Genetic variation in the Australian Dingo.


Alan Wilton
School of Biocemistry and Molecular Genetics
University of New South Wales

The dingo is in danger of extinction in the wild.
   Hybridisation with domestic dogs is common. Hybrids are difficult to physically distinguish from pure dingoes. Conservation groups have begun breeding programs to preserve the dingo. The purity of their breeding stock is unknown. The DNA tests developed in this project will be used to identify pure dingoes. This will allow zoos and wildlife parks to ensure the survival of Australia's native dog. This is the main thrust of this proposal. Two other consequences of the work follow.

Effects of hybridisation:

There has been little research on the effect of hybridisation of closely related groups on the ability of the population to survive or the effect on its place in the local ecology. The best studied case is perhaps the red wolf which was a target for a concerted conservation effort until it was found to be a hybrid between the gray wolf and the coyote (Lehman et al, 1991). Hybridisation may cause genetic change in the population and, although there is strong strong social pressure to avoid it because of the damage it may cause, there are no long term studies on whether it would actually have an adverse effect. The dingo and the dog are quite different in behaviour and patterns of reproduction and it is possible that changes in hunting behaviour of hybrids and increase in the numbers of litters per year would adversely affect populations of other native species. However, whether hybridisation would have a long term effect on fertility or the survival of a wild canine population is not known. The production of markers to differentiate dog from dingo genetic material is a first step in a long term study of the effects of hybridisation. Without the ability to determine the extent of hybridisation such a study would not be possible. We are not proposing to do such a study here but only develop the tools to make it possible.

Dogs are useful vehicles for the study of genetic disease and thus gene function. Gene mapping in dogs to identify disease genes is in its infancy relative to many species. Maps of highly polymorphic mcrosatellite loci are under construction (Lingaas et al, 1997) but these contain no information on functional genes. The location of blocks of genes on these maps would make these more useful for disease gene identification and gene cloning. The high conservation of syntenic groups of genes among mammals makes this possible by localising a small number of genes. However polymorphisms are required to place genes on the map. It is often difficult to find variation in a functional gene within a species but crosses between related species that produce fertile offspring are often extremely useful because accumulation of genetic differences often produces fixed differences at large numbers of loci. Hybrids will be heterozygous at these loci and data from offspring out of backcrosses to either parent can be used to map the functional genes. This has been successfully applied to mapping genes in tammar wallaby (McKenzie et al, 1993). We intend to identify markers useful for localising functional genes by examining differences between dingoes and dogs at intronic sequences of conserved genes.

Dingoes are a type of Asian dog which is possibly derived from the Indian or Arabian wolf by domestication not more than 10,000 year before present (Corbett, 1995, Ch1). They have been spread throughout South-East Asia and the Pacific by man. They first arrived in Australia less than 5,000 BP.
Although they have many distinct physical and behavioural characteristics that differentiate them from domestic dogs, such as an annual breeding cycle, the two species interbreed and produce fertile offspring as do many wolf-like species.
Western influences have lead to the introduction of domestic dogs throughout the dingoes range. The proportion of pure dingoes in the wild canid populations is steadily decreasing with large proportions (>80%) known only in Thailand and Australia (Corbett, 1995, Ch10). Within Australia hybridisation occurs most readily in highly populated areas with populations from the East coast and South-East and South-West containing mostly hybrids. The more isolated central and Northern areas have the largest proportions of pure dingoes but with increasing numbers of domestic dogs on properties and in aboriginal camps, the dingo is under threat in even these areas.
If the dingo is to be preserved as a native animal, a conservation program needs to be undertaken. Public education is probably the best way to reduce the numbers of dogs available for hybridisation in the wild and several groups are working towards this. Captive breeding programs are the best way to ensure the long-term future of the Australian dingo and zoos, wild-life parks and dingo associations are undertaking this. However, how can the genetic purity of the breeding stock be assured? Many animals in captivity come from South-East Australia where the proportion of pure dingoes is estimated to be as little as 22% in one population with a maximum of 65% (Corbett, 1995, p166, ).

Why Conserve the dingo?

The Australian dingo (Canis lupus dingo) is recognised as a subspecies of the wolf-like canids. It is biologically distinct from dogs with differences in reproduction (dingoes breed only once a year), coat colour, and other physical and physiological characters (Corbett, Ch3). To allow the dingo to disappear due to interference from modern society would be unethical. This is recognised by the many groups who have established captive breeding programs, such as the Australian Native Dog Conservation Society and the Australian Dingo Conservation Association. In some areas of Australia dingoes are being housed with families and kept as pets. The dingo is currently a recognised dog breed and is being bought and sold.

