Mimivirus Circulation among Wild and Domestic Mammals, Amazon Region, Brazil

Fábio P. Dornas, Felipe P. Rodrigues, Paulo V.M. Boratto, Lorena C.F. Silva, Paulo C.P. Ferreira, Cláudio A. Bonjardim, Giliane S. Trindade, Erna G. Kroon, Bernard La Scola, and Jônatas S. Abrahão

Author affiliations: Universidade Federal de Minas Gerais, Belo Horizonte, Brazil (F.P. Dornas, F.P. Rodrigues, P.V.M. Boratto, L.C.F. Silva, P.C.P. Ferreira, C.A. Bonjardim, G.S. Trindade, E.G. Kroon, J.S. Abrahão); URMITE CNRS UMR 6236, IRD 3R198, Aix Marseille Université, Marseille, France (B. La Scola)

Abstract

To investigate circulation of mimiviruses in the Amazon Region of Brazil, we surveyed 513 serum samples from domestic and wild mammals. Neutralizing antibodies were detected in 15 sample pools, and mimivirus DNA was detected in 9 pools of serum from capuchin monkeys and in 16 pools of serum from cattle.

The group of nucleocytoplasmic large DNA viruses includes viruses that are able to infect different hosts, such as animals, green algae, and unicellular eukaryotes (1). Several members of this group are widely distributed in various environments, actively circulate in nature, and are responsible for outbreaks of medical importance (2,3). Mimiviridae, the newest family in this group, has been researched as a putative pneumonia agent and found in different biomes worldwide (3,5–9). The ubiquity of freeliving amebas and their parasitism by mimiviruses enhances the prospect that diverse environments could shelter these giant viruses (8–10). Mimiviruses can induce infection in a murine model, have had antibodies detected in patients with pneumonia, and can replicate in murine and human phagocytes (11,12). Moreover, although some authors suggest that mimivirus is a not frequent pneumonia agent (4), mimivirus has been isolated from a human with pneumonia (3).

The biomes in Brazil, particularly in the Amazon region, provide the diversity, species richness, and ecologic relationships ideal for identifying circulation of mimiviruses. Preliminary studies found Acanthamoeba polyphaga mimivirus (APMV) genomes in samples of bovine serum from Germany (13,14), indicating that the analysis of samples from vertebrates could be a way to explore and understand the circulation of this group of viruses in nature. We describe the detection of mimivirus antibodies and DNA in 2 mammalian species in the Amazon region of Brazil.

The Study

Figure 1

Thumbnail of States in Brazil where serum samples were collected for study of mimivirus in mammals. Dots indicate collection sites. ES, Espírito Santos; BA, Bahia; PA, Pará; TO, Tocantins; MA, Maranhão.

Figure 1. . States in Brazil where serum samples were collected for study of mimivirus in mammals. Dots indicate collection sites. ES, Espírito Santos; BA, Bahia; PA, Pará; TO, Tocantins; MA, Maranhão.

We selected 321 serum samples collected from wild monkeys from the Amazon region of Brazil during 2001–2002: 91 from black howler monkeys (Alouatta caraya) and 230 from capuchin monkeys (Cebus apella). Samples were collected in an overflow area of a fauna rescue program during the construction of a hydroelectric dam in Tocantins State (Figure 1, Appendix). The monkeys had no previous contact with humans. After blood collection, the animals were released into areas selected by environmental conservation programs. We also collected serum samples from cattle (Bos taurus): 147 samples from Pará and Maranhão States in the Amazon region and 45 from Bahia and Espírito Santo States in the Caatinga and Mata Atlântica biomes.

Figure 2

Thumbnail of Consensus bootstrap phylogenetic neighbor-joining tree of helicase gene from nucleocytoplasmic large DNA viruses showing alignment of mimivirus and megavirus isolates obtained from Cebus apella (CA) and bovids (Bos) in Brazil. Tree was constructed by using MEGA version 4.1 (www.megasoftware.net) on the basis of the nucleotide sequences with 1,000 bootstrap replicates. Bootstrap values >90% are shown. Nucleotide sequences were obtained from GenBank. Scale bar indicates rate of evo

Figure 2. . Consensus bootstrap phylogenetic neighbor-joining tree of helicase gene from nucleocytoplasmic large DNA viruses showing alignment of mimivirus and megavirus isolates obtained from Cebus apella(CA) and bovids (Bos) in Brazil....

All samples underwent serologic and molecular testing for mimivirus (Figure 2). Because total serum volumes were low, the specimens were grouped into pools of 2–5 serum samples (20 μL for each sample) from animals belonging to the same species that were from the same collection area. A total of 210 pools were compiled (Table). Pools were tested by real-time PCR targeting the conserved helicase viral gene (primers 5′-ACCTGATCCACATCCCATAACTAA-3′and 5′-GCCTCATCAACAAATGGTTTCT-3′). DNA extractions were performed by using phenol-chloroform-isoamyl alcohol, and DNA quality and concentration were checked by using a nanodrop spectrophotometer (Thermo Scientific, Waltham, MA, USA). PCRs were performed by using the One Step SYBr Green Master Mix (Applied Biosystems, Foster City, CA, USA), and real-time PCR quality and sensitivity parameters were adjusted, including efficiency (102.6%) and R2 (0.992). APMV (kindly provided by Didier Raoult, Marseille, France) was used as a positive control. The serum pools were manipulated in a laminar flow cabinet, separate from any virus samples, to avoid cross-contamination.

Of the 210 pools, 25 (11.9%) were positive for APMV (viral loads 1.4 × 103 to 2.3 × 106 copies/mL); 9 (4.3%) pools were capuchin monkey serum and 16 (7.6%) were bovine serum, all from the Amazon region. Mimivirus DNA was not detected in serum from black howler monkeys or cattle from Bahia and Espírito Santo States (Table). Using external primers 5′-ACCTGATCCACATCCCATAACTAAA-3′ and 5′-ATGGCGAACAATATTAAAACTAAAA-3′, we amplified a larger fragment of the helicase gene (390 bp) from all 25 positive samples; 12 positive serum pools, 4 from capuchin monkeys and 8 from cattle, were chosen for helicase gene sequencing and analysis. The helicase fragments were directly sequenced in both orientations and in triplicate (MegaBACE sequencer; GE Healthcare, Buckinghamshire, UK). The sequences were aligned with previously published sequences from GenBank by using ClustalW (www.clustal.org) and manually aligned by using MEGA software version 4.1 (www.megasoftware.net). Modeltest software (www.molecularevolution.org/software/phylogenetics/modeltest) was used determine which model of evolution was most appropriate for our analysis.