Molecular epidemiology of rabies in Colombia 1994–2005 based on partial nucleoprotein gene sequences

Available online at www.sciencedirect.com

Virus Research 130 (2007) 172–181

Molecular epidemiology of rabies in Colombia 1994–2005 based on partial nucleoprotein gene sequences夽 Andr´es P´aez a,b,∗ , Andr´es Velasco-Villa c , Gloria Rey a , Charles E. Rupprecht c a

Laboratorio de Virolog´ıa, Instituto Nacional de Salud (INS), Av. El Dorado Cra 50 CAN, Bogot´a D.C., Colombia b Departamento de Ciencias B´ asicas, Universidad de La Salle, Sede de la Candelaria, Bogot´a D.C., Colombia c Division of Viral and Rickettsial Diseases, Centers for Disease Control and Prevention, 1600 Clifton Road, Mailstop G33, Atlanta, GA 30333, USA Received 6 February 2007; received in revised form 25 May 2007; accepted 11 June 2007 Available online 23 July 2007

Abstract One hundred and twenty-four rabies viruses (RABV) were isolated from humans and eight species of mammals in Colombia during 1994–2005. To determine the genetic and reservoir-associated diversity cDNA fragments encoding 88 amino acids at the carboxyl terminus of the nucleoprotein were sequenced and used in phylogenetic analyses. Eight genetic lineages (GL) were characterized. GL1, GL2 and GL3 consisted of dog-associated antigenic variant (AV) 1 RABV, isolated in the centre-east, north and southwest of Colombia, respectively. GL1 is apparently extinct in Colombia. The GL4 were AV3, AV8 and non-determined (ND) AV viruses associated with hematophagous bats. The GL5 and GL6 consisted of AV4 viruses. GL6 isolate was found associated with Tadarida brasiliensis bats. GL5 segregated independently. The GL7 and GL8 segregated independently within clades associated with colonial insectivorous and solitary bats, respectively. Both of these were represented by NDAV viruses. Viruses isolated from humans grouped within GL2, GL3 and GL4, which in turn corresponded to AV1, 3, 8 and ND. Dogs and D. rotundus are the two major rabies reservoirs and vectors in Colombia. Insectivorous bats may also be important rabies reservoirs but spillovers to other species are rare. Our data were consistent with previous studies in which partial Psi, G and L gene sequences were analyzed. Our results confirmed the existence of RABV of unclassified AV in Colombia. © 2007 Elsevier B.V. All rights reserved. Keywords: Rabies; Molecular epidemiology; Nucleoprotein; Colombia; Virus; Zoonosis; Phylogeny

1. Introduction Rabies is caused by highly neurotropic viruses belonging to Lyssavirus genus, family Rhabdoviridae (Beran and Steele, 1994). The virion contains a single-stranded, non-segmented RNA genotype of 11–12 kb that encodes five structural proteins (Wunner et al., 1988; Tordo and Kouknetzoff, 1994). Rabies virus (RABV) reservoirs belong mainly to Carnivora and Chi夽 The findings and conclusions in this report are those of the authors and do not necessarily represent the view of the funding agencies. Use of trade names and commercial sources are for identification only and do not imply endorsement by the USA Department of Health and Human Services. ∗ Corresponding autor at: Laboratorio de Virolog´ıa, Instituto Nacional de Salud (INS), Av. El Dorado Cra 50 CAN, Bogot´a D.C., Colombia. Tel.: +57 1 2207700x442/549; fax: +57 1 2200928; Departamento de Ciencias B´asicas, Universidad de La Salle, Bogot´a D.C., Colombia. Tel.: +57 1 3535360x2501; fax: +57 1 3535360x2501. E-mail addresses: [email protected], [email protected] (A. P´aez).

0168-1702/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2007.06.008

roptera species, which may transmit the disease to cul-de-sac mammals, including humans. Rabies occurs in two different epidemiological forms: urban rabies, with the domestic dog as the main reservoir and transmitter, and sylvatic rabies, with different wildlife species acting as reservoirs. Urban and sylvatic rabies is an important public health and economic problem in most Latin American countries. Rabies antigenic typing and molecular epidemiology studies have recently gained relevance in Latin America as a way to investigate the dynamics of transmission both in the geographical and inter species area of knowledge (D´ıaz et al., 1994; De Mattos et al., 1996, 1999; Ito et al., 2001; Favoretto et al., 2002; P´aez et al., 2003, 2005; Hughes et al., 2004; Cisterna et al., 2005; Kobayashi et al., 2005; Velasco-Villa et al., 2005, 2006; Carnieli et al., 2006). Both, urban and sylvatic rabies are present in Colombia, and spillover of urban rabies into the wild and the opposite has been reported (P´aez et al., 2005). According to official reports of the National Institute of Health in Colombia, of 1139 cases of urban rabies registered to the Min-

A. P´aez et al. / Virus Research 130 (2007) 172–181

173

Table 1 Epidemiological information for 124 RABV isolates from Colombia 1994–2005 Sample

