequenced (Institutional Review Board approval UW20–168). Until admission to the quarantine centre or hospital on 26 February, the two domestic helpers shared responsibility for daily care of the dog. (Figure 1).
Nasal and oral swabs collected from the dog on the 26th February were positive by quantitative RT-PCR for SARS-CoV–2 RNA while rectal and faecal specimens were negative. Initial results obtained at AFCD’s Tai Lung laboratory were confirmed by additional tests on the same samples done at the School of Public Health, The University of Hong Kong (HKU), a World Health Organization reference laboratory for SARS-CoV–2 diagnostics. The positive samples were RT-PCR positive for six gene-sequence targets of SARS-CoV–2 (see Table 1). AFCD used assays for the E and rdrp gene sequences (TIB Molbiol Lightmix® Modular Assays). HKU assays for nsp14 and N gene that detect SARS-CoV, SARS-CoV–2 and bat SARS- CoV (2) and assays for nsp16 and M gene that are specific for nCoV with no reaction with SARS-CoV.4 On subsequent sampling occasions, the nasal swabs remained positive throughout, the oral swabs tested positive on the first two samplings (on 2nd March only E gene was positive). Rectal swabs and faecal specimens remained negative throughout. Nasal swabs had higher viral load (lower CT values) than the oral swabs.
Attempts to culture the RT-PCR positive nasal swab samples collected on three occasions (28th February, 2nd & 5th March) on Vero-E6 cells gave negative results. Given the low viral load (range 7.5xE2 to 2.6 x10E4 RNA copies per mL of specimen) it was unlikely that virus culture would be successful. In human patients with COVID19, virus isolation was not successful when viral load in the specimen was <10E6 per mL.5 The serum sample collected from the dog on 3 March 2020, together with 20 control samples from unaffected dogs were tested in micro-neutralization test for SARS-CoV–2. All samples, including that of the infected dog tested negative for neutralizing antibody. In the absence of invasive disease in the dog, it is not surprising that the dog remained sero-negative. It is known that patients with mild or asymptomatic MERS-CoV infection fail to develop antibody responses.6
Viral RNA from the nasal swab specimen collected from the dog on 26th February was genetically sequenced and compared with the virus found in clinical specimens from the owner and the secondary case A. Full virus genome sequence (29,764 nucleotides) was obtained from the index case and secondary case A and 27,871 nucleotides (94% of genome; from nucleotide position 1933 to 19,803) was obtained from the initial nasal swab of the dog (Table 2) (Supplementary Fig 1). The viral sequences from the index case and the first domestic helper were identical across the full genome. The viral sequence from the dog differed at 3 nucleotide positions, viz. A6,567G within nsp3 gene, A18,950G within nsp14 and C29,730T within 3’UTR region. The mutations at positions 6567 and 18,950 were non- synonymous leading to amino acid changes D6567G and K18950R respectively. Interestingly, there was nucleotide heterogeneity at position 6567 in the dog virus sequence with 41% of sequence reads being A as found in the index case while 59% were G (sequence coverage >500 fold). Partial gene sequencing was conducted on the sample from secondary case B covering the region of the nucleotide changes seen in the canine sample but the nucleotide residues were identical to those found in the index case rather than the dog.