The dingo or what appears to be dingoes will be preserved by these groups. With DNA testing we would be able to ensure that it is the true dingo and not a dog hybrid population.
Is there only one type of dingo to conserve or are there several genetically different races of the dingo that should be conserved separately? Physical differences between southern and northern dingoes suggest that there may be more than one. Through DNA testing we will be able to determine if there is a sufficient genetic differences to consider the populations from different regions as races.

Current Methods of Assessing Dingo Purity

Presently the purity of dingoes ancestry is assessed on the basis of skull measurements or physical appearance (Newsome et al, 1980, Newsome and Corbett, 1982, 1985). According to Corbett (1995, p173) external body characteristics are unreliable for classification even when applied by the most experienced dingo experts. Skull measurements are reliable for distinguishing pure dingoes from pure dogs but it may take many years before a wild caught animal used for breeding dies and skull measurements can be taken to assess its purity. Alternative methods of taking measurements such as X-rays or CAT scans of the skull are possible but impractical. Also skull measurements are less useful for detecting backcross dingoes, eg 3/4 dingoes etc.

Alternative Methods of Assessing Dingo Purity

A number of diagnostic DNA markers can be used to assess the level of purity of a dingo. The larger the number of markers the smaller the level of introgression of dog genes that can be detected.
Development of diagnostic DNA markers will allow the maintenance of the dingo as a distinct subspecies. Attempts to use isozyme markers for this purpose were not successful (Cole et al, 1977). It is mainly the recent availability of microsatellite markers in dogs that make it practical now.
Genetic Markers.   Highly variable markers are required so that the chances of finding differences between dingoes and dogs is greatest. Microsatellites or simple sequence repeats are highly variable. There are a large number of microsatellite markers available for the dog (300-400) (Ostrander et al, 1993, Holmes et al, 1993). Dog microsatellites have been successfully used to analyse population structure of other canids such as wolves and coyotes (Roy et al, 1994). We have tested 10 microsatellites in the dingo and all amplify fragments of similar size to the dog and all are polymorphic. Four have been tested on 16 crossbred dogs and 16 dingoes which are not suspected of any dog ancestry. All microsatellites show different distributions of alleles in the two groups with the variation in the dingoes noticeably lower. For one locus there is no allele that is common to both groups. For this dinucleotide repeat dingoes have odd numbered allele sizes while dog alleles are of even number sizes. If this difference holds up with a larger sample size the locus would make a good diagnostic marker for introgression of dog genes. Since only 4 loci have been tested and one looks promising as a diagnostic marker it suggests that many other microsatellite loci will be too.
Microsatellites loci with overlapping distributions of allele sizes in dingoes and dogs can also be used to determine whether an animal is more likely to be a dog, dingo or hybrid. The fewer the alleles shared, the more useful the marker.
Variation in mitochondria   has been useful in the study of closely related species including canids (Lehman and Wayne, 1991, Lehman et al, 1991). There are likely to be differences in the mitochondrial DNA sequences of dogs and dingoes that can be exploited. Because of their maternal inheritance, mitochondrial markers are useful in determining the types of matings that occur to create hybrids.
Sequence variation in introns.   There is likely to be little sequence differences between dogs and dingoes for coding regions of most genes. However sequences not under stringent selection such as many intron sequences may provide a source of variation in which drift has lead to fixation of different sequences in dogs and dingoes. The Conserved Amplifiable Tag Sequences (CATS markers) developed in Steve O'Brien's lab are a series of primers designed to the conserved sequences of genes that will amplify across introns in most mammalian species. Primers are available for a large number of loci and could be used to search for diagnostic differences between dog and dingo genetic material.

Research Plan

1.    Type a number of dingoes and dogs for highly variable markers to develop several markers which are diagnostic for dog genetic material and for dingo genetic material.

