Infection, Genetics and Evolution 4 (2004) 45–51 A phylogenetic reconstruction of the epidemiological history of canine rabies virus variants in Colombia Gareth J. Hughes a , Andrés Páez b , Jorge Bóshell b , Charles E. Rupprecht a,∗ a Rabies Section, Centers for Disease Control and Prevention, 1600 Clifton Road, Mail-Stop G33, Atlanta, GA 30333, USA b Laboratorio de Virologia, Instituto Nacional de Salud (INS), Av. El Dorado Cra 50, Bogotá, Colombia Received 14 November 2003; received in revised form 4 December 2003; accepted 5 December 2003 Abstract Historically, canine rabies in Colombia has been caused by two geographically distinct canine variants of rabies virus (RV) which between 1992 and 2002 accounted for ∼95% of Colombian rabies cases. Genetic variant 1 (GV1) has been isolated up until 1997 in the Central Region and the Department of Arauca, and is now considered extinct through a successful vaccination program. Genetic variant 2 (GV2) has been isolated from the northern Caribbean Region and continues to circulate at present. Here we have analyzed two sets of sequence data based upon either a 147 nucleotide region of the glycoprotein (G) gene or a 258 nucleotide region that combines a fragment of the non-coding intergenic region and a fragment of the polymerase gene. Using both maximum likelihood (ML) and Markov chain Monte Carlo (MCMC) methods we have estimated the time of the most recent common ancestor (MRCA) of the two variants to be between 1983 and 1988. Reconstructions of the population history suggest that GV2 has been circulating in Colombia since the 1960s and that GV1 evolved as a separate lineage from GV2. Estimations of the effective population size at present show the GV2 outbreak to be ∼20 times greater than that of GV1. Demographic reconstructions were unable to detect a decrease in population size concurrent with the elimination of GV1. We find a raised rate of nucleotide substitution for GV1 gene sequences when compared to that of GV2, although all estimates have wide confidence limits. We demonstrate that phylogenetic reconstructions and sequence analysis can be used to support incidence data from the field in the assessment of RV epidemiology. © 2003 Elsevier B.V. All rights reserved. Keywords: Rabies; Epidemiology; Population dynamics; Evolution; Phylogenetics; Colombia 1. Introduction Rabies virus (RV) is the type member of the genus Lyssavirus, family Rhabdoviridae. RV has a negative-sense RNA genome that is ∼12 kb in length, and consists of five genes: nucleoprotein, phosphoprotein, matrix protein (M), glycoprotein (G), and polymerase (Tordo et al., 1988). The G seems to control all major aspects of host cell infection such as receptor binding, antigenicity and host adaptation (Badrane and Tordo, 2001; Wunner, 2002). RNA viruses are characterized by a high mutation rate during replication due to the lack of proofreading and post-replication error correction by RNA polymerases (Domingo and Holland, 1994). This genetic diversity seems to provide an adaptive potential for RV that can vary according to natural history (Morimoto et al., 1998; Kissi et al., 1999). ∗ Corresponding author. Tel.: +1-404-639-1050; fax: +1-404-639-1564. E-mail address: cyr5@cdc.gov (C.E. Rupprecht). In South America, rabies exists as two distinct epidemi- ological forms: urban rabies (in which the domestic dog is the principal reservoir) and sylvatic rabies (where the principal reservoir can be any one of a number of wildlife species, e.g., bats, foxes, etc.). Urban dog rabies presents the greatest risk to human health and there has been consid- erable effort to eliminate urban dog rabies from a number of Latin American countries. In Colombia, rabies reservoirs exist in domestic dog, bat (Páez et al., 2003) and gray fox populations (Páez et al., in preparation). From molecular epidemiological analysis, Colombian isolates of canine RV can be divided into two distinct variants: genetic variant 1 (GV1) and genetic variant 2 (GV2). These are separated spatially (GV1 in the more southern Central Region and the Department of Arauca; GV2 in the northern Caribbean Region) and temporally (albeit with considerable overlap). GV1 was last isolated in 1997 and is considered to be eliminated (Páez et al., 2003). During 1992–2002, infection with GV2 accounted for ∼70% of rabid dogs in Colombia (Páez et al., 2003). 1567-1348/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.meegid.2003.12.001