Molecular epizootiology of rabies associated with terrestrial carnivores in Mexico

Virus Research 111 (2005) 13–27

Molecular epizootiology of rabies associated with terrestrial carnivores in Mexico Andr´es Velasco-Villa a,b,∗ , Lillian A. Orciari a , Valeria Souza c , V´ıctor Ju´arez-Islas b , Mauricio Gomez-Sierra b , Amanda Castillo c , Ana Flisser d , Charles E. Rupprecht a a

Viral and Rickettsial Zoonosis Branch, Division of Viral and Rickettsial Diseases, Centers for Disease Control and Prevention, 1600 Clifton Road, Mailstop G33, Atlanta, GA 30333, USA b Laboratorio de Rabia, Instituto de Diagn´ ostico y Referencia Epidemiol´ogicos, SSA Carpio 470, Colonia Santo Tom´as, 11340 M´exico, DF, M´exico c Laboratorio de Evoluci´ on Molecular y Experimental, Departamento de Ecolog´ıa Evolutiva. Instituto de Ecolog´ıa, AP 70-275, UNAM, 04510 M´exico, DF, M´exico d Departamento de Microbiolog´ıa y Parasitolog´ıa, Facultad de Medicina, UNAM, 04510 M´ exico, DF, M´exico

Abstract Epizootiological patterns of rabies are described, using antigenic and genetic analysis of samples obtained from infected domestic and wild mammals in 20 Mexican states during 1976–2002. Two independent origins are suggested for rabies in Mexican carnivores. One group shares ancestry with canine rabies, while the other group appears to share a common origin with bat rabies in North America. More than 12 sublineages were found in rabid dog populations, suggesting at least six major spatio-temporal foci. Coyote rabies was found as independent enzootic foci that probably emerged via spillover from dog rabies, translocated from major foci in the southcentral and western regions of Mexico. One focus of gray fox rabies was widely distributed in northwestern Mexico, overlapping with a focus in the same species in the southwestern United States. A skunk rabies focus distributed in the northcentral Mexican states appears to share a common origin with bat rabies foci in North America, and is a close relative of southcentral skunk and raccoon rabies in the United States. Two other skunk foci share a common ancestor with canine rabies and were distributed in northwest Mexico and Yucatan. © 2005 Elsevier B.V. All rights reserved. Keywords: Epizootiology; Rabies; Terrestrial carnivores

1. Introduction Rabies has been recorded since the first human civilizations, with the dog as the main transmitter (Wilkinson, 2002). In developed countries, traditional descriptive approaches have been applied to infer patterns of disease transmission to humans (Krebs et al., 2003). This approach has been reinforced by the process of rabies virus typing at the antigenic and genetic level (Bourhy ∗ Corresponding author at: Rabies Unit, Viral and Rickettsial Zoonoses Branch, Division of Viral and Rickettsial Diseases, Centers for Disease Control and Prevention. Mailstop G33, Atlanta, GA 30333, USA. Tel.: +1 404 639 1055; fax: +1 404 639 1564. E-mail address: [email protected] (A. Velasco-Villa).

0168-1702/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2005.03.007

et al., 1993; Rupprecht et al., 1987; Smith et al., 1992). The detection and analysis of subtle differences within the rabies virus proteins and genes have permitted the identification of viral variants specifically maintained by different animals, such as dogs, foxes, raccoons, skunks, and bats. The combined use of those latter approaches, together with an efficient surveillance system for disease detection in animal populations, have allowed detailed descriptions of the distributions of major rabies foci, as well as the likely hosts responsible for maintenance (Bourhy et al., 1999; Nadin-Davis et al., 1999; Smith et al., 1995). Data provided by such molecular approaches have permitted insights to virus-reservoir relationships, patterns of transmission and dissemination, as well as viral evolution (Badrane and Tordo, 2001; Bourhy et al., 1993, 1999; Holmes et al., 2002; Nadin-Davis et al., 1999;

14

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

Table 1 Standard antigenic reactivity of monoclonal antibodies against rabies viruses in Latin America MAba identification number reservoir associated

C1

C4

C9

C10

C12

C15

C18

C19

AgVb

CVS/ERA/SAD/PASTc Dog/Mongoose Dog Vampire bat Tadarida brasiliensis Vampire bat Lasiurus cinereus Arizona fox South central skunk Tadarida brasiliensis mexicana Baja California Sur skunk Vampire bat

+ + + − − − +/− + − + + −

+ + + + + + + + + + + +

+ + − + + +/− + + + + + +

+ + + + + + + − + + + +

+ + + + + + + + + + − −

+ + + − − +/− − + + − + −

+ − − − − − − − + − − −

+ + + + − +/− − + + − + +

Controld 1 2 3 4 5 6 7 8 9 10 11

a b c d

Antigenic variant (specific variant type). Monoclonal antibody identification number. Laboratory fixed strains. Positive control for monoclonal antibody reactivity.

Smith et al., 1995). In contrast, developing countries are faced with less than ideal surveillance in animal populations. The reduced resources available are prioritized for diseases with overwhelming human morbidity and mortality. Often, sample availability is limited and wholly dependent upon human rabies occurrence or the perception of major outbreaks. During the past decade, different analytical approaches have been applied to Mexican rabies virus samples. These approaches have provided insights regarding rabies virus diversity and the likely reservoirs responsible for transmission and maintenance (De Mattos et al., 1999; Loza-Rubio et al., 1999; Velasco-Villa et al., 2002). However, previous studies have been limited by sample sizes or the approach applied, to address the deeper descriptive epizootiology of rabies in Mexico. In the present work, we incorporate 138 sequences comprising a 30-year period. The objective of this study was the molecular analysis of rabies viruses associated with terrestrial carnivores in Mexico, describing updated disease distributions, suggested trends of interspecies transmission, and predicted patterns of dissemination, by using 88 amino acids of the nucleoprotein C terminus. 2. Materials and methods 2.1. Antigenic characterization The rabies virus N protein was characterized with a panel of eight monoclonal antibodies, previously used to infer rabies virus reservoir species associations in Latin America and the Caribbean (Diaz et al., 1994). This reduced panel was able to identify 11 reactivity patterns associated with different animals involved with rabies virus maintenance and transmission in Mexico and South America (Table 1). 2.2. Samples and sequences In the present study, 138 sequences were analyzed, including the outgroup represented by the Lyssavirus species

European Bat Lyssavirus 2 (EBL2) and Duvenhage. The analysis included 61 new samples gathered over a 4-year period (1999–2002), from 20 Mexican states (Table 2). These were chosen on the basis of the antigenic variant to encompass the greatest Mexican rabies virus diversity as possible. The remaining 77 correspond to historical samples, already sequenced and antigenically typed, used in context to support a greater analysis in space and time (Table 3 ). All samples represent major terrestrial carnivore rabies enzootic foci, associated with raccoons, skunks, foxes, dogs and coyotes in the United States (US), Canada and Mexico (De Mattos et al., 1999; Nadin-Davis et al., 1993, 1999; Smith et al., 1992, 1995). Within the 138 sequences, five sequences representing fox rabies in Yugoslavia and Israel were included (David et al., 2000), plus nine sequences associated with rabid bats, to make a distinction between Old World and New World fox rabies, as well as between bat and terrestrial carnivore rabies, respectively (De Mattos et al., 1999; Smith et al., 1992). 2.3. Partial amplification of the nucleoprotein gene and its sequencing Total RNA was extracted from infected brain tissue by using Trizol (Invitrogen, San Diego, CA, formerly, GIBCOBRL, Inc.), according to the manufacturer’s instructions. Complementary DNA was produced by RT-PCR, using primers 304 sense and 1066 anti-sense, as described (Smith, 2002). Purified amplicons were sequenced on an Applied Biosystems 377 DNA automated sequencer, as described (De Mattos et al., 1999; Smith et al., 1995). 2.4. Nucleoprotein region used for construction of the phylogeny Fragments of 264 bp of the rabies virus nucleoprotein genes from nucleotide 1157 to 1420, with respect to nucleotide positions in SAD B19 nucleoprotein gene, were used (Conzelmann et al., 1990; De Mattos et al., 1999; Smith et