Host

Year

Geographical location

AV

GL

Genebank accession

H02/94 CT1/94 H02/95 H03/95 H04/95 H05/95 H06/95 CT1/95 B01/95 BV1/95 BV2/95 BV3/95 SH1/95 HR1/95 PG1/95 C14/95 C15/95 C16/95 C17/95 C18/95 C19/95 CT1/96 CT2/96 CT3/96 CT4/96 BV1/96 BV2/96 BV3/96 BV4/96 BV5/96 BV6/96 HR1/96 C01/96 C03/96 C04/96 C05/96 C06/96 C07/96 C08/96 C13/96 H03/97 H05/97 H06/97 CT1/97 CT2/97 BV1/97 BV2/97 BV3/97 BV4/97 BV5/97 BV6/97 BV7/97 BV8/97 BV9/97 BV0/97 HR1/97 C01/97 C02/97 C03/97 C04/97 C05/97 C24/97 C30/97 C31/97

Human Cat Human Human Human Human Human Cat Bat (Molossus molossus) Cow Cow Cow Goat Horse Swine Dog Dog Dog Dog Dog Dog Cat Cat Cat Cat Cow Cow Cow Cow Cow Cow Horse Dog Dog Dog Dog Dog Dog Dog Dog Human Human Human Cat Cat Cow Cow Cow Cow Cow Cow Cow Cow Cow Cow Horse Dog Dog Dog Dog Dog Dog Dog Dog

1994 1994 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997 1997

Sasaima, Cundinamarca Neiva, Huila Sincelejo, Sucre Valledupar, Cesar Mar´ıa la Baja, Bol´ıvar Ca˜naveral, Bol´ıvar Barranquilla, Atl´antico Guamo, Tolima Cali, Valle Chocont´a, Cundinamarca Tunja, Boyac´a Chiquinquir´a, Boyac´a Tunja, Boyac´a Subachoque, Cundinamarca Tunja, Boyac´a Tunja, Boyac´a Samac´a, Boyac´a Tunja, Boyac´a Arjona, Bol´ıvar Ventaquemada, Boyac´a Chiquinquir´a, Boyac´a Mesitas, Cundinamarca Arauca, Arauca La Mesa, Cundinamarca Viot´a, Cundinamarca Chiquinquir´a, Boyac´a Chiquinquir´a, Boyac´a Tunja, Boyac´a Tame, Arauca Chiquinquir´a, Boyac´a Arauca, Arauca Chiquinquir´a, Boyac´a Chiquinquir´a, Boyac´a Ubat´e, Cundinamarca Arauca, Arauca Saboya, Boyac´a Campo de la Cruz, Atl´antico Armenia, Quindio Arauca, Arauca Sop´o, Cundinamarca Pto. Rico, Bol´ıvar Monter´ıa, Cordoba Ceret´e, Cordoba Zipaquir´a, Cundinamarca Sn. Pelayo, Cordoba Chiquinquir´a, Boyac´a Tabatinga, Amazonas Pto. L´opez, Meta Mahates, Bol´ıvar Chiquinquir´a, Boyac´a Neira, Caldas Curuman´ı, Cesar Leticia, Amazonas Pijao, Quind´ıo Mocoa, Putumayo Viot´a, Cundinamarca Turbaco, Bol´ıvar Arauca, Arauca Silvana, Cundinamarca Arauca, Arauca Mahates, Bol´ıvar Arauca, Arauca Arauca, Arauca Ibagu´e, Tolima

ND 8 1 1 3 1 1 ND. 4 1 1 1 1 1 1 1 1 1 1 1 1 3 1 ND ND 1 1 1 3 1 3 1 1 1 1 1 1 3 1 1 1 1 1 3 1 1 3 3 1 1 ND 3 3 1 3 1 1 1 3 1 1 1 1 ND

4 4 2 2 4 2 2 4 6 1 1 1 1 1 1 1 1 1 2 1 1 4 1 4 4 1 1 1 4 1 4 1 1 1 1 1 2 4 1 1 2 2 2 4 2 1 4 4 2 1 4 4 4 1 4 1 2 1 4 1 2 1 1 4

EF377678 EF377676 EF377619 EF377613 EF377686 EF377637 EF377612 EF377675 EF377701 EF377673 EF377652 EF377654 EF377645 EF377647 EF377651 EF377643 EF377644 EF377646 EF377650 EF377648 EF377681 EF377671 EF377694 EF377674 EF377677 EF377693 EF377688 EF377649 EF377668 EF377695 EF377662 EF377636 EF377683 EF377689 EF377672 EF377682 EF377631 EF377661 EF377691 EF377684 EF377585 EF377633 EF377634 EF377696 EF377616 EF377680 EF377698 EF377663 EF377655 EF377635 EF377657 EF377665 EF377697 EF377690 EF377699 EF377626 EF377615 EF377670 EF377640 EF377687 EF377642 EF377653 EF377685 EF377658

174

A. P´aez et al. / Virus Research 130 (2007) 172–181

Table 1 (Continued) Sample

Host

Year

Geographical location

AV

GL

Genebank accession

H06/99 C02/99 H01/00 CT1/00 B01/00 F01/00 F02/00 F03/00 F04/00 F05/00 F07/00 F08/00 CT1/01 CT2/01 BV1/01 F10/00 F11/01 F12/01 F13/01 C01/01 C02/01 C03/01 C04/01 C05/01 C08/01 C09/01 C11/01 C12/01 C14/01 C15/01 C57/01 C58/01 C61/01 B02/02 HR1/02 F15/01 F16/02 C01/02 C02/02 C03/02 C04/02 C05/02 H01/03 CT1/03 BV1/03 BV2/03 F18/02 F19/02 F21/03 F22/03 C01/03 C02/03 H01/04 B01/04 F23/04 C01/04 C02/04 C03/04 C04/04 H01/05

