Figure 2

Figure 2. Chronologic detection of vaccinia virus in South America. Zoomed-in map shows location of Durazno County, Uruguay, where serum samples were collected from dairy cattle in 2009 to test for the presence...

We analyzed serum samples that were collected in May 2009 from 125 dairy cows in Durazno County (33°23′0′′S, 56°31′0′′W), Durazno State, Uruguay (Figure 2). The cattle herds had no clinical sings of disease at the time of serum collection. To determine the presence of neutralizing antibodies in the serum samples, we used an orthopoxvirus plaque-reduction neutralization test as previously described (9). The serum titer was defined as the highest dilution that inhibited >70% of virus plaques compared with negative controls (4).

Because previous studies have detected viral DNA in the serum of animals and humans with and without clinical manifestations (4,10,11), we performed a molecular investigation to identify orthopoxvirus. We used quantitative PCR (qPCR) to amplify VACV growth factor gene (C11R) DNA. This qPCR tool has high sensitivity and specificity and, thus, has been routinely used as an orthopoxvirus diagnostic tool by our group (12). For molecular characterization, we used the nonconserved orthopoxvirus hemagglutinin gene (A56R) (13). We used VACV–Western Reserve as the PCR-positive control for amplification and characterization.

The PCR A56R products obtained from C11R PCR–positive samples were sequenced in both orientations and subjected to capillary electrophoresis (3130 Genetic Analyzer, BigDye Terminator Cycle Sequencing Kit v3.1; Applied Biosystems, Foster City, CA, USA). We used the ClustalW (http://www.clustal.org/) method to align sequences with previously published orthopoxvirus sequences from GenBank; alignments were manually checked with MEGA6 (http://www.megasoftware.net/). We constructed phylogenetic trees using the neighbor-joining method with 1,000 bootstrap replicates and the Tamura 3-parameter model in MEGA6. All field and laboratory clinical samples were processed separately to avoid cross-contamination. Serologic and molecular tests were performed in 2 independent experiments and in duplicate.

We detected neutralizing antibodies against orthopoxvirus in 28 (22.4%) of 125 serum samples from cattle in Uruguay; titers were 100 neutralizing units (NU)/mL in 10 (35.7%) samples, 200 NU/mL in 11 (39.3%) samples, 400 NU/mL in 5 (17.9%) samples, and 800 NU/mL in 2 (7.1%) samples (Technical Appendix Figure 1, panel A). To confirm that the orthopoxvirus seropositivity represented seropositivity to VACV, we used the C11R gene to molecularly test DNA in the serum samples; 2 (1.6%) samples were positive by qPCR, and 1 of those was also positive by the plaque-reduction neutralization test.

One of the 2 C11R PCR–positive isolates was also positive for the A56R gene; we sequenced the gene and named the strain VACV Uruguay. Alignment of the A56R nucleotide sequence showed the presence of an 18-nt signature deletion, which is also present in sequences of Brazilian-VACV group I, but not group II, viruses. Unlike sequences for other VACVs, the sequence for VACV Uruguay had an A→T polymorphism (Technical Appendix Figure 2, panel A). VACV Uruguay exhibited higher identity with group I (98.6% identity) viruses from Brazil and Argentina than to group II (97.3% identity) viruses from Brazil (Technical Appendix Figure 1, panel B). Furthermore, in the phylogenetic tree based on A56R nucleotide sequences, the VACVs from Uruguay clustered with group I VACVs that had been detected during outbreaks in Brazil and with viruses from Argentina (Technical Appendix Figure 2, panel B).

Since 1999, VACV has been isolated from symptomatic and asymptomatic cattle, humans, and wildlife from the north to the extreme south of Brazil (5,6,8), and in 2014, VACV was described in bovine serum samples from Argentina (4). Although no exanthematous VACV outbreaks have been reported among cattle in Uruguay, we detected orthopoxvirus antibodies and VACV DNA in serum samples from dairy cattle in the country, indicating they have been exposed to VACV. Uruguay shares a border with Brazil, and its western border is shared with Entre Rios Province in Argentina, where VACV DNA has been detected. In addition, Uruguay shares its northern and eastern borders with Rio Grande do Sul State, Brazil, where Pelotas VACV has been isolated from horses (4,5).

Our finding of orthopoxvirus antibodies and VACV DNA indicates a possible undetected or silent circulation of VACV in Uruguay. Considering the importance of the livestock sector in all countries of South America, concern exists about the possible spread of VACV beyond Brazil, Argentina, and Uruguay (4–6,14). Despite surveillance by veterinarians, efforts to stop the spread of VACV at borders may be hampered by the movement of infected rural workers, the marketing of asymptomatic live animals, and the misdiagnosis of VACV infection. Furthermore, VACV has been shown to circulate in wild environments, and it has been hypothesized that rodents may serve as VACV hosts, as they do for other orthopoxviruses, and facilitate the spread of VACV in border areas (15).

Bovine vaccinia outbreaks in South America were first reported in Brazil, but we cannot rule out the possibility of autochthonous circulation of VACV in Uruguay and other countries in South America. Additional studies are needed to elucidate VACV seroprevalence in other countries in South America, and further research is needed to clarify the transmission pathways related to the spread of VACV in South America.