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

15

Table 2 Recent rabies virus samples from Mexico Identification number Lineagea

Animal

AgVb

State

Year

GeneBank number

247coahcy01 (D1) 4508tampct02 (D1) 4506tampct02 (D1) 3288michdg00 (D2) T 4823slpdg96 (D4) T 3269zacbov00 (D4) 3284mxdg99 (D5) T 3260dfdg00 (D5) 3267mxdg01 (D5) 4516mxdg02 (D5) 4511mxdg02 (D5) 4514mxdg02 (D5) T 3325hgodg02 (D5) 3244mxdg98 (D6) 4822puedg96 (D7) T 3278grodg99 (D7) T 3299dfdg99 (D8) 3281puedg00 (D8) 3279puedg01 (D8) 3302tlxdg01 (D8) T 3275gtoct01 (D8) T 3313jaldg02 (D8) 4491mxdg02 (D8) 4513puedg02 (D8) 4519puedg02 (D8) 3303tlxdg02 (D8) T 4815slpdg99 (D9) 3251grodg00 (D9) 3285grodg01 (D9) 3287grodg01 (D9) 4476yucdg99 (D10) 4485yucdg01 (D10) 4481yucdg01 (D10) 4480yucpg02 (D10) 4486chisdg02 (SI) 641sinsk01 (W1) 231sinsk01 (W1) 1535dgosk01 (W1) 3280zacfx99 (W2) 3253sonfx01 (W2) 3282chihpm01 (W2) 3264sinbct01 (W2) 4510sonbct02 (W2) 4498sonfx02 (W2) T 3304mxdg02 (W2) 4475yucdeer9 (W7) 4463yuctepz99 (W7) 4464yuctepz99 (W7) 4466yucuknW5 (W7) 4471yucsk01 (W7) 3265bcssk01 (W9) 3290bcssk02 (W9) 3259bcsbv02 (W9) 3291chihbv99 (W11) 3289jalct00 (W13) 3247jalsk01 (W13) 3296jalsk01 (W13) 3248zacsk01 (W13) 4489dgosk01 (W13) 4502slpsk02 (W13) 4505slpsk02 (W13)

Coyote Cat Cat Dog Dog Cow Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Cat Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Pig Dog Skunk Skunk Skunk Gray fox Gray fox Cougar Bobcat Bobcat Gray fox Dog Deer Agouti Agouti No data Skunk Skunk Skunk Cow Cow Cat Skunk Skunk Skunk Skunk Skunk Skunk

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ND ND ND ND 1 1 1 1 7 7 7 7 7 7 1 ND ND ND ND ND 10 10 10 8 8 8 8 8 8 8 8

Coahuila Tamaulipas Tamaulipas Michoac´an San Luis Potosi Zacatecas Estado de M´exico M´exico DF Estado de M´exico Estado de M´exico Estado de M´exico Estado de M´exico Hidalgo Estado de M´exico Puebla Guerrero M´exico DF Puebla Puebla Tlaxcala Guanajuato Jalisco Estado de M´exico Puebla Puebla Tlaxcala San Luis Potosi Guerrero Guerrero Guerrero Yucatan Yucatan Yucatan Yucatan Chiapas Sinaloa Sinaloa Durango Zacatecas Sonora Chihuahua Sinaloa Sonora Sonora Estado de M´exico Yucat´an Yucat´an Yucat´an Yucat´an Yucat´an Baja California Sur Baja California Sur Baja California Sur Chihuahua Jalisco Jalisco Jalisco Zacatecas Durango San Luis Potosi San Luis Potosi

2001 2002 2002 2000 1996 2000 1999 2000 2001 2002 2002 2002 2002 1998 1996 1999 1999 2000 2001 2001 2001 2002 2002 2002 2002 2002 1999 2000 2001 2001 1999 2001 2001 2002 2002 2001 2001 2001 1999 2001 2001 2001 2002 2002 2002 1999 1999 1999 No data 2001 2001 2002 2002 1999 2000 2001 2001 2001 2001 2002 2002

AY561764 AY561762 AY561763 AY561770 AY561766 AY561765 AY561774 AY561772 AY561775 AY561778 AY561779 AY561776 AY561788 AY561773 AY561780 AY561789 AY561771 AY561781 AY561782 AY561786 AY561769 AY561768 AY561783 AY561784 AY561785 AY561787 AY561767 AY561790 AY561791 AY561792 AY561794 AY561795 AY561796 AY561797 AY561793 AY561804 AY561805 AY561806 AY561798 AY561799 AY561802 AY561803 AY561800 AY561801 AY561777 AY561818 AY561819 AY561820 AY561821 AY561822 AY561807 AY561808 AY561809 AY561810 AY561813 AY561811 AY561812 AY561814 AY561815 AY561816 AY561817

SI, segregates independently; the sampled did not cluster within any lineage or sublineage. ND, the antigenic variant was not determined. T Superscript bold letter on the left side of the identification number indicates evidence of translocation on the basis of the epidemiological questionnaire or history available. a Lineage as observed in the phylogenetic tree is denoted in bold with one letter and one number below the identification number. b AgV. Antigenic variant determined with the reduced panel of monoclonal antibodies.

16

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

Table 3 Historical rabies virus sequences representing major enzootic foci of terrestrial rabies in Mexico and the United States Identification number Lineagea

Animal

AgVb

State

Year

Identification number in the referencec

835txhm76 (D1) 834txhm79 (D1) 2193txhm79 (D1) 2160txdg85 (D1) 2361txdg94 (D1) 2587txcy94 (D1) 848ormichhm89 (D2) 555dfdg78 (D3) 3124puedg91 (D3) 3127tlxgt91 (D3) 3126puepg93 (D3) 3147qrodg95 (D3) 3151Mxuk (D3) 3138michdk91 (D4) 2184camrhm93 (D4) 3142dffrr90 (D6) 3121dfhm91 (D6) 3134dfpg91 (D6) 1278mxdg91 (D6) 1282mxdg91 (D6)

Human Human Human Dog Dog Coyote Human Dog Dog Goat Pig Dog No data Donkey Human Ferret Human Pig Dog Dog

1 1 1 1 1 1 1 1 1 10 1 1 1 1 1 1 1 1 1 1

Texas Texas Texas Texas Texas Texas Oregon Mexico DF Puebla Tlaxcala Puebla Queretaro No data Michoacan California Mexico DF Mexico DF Mexico DF Mexico DF Mexico DF

1976 1979 1979 1985 1994 1994 1989 1978 1991 1991 1993 1995 No data 1991 1993 1990 1991 1991 1991 1991

1

Txushm835 Txushm834 1 Txushm2193 1, 2 Txusdg2160 1 Txusdg2361 1 Txuscy2587 1, 2 Orushm848 1, 2 Mxmxdg555 1 Mxdg3124 1 Mxgt3127 1 Mxpg3126 1 Mxdg3147 1 Mxuk3151 1 Mxdk3138 1 Caushm2184 1 Mxfrr3142 1 Mxhm3121 1 Mxpg3134 1, 2 Mxmxdg1278 1, 2 Mxmxdg1282 1