Proposed marker loci to be tested

    - 50 microsatellite markers will be used to type 20 dogs and 20 unrelated dingoes. They should provide 10 diagnostic loci for dog versus dingo genetic material at the current rate of detection of informative markers. (A number of loci are needed to detect dog genes in a dingo background arising from a hybridisation event several generations ago since the chance the dog allele is passed on in each generation is 0.5. We would like to get at least one diagnostic marker per chromosome if possible.)
    - mitochondrial D-loop variation detected by sequencing products from dogs and dingoes.
    - If there are insufficient diagnostic markers among the 50 microsatellite loci, intron sequences for 20 genes will be examined for sequence differences between dingoes and dogs before more microsatellites are tested.
2.     Since uncertainty surrounds the purity of any modern dingo, loci that have only a small overlap in allele sizes will be tested on DNA extracted from preserved specimens such as skins collected in the 1800s or early 1900s. At least 20 samples will be tested, more if they are available.
3.    We are confident that sufficient markers will be found from the above. However, if expectations are not met and more markers are needed they will be developed. The Representational Difference Analysis (RDA) technique is specifically designed to amplify differences between two pools of DNA by subtracting any common sequences. Pools of dog and pools of dingo genomic DNA, which has been digested with a specific restriction enzyme and subjected to RDA, will provide DNA fragments containing restriction site differences between the species. PCR primers will be designed to the sequences flanking the restriction site so it can be amplified and tested in dogs and dingoes.
4.    Type all presumed dingoes for the diagnostic loci to determine which are pure dingoes and which hybrids. Compare levels of hybrid occurrence in different regions of the country.
5.    Type a large number of pure dingoes from each of several populations from different regions of the country for 20 microsatellite loci. Compare the allele frequencies at these loci to determine to determine whether differences exist that suggest the populations genetically distinct. Fifty animals from each of 3 regions including alpine, desert and tropical will be typed. A small number of New Guinea Singing Dogs (NGSD) which are closed related to the Australian dingo will also be typed for comparison.
6.    Type Australian cattle dogs (which have dingo ancestry in the breed) to determine whether the dingo ancestry can be detected after many generations.


The easiest samples to collect are those from animals in captivity. However, the purity of these and dingoes sampled in the wild is unknown. The animals in captivity have been classed as dingoes on the basis of external physical characteristics and, for many that were born in captivity, their status has been supported by the classification of their parents as pure dingoes based on skull measurements. Preliminary microsatellite data on 16 of the captive dingoes and 16 dogs shows a much lower heterozygosity in dingoes and fewer alleles at each locus. Gene frequencies are very different with many new alleles in the dingoes as well. At the most discriminating locus there was one allele in one individual that was the same size as the alleles found in the dog sample. This suggests that at least the majority if not all of this group of captive dingoes are pure bred.
Pre- or early settlement base Although data collected to date does not support it, the possibility exists that many living dingoes in captivity are not pure bred. To determine the genotypes at the test loci in the dingo before the introduction of dogs we plan to extract DNA from preserved specimens for typing. DNA can be extracted from small amounts of skin from tanned hides or mounted specimens. A large collection to which we hope to gain access exists at the British Museum and other museums will be approached.
Source of Contemporary Samples. We do not intend to trap and bleed or kill any wild dingoes specifically for the population comparisons. We will use captive dingoes of known origin and tissue or blood collected from animals killed or trapped for other reasons. Blood samples will be taken by qualified veterinary staff.
Captive animals.
There are several conservation groups that have expressed strong interest in this program. David Steward from Australian Native Dog Conservation Society is doing his MSc on the project. He cares for a colony of about 40 dingoes at the Merigal Dingo Education Centre at Bargo. The Australian Dingo Conservation Association is also extremely interested in the project. They have already supplied 28 samples. They have access to a large number of animals that have been fostered out to families as pets and will arrange blood samples to be collected from these animals as well as provide some financial support for the project. We have samples from Northern Territory Wildlife Park (6), offers of samples from Taronga Zoo and Healesville Sanctuary, and expressions of interest from Perth Zoo and Western Plains Zoo. All are keen to determine the status of the animals in their care. We will be contacting other zoos to request specimens. We currently hold tissue or blood samples from 36 dingoes at UNSW and 16 at VIAS with Associate Investigator Nick Robinson. About 40 animals are waiting to be bled at Merigal, Taronga and Healesville.
Wild animals.
We also intend to obtain DNA from preserved specimens. Barry Oakman of Australian Dingo Conservation Association has access to a large collection of tanned dingo skins collected by trappers that we will extract DNA from. We have 20 skin samples in hand. Through contacts on various Rural Land Protection programs which are culling dingoes, he is also arranging for samples of ear tissue to be collected from new kills. David Jenkins of Hydatid Control and Epidemiology Program has offered his a collection of formaldehyde preserved tissue from dogs collected in the wild. We also intend to obtain DNA from preserved specimens in museums. Besides Australian Museums, the British Museum has a large collection and we will approach them for samples when funding is available. DNA can also be extracted from bones and teeth and if a source of these is located we will collaborate with anthropologists like Lisa Matisoo-Smith, Univ Auckland, who is working on Pacific Island dingoes from bone remains.
New Guinea Singing Dogs are the New Guinea equivalent of the dingo and are closely related to it. We currently have a single sample from an animal originally at Taronga Zoo. It is indistinguishable from the Australian dingo samples at the microsatellites analysed to date.
Domestic dog samples.
From our work on dog genetic diseases we have over 200 samples from border collies but for comparison to dingo we need dogs of mixed breed. Medical Research Institute, Prince of Wales Hospital, has agreed to provide us with samples from the dogs of mixed breed. We are also collecting a number of samples from Australian cattle dogs through Noreen Clark of Tirtla Kennels. Cattle dogs were a breed with deliberate dingo crosses in their ancestry. Seventeen samples have been collected. Besides David Steward who is working on the project we have several veterinarians collaborating, particularly Marilyn Gill, who is prepared to collect blood samples from dogs for us if the owners give permission.


