INTRODUCTION

Analyses of DNA variation offer a unique perspective into the domestication of the dog (Canis familiaris) and into the origin of the American dogs in particular. People in America have had dogs for at least 8-9,000 years(1), much longer than any other domesticate. Did Americans independently come up with the idea to domesticate dogs? If this was the case, it would imply that dogs might have played an important role in primitive societies since they were also the first species to be domesticated in Eurasia. Alternatively, did the people who colonized America bring already domesticated dogs with them? In this case, perhaps the dogs were one of the factors that helped stone-age humans move into the Siberian Arctic and into the New World. This would also imply that dogs were domesticated earlier than previously thought based on the archaeological record, because they would have needed to spread across Europe and Asia as well as into America almost instantly. The archaeological record has suggested that dogs were domesticated about 14,000 years ago(2, 3) and this is very close to the presumed first arrival of humans to the Americas(4).

METHODS

Here we will discuss results from studies of mitochondrial DNA (mtDNA) variation. This DNA molecule is located in an organelle, the mitochondrion, and not the nucleus of the cell. For this reason it is inherited only from the mother, and therefore is inherited as a unit because there is no opportunity for it to recombine with the father's mitochondrial DNA. This makes these sequences easier to analyze, and make phylogenies with. Also, there are from tens to thousands of mitochondria in each cell, although there is only one nucleus. The relatively higher copy number of mitochondrial DNA in cells means that even long after an organism has died, the chances of recovering genetic material are much larger for mtDNA than for DNA coming from the nucleus. The DNA retrieved from old remains is called "ancient DNA" and studies of this ancient genetic material rely heavily on the mtDNA. Using these techniques it is possible to isolate and analyze DNA from animal remains that are up to approximately 100,000 years old (5), depending on the preservation of the specimen.

In one early sequence based study, mtDNA sequences (haplotypes) were obtained from 140 dogs from 67 breeds and from 162 gray wolves (Canis lupus) from across their range by Vilà et al.(6). They compared the dog and wolf sequences with each other and another closely related canid, the coyote (Canis latrans). All of the sequences in dogs were very similar to sequences in gray wolves, and less similar to the coyote. This indicated that the gray wolf was the ancestor of modern dogs. When Vilà et al. constructed a tree to look at how each individual sequence was related to the others, they found that the dogs formed four clades (groups of most related sequences) inside the genetic diversity of gray wolves (Figure 1). They named these clades I-IV. Most of the dogs that were studied had a clade I sequence.

Figure 1. Phylogeny of modern gray wolves and domestic dogs. Neighbor-joining phylogeny of modern domestic dog and gray wolf partial mitochondrial control region sequence with four main clades of dogs labeled I-IV after Vilà et al. (6). Dog lineages are marked in gray. The phylogeny is rooted with a coyote.

ORIGIN OF DOMESTIC DOGS

When European explorers arrived to the New World, they found continents filled with people, and their dogs. What was the origin of these dogs? The ancestor of domestic dogs, the gray wolf, naturally ranges over most of Europe, Asia and North America and the same process that had led to the domestication of the dog in Eurasia could have taken place in the New World. However, all four of the dog clades identified by Vilà et al.(6) were more closely related to Eurasian gray wolf sequences than to American gray wolf sequences. This supports the archeological view that suggests that dogs were domesticated in Eurasia(2). Some dogs from American breeds, such as the Alaskan Husky, the Eskimo dog, the Newfoundland and the Mexican hairless dog, were included in the study. These dogs shared haplotypes (mtDNA sequences) with other breeds of dogs from Europe and Asia, and were not particularly closely related to one another. This could indicate that American dogs had a Eurasian origin. However, the results were not conclusive. Very few American breeds had been included in the study, and some of those had a very recent origin, likely from dogs of recent European origin (such as the Chesapeake Bay retriever), or have likely been interbred with dogs of recent Eurasian ancestry.