Identification number Lineagea

Animal

AgVb

State

Year

Identification number in the reference or GeneBank numberc

3143puebv94 (D7) 3125oaxdg92 (D9) 726cahm61 (D11) 953cachihhm79 (D11) 947azsonhm81 (D11) 1274sondg88 (D11) 672sndg88 (D11) 671sndg88 (D11) 673sndg88 (D11) 2520cacy90 (D11) 1303orchihdg91 (D11) 3128dgodg91 (D11) 3144chihbv94 (D11) 2830txcy95 (D11) 559mxdg78 (SI) 674cagroct87 (SI) 3122michct90 (SI) 3123dfbv91 (SI) 398azfx79 (W2) 186azfx86 (W2) 3148chihbct90 (W2)

Cattle Dog Human Human Human Dog Dog Dog Dog Coyote Dog Dog Cattle Coyote Dog Cat Cat Cow Fox Fox Bobcat

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 7 7

Puebla Oaxaca US-Mexico border California Arizona Sonora Sonora Sonora Sonora California Oregon Durango Chihuahua West Texas Mexico DF California Michoacan Michoacan Arizona Arizona Chihuahua

1994 1992 1961 1979 1981 1988 1988 1988 1988 1990 1991 1991 1994 1995 1978 1987 1990 1991 1979 1986 1990

1

Identification number Lineagea

Animal

AgVb

Location

Year

Identification number in the reference or GeneBank numberc

3140bsonbct94 (W2) 294txfx83 (W3) 3700txfx97 (W3) 732cask87 (W4) 963arsk84 (W5) 3789wisk98 (W5) 127cask74 (W6) 2521cask94 (W6) 223Israjackal (W8) 210Israfx (W8) 310Israfx (W8), 3813ygfx (W8) 3815ygfx (W8) 3141bcssk90 (W9) 3136bcssk92 (W9) 3145bcssk95 (W9) 1578canarctfx (W10)

Bobcat Fox Fox Skunk Skunk Skunk Skunk Skunk Jackal Fox Fox Fox Fox Skunk Skunk Skunk Arctic fox

7 1 1 1 1 1 1 1 NT NT NT NT NT 10 10 10 NT

Sonora Texas Texas California Arkansas Wisconsin California California Israel Israel Israel Yugoslavia Yugoslavia Baja California sur Baja California Sur Baja California Sur Canada

1994 1983 1997 1987 1984 1998 1974 1994 1998 1995 1993 2000 2000 1990 1992 1995 1991

1

Mxbv3143 Mxdg3125 1, 2 Caushm726 1 Caushm953 1, 2 Azushm947 1, 2 Mxsndg1274 1, 2 Mxsndg672 1, 2 Mxsndg671 1, 2 Mxsndg673 1 Causcy2520 1, 2 Orusdg1303 1 Mxdg3128 1 Mxbv3144 1 Txuscy2830 1, 2 Mxmxdg559 1, 2 Causct674 1 Mxct3122 1 Mxbv3123 1, 3 Azusfx398 1, 3 Azusfx186 1 Mxgm3148 1

Mxgm3140 Txusfx294 1 Txusfx3700 1, 3 Caussk732 1, 3 Arussk963 1, 2 Wiussk3789 1, 3 Caussk127 1, 3 Caussk2521 AF162817 AF162827 AF162832 AJ296195 AJ296194 1 Mxsk3141 1 Mxsk3136 1 Mxsk3145 L20675 1, 2

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

17

Table 3 (Continued ) Identification number Lineagea

Animal

AgVb

Location

Year

Identification number in the reference or GeneBank numberc

2756canarctfx (W10) 964arsk84 (W11) 3436kssk87 (W11) USskArizona (W11) GAH1 (W12) GAH2 (W12) FH45 (W12) FH48 (W12) 3150dftb95 807catb76 2075txtb84 3130chisbv93 3129chishr93 3131verbv95 3146verbv95 3137chisbv94 2320caefbt93 EBL2 Duvenhage

Arctic fox Skunk Skunk Skunk Raccoon Raccoon Raccoon Raccoon Free tail bat Cat Free tail bat Cow Horse Cow Cow Cow Big brown bat European house bat Human

NT 8 8 8 NT NT NT NT 9 9 9 3 3 3 11 11 NT NT NT

Canada Arkansas Kansas Arizona Georgia Georgia Florida Florida Mexico DF California Texas Chiapas Chiapas Veracruz Veracruz Chiapas California France South Africa

1993 1984 1997 2001 1983 1983 1987 1987 1995 1976 1984 1993 1993 1995 1995 1994 1993 1989 1986

L20675 1, 3 Arussk964 1 Ksussk3436 AY170226 Unpublished data. Unpublished data. Unpublished data. Unpublished data. 1 Mxtbbt3150 1 Caustbbt807 1 Txustbbt2075 1 Mxbv3130 1 Mxhr3129 1 Mxbv3131 1 Mxbv3146 1 Mxbv3137 1 Causefbt2320 Outgroup (Bourhy et al., 1993) Outgroup (Bourhy et al., 1993)

SI, segregates independently, the sample did not cluster within any lineage or sublineage; NT, the genetic variant has not been identified with the reduced panel of monoclonal antibodies. 1 De Mattos et al., 1999; 2 Smith et al., 1995; 3 Orciari, 1995. a Lineage as observed in the phylogenetic tree is denoted in bold with one letter and one number below the identification number. b AgV. Antigenic variant determined with the reduced panel of monoclonal antibodies. c Contains sample numbers as they appeared in the original reference or in GenBank. In the same column, superscript numbers on the left side of every sample number or GenBank designation number indicates the original reference where epidemiological relevant information may be found.

al., 1992). Within these fragments, 88 amino acids were encoded, which were localized from amino acid positions 363 to 450, according to the fixed laboratory strains SAD B19, ERA and FLURY. 2.5. Phylogenetic tree construction Multiple alignments were performed by using CLUSTALW (http://www.ebi.ac.uk/clustalw/index.html). The 136 sequences were formerly edited to a common 264 base fragment using BioEdit (Hall, 1999). Molecular analyses were conducted using MEGA 2.1 (Kumar et al., 2001). Distance matrix (Neighbor-Joining) and maximum parsimony methods were used for phylogenetic analysis of Mexican rabies virus taxa. For neighbor joining, corrected nucleotide substitutions were calculated using Kimura’s two-parameter method. The confidence limits, for both methods, were estimated by a bootstrap algorithm applying 1000 iterations (Felsenstein, 1985). To provide a root to the phylogenetic trees, the Lyssavirus species Duvenhage and EBL2 were included as outgroups (Figs. 1 and 2). For clarity, an operational definition of a lineage was considered a group of related samples associated with the same putative reservoir. Similarly, an operational definition for a sublineage was a group of clusters within a major lineage, differing in spatio-temporal characteristics (Fig. 1). Nucleotide and amino acid pairwise p-distances were calculated within (Table 4) and among (Table 5) all lineages and sublineages using MEGA 2.1 software (Kumar et al., 2001). A consensus sequence for each lineage and sublineage was generated using the program BioEdit (Hall, 1999). To observe with

greater clarity any possible time-space relationships among lineages and sublineages, new phylogenies (with the same methods described above) were constructed using consensus sequences only. Sequences that did not cluster within any lineage or sublineage were used as independent groups in the construction of the consensus phylogeny (Fig. 2).