We are currently typing large numbers of dog microsatellites for the mapping of disease genes in domestic dogs. We are sizing PCR products on ABI 377 automated DNA sequencers after labelling them with fluorescent dyes by adding fluoro-dUTP to the PCR reaction and analysing results with Genotyper software. This method labels both DNA strands so the two strands are often visible running at different rates. Allele calling is possible but not as simple as with fluoro-labelled primers where only one strand is labelled. However, making fluoro-labelled primers is not cost effective for a small number of samples such as required when testing a new microsatellite (30-40 samples). Associate Investigator Nick Robinson at VIAS has approximately 15 fluoro-labelled primers used for dog parentage testing.
Sequence variation at mitochondrial D-loop or introns. Initially a small number (~3) of dogs and dingo samples will be amplified and sequenced. The sequences will be compared and if different will be analysed by Single Strand Conformation Polymorphism (SSCP) on the ABI 377. If the different sequences cannot be detected by SSCP other methods for detecting single base differences such as heteroduplex analysis or mismatch enzyme cleavage (Cotton, 1993) will be used to type other samples without sequencing.

Cole SR, Baverstock PR, Green B. Lack of genetic differentiation between domestic dogs and dingoes at a further 16 loci. Aust J Exp Biol Med Sci 22:229-232, 1977
Corbett, LK. The dingo in Australia and Asia. University of New South Wales Press Ltd, Sydney, 1995
Cotton, RGH. Current methods of mutation detection. Mutat Res 285: 125-144, 1993
Holmes NG, Mellersh CS, Humphreys SJ et al. Isolation and characterization of microsatellites from canine genome. Animal Genet 24: 289-292, 1993
Lehman NA, Eisenhawer , Hansen LD, et al. Introgression of coyote mitochondrial DNA into sympatric North American gray wolf populations. Evolution 45: 104-119, 1991
Lehman N, Clarkson P Mech LD et al. The use of DNA fingerprinting and mitochondrial DNA to study genetic relationships within and among wolf packs. Behav Ecol Sociobiol 30: 83-94, 1992
Lehman N, Wayne RK. Analysis of coyote mitochondrial DNA genotype frequencies: estimation of effective number of alleles. Genetics 128: 405-416, 1991
Lingaas, F, Soerensen, A, Dolf, G, Binns, M, et al. Towards construction of a canine linkage map. Establishment of 16 linkage groups. Mammalian Genome (in press), 1997
McKenzie, L.M., Collet, C. and Cooper, D.W. (1993). Use of a sub-species cross for efficient development of a linkage map for a marsupial mammal , the tammar wallaby (Macropus eugenii). Cytogenet. Cell Genet. 64: 264-267.
Newsome AE, Corbett LK, Carpenter. The identity of the dingo. I. Morphological discrimination of the dingo and dog skulls. Aust J Zool 28: 615-625, 1980
Newsome AE, Corbett LK. The identity of the dingo. II. Hybridisation with domestic dogs in captivity and in the wild. Aust J Zool 30: 365-374, 1982
Newsome AE, Corbett LK. The identity of the dingo. III. The incidence of hybrids and their coat colours in remote and settled regions of Australia. Aust J Zool 33: 363-375, 1985
Ostrander EA, Sprague GF, Rine J. Identification and characterization of dinucleotide repeats (CA)n markers for genetic mapping in dog. Genomics 16: 207-213, 1993
Roy MS, Geffen E, Smith D, Ostrander EA, Wayne RK. Patterns of differentiation and hybridization in North American wolflike canids, revealed by analysis of microsatellite loci. Mol Biol Evol 11: 553-570, 1994

Submitted by John Chandler
Back to Home Page
Back to Articles

Back to The Australian Dingo