ORIGIN OF MODERN AMERICAN DOGS

The Mexican hairless dog, or Xoloitzcuintle, has a long archeological record in America. It is represented on many clay figures from the Colima culture (250 BC to 450 AD) in western Mexico(7) and by many osteological remains, and is thought to have been protected from the Spanish conquistadors in isolated areas because of its religious value (8). This breed was present in America long before the arrival of historic Eurasians with their dogs, and may represent the dogs that were present in America in pre-historic times. For these reasons Vilà et al.(9) surveyed the genetic diversity in the Mexican hairless dog in order to determine whether or not there had been a separate domestication of dogs in America. In 19 individuals seven different haplotypes belonging to three of the four main dog clades were identified. These haplotypes were not particularly closely related to one another, were similar to haplotypes in modern European breeds, and were not at all closely related to American wolves. This result supported the earlier conclusion that American dogs shared a common origin with Eurasian dogs. However, it has been shown that the peculiar traits characteristic of Mexican hairless dogs, including hairlessness and reduced dentition, are dominant traits(7). As a result of this, a large proportion of the offspring of the cross between a Mexican hairless dog and any other dog may also lack hair and show reduced dentition, and could thus be mistaken by a pure Mexican hairless dog. Because these characteristics are dominant and several hundred years have passed since the explorers and their dogs arrived to Mexico, it is possible that there has been mixing between these and native dogs, perhaps quite extensively, and the results of crosses, sharing morphologic characteristics with pure individuals, could be misidentified. Perhaps many modern Mexican hairless dogs do not genetically resemble the dogs that were in America before the arrival of the Spanish explorers.

The large number of Old World dogs that have been brought to America could have mixed with or replaced local dogs, obscuring the history of the American domestic dogs. For this reason, to really understand the origin of the pre-contact dogs in America it is necessary to study them directly. Therefore, using ancient DNA techniques Leonard et al.(10) genetically analyzed 38 pre-Columbian dog remains from Latin America (Mexico, Bolivia and Peru) and 12 dog remains from Alaska from before its discovery by European explorers. Since the preservation of all the samples had not been equally good, of the total of 50 samples studied, Leonard et al. obtained mtDNA sequences from only 24. Seventeen of the 24 samples successfully analyzed yielded haplotypes that had never before been observed in dogs throughout the world, and six had haplotypes previously identified in domestic dogs. In all cases the haplotypes were very closely related to other dog haplotypes, and not closely related to American wolf haplotypes. This strongly supports the hypothesis that American and Eurasian domestic dogs share a common origin. It follows that people brought fully domesticated dogs with them when they colonized the New World. No evidence of a separate domestication of dogs from North American gray wolves or of backcrossing between domestic dogs and North American gray wolves was found. However, although the haplotypes found in American dogs were closely related to Eurasian dogs, some of them formed a clade within the main group (clade I) of domestic dogs (called clade a in Figure 2). This group of haplotypes found only in America indicates that the dogs were present and isolated in the New World for a considerable amount of time. This long period of isolation led to the appearance of a group of sequences (haplotypes) that are similar to each other but easily distinguishable from dogs from other parts of the world.

Figure 2. Minimum spanning network of clade I dog haplotypes. Modern haplotypes are represented as white circles, and ancient American haplotypes are represented in gray. Each bar between haplotypes represents a one base pair change. Based on (10).

ORIGIN OF ANCIENT AMERICAN DOGS

These studies strongly support the hypotheses that dogs were only domesticated in Eurasia (outside of India,[11]), and that humans brought already domesticated dogs with them when they colonized America from Asia. This could suggest that dogs were domesticated in the Old World more than 15,000 years ago, to give them time to spread across Eurasia and into America. Alternatively, the spread of the dogs between ancient human groups could have been very rapid despite the absence of transmission of other cultural artifacts. Further, this suggests that dogs have accompanied humans throughout their expansion in America.