Human Dog Human Cat Bat (Eptesicus brasiliensis) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Cat Cat Cow Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Bat (E. brasiliensis) Horse Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Dog Dog Dog Dog Dog Human Cat Cow Cow Fox (U. cinereoargenteus) Fox (U. cinereoargenteus) Fox (ND) Fox (U. cinereoargenteus) Dog Dog Human Bat (ND insectivorous) Fox (U. cinereoargenteus) Dog Dog Dog Dog Human

1999 1999 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2001 2001 2001 2000 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2001 2002 2002 2002 2002 2002 2002 2002 2002 2002 2003 2003 2003 2003 2003 2003 2003 2003 2003 2003 2004 2004 2004 2004 2004 2004 2004 2005

Ci´enaga, Magdalena Mar´ıa la Baja, Bol´ıvar Orito, Putumayo Cali, Valle Cali, Valle Manat´ı, Atl´antico Aracataca, Magdalena Polo Nuevo, Atl´antico Zona Bananera –Magdalena Sto. Tom´as, Magdalena Fundaci´on, Magdalena Sto. Tom´as, Magdalena Zona Bananera, Magdalena Pivijay, Magdalena Chin´acota, Nte. Stder Ponedera, Atl´antico Polo Nuevo, Atl´antico Palmar, Magdalena Palmar, Magdalena Cartagena, Bol´ıvar Sn. Ju´an, Bol´ıvar Barranquilla, Atl´antico Sn. Juan, Bol´ıvar Calamar, Atl´antico Ci´enaga, Magdalena Ret´en, Magdalena Soledad, Atl´antico Pto. Colombia, Atl´antico Ci´enaga, Magdalena Ci´enaga –Magdalena Sierra Nevada, Magdalena Soledad, Atl´antico Barranquilla, Atl´antico Cartago, Valle Pto. Carra˜no, Vichada Pinto, Magdalena Aracataca, Magdalena Malambo, Atl´antico Plato, Magdalena Zona Bananera, Magdalena Polo Nuevo, Atl´antico Manat´ı, Atl´antico Quipile, Cundinamarca Plato, Magdalena Sardinata, Nte.Stder Sardinata, Nte.Stder Fundaci´on, Magdalena Fundaci´on, Magdalena Sardinata, Nte.Stder Pivijay, Magdalena Pivijay, Magdalena Ci´enaga, Magdalena Bajo Baud´o, Choc´o Pereira, Risaralda Pivijay, Magdalena Chivolo, Magdalena Chivolo, Magdalena Fundaci´on, Magdalena Pte. Nacional, Santander Alto Baud´o, Choc´o

1 1 1 3 4 1 1 1 1 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ND 3 1 1 1 1 1 1 1 8 1 3 3 1 1 ND 1 1 1 3 4 1 1 1 1 8 3

2 2 3 4 5 2 2 2 2 2 2 2 2 2 4 2 2 2 2 2 2 2 2 2 3 2 2 2 2 2 2 2 2 8 4 2 2 2 2 2 2 2 4 2 4 4 2 2 7 2 2 2 4 5 2 2 2 2 4 4

EF377591 EF377628 EF377679 EF377659 EF377704 EF377624 EF377594 EF377621 EF377583 EF377614 EF377597 EF377623 EF377593 EF377586 EF377667 EF377630 EF377692 EF377620 EF377632 EF377627 EF377610 EF377629 EF377603 EF377638 EF377598 EF377589 EF377622 EF377641 EF377607 EF377608 EF377588 EF377618 EF377617 EF377705 EF377664 EF377609 EF377584 EF377599 EF377587 EF377604 EF377639 EF377625 EF377669 EF377606 EF377666 EF377700 EF377600 EF377602 EF377702 EF377605 EF377595 EF377601 EF377660 EF377703 EF377611 EF377592 EF377596 EF377590 EF377656 EF377706

For each of the 124 RABV isolates the host, year, geographical origin (town and department in Colombia), antigenic variant, genetic lineage are shown, as well as GenBank accession numbers for nucleotide sequences of region 1157–1420 with respect to nucleotide positions in the rabies virus SAD B19 nucleoprotein gene, coding for 88 amino acids in position 363–450 of the nucleoprotein. For clarity the prefixes were added to the code: C (canid dog), H (human), CT (cat), B (bat), HR (horse), F (fox), BV (bovine), SH (goat) and PG (swine). Numbers following a solidus indicate the year of isolation. Abbreviations: ND: non-determined, AV: antigenic variant, GL: genetic lineage. Twenty-eight strains analyzed in previous studies are shown underlined.