3. Results Analysis of the 136 sequences indicated the occurrence of 142 conserved sites, 178 variable sites, and 160 parsimonious informative sites. The nucleotide identity in the 136 sequences ranged from 78.7 to 100% with an average of 90.3%, whereas the amino acid identity ranged from 89 to 100% with an average of 97.2%. Similarly, consensus sequences varied from 79.3 to 97.7% with an average of 89.5% for nucleotide identity, and from 90.8 to 100%, with an average of 97% for amino acid identity (Table 5). The pattern of nucleotide variation within the samples defined 23 statistically supported lineages and sublineages, divided into major clusters (B and C) from node A (Figs. 1 and 2). Cluster B contained all sequences associated with canine species and skunks, whereas cluster C contained sequences associated with skunks, raccoons and bats. Node D grouped several sublineages specifically associated with dog rabies, D1–D11, which were differentiated from the remaining lineages related with wildlife rabies foci, specifically foxes and skunks, W1–W10 (Figs. 1 and 2). Most lineages and sublineages were identified as antigenic variant (AgV) 1, associated with dogs, skunks and foxes, whereas variants 7 and 8–10 were reported in gray foxes and

18

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

19

Fig. 2. Phylogenetic relationships among major lineages and sublineages observed in Fig. 1. Individual sequences that did not segregate within any lineage were included in the trees. (a) Maximum parsimony tree, consensus tree obtained from 135 trees; (b) Distance matrix tree obtained by neighbor joining method numbers. For both trees, the number at the left of the nodes indicates the statistical support obtained by the bootstrap method, 1000 repetitions and the letters denote nodes of major interest.

Fig. 1. Distance matrix phylogenetic tree, constructed by the neighbor joining method. The scale indicates relative evolutionary distance based upon the number of changes among taxa. Taxa in bold font indicates historical sequences already reported, taxa with regular font indicates new sequences and taxa underlined denotes imported cases. Arbitrary names with a capital letter and number were assigned to clusters with high statistical support at the left side. Bootstrap values were written with bold numbers at the left of each node.

20

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

Table 4 Intralineage or intrasublineage description of rabies virus used to obtain consensus sequences Lineage (AgV)

Reservoir host

Distribution

D1 (V1) D2 (V1) D3 (V1) D4 (V1) D5 (V1) D6 (V1) D7 (V1) D8 (V1) D9 (V1) D10 (V1) D11 (V1)

Coyote-dog Dog Dog Dog Dog Dog Dog Dog Dog Dog Dog

W1 (V1) W2 (V7)

Skunk Gray fox

Lineage (AgV)

Animal

W3 (V1) W5 (V1)

Fox Skunk

Texas North central region US

W6 (V1)

Skunk

W7 (ND) W8 (NT)

Unidentified reservoir Fox

W9 (V10))

Skunk

W10 (NT)

Arctic fox

W11 (V8)

Skunk

W12 (NT) W13 (V8)

Raccoons Skunks

a b c d

Period

a Nucleotide

b Amino

identity

acid identity

Non consensus sites

c Amino

acid

of change

East Texas-Mx border Michoacan Puebla and Tlaxcala Michoacan and Morelos Estado de Mexico Estado de Mexico Puebla and Tlaxcala Puebla and Tlaxcala Guerrero and Oaxaca Yucatan West US-Mx Border

76–02 89–00 78–95 91–93 91–02 91–98 94–99 99–02 92–01 99–02 79–95

0.006 (99.4) 0.009 (99.1) 0.009 (99.1) 0.003 (99.7) 0.002 (99.8) 0.006 (99.4) 0.008 (99.2) 0.002 (99.8) 0.006 (99.4) 0.006 (99.4) 0.014 (98.6)

100 100 100 100 100 100 100 100 100 100 100

Consensus achieved 1186 C/T, 1258 C/T 1384 T/C Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved

Consensus achieved A372 , S396 F438 Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved Consensus achieved

Sinaloa and Durango South Arizona North west Mexico Distribution Period

01 79–02

0.011 (98.9) 0.006 (99.4)

100 100

Consensus achieved 1420 G/A

Consensus achieved S450

a Nucleotide

d Type

Silent Silent Silent Silent Silent Silent Silent Silent Silent Silent NS sample specific Silent Silent

b Amino Non consensus sites acid identity

c Amino

identity 83–97 84–98

0.004 (99.6) 0.067 (93.3)

100 98.9

G413 E365 , A371 , A372 , L381 , D383 , D391 , E398 , T399 , I/M410 , S429 , Y446 , S450,

Silent One NS Amino acid 410

California, US

74–94

0.016 (98.4)

98.9

1309 A/G 1165 A/G, 1183 G/A, 1186 C/T, 1213 A/G, 1219 T/C, 1243 C/T 1264 A/G, 1267 T/C 1300 A/G 1357 T/A 1408 C/T, 1420 A/G 1170G/A, 1255 C/T 1261 G/T, 1315 C/A

Q/R367 , F395 , G397 , R415

Yucatan

99–01

0.002 (99.8)

100

Consensus achieved

Consensus achieved

One NS Amino acid 367 Silent

Yugoslavia and Israel Baja California Sur Canada

93–00

0.024 (97.6)

98.9

Consensus achieved

Consensus achieved

Silent

90–02

0.031 (96.9)

100

Consensus achieved

Consensus achieved

Silent

91–93

0.016 (98.4)

100

K364 , T375 , S389 , A433

Silent

US South Central region

84–99

0.06 (94)

100

A380 , D384 E392 , F395 , E398 , P402 , H431 , N443 , K444

Silent

US Eastern Coast Mex North Central Region

83-87 0.005 (99.5) 2000-2002 0.034 (96.6)

1162 A/G, 1195G/A 1237 G/A, 1369 T/C 1210 C/A, 1222 T/C 1246 A/G, 1255 C/T 1264 G/A, 1276 G/A 1363 C/T, 1399 C/T 1402 A/G 1180 A/G, 1375 C/T 1255 T/C, 1420 T/G

E370 , P435 F395 , S450

Silent Silent

100 98.9

acid

d Type of change

Average nucleotide p-distance and nucleotide identity. average amino acid identity. amino acid encoded at non consensus sites. Type of mutation observed at non consensus sites; Aa, amino acid.

skunks, respectively (Tables 1 and 4). Several amino acid changes appeared to be lineage–sublineage specific, and were consistently conserved in excess of 2 years (Fig. 3). Most amino acid changes had similar physicochemical properties, with the exception of amino acid substitutions in lineages W7, W11, W12 and W13 with suggested structural changes in the N protein (Fig. 3). Findings for each of the major domestic animal and wildlife groups are presented separately.

