Swinepox Virus Outbreak, Brazil, 2011

To the Editor: Swinepox virus (SWPV), which replicates only in swine, belongs to the Suipoxvirus genus of the Poxviridae family. It is the etiologic agent of a skin disease of pigs, characterized by generalized pustular lesions and associated with high rates of illness (occasionally >80%). It occurs mainly on farms with poor management and housing conditions and affects primarily pigs <3 months of age; adult pigs show milder signs. The disease is mechanically transmitted by pig lice or through direct animal contact (1). Vaccinia virus (VACV; Orthopoxvirus genus) also causes a similar pustular disease in pigs that is difficult to distinguish clinically from SWPV infections. VACV infections were common during smallpox vaccination campaigns, when VACV was transmitted to domestic animals from lesions of vaccinees (1,2).

. Therefore, distinguishing between SWPV and VACV infections during outbreaks of pustular disease in pigs is essential for evaluating whether VACV infection might have spread to pigs and whether SWPV could be detected in Brazil.
We describe the molecular identifi cation of SWPV as the etiologic agent of an outbreak of pustular disorder in pig herds. In November 2010 and January 2011, ≈850 of 3,460 animals on 3 pig farms in Holambra, São Paulo, Brazil, had generalized pustular lesions on the body, fever (38.0°C-39.7°C) and mild weight loss. Lesions evolved from macules or papules to umbilicated lesions with pustular content, followed by crusting (online Appendix Figure, www.cdc. gov/EID/content/17/10/11-0549-appF.htm). Secondary dermatitis was also noticed. Healing occurred after 3-4 weeks, but the disease started subsequently in previously healthy animals. Although the fi rst clinical signs of disease started in the nursery units (pigs 40-50 days old), nearly 70% of the sick pigs were at the fi nishing units (pigs 127-134 days old), where elevated animal density and defi cient sanitation conditions were observed. These fi ndings may account for the high attack rate (nearly 50%) in fi nisher pigs, although overall illness was moderate (nearly 25%) when animals from all units were analyzed together. No deaths were associated with the outbreak, in concordance with the low death rates reported for SWPV infections (<5%) (1). The affected farms belonged to the same owner, who reported frequent movement of animals between the farms. Scabs from 7 animals were used for DNA extraction (4), followed by PCR detection of poxvirus DNA (7). We used primers designed to anneal to gene regions conserved in different poxviruses: FP-A2L, 5′-TAGTTTCAGAACAAGGATA TG-3′ and RP-A2L, 5′-TTCCCATAT TAATTGATTACT-3′ directed the amplifi cation of a 482-bp fragment of the virus late transcription factor-3 (www.poxvirus.org); primer sets for the DNA polymerase gene (543-bp fragment) and DNA topoisomerase gene (344-bp fragment) were previously described (7). Amplicons were directly sequenced as described (4,5). Consensus primers that specifi cally detect the full-length hemagglutinin gene of Eurasian-African orthopoxviruses were used to investigate VACV in the samples (4).
The nucleotide sequences obtained for the fragments of the DNA polymerase, DNA topoisomerase, and virus late transcription factor-3 of the clinical specimens were aligned with sequences from other poxviruses available in the public database (GenBank). They showed 100% nt identity with their orthologs of SWPV Nebraska strain. Concatenated amino acid alignments were used for phylogenetic inference (Figure). The clinical isolates and SWPV branched together in the phylogenetic tree with high bootstrap support. No amplifi cation of the hemagglutinin gene was obtained, demonstrating that the animals were not infected with VACV. Samples were also negative for Erysipelothrix spp. (by PCR and ELISA) and porcine circovirus-2 (by PCR).
Outbreaks of swinepox disorders have been frequently reported in Europe, North America, and Oceania, and special attention has been given to congenital cases, which usually lead to high case-fatality rates (2,8,9). Our data identifi ed SWPV as the cause of a recent outbreak in Brazil and suggest that previous outbreaks in the neighboring municipality of Campinas in 1976 and 1980 (3) may have been caused by SWPV as well because pigs are the only host and reservoir of the virus. Further sequencing analysis of the virus isolates will be necessary to characterize the strain of SWPV circulating in Brazil.
Recently, an outbreak of VACVrelated disease in horses was reported in southern Brazil, which alerted the scientifi c community to the possible spread of this disorder to animal hosts other than dairy cattle (10). However, our data clearly demonstrate that this outbreak in pigs does not represent a spread of VACV infection, despite frequent reports of VACV-related outbreaks in dairy cows in São Paulo State (6). Therefore, the differential diagnosis of skin diseases of pigs might be a useful tool in epidemiologic surveys to access VACV spread and host range in Brazil. To the Editor: Having worked with numerous species of research nonhuman primates over the past 26 years, I have a keen interest in related occupational health and safety. In this regard, I was quite interested in the recent report by Li et al. (1) and have some comments and questions relative to this article.
The occurrence of Plasmodium spp. infection in feral primates, feral source captive primates, or primates