A. P´aez et al. / Virus Research 130 (2007) 172–181

istry of Health between January 1992 and December 2006, 1088 have been dog rabies and 51 have been human rabies cases. Of these cases, 93% can be grouped into three main geographical regions: Arauca (61 total cases), the Central Region (248 total cases) and the Caribbean Region (753 total cases). The additional 77 cases are distributed throughout the rest of the country. The rabies outbreaks in Arauca and the Central Region ended in 1997 due to efficient vaccination campaigns and sustained rabies elimination programs. The outbreak in the Caribbean Region is ongoing with five dog cases and one human case reported in 2006 (National Institute of Health, Colombia). Sylvatic rabies has recently gained increasing relevance as a public health issue with the death of a person infected with AV8 RABV in central Colombia and at least 17 indigenous people infected with AV3 RABV, bitten by rabid vampire bats in a western area of the country (Valderrama et al., 2006). The study here consisted of 124 samples obtained within the period 1994–2005, in 14 humans and 8 species of wild and domestic mammals (Table 1). Twenty-eight of the samples had been analyzed in previous molecular epidemiology projects based on partial Psi, glycoprotein and polymerase gene nucleotide sequences (Table 1) (P´aez et al., 2003, 2005; Hughes et al., 2004). The aim of this study was to use partial nucleoprotein gene nucleotide sequences to derive phylogenetic relationships between RABV isolated throughout Colombia, and with RABV related with major rabies enzootics associated with terrestrial carnivores and bats in the Americas. 2. Materials and methods 2.1. The study sample This study included 124 urban and sylvatic RABV samples isolated from dogs (n = 49), humans (n = 14), cats (n = 11), foxes (n = 18), cows (n = 22), insectivorous bats (n = 4), horses (n = 4), swine (n = 1) and goat (n = 1) (Table 1). The samples were isolated in the period 1994–2005 from rabies outbreaks in Arauca, eastern Colombia (n = 9), the Central Region (n = 22), the Caribbean Region (n = 60), Choc´o, western Colombia (n = 2), and other sites in Colombia (n = 31). 2.2. Rabies diagnosis Primary rabies diagnosis was achieved by direct immunofluorescence tests using polyclonal antibodies and virus isolation in Institute of Cancer Research (ICR) mice. RABV isolates were stored at −80 ◦ C in the form of frozen mouse brain material. 2.3. Antigenic analysis Antigenic characterization was performed by indirect immunofluorescence of acetone-fixed impressions of RABVinfected brain material as described previously, using a panel of eight monoclonal antibodies produced against the RABV nucleoprotein at the Centers for Disease Control and Prevention in Atlanta, GA, USA (CDC). This panel was previously used to

175

infer RABV reservoir species associations in Latin America and the Caribbean (D´ıaz et al., 1994). For clarity, we referred as nondetermined antigenic variants (NDAV), those isolates in which the reaction pattern obtained with the panel of eight monoclonal antibodies does not match with none of the 11 antigenic patterns or profiles so far reported by CDC. 2.4. RNA extraction and PCR RNA was extracted as previously described (P´aez et al., 2002). Briefly, 100 mg of frozen mouse brain-passaged material was dissolved in 0.75 ml Trizol (Gibco-BRL, Gaithersburg, MD, USA) and extracted once with 0.25 ml chloroform. Total RNA was precipitated with one volume of 100% iso-propyl alcohol, washed with 70% ethanol and made up to 50 ␮l with 1% diethyl-pyrocarbonate (DEPC) in double-distilled water. Complementary DNA was produced by a reverse transcription using rabies RNA as template. The amplicons were produced by RT-PCR, using primer 304 sense of sequence 5 TTGACGAAGATCTTGCTCAT-3 and primer 1066 anti-sense of sequence 5 -TCCCTGAAGAATCTTCTCTC-3 as described (Smith, 2002). 2.5. DNA sequencing and phylogenetic analysis Amplicons were purified (GFX PCR DNA and Band Purification System, Amersham Pharmacia Biotech Inc., New Jersey, USA) and sequenced on an Applied Biosystems 377 as described (De Mattos et al., 1999; Smith et al., 1995). Fragments of 264 bp of the rabies virus nucleoprotein genes from nucleotide 1157 to 1420 (with respect to nucleotide positions in the rabies virus SAD B19 nucleoprotein gene) were used (Conzelmann et al., 1990). Within these fragments, 88 amino acids were encoded and localized in position 363–450, according to the fixed laboratory strains SAD B19 and ERA. Multiple alignments were performed by using CLUSTAL W (http://www.ebi.ac.uk/clustalw/index.html). The 124 sequences were formerly edited to a common 264-base fragment using BioEdit (Hall, 1999). Phylogenetic analyses were conducted using MEGA 2.1 (Kumar et al., 2001). Distance matrix (neighbor-joining) method was used for phylogenetic analysis of Colombian versus RABV from the Americas (Figs. 2 and 3). Corrected nucleotide substitutions were calculated using Kimura’s two-parameter method. The significance of each clade in the phylograms was estimated by a bootstrap algorithm applying 1000 iterations (Felsenstein, 1985). Pairwise genetic distances were calculated using MEGA 2.1 software. To provide a root to the phylogenetic trees, the Lyssavirus species Duvenhage virus (DUVV) and European bat lyssavirus 2 (EBLV-2) were included as outgroups. 3. Results Eight GL were characterized in our study sample (Figs. 1 and 2). The GL1, GL2 and GL3 were found associated with dogs. GL4 was associated with hematophagous bats (D.

176

A. P´aez et al. / Virus Research 130 (2007) 172–181

Fig. 1. Map of Colombia showing the geographical origin of the rabies samples. For clarity the prefixes were added to the code: C (canid dog), H (human), CT (cat), B (bat), HR (horse), F (fox), BV (bovine), SH (goat) and PG (swine). Numbers following a solidus indicate the year of acquisition. Symbols identify host species and colors identify RABV GL.

rotundus). The GL5 was found independently segregated within a RABV clade associated with colonial insectivorous bats in the Americas, whereas the GL6 was strongly associated with RABV associated with Tadarida brasiliensis bats in South America. GL7 and GL8 segregated independently within RABV clades associated with colonial insectivorous bats and solitary bats in the Americas, respectively (Fig. 3).