3.1. Dog rabies More than 10 different sublineages (D2–D11) were associated with at least six major dog rabies foci throughout Mexico, during the period 1976–2002 (Figs. 1 and 2). They presented an average nucleotide and amino acid identity of 96.7 and 99.8 %, respectively (Table 4). These included: a focus in Tlaxcala-Puebla encompassing sublineges D3, D7,

Table 5 Interlineage amino acid and nucleotide identities as well as its respective p-distance values Nucleotide Amino acid D1 D1–D12 W1

W3 W4 W5 W6 W7 W8 W9 W10 W11 W12 W13 Vamp Tdbr

D2–D12

W1

W2

W3

W4

W5

W6

W7

W8

W9

W10

W11

W12

W13

Vamp

Tdbr

97.7 0.023

92 0.08 94.3 0.057

89.3 0.107 91.6 0.084 90.4 0.096

92.7 0.073 95 0.050 93.1 0.069 91.2 0.088

91.2 0.088 92.7 0.073 90.4 0.096 88.5 0.115 92 0.080

94.3 0.057 95 0.050 92.3 0.077 91.6 0.084 93.9 0.061 93.1 0.069

91.2 0.088 93.5 0.065 91.2 0.088 89.7 0.103 92.3 0.077 93.1 0.069 93.5 0.065

91.2 0.088 92.7 0.073 91.6 0.084 89.3 0.107 91.6 0.084 90.4 0.096 92.3 0.077 93.1 0.069

90.4 0.096 92.3 0.077 88.5 0.115 88.1 0.119 90.4 0.096 89.7 0.103 92 0.088 91.2 0.088 90.8 0.092

89.7 0.103 91.2 0.088 90 0.100 88.5 0.115 90.8 0.092 89.7 0.103 92.3 0.077 91.6 0.084 91.2 0.088 88.9 0.11

85.8 0.142 87.4 0.126 87.7 0.123 86.2 0.138 88.9 0.111 84.7 0.153 88.9 0.111 86.2 0.138 88.5 0.115 87 0.130 87.4 0.126

82.4 0.176 83.1 0.169 83.5 0.165 82.4 0.176 83.1 0.169 82 0.180 81.6 0.184 82.4 0.176 80.5 0.195 83.5 0.165 82.8 0.172 79.3 0.207

82 0.180 82.8 0.172 80.8 0.192 82 0.180 83.9 0.161 80.8 0.192 81.6 0.184 82.4 0.176 80.1 0.99 83.1 0.169 80.5 0.195 81.2 0.188 88.9 0.111

84.7 0.153 85.4 0.146 83.1 0.169 83.5 0.165 83.9 0.161 84.7 0.153 86.2 0.138 83.9 0.161 82.8 0.172 85.1 0.149 85.4 0.146 82.8 0.172 89.3 0.107 90 0.100

83.5 0.165 82.8 0.172 80.8 0.192 79.3 0.207 84.3 0.157 80.8 0.192 83.9 0.161 81.2 0.188 82.4 0.176 83.9 0.161 82 0.180 85.1 0.149 83.9 0.161 83.5 0.165 85.4 0.146

83.9 0.161 83.1 0.169 82 0.180 81.2 0.188 85.1 0.149 82 0.180 83.1 0.169 81.2 0.188 82.4 0.176 84.7 0.153 81.6 0.184 83.1 0.169 83.1 0.169 83.5 0.165 84.7 0.153 93.5 0.065

98.9 0.011 97.7 98.9 0.023 0.011 95.4 96.6 0.046 0.034 98.9 100 0.011 0 96.6 97.7 0.034 0.023 98.9 100 0.011 0 96.6 97.7 0.034 0.023 96.6 97.7 0.034 0.023 96.6 97.7 0.034 0.023 96.6 97.7 0.034 0.023 94.3 95.4 0.057 0.046 96.6 97.7 0.0324 0.023 94.3 95.4 0.057 0.046 95.4 96.6 0.046 0.034 93.1 94.3 0.069 0.057 94.3 95.4 0.057 0.046

95.4 0.046 98.9 0.011 96.6 0.034 98.9 0.011 96.6 0.034 96.6 0.034 96.6 0.034 96.6 0.034 94.3 0.057 96.6 0.034 94.3 0.057 95.4 0.046 93.1 0.069 94.3 0.057

96.6 0.034 94.3 0.057 96.6 0.034 94.3 0.057 94.3 0.057 94.3 0.057 96.6 0.034 94.3 0.057 94.3 0.057 92 0.08 93.1 0.069 90.8 0.092 92 0.080

97.7 0.023 100 0 97.7 0.023 97.7 0.023 97.7 0.023 97.7 0.023 95.4 0.046 97.7 0.023 95.4 0.046 96.6 0.034 94.3 0.057 95.4 0.046

97.7 0.023 96.6 0.034 95.4 0.046 95.4 0.046 95.4 0.046 93.1 0.069 95.4 0.046 93.1 0.069 94.3 0.057 92 0.080 93.1 0.069

97.7 0.023 97.7 0.023 97.7 0.023 97.7 0.023 95.4 0.046 97.7 0.023 95.4 0.046 96.6 0.034 94.3 0.057 95.4 0.046

95.4 0.046 95.4 0.046 96.6 0.034 93.1 0.069 95.4 0.046 93.1 0.069 94.3 0.057 93.1 0.069 94.3 0.057

95.4 0.046 95.4 0.046 95.4 0.046 95.4 0.046 93.1 0.069 94.3 0.057 93.1 0.069 94.3 0.057

95.4 0.046 93.1 0.069 97.7 0.023 95.4 0.046 94.3 0.057 92 0.080 93.1 0.069

95.4 0.046 95.4 0.046 93.1 0.069 94.3 0.057 94.3 0.057 95.4 0.046

93.1 0.069 90.8 0.092 94.3 0.057 90.8 0.092 92 0.080

97.7 0.023 96.6 0.034 94.3 0.057 93.1 0.069

94.3 0.057 92 0.080 90.8 0.092

94.3 0.057 93.1 0.069

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

W2

D1

98.9 0.011

Lower matrix, in bold numbers, shows interlineage amino acid identity (percentage) and p-distance values; upper matrix shows interlineage nucleotide identity (percentage) and p-distance values. Vamp, vampire; Tdbr, Tadarida brasiliensis.

21

22

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

Fig. 3. Consensus amino acid sequences from major rabies virus lineages and sublineages found in Mexico and in the United States. Amino acid positions were identified from 363 to 450, according to laboratory strains SAD B19, ERA and FLURY. Solid line boxes indicate two overlapping epitopes at antigenic site IV related to putative B cell activation. Dash line boxes show the position of a putative immunodominant epitope, 31 D, related with T cell activation in vitro. Arrow shows the putative phosphorylation site. *Stop codon.

D8; a focus in Guerrero-Oaxaca that included sublineage D9, and single samples 559dfdg78 and 674cagroct87; an focus in Estado de Mexico that comprised sublineages D5 and D6; a focus in Michoacan that contained sublineage D2, and single samples 3122michct90 and 3123michbv91; a focus in Yucatan that contained sublineage D10; and a focus in Northwest Mexico-US border encompassing sublineage D11, which contained historical samples from an apparently extinct dog rabies enzootic that occurred in Durango, Sonora and Chihuahua during 1976–1994 (Table 4; Figs. 2 and 4). The phylogenetic links between sublineages, for instance, D7–D8, D9-559dfdg78, D4–D5, D3–D2–D1 and D11-single samples from Michoacan, Chiapas and Guerrero, belonging

to different foci, suggest either historical or recent dissemination events (Fig. 2). Evidence of dog rabies dissemination was also observed within sublineages D3, D5, D7, D8 and D9, which grouped samples from other Mexican states. Movements from the Puebla Tlaxcala dog enzootic focus (D7, D8 and D3) to the states of Guerrero, Distrito Federal, Guanajuato, Jalisco and Estado de M´exico were detected. Similarly, a focus at Estado de Mexico (D5) contained recent rabies cases from Hidalgo, Zacatecas and Distrito Federal. A focus in Guerrero (D9) presented association with one historical case from San Luis Potosi (Table 2; Fig. 1). Conversely, the D11 focus displayed spread among Northern states of Mexico (Sonora, Durango, Chihuahua) and Michoacan, related to

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

23

Fig. 4. Geographic distribution of rabies virus major lineages found within Mexico.