HYBRIDIZATION

Hybridization between dogs and American gray wolves and/or coyotes in prehistoric and recent times has been suggested based on morphological characters and accounts by early explorers(1,12). Genetic surveys of dogs in America have not found any evidence of hybridization between dogs and native American wild canids(6,9,10, however, see [13]). A genetic survey of the coyotes in the southeast United States did find evidence of hybridization between coyotes and a female dog, although the offspring of this cross were incorporated into the coyote population and may have not affected the dog population(14). However, the studies mentioned here are all based on mitochondrial DNA, which is inherited only through the mother. Therefore, if the offspring of a hybridization event between a female domestic dog and a male wolf or coyote were incorporated into the domestic dog population, it would not have been detected by these studies. Studies with markers which are also inherited through the paternal line might in the future provide a more complete view of the origin of the American dogs and their interaction with other wild canids.

CONCLUSIONS

The genetic data have shown us that dogs were domesticated from gray wolves in Europe or Asia (but not India). There is no evidence of a separate domestication event from North American gray wolves, or any other North American canid. These data indicate that people brought already domesticated dogs with them when they colonized America from Asia. The dogs brought to America by the early Americans were isolated in the New World from Old World dogs, so they diverged and evolved new sequences, and a whole new sub-clade of sequences appeared (clade a in Figure 2). These divergent and identifiable American dogs are not represented in any modern breed surveyed so far. It is possible that these dogs were replaced by the dogs of the invading conquistadors and later colonists. However, it is also possible that representatives of the American clade of dogs survive in dogs in America which are not members of any recognized breed. Further research will contribute to see if these lineages still survive and will provide a better understanding of the relationships between American dogs and other wild canids.

References

1.  Schwartz J. A history of dogs in the early Americas. Yale University Press, New Haven, 1997.

2.  Clutton-Brock, J. A natural history of domesticated mammals. 2nd edition. Cambridge University Press, Cambridge, 1999.

3.  Sablin M.V. and G.A. Khlopachev. The earliest Ice Age dogs: evidence from Eliseevichi. Current Anthropology 2002; 43:795-799.

4.  Elias S. A, S. K. Short, C. H. Nelson and Birks H. H. Life and times of the Bering land bridge. Nature 1996; 382:60-63.

5.  Wayne R. K., J. A. Leonard and A. Cooper. Full of sound and fury: the recent history of ancient DNA Annual Review of Ecology and Systematics1999; 30:457-477.

6.  Vilà C., P. Savolainen, J. E. Maldonado, I. R. Amorim, J. E. Rice, R. L. Honeycutt, K. A. Crandall, J. Lundeberg and R. K. Wayne. Multiple and ancient origins of the domestic dog. Science 1997; 276:1687-1689.

7.  Cordy-Collins, A. An unshaggy dog history. Natural History 1994; 2:34-40,.

8.  Valadez R. El perro Mexicano. Instituto de Investigaciones Antropológicas, Universidad Nacional Autónoma de México, 1995.

9.  Vilà C., J. Maldonado and R. K. Wayne. Phylogenetic relationships, evolution, and genetic diversity of the domestic dog. Journal of Heredity 1999; 90:71-77.

10. Leonard J. A., R. K. Wayne, J. Wheeler, R. Valadez, S. Guillén and C. Vilà. Ancient DNA evidence for Old World origin of New World dogs. Science 2002; 298:1613-1616.

11. Sharma D. K., J. E. Maldonado, Y. V. Jhala and Fleischer R. C. Ancient wolf lineages in India. Proceedings of the Royal Society of London, Series B. 271 Suppl. 3, S1-S4, 2004.

12. Valadez R., B. Rodríguez, R. Cabrera, G. Cowgill and S. Sugiyama. Híbridos de lobos y perros (tercer acto): hallazgos en la pirámide de Quetzalcoatl de la antigua ciudad de Teotihuacan. AMMVEPE 2002; 13(5-6):165-231.

13. Koop B. F., M. Burbidge, A. Byun, U. Rink and Crockford S. J. Ancient DNA evidence of a separate origin for North American indigenous dogs. In: Dogs Through Time: An Archaeological Perspective, Crockford S. J. (Editores). British Archaeological Reports, Oxford, 2000:271-285.

14. Adams J. R., J. A. Leonard and L. P. Waits. Widespread occurrence of a domestic dog mitochondrial DNA haplotype in southeastern US coyotes. Molecular Ecology 2003; 2:541-546.