3.1. Dog rabies RABV associated with dogs in Colombia segregated into three genetic lineages GL1, GL2 and GL3 related to same number of geographic regions (Figs. 1 and 2). All three GLs were AV1, with an overall genetic divergence of 3%. GL1 consisted of 31 samples isolated in central and eastern Colombia

A. P´aez et al. / Virus Research 130 (2007) 172–181

177

Fig. 2. Phylogenetic relationships among 124 RABV isolates from Colombia during 1994–2005. Analysis was based on partial nucleoprotein gene nucleotide sequences coding for amino acids 363–450 of the protein. Neighbor-joining (Kimura’s two-parameter method) was used for the analysis. The significance of each clade was estimated by a bootstrap algorithm applying 1000 iterations. Numbers at nodes indicate bootstrap values greater than 70%. Arrows indicate the genetic lineage, year of isolation and hosts. The Lyssavirus species Duvenhage virus (DUVV) and European bat lyssavirus 2 (EBLV-2) were included as outgroups.

178

A. P´aez et al. / Virus Research 130 (2007) 172–181

Fig. 3. Phylogenetic relationships among RABV isolated in the Americas. GenBank accession numbers were used for the identification of isolates, except for Colombian isolates which were identified by their original sample code. Consecutive accession numbers were separated by a dash, non-consecutive numbers were separated by a comma. Bibliographic references were cited when no GenBank accession numbers available. Analysis was based on partial nucleoprotein gene nucleotide sequences coding for amino acids 363–450 of the protein. Neighbor-joining (Kimura’s two-parameter method) was used for the analysis. The significance of each clade was estimated by a bootstrap algorithm applying 1000 iterations. Numbers at nodes indicate the percentage of the confidence limits. The Lyssavirus species Duvenhage virus (DUVV) and European bat lyssavirus 2 (EBLV-2) were included as outgroups.

A. P´aez et al. / Virus Research 130 (2007) 172–181

during years 1994–1997 from six species that included dogs, cats, cows, horses, a goat and a pig. The genetic divergence in GL1 over this period and for the spectrum of its hosts it was 0.7% (Fig. 2). The GL2 consisted of 59 samples from northern Colombia isolated in years 1994–2004 from five species that included dogs, cats, humans, cows and gray foxes Urocyon cinereoargenteus—Schreber 1775 (Eisenberg, 1989). The average pairwise genetic similarity among isolates in GL2 was 99.7%. The GL3 was represented in our study sample by only one sample from a human bitten by a rabid dog in the southwest of Colombia in 2000 (Saad et al., 2001). Genetic similarity of GL3 with GL1 and GL2 was 94.9% and 97.5%, respectively. Within the cluster associated with terrestrial carnivores, GLs 1–3 grouped closely to each other, near dog rabies from Mexico and distant to dog rabies from Argentina, Bolivia, Brazil, and Peru (Fig. 3).

179

Table 2 Pattern of reactivity of GL4, GL7 and GL8 RABV with the panel of eight monoclonal antibodies used for rabies virus antigenic characterization Genetic lineage

GL4 GL7 and GL8

Monoclonal antibodies C1

C2

C9

C10

C12

C15

C18

C19

– –

+ +

+ +

+ +

+ –

– –

+ –

+ –

RABV of GL4 (H02/94, CT1/95, CT3/96, CT4/96, BV6/97, C31/97), GL7(F21/03) and GL8(B02/02) were recognized (+) or not recognized (−) by each of the eight monoclonal antibodies (C1–C19) of the panel used for rabies virus antigenic typing.

cow (BV6/97), and a dog (C31/97) (Table 1). CT1/95, CT3/96, CT4/96 and H02/94 grouped closely together, and distant from the subclade formed by BV6/97 and C31/97, within the cluster of RABV associated with hematophagous bats in Fig. 2.

3.2. Fox rabies 3.4. Insectivorous bat rabies Seventeen RABV in this study were isolated from U. cinereoargenteus gray foxes in northern Colombia. These consisted of AV1 and grouped within GL2 which was associated with dogs. The extent of similarity between gray fox and dog isolates within GL2 was 99.7%. GL2 gray fox isolates were compared with enzootic rabies associated with gray foxes in Texas and Arizona, USA, and in Northern Mexico; red foxes in Europe; and dog rabies in South America related to AV1 and AV2. Similarity values of 92.6%, 90%, 91%, 90.8%; 91.8% were obtained, respectively. Rabies in gray foxes in Colombia was not closely related to that in gray foxes from Mexico and USA (Fig. 3). F21/03 was a NDAV isolated in a non-determined species of fox near the city of Sardinata in the northeast of Colombia. F21/03 segregated independently within the clade associated with colonial insectivorous bats in the Americas (Fig. 3). F21/03 had its closest genetic identity (although relatively far relatedness) with RABV associated with Myotis genus in South America (89.6%) (Fig. 3). F21/03 formed GL7 in our study. 3.3. Hematophagous bat rabies The GL4 consisted of mainly AV3 and a few AV8 RABV, all of these associated with hematophagous bats (D. rotundus) in South America. The GL4 consisted of 28 samples isolated in the period 1994–2005 throughout Colombia from five species that included humans, dogs, cats, cows and horses. The average pairwise genetic similarities among GL4 RABV were 97.1%, 98.2% and 98.6% for geographic regions A, B and C, respectively. Three groups of GL4 RABV were formed within the cluster of RABV associated with hematophagous bats in South America (Fig. 3). Those groups were found closely correlated with their geographical origin in Colombia (A–C in Fig. 1). Six RABV of NDAV were part of GL4, which produced a unique and unusual pattern of reactivity with the panel of eight monoclonal antibodies (Table 2). These were geographically circumscribed to central Colombia departments of Cundinamarca, Tolima and Caldas (regions B and C in Fig. 1), and isolated from three cats (CT1/95, CT3/96 and CT4/96), a human (H02/94), a