dog rabies cases in Texas and California. This latter focus was deemed extinct because there were no recent reports in the region identified with that variant. Canine rabies foci detected in Yucatan did not appear to have phylogenetic connections to any other individual sample or sublineage (Figs. 1 and 2). 3.2. Dog-coyote rabies The samples obtained from two cats and one coyote (Canis latrans) in the border states of Coahuila and Tamaulipas clustered together with several historical samples pertaining to a long-term rabies enzootic focus (1976–1994) in coyotes distributed in Southeast Texas (Table 4; Fig. 1). This D1 sublineage showed a consensus amino acid change (Ser 426 change to Thr 426) with respect to all other dog-related rabies foci D2–D11 (Fig. 3). Additionally, D1 presented a phylogenetic link with sublineages D2 and D3, which corresponded to rabies foci located in Michoacan during 1989–2002 and Puebla during 1978–1995 (Figs. 1 and 2). 3.3. Fox rabies Lineage W2 had an internal average pairwise identity of 99.4%, comprising samples obtained from gray foxes, wild

cats and one dog during 1990 to 2002 (Table 4; Fig. 1). The cluster was associated distinctively with the gray fox (Urocyon cinereoargenteus), AgV 7, with a distribution that encompassed the Northwest states, such as Sonora, Chihuahua and Sinaloa, as well as the northcentral state of Zacatecas (Table 2; Fig. 4). Lineage W2 segregated independently from those monophyletic clusters associated with fox rabies in Texas (Texas fox variant) and Canada (Arctic fox variant), as well as samples from Yugoslavia and Israel (Figs. 1 and 2). In addition, W2, W3 and W10 had at least one consensus amino acid difference with respect to each other and with all other lineages (Fig. 3). 3.4. Skunk rabies At least four lineages, identified within branches with bootstrap values higher than 90, were associated with skunk rabies in Mexico (Table 2; Fig. 1). Their distributions comprised the States of Sinaloa and Durango for W1; Baja California Sur for W9; San Luis Potosi, Zacatecas, Durango and Jalisco for W13; and Chihuahua for W11 (Fig. 4). The latter two were identified as AgV 8 and presented a consensus amino acid change at position 378, Asp-Glu, shared with raccoons, vampire and big brown bats. These also had some

24

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

amino acid substitutions that were lineage-specific (Fig. 3; Table 5). Conversely, W1 and W9 lineages clustered together with samples associated with canine species, which derived at node B (Figs. 1 and 2). They were identified as AgV 1 and 10 and had an average intralineage pairwise nucleotide identity of 98.9 and 96.9% (100% of amino acid identity), respectively (Table 4). The W9 group presented two substitutions at amino acids 377 and 407, sharing the latter with gray and Arctic foxes, whereas W1 had a change in amino acid 448, similar to lineages W11 and W12 (Fig. 3). 3.5. Unknown reservoir lineage W7 This lineage was composed of five samples obtained from one deer (Odocoileus virginianus), two agoutis (Aguti paca) maintained in captivity, and one spotted skunk (Spilogale putorius) in Merida, Yucatan (Table 2; Fig. 4). This group segregated independently from dog rabies foci (node D) but within the node B associated with rabies in wild terrestrial carnivores (Figs. 1 and 2). The intralineage nucleotide and amino acid identities were 99.8 and 100%, respectively. Amino acid substitutions were detected at Gln 367 (changed to Lys) and at Val 379 (changed to Ala) when interlineage comparisons were performed (Fig. 3). The latter amino acid change was also observed in lineage W13 (Fig. 3).

4. Discussion Overall, a relatively high degree of variation was found within the last 264 nucleotides encoding for 88 amino acids at the C terminus of the rabies virus nucleoprotein. Most regions contributing to putative B and Th cell epitopes were retained, as was the putative phosphorylation site, at Ser 389 (Dietzschold et al., 1987; Ertl et al., 1991). Some amino acid positions conserved along time were found in association with specific lineages, suggesting that traces of positive selection may also be found within variable regions of the nucleoprotein gene, provided that a representative spatiotemporal spectrum of variation was included within the set of samples analyzed. Although the degree of nucleotide and amino acid identity depends on the precise genomic region under study, the use of relatively high informative regions within a greater context of comparison (which includes several clades associated with several putative rabies reservoir so far reported in the region) have allowed not only to find out consistent epidemiologic relevant associations (such as species specific clusters and lineages linked with spatiospecific rabies foci) but also, may contribute to have more accurate figures on intragenotype and intergenotype diversity (Bourhy et al., 1999; Johnson et al., 2002). By comparison, values of nucleotide and amino acid identity were lower than those proposed for genotype classification (Kissi et al., 1995). These results corroborate that the intragenotype variation (and, in consequence, rabies virus diversity) has been underestimated, as predicted in earlier studies (Bourhy et al.,

1993). Conversely, other variable and highly informative regions within the rabies virus genome should be explored in order to make unequivocal epidemiological or evolutionary assertments whenever a restricted set of samples (in terms of the spatio-temporal and putative reservoir diversity it encompasses) is being studied (Bourhy et al., 1993, 1995, NadinDavis et al., 1999, 2001, 2002, 2003; Badrane and Tordo, 2001; Holmes et al., 2002). Provided that infrastructural or technological resources do not constitute a major constrain most of the new and historical rabies positive sets of samples should be sequenced on a global basis targeting all those informative regions previously reported, so that more accurate regional or global comparisons can be made. The diversity of rabies viruses associated with terrestrial carnivores in Mexico rivals the complexity reported in the US (Smith, 1996; Smith et al., 1992, 1995). Such variation is not unexpected, if the size and biodiversity of Mexico is considered, because the country comprises one of the five biological megadiversities of the Americas, and contains 10% of the world’s flora and fauna (Fa and Morales, 1998). The phylogenetic relationships of Mexican and US samples, even though they only pertained to genotype 1, are generally consistent with the trends reported earlier. (Badrane and Tordo, 2001; Holmes et al., 2002). The Chiroptera group contained lineages associated with skunk and raccoon rabies viruses, detected in both Mexico and the US. The Carnivora group contained canine (dog, fox and coyote) and skunk-related viruses. The latter data may suggest that rabies in Mexico may have an independent autochthonous origin maintained by bats, with spillover to skunks and an unrelated ancestor mainly found in dogs, which derived from the Old World through colonization, with subsequent spillover to foxes, skunks and coyotes (Badrane and Tordo, 2001; Smith et al., 1992). Dog rabies cases in Mexico have decreased dramatically over the last 10 years as a consequence of intensified vaccination campaigns (Velazquez-Monroy et al., 2003). Nonetheless, six active dog rabies foci were found. Two were involved with human rabies cases in Puebla during 2000 and in Estado de Mexico during 2001 (Secretar´ıa de Salud, 2003). Two additional human rabies cases were reported recently in the states of Oaxaca during 2002 and in Chiapas during 2003 (Secretar´ıa de Salud, 2004). These cases provide evidence that dog rabies foci persist in the presence of low or under reported activity. The relationships among individual samples and sublineages, observed in the phylogenetic trees, suggested both recent and historical dog rabies dissemination. These included at least five major rabies foci in different states within Mexico and at the US-Mexico border. Canine rabies patterns do coincide with trade routes and immigration paths to major cities within Mexico and towards the US, as previously noted (Smith et al., 1992; De Mattos et al., 1999). Both long and short distance patterns of dissemination were observed, suggesting spread by individual dog movements as well as dissemination by human translocation.