The GL5 was represented by two samples (B01/04 and B01/00) isolated from insectivorous bats in the cities of Cali and Pereira in the southwest and center of Colombia in 2000 and 2004, respectively. GL5 viruses segregated independently within the cluster related with colonial insectivorous bats (Fig. 3). The highest identity values for GL5 were obtained by comparisons with RABV isolated from Eptesiscus fuscus in North America (93.9%). The GL6 was represented by sample B01/95 isolated from a Molossus molossus insectivorous bat in the city of Cali in 1995. This segregated along with RABV associated with T. brasiliensis bats in South America (Fig. 3), with an average pairwise identity value of 96.6%. GL6 got similarity values in the range of 90 to 94% with RABV associated with colonial insectivorous bats, such as E. fuscus from North America (93%) and GL5 from Colombia (93.4%). However, the relationships of GL6 with these two lineages were not monophyletic (Fig. 3). The GL8 consisted of isolate B02/02 obtained from an Eptesicus brasiliensis insectivorous bat near the city of Cartago in the southwest of Colombia in 2002. The GL8 produced an unusual pattern of reactivity with the panel of eight monoclonal antibodies (Table 2), therefore was classified as of NDAV. GL8 segregated independently within the cluster associated with solitary bats (Fig. 3). However, its highest identity values occurred by comparisons with RABV associated with Sagui monkeys from Brazil, Lasionycteris noctivagans/Pipistrellus subflavus cluster (93.4%) and with Lasiurus cinereus (90.8%) in the Americas (Fig. 3). 4. Discussion Previous studies on the molecular epidemiology of rabies in Colombia had focused on major urban rabies epizootics, with only a few cases from outside epizootic areas included, and with smaller and less representative study samples than the one used here (P´aez et al., 2003, 2005; Hughes et al., 2004). The aim of our study was to perform a robust molecular epidemiology study of RABV transmitted in Colombia in years 1994–2005, focusing

180

A. P´aez et al. / Virus Research 130 (2007) 172–181

not only in major rabies epizootics, but also on cases of urban and sylvatic rabies from throughout the country. cDNA fragments of 264 nt encoding 88 amino acids at the carboxyl terminus of the nucleoprotein were obtained by RT-PCR, sequenced and used in phylogenetic analyses. Our study included 124 rabies samples obtained from nine species of mammals including 14 humans (Table 1). The existence of at least eight rabies virus genetic lineages in this study was established. GLs 1–3 were associated with dogs, however, these were isolated from other six species including cats, cows, horses, goat, swine and gray foxes. Our data is consistent with previous studies in that GL1 is apparently extinct in Colombia since 1997, for which efficient rabies vaccination campaigns and elimination programs may have played a crucial role. Phylogenetic analysis showed that GL2 RABV isolated in humans, dogs and gray foxes had no defined subvariants and no clear tendencies of variant subgrouping by geographical region. The GL2 RABV isolated in gray foxes from northern Colombia seems to be a spillover originating from rabid dogs. However, fox to fox transmission is suspected due to the sustained number of cases in foxes despite the overall decrease in canine rabies in the region. Although no transmission of rabies has been confirmed from gray foxes to humans, this may easily happen in natural conditions in northern Colombia due to the frequent contact with farmers and domestic animals such as dogs, cats and cattle. Gray foxes may therefore pose a threat for public health in this region. Urban RABV in Colombia seems to be looking in gray foxes a new host in which these may not be vulnerable to the current rabies control strategies in dogs used by the Ministry of Health in Colombia. Novel rabies control strategies need to be evaluated for use in Colombia, not only for wild rabid gray foxes but for other wild carnivore species which may become a threat for public health. The GL3 became a new GL in our study supported by a high value of bootstrap (82%) that differentiates this from GL1. The GL3 in Colombia was represented by one AV1 strain isolated from a human bitten by a rabid dog in the southwest of the country in 2000 (Saad et al., 2001), and it is presumably transmitted in the neighboring country of Ecuador. According to our study, no successful colonization of host species other than dogs was confirmed for GLs 1–3 RABV in Colombia. The topologies of the hematophagous and insectivorous bat RABV clusters (Fig. 2) indicated that these are genetically diverse groups, suggesting that in Colombia several RABV variants circulate independently in bats. The high bootstrap values in the GL4 cluster that correlates with geographical regions A–C (Figs. 2 and 3) suggested the existence of a geographical structure in the rabies population associated with vampire bats in Colombia, and the existence of multiple rabies foci associated with these reservoirs. We found RABV associated with colonial and solitary insectivorous bats in the Americas (Fig. 3). The GL6 was represented by a rabies strain isolated from a Molossus molossus insectivorous bat in the southwest of Colombia in 1995. The GL6 was found strongly related to T. brasiliensis from North America and South America. The GL5, GL7, GL8 had high identity values with RABV associated with bat rabies reservoir species reported in the North America such as E. fuscus, L. noctivagans and P. subflavus. However the relations among these are not monophyletic, suggesting that GL5,

GL7 and GL8 are associated with not yet identified species of insectivorous bats in Colombia. Transmission of rabies from insectivorous bats to terrestrial carnivores is suggested here, as it is the case of strain F21/03 (GL7), isolated from an unknown species of fox, and associated with colonial insectivorous bats of the Myotis genus in South America (Fig. 3). RABV genetic diversity in Colombia was found to be larger and more complex than inferred in previous studies, however our results are consistent with those as both were able to identify GL1 and GL2 of dog origin, the association between dog rabies and gray foxes in northern Colombia and the group of RABV GLs associated with bats. Overall the main breakthroughs of this study are the identification of GL3 RABV of dog origin. Secondly, we observed the differentiation of two groups within the cluster of rabies viruses associated with Chiroptera: those associated with hematophagous and those with insectivorous bats. Our data indicated that there are two main reservoirs of rabies in Colombia. Those are dogs and hematophagous bats (D. rotundus). Other less important rabies reservoirs are insectivorous bats in which T. brasiliensis species are included as far as for GL6 concerns. No RABV were found associated with terrestrial rabies reservoirs other than dogs, although the epidemiology of rabies in foxes warrants further investigation. Eight rabies isolates were classified as of NDAV. Six of these grouped into GL4 associated with D. rotundus hematophagous bats, and were circumscribed to central Colombia, departments of Cundinamarca, Tolima and Caldas (regions B and C in Fig. 1). According to the phylogenetic analysis, these six viruses may be divided into two groups, suggesting the existence of at least two foci of GL4 NDAV RABV in this region: firstly CT1/95, CT3/96, CT4/96 and H02/94, and secondly BV6/97 and C31/97. The remaining two viruses from the group of eight NDAV RABV, formed GL7 and GL8 (one each), and were associated with colonial insectivorous and solitary bats, respectively (Fig. 3). The pattern of reactivity with the panel of eight monoclonal antibodies produced by the six NDAV RABV in the first group was unique and different to that produced by the two NDAV RABV in the second group formed by GL7 and GL8 (Table 2). These results together showed the existence of two antigenic and four genetic groups of NDAV RABV in the study sample and confirmed the diversity of RABV circulating in the Americas is far larger than expected when the panel of eight monoclonal antibodies was first standardized. Previous studies had also shown the existence of NDAV RABV in the Americas (Favoretto et al., 2002; Velasco-Villa et al., 2005; Velasco-Villa et al., 2006). Partial or complete genes for the nucleoprotein, glycoprotein, and polymerase, as well as the Psi region, have been the most widely used markers for RABV phylogeny, genetic diversity and molecular epidemiology studies. We used a fragment of the nucleoprotein gene as the genomic marker, and the resulting phylogenetic relationships and genetic diversity were consistent with previous studies in which the other genes were used (P´aez et al., 2003; P´aez et al., 2005; Hughes et al., 2004). This confirmed the suitability of any of these genomic regions for studies on the phylogeny and molecular diversity of RABV. However, using our fragment of 264 nt of the nucleoprotein gene may become a less expensive and more effective approach due to its shorter length, which makes it more suit-

A. P´aez et al. / Virus Research 130 (2007) 172–181

able for the analysis of decomposed samples. In addition this has been the preferred genomic region for lyssavirus phylogeny, evolution and molecular epidemiology studies in the Americas and worldwide. It is sensitive enough for the separation of the different lineages of viruses according to geographical, temporal and species-specific virus transmission dynamics (Bourhy et al., 1999; Smith, 2002; De Mattos et al., 1999; Velasco-Villa et al., 2005; Cisterna et al., 2005; Carnieli et al., 2006). Acknowledgements This study was sponsored by Instituto Nacional de Salud (INS) in Bogot´a D.C., Colombia, the Centers for Disease Control and Prevention (CDC) in Atlanta GA, USA, and the American Fellows Program—Partners of the Americas of the USA Government. Authors thank Departamento de Investigaciones—Universidad de La Salle in Bogot´a D.C. for its valuable support. Authors also thank Mrs. Rebecca Alvarado, Program Officer of the American Fellows Program for her valuable help and collaboration, and Mr. Gabriel Perdomo at Instituto Nacional de Salud for the production of the figures of this paper. References Beran, G.W., Steele, J.H., 1994. Handbook of Zoonoses, Section B. CRC Press, Boca Rat´on, FL, p. 307. Bourhy, H., Kissi, B., Audry, L., Smreczak, M., Sadkowska-Todys, M., Kulonen, K., Tordo, N., Zmudzinski, J.F., Holmes, E.C., 1999. Ecology and evolution of rabies virus in Europe. J. Gen. Virol. 80, 2545–2557. Carnieli, P., Brandao, P.E., Carrieri, M.L., Castilho, J.G., Macedo, C.I., Machado, L.M., Rangel, N., de Carvalho, C., de Carvalho, V.A., Montebello, L., Wada, M., Kotait, I., 2006. Molecular epidemiology of rabies virus strains isolated from wild canids in Northeastern Brazil. Virus Res. 120 (1/2), 113–120. Cisterna, D., Bonaventura, R., Caillou, S., Pozo, O., Andreau, M.L., Fontana, L.D., Echegoyen, C., de Mattos, C., de Mattos, C., Russo, S., Novero, L., Elberg, D., Freire, M.C., 2005. Antigenic and molecular characterization of rabies virus in Argentina. Virus Res. 109 (2), 139–147. Conzelmann, K.K., Cox, J.H., Schneider, L.G., Thiel, H.J., 1990. Molecular cloning and complete nucleotide sequence of the attenuated rabies virus SAD B19. Virology 175, 485–499. De Mattos, C.A., De Mattos, C.C., Smith, J.S., Miller, E.T., Papo, S., Utrera, A., Osburn, B.I., 1996. Genetic characterization of rabies field isolates from Venezuela. J. Clin. Microbiol. 34 (6), 1553–1558. De Mattos, C.C., De Mattos, C.A., Loza-Rubio, E., Aguilar-Setien, A., Orciari, L.A., Smith, J.S., 1999. Molecular characterization of rabies virus samples from M´exico: implications for transmisi´on dynamics and human risk. Am. J. Trop. Med. Hygiene 61, 587–597. D´ıaz, A.M., Papo, S., Rodriguez, A., Smith, J.S., 1994. Antigenic analysis of rabies-virus samples from Latin America and the Caribbean. Zentralbl Veterinarmed B 41, 153–160.