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

Major foci contained more than one sublineage, which did not group in monophyletic clusters, suggesting co-existing or co-emerging viruses. Taken together, these findings suggest that multiple genetic viral variants may appear randomly by drift in different dog subpopulations separated by geographical barriers (highways, mountains), followed by their spread throughout a local dog subpopulation (infected with a particular variant) and dissemination thereafter, as proposed for fox rabies (Nadin-Davis et al., 1999). According to current phylogenetic evidence and as suggested previously, coyote rabies, maintained as a stable enzootic for more than 26 years, is the product of historical dog rabies spillover events (Smith et al., 1992, 1995). An enzootic focus of dog rabies may have been maintained for several years in the northeast border states of Mexico before 1976, when coyote rabies cases were reported in eastern Texas. A link of D1 with an older sublineage was not identified, but its relationship with D3 (the oldest lineage found in Mexico to date) suggests that the coyote rabies virus variant may have emerged via dog translocation from long-term canine rabies enzootic foci within Mexico. Strikingly, the coyote rabies virus variant did not present any direct ties with the D11 sublineage, which represented the major dog enzootic focus along the Mexico-US border. Health authorities should devote more emphasis to control both canine rabies and dog translocation as actions to prevent future spillover infection to coyotes and other wild mammals. Additionally, although coyote rabies virus variants have not been identified from stable dog enzootic foci, local health authorities should strengthen wildlife rabies surveillance to test the dog-coyote transmission hypothesis. The occurrence of gray fox rabies in Mexico had been reported since 1999, but the rabies virus variant has only been sampled from bobcats, Lynx rufus (De Mattos et al., 1999). Here, we report three sequences obtained from gray foxes in the states of Sonora and Zacatecas. These also include sequences from wild felids (such as bobcats in Sonora, Sinaloa and from the cougar, Puma concolor in Cihuahua). The latter samples had been reported previously (Velasco-Villa et al., 2002). Mexican gray fox rabies virus variants obtained from 1990 to 2002 grouped within a monophyletic cluster, which contrast drastically with the huge diversity observed in dog-related samples. These dissimilar trends observed may suggest that temporal effects may be largely involved in the evolution of rabies virus within this two canid species, given that rabies has been reported enzootic in dog populations of Mexico since the beginning of the 18th century, unlike gray fox rabies which started to be reported enzootic in early 20th century (Johnson, 1971; Secretar´ıa de Salud, 2001; Wilkinson, 2002). Also differences in the biology of both canid species may account in the degree of variation found in rabies virus (Anderson et al., 1981). Additionally, this pattern suggests that gray fox rabies (at least in Mexico) is the product of a recent spread of a single gray fox rabies focus. This inference not only correlates with gray fox ecology but particularly with its ability to move over large areas and par-

25

tially explains the present distribution of rabies cases (Fa and Morales, 1998). The W2 lineage was also found closely related to the enzootic gray fox rabies foci in Arizona (Krebs et al., 2003; Smith et al., 1995). Lineages W2 and W3 have been found co-circulating in the southeast corner of Arizona, but the W3 lineage, the so-called Texas fox, have not been detected in Mexico. In spite of their spatial-temporal overlap, W2 and W3 are not closely related, which supports the idea they are fox rabies enzootic foci with independent origins, with probable emergence from different fox populations or species (Smith et al., 1995; Smith, 1996). Rabies in skunks has been reported in Mexico by different groups (Aranda and L´opez de Buen, 1999; De Mattos et al., 1999; Loza-Rubio et al., 1999; Velasco-Villa et al., 2002). Like canine rabies, skunk rabies harbors a similarly high genetic diversity. Skunk rabies in Mexico presents two different ancestries, one common with canine rabies for W1 and W9, and the other common to that presented by US and Mexican bat rabies, for W11 and W13. A similar pattern has been reported for raccoon and skunk rabies in North America, which suggests multiple independent emergences of skunk-adapted rabies from both canid and chiropteran reservoirs (Badrane and Tordo, 2001; Nadin-Davis et al., 2001; Holmes et al., 2002). The occurrence of sublineages was clearly observed within W9 and W13, despite the low number of samples analyzed. These results suggest that rabies virus samples are spatio-temporally structured in skunk populations, most likely influenced by skunk biology (Anderson et al., 1981). Additionally, the phylogenetic patterns observed for skunk rabies also suggest the random emergence of new variants in different subpopulations of skunks, as suggested for fox rabies in Canada (Nadin-Davis et al., 1993, 1999). Human rabies transmitted by spotted skunks (lineage W9) occurred in Baja California Sur (Velasco-Villa et al., 2002). Although it has not been documented in the northcentral region of Mexico, the potential risk has been presented in previous studies (Aranda and L´opez de Buen, 1999). One striking difference between skunk rabies in the US and Mexico was the involvement of different species. Mexico has its major reservoir in spotted skunks, Spilogale putorius, whereas the US reservoir is primarily the striped skunk Mephitis mephitis, though the two species are sympatric in both countries. The phylogenetic link found between skunk and raccoon rabies in the US may be a point of concern, given that both species are in contact, with transient events of raccoon rabies dissemination to skunks, representing a risk of raccoon rabies spillover to skunks (Guerra et al., 2003). Spotted skunks and raccoons are sympatric in Mexico, but raccoon rabies has not been reported to date. One Mexican sample, 3291chihbv99, obtained from a cow in Ciudad Juarez Chihuahua, was genetically related with the US Southcentral skunk variant, specifically with samples from Flagstaff, Arizona. This finding suggests that the distribution of the variant previously reported by other authors (Smith, 1996) is much wider, encompassing part of the Mexican territory.

26

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27

The Yucatan lineage W7 clustered within the wildlife rabies group associated with terrestrial carnivores. A reservoir could not be identified with certainty, due to the low number of samples and because of its lack of a link with any other known reservoir. Later results suggest an ancient spillover occurred from canine rabies, just as in the case W1, and the rest of the rabies foci associated with terrestrial carnivores that were observed within this group. This genetic variant was obtained from two rodents (Aguti paca), one deer (Odocoileus virginianus) and one skunk (Spilogale putorius). Although the skunk may be the most likely reservoir for the new variant, more studies will be necessary to corroborate the latter inference. The discovery of patterns of rabies involvement with different animals, reservoir diversity, interspecies transmission, altered distribution and translocation of major hosts, as well as the detection of minor enzootic or stable foci, was made possible by the use of molecular analyses paired with passive public health surveillance data. These data suggest an extraordinary complexity of rabies virus associated with terrestrial carnivores heretofore not described for Mexico. Moreover, use of the last 88 amino acid of the nucleoprotein C terminus appears to be a useful tool for investigations of the molecular epidemiology of rabies, due to its high non-random pattern of neutral change which occurs when rabies virus is maintained within host populations. Ultimately, the emergence and maintenance of distribution patterns of rabies enzootic cycles depend primarily upon the biology and ecology of their respective reservoirs, coupled with a plethora of RNA viruses intent upon expansion of niche space. Additional studies will focus upon bat rabies virus samples, the sequencing of larger regions within the genome and a sampling scheme trying to encompass wider spatio-temporal ranges within rabies foci previously detected. Acknowledgments We thank our colleagues working at the National Network of Public Health Rabies Laboratories in Mexico and in the US as well as the epidemiologists and personnel of the rabies control program in Mexico, who provided samples and epidemiological information about them; Jose Luis Marrufo Olivares and Isaias Sauri from the Laboratorio Central Regional of Merida, Yucatan for providing information about Yucatan samples; and to Caroline J. Henderson, Jennifer M. Snaman and Leslie Real who kindly provided several raccoon rabies virus sequences. This work was submitted in partial fulfillment of the requirements for the D. Sc. degree for Andres Velasco-Villa at Doctorado en Ciencias Biomedicas, Universidad Nacional Autonoma de Mexico and CONACyT. References Anderson, R.M., Jackson, H.C., May, R.M., Smith, A.M., 1981. Population dynamics of fox rabies in Europe. Nature 289, 765–771.