181

Eisenberg, J.F., 1989. Mammals of the Neotropics. The Northern Tropics, vol. 1. University of Chicago Press, pp. 262–289. Favoretto, S.R., Carrieri, M.L., Cunha, E.M., Aguiar, E.A., Silva, L.H., Sodre, M.M., Souza, M.C., Kotait, I., 2002. Antigenic typing of Brazilian rabies virus samples isolated from animals and humans, 1989–2000. Rev. Inst. Med. Trop. Sao Paulo 44 (2), 91–95. Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783–791. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignme editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98. Hughes, G.J., P´aez, A., B´oshell, J., Rupprecht, C.E., 2004. A phylogenetic reconstruction of the epidemiological history of canine rabies virus variants in Colombia. Infect. Gen. Evol. 4, 45–51. Ito, M., Arai, Y.T., Itou, T., Sakai, T., Ito, F.H., Takasaki, T., Kurane, I., 2001. Genetic characterization and geographic distribution of rabies virus isolates in Brazil: identification of two reservoirs, dogs and vampire bats. Virology 284 (2), 214–222. Kobayashi, Y., Sato, G., Shoji, Y., Sato, T., Itou, T., Cunha, E.M., Samara, S.I., Carvalho, A.A., Nociti, D.P., Ito, F.H., Sakai, T., 2005. Molecular epidemiological analysis of bat rabies viruses in Brazil. J. Vet. Med. Sci. 67 (7), 647–652. Kumar, S., Tamura, K., Jakobsen, B., Nei, M., 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17, 1244–1245. P´aez, A., Garc´ıa, C., B´oshell, 2002. Standardization of rabies virus genome amplification for molecular epidemiology studies (article in Spanish). Biom´edica 22 (1), 71–76. P´aez, A., N´un˜ ez, C., Garc´ıa, C., B´oshell, J., 2003. Molecular epidemiology of rabies epizootics in Colombia: evidence for human and dog rabies associated with bats. J. Gen. Virol. 84, 795–802. P´aez, A., Saad, C., N´un˜ ez, C., B´oshell, J., 2005. Molecular epidemiology of rabies in northern Colombia 1994–2003: evidence for human and fox rabies associated with dogs. Epidemiol. Infect. 133, 529–536. Saad, C., Rodr´ıguez, G., Revelo, D., 2001. Paralytic rabies in a six year old girl (article in Spanish). Biom´edica 21, 86–91. Smith, J.S., Orciari, L.A., Yager, P.A., 1995. Molecular epidemiology of rabies in the United States. Semin. Virol. 6, 387–400. Smith, J.S., 2002. Molecular epidemiology. In: Jackson, A.A., Wunner, W.H. (Eds.), Rabies. Academic Press, San Diego, pp. 79–111. Tordo, N., Kouknetzoff, A., 1994. The rabies virus genome: an overview. Onderspoort J. Vet. Res. 67, 45–49. Valderrama, J., Garc´ıa, I., Figueroa, G., Rico, E., Sanabria, J., Rocha, N., Parra, E., Saad, C., P´aez, A., 2006. Outbreaks of human rabies transmitted by vampire bats in Alto Baud´o and Bajo Baud´o municipalities, department of Choc´o, Colombia, 2004–2005 (article in Spanish). Biom´edica 26, 387–396. Velasco-Villa, A., Orciari, L.A., Souza, V., Ju´arez-Islas, V., Gomez-Sierra, M., Castillo, A., Flisser, A., Rupprecht, C.E., 2005. Molecular epizootiology of rabies associated with terrestrial carnivores in M´exico. Virus Res. 111, 13–27. Velasco-Villa, A., Orciari, L.A., Juarez-Islas, V., Gomez-Sierra, M., PadillaMedina, I., Flisser, A., Souza, V., Castillo, A., Franka, R., Escalante-Mane, M., Sauri-Gonzalez, I., Rupprecht, C.E., 2006. Molecular diversity of rabies viruses associated with bats in Mexico and other countries of the Americas. J. Clin. Microbiol. 44, 1697–1710. Wunner, W.H., Larson, J.K., Dietzchold, B., Smith, C.L., 1988. The molecular biology of rabies virus. Rev. Infect. Dis. 10, 771–784.