Aranda, M., L´opez de Buen, L., 1999. Rabies in skunks from M´exico. J. Wildl. Dis. 35, 574–577. Badrane, H., Tordo, N., 2001. Host switching in Lyssavirus history from the chiroptera to the carnivora orders. J. Virol. 75, 8096–8104. Bourhy, H., Kissi, B., Tordo, N., 1993. Molecular diversity of Lyssavirus genus. Virology 194, 70–81. 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. 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. David, D., Yakobson, B., Smith, J.S., Stram, Y., 2000. Molecular epidemiology of rabies virus samples from Israel and other middle- and near-eastern countries. J. Clin. Microbiol. 38, 755–762. De Mattos, C.C., De Mattos, C.A., Loza-Rubio, E., Aguilar-Seti´en, A., Orciari, L.A., Smith, J.S., 1999. Molecular characterization of rabies virus samples from Mexico: implications for transmission dynamics and human risk. Am. J. Trop. Med. Hyg. 61, 587–597. Diaz, A.M., Papo, S., Rodr´ıguez, A., Smith, J.S., 1994. Antigenic analysis of rabies-virus samples from Latin America and the Caribbean. J. Vet. Med. Series B 41, 153–160. Dietzschold, B., Lafon, M., Wang, H., Otvos Jr., L., Celis, E., Wunner, W.H., Koprowski, H., 1987. Localization and immunological characterization of antigenic domains of the rabies virus internal N and NS proteins. Virus Res. 8, 103–125. Ertl, H., Dietzschold, B., Otvos, L., 1991. T helper cell epitope of rabies virus nucleoprotein defined by tri- and tetrapeptides. Eur. J. Immunol. 21, 1–10. Fa, J.E., Morales, M.L., 1998. Patrones de Diversidad de Mam´ıferos en M´exico. In: Ramamoorthy, T.P., Bye, R., Lot, A., Fa, J. (Eds.), Diversidad Biol´ogica de M´exico: Or´ıgenes y Distribuci´on. Instituto de Biolog´ıa. Universidad Aut´onoma de M´exico, pp. 315–352. Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783–791. Guerra, M.A., Curns, A.T., Rupprecht, C.E., Hanlon, C.A., Krebs, J.W., Childs, J.E., 2003. Skunk and raccoon rabies in the eastern United States: temporal and spatial analysis. Emerg. Infect. Dis. 9, 1143–1150. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41, 95–98. Holmes, C.E., Woelk, H.C., Kassis, R., Bourhy, H., 2002. Genetic constraints and adaptive evolution of rabies virus in nature. Virology 292, 247–257. Johnson, H.N., 1971. General epizootiology of rabies. In: Nagano, Y., Davenport, F.M. (Eds.), Rabies. Proceedings of Working Conference on Rabies Sponsored by Japan-United States Cooperative Medical Science, Program, University of Tokyo Press, pp. 237–251. Johnson, N., McElhinney, M.L., Smith, J., Lowings, P., Fooks, R.A., 2002. Phylogenetic comparison of the genus Lyssavirus using distal coding sequences of the glycoprotein and nucleoprotein genes. Arch. Virol. 147, 2111–2123. Kissi, B., Tordo, N., Bourhy, H., 1995. Genetic polymorphism in the rabies virus nucleoprotein gene. Virology 209, 526–537. Krebs, J.W., Wheeling, J.T., Childs, J.E., 2002. Rabies surveillance in the United States during. J. Am. Vet. Med. Assoc. 223, 1736–1748. Kumar, S., Tamura, K., Jakobsen, I.B., Nei, M., 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17, 1244–1245. Loza-Rubio, E., Aguilar-Seti´en, A., Bahloul, C., Brochier, B., Pastoret, P.-P., Tordo, N., 1999. Discrimination between epidemiological cycles of rabies in Mexico. Arch. Med. Res. 30, 144–149. Nadin-Davis, S.A., Casey, G.A., Wandeler, A.I., 1993. Identification of regional variants of the rabies virus within the Canadian province of Ontario. J. Gen. Virol. 74, 829–837.

A. Velasco-Villa et al. / Virus Research 111 (2005) 13–27 Nadin-Davis, S.A., Sampath, M.I., Casey, G.A., Tinline, R.R., Wandeler, A.I., 1999. Phylogeographic patterns exhibited by Ontario rabies virus variants. Epidemiol. Infect. 123, 325–336. Nadin-Davis, S.A., Huang, W., Armstrong, J., Casey, G.A., Bahloul, C., Tordo, N., Wandeler, A., 2001. Antigenic and genetic divergence of rabies viruses from bat species indigenous to Canada. Virus Res. 74, 139–156. Nadin-Davis, S.A., Abdel-Malik, M., Armstrong, J., Wandeler, A., 2002. Lyssavirus P gene characterisation provides insights into the phylogeny of the genus and identifies structural similarities and diversity within the encoded phosphoprotein. Virology 298, 286–305. Nadin-Davis, S.A., Simani, S., Armstrong, J., Fayaz, A., Wandeler, A., 2003. Molecular and antigenic characterization of rabies viruses from Iran identifies variants with distinct epidemiological origins. Epidemiol Infect. 131, 777–790. Orciari, A.L., 1995. Master thesis, University of Georgia, Athens, Georgia, US. Rupprecht, C.E., Glickman, L.T., Spencer, P.A., Wiktor, T.J., 1987. Epidemiology of rabies virus variants. Differentiation using monoclonal antibodies and discriminant analysis. Am. J. Epidemiol. 126, 298–309. Secretar´ıa de Salud, SSA, 2001. Programa de acci´on: rabia. Mexico DF. Secretar´ıa de Salud, pp. 13–33s. Secretar´ıa de Salud, SSA, 2003. Direccion General de Epidemiologia. Anuarios de morbilidad, http://www.epi.org.mx/infoepi/index.htm.

27

Secretar´ıa de Salud, SSA, 2004. Sistema Nacional de Vigilancia Epi´ demiol´ogica. Epidemiolog´ıa Sistema Unico de Informaci´on, vol. 53, p. 3. Smith, J.S., Orciari, L.A., Yager, P.A., Seidel, D.H., Warner, C.K., 1992. Epidemiologic and historical relationships among 87 rabies virus samples as determined by limited sequence analysis. J. Infect. Dis. 166, 296–307. 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., 1996. New aspects of rabies with emphasis on epidemiology, diagnosis and prevention of the disease in United States. Clin. Microbiol. Rev. 9, 166–176. Smith, J.S., 2002. Molecular epidemiology. In: Jackson, A.C., Wunner, W.H. (Eds.), Rabies. Academic Press, San Diego, pp. 79–111. Velasco-Villa, A., G´omez-Sierra, M., Hern´andez-Rodr´ıguez, G., Ju´arezIslas, V., Mel´endez-F´elix, A., Vargas-Pino, F., Vel´azquez-Monroy, O., Flisser, A., 2002. Antigenic diversity and distribution of rabies virus in Mexico. J. Clin. Microbiol. 40, 951–958. Velazquez-Monroy, O., Vargas-Pino, F., Gutierrez-Cedillo, V., LecuonaOlivares, L., 2003. Advances in canine rabies control in Mexico. In: The XIV international conference “Rabies in the Americas”. Thomas Jefferson University, Philadelphia, Pennsylvania, USA, p. 78. Wilkinson, L., 2002. History. In: Jackson, A.C., Wunner, W.H. (Eds.), Rabies. Academic Press, San Diego, pp. 1–22.