A Summary of Theories Concerning the Harlequin Variant in the Great Dane

by JP Yousha

This article presents some theories proposed to explain harlequin coloration. A glossary and references are provided at the end of the article. A short summary of the loci considered to be involved in producing harlequins precedes the discussion on proposed theories. For a more in depth understanding of the genes involved in coat color, the most complete reference is still C.C. Little's The Inheritance of Coat Color in Dogs. (see references).

This compilation does not claim to be definitive, nor does the author claim authority on harlequin breeding. Scientific jargon is consciously kept to a minimum. It is offered to bring together several sources of theory on the harlequin variant and in the hopes that it may stimulate further discussion on the question of harlequin coloration in the Great Dane.

The M locus has two proposed alleles, M and m (Little, 1957). Most breeds are homozygous mm, and neither merle nor harlequin coloration is produced. Dachshunds and some collie and shepherd breeds, as well as the Great Dane and the Catahoula Leopard Dog can carry the dominant M allele. In all other documented extant breeds besides the Great Dane, the heterozygote (Mm) produces merle coloration. The homozygote (MM) is a white or near white dog, with the exception of some SS dachshunds (ibid). Ear, eye, urogenital and gastrointestinal abnormalities are reported in MM animals (et al, Willis, 1989). There are reports of eye defects in Mm heterozygote Dachshunds (Klinckmanm, 1986), which has prompted the FCI to consider a ban on breeding dogs carrying the M dominant allele (Willis, 1989). The Great Dane, in addition to the merle pattern, produces a harlequin variant, assumed to require the action of the M locus. Mm dogs are either merles or harlequins, and MM dogs are white or near white in coloration and may have eye, ear and/or reproductive abnormalities (Little, 1957). In the Great Dane, harls and merles are presumed AsAs, although the ay allele exists in Great Danes and some harls and merles may be Asay. In other breeds described, merle coloration is produced in the atat (tan-point) homozygote (ibid). In the ayat (shaded sable) dog, or ayay (fawn/sable) dog, Mm heterozygote phenotype is generally indistinguishable from mm homozygote phenotype (ibid). Whether the seemingly exclusive conjunction of the As and M alleles in the Great Dane has any bearing on the production on the harlequin variant is unknown.

The S series contains four alleles: S si sp and sw; and produces recessive white spotting in a somewhat predictive pattern beginning at the extremities and extending to the torso and head. S is presumed dominant over si; SS and Ssi animals have no more than 10% white, confined to the feet, chest, and/or belly. Ssp and Ssw dogs may appear as "pseudo-irish" in patterning (ibid); North American Dane breeders refer to this pattern as "Boston". Si produces an increase in white which may extend to a blaze and collar and the tail tip, as well as to the hocks and pasterns, belly and chest. Si is presumed dominant to sp; sisi and sisp dogs appear as "mismarked" or "Boston" in pattern due to incomplete dominance. This pattern has been standardized in such breeds as the Basenji and the Collie, and this allele tends to symmetry in its markings (ibid). Sisw may appear, phenotypically, as piebald due to incomplete dominance. Sp is the piebald spotting allele and produces a wide variation in the amount and location of white, which may also vary by breed. Spsp and spsw dogs may demonstrate such a wide variation in coat color that they appear, phenotypically, as SS, Ssi, sisi, sisp or even swsw dogs. The sp allele is also assumed to demonstrate more asymmetry in patterning than the si allele. The variation in Beagle coloration is the usual example given (ibid). Sw is "extreme white piebald" and swsw dogs are represented in white or near-white breeds such as the Great Pyrenees and Bichon Frise. Color, if present, is confined to the tail root and head (ibid).

The variation in phenotype seen is conjectured to result from plus and minus modifier genes, of undetermined number, which result in a wide variety of spotting patterns in dogs (ibid). Not all breeds may have the same type and number of modifiers, so different breeds may standardize in varying phenotypes off each allele (ibid). Other loci offered as contributing to the harlequin variant are discussed with the relevant theory below.

Little (1957) proposed harlequin and merle were produced as a result of interaction between the M and S loci. Most harlequins are Mmspsp, as are many merles. Some more heavily marked harlequins and merles may be MmSsp. Merles with very little white may be MmSS. Expected litter ratios in harl to harl (or merle x merle) breedings would be 1/4 black (mmspsp), 1/2 merle and/or harl (Mmspsp) and 1/4 white (MMspsp). This theory assumes, although a phenotypic difference exists between harls and merles, they are genotypically identical and it would therefore be impossible to increase the percentage of harls over merles in a litter. In fact Little himself says the harlequin breeder is "working against himself" in the attempt to produce harlequins "by seeking the simultaneous presence of two mutually incompatible characteristics" (ibid).

Burns and Fraser (1966) proposed harlequins are EEMm and merles either EeMm or EebrMm. This would not account for the regularity of merles born to harl x harl breedings. (i.e. EE x EE could not produce Ee or Eebr). Ironically, such a theory would allow for the regular production of harls from merle x merle breedings (i.e. Ee x Ee, etc. can produce EE).

Some authors have suggested more that one merle gene or allele exists (Schaible, 1976; Robinson, 1982; Sponenberg, 1985). Harlequin coloration is produced by the action of this second gene or allele. Sponenberg (1985) suggested Mhm produces harls, while Mm produces merles. He proposes MhMh homozygosity is embryonically lethal, as it is in approximately 50% of MhM heterozygotes conceived. Again this would not account for the regular appearance of merles in harl x harl breedings. Litter number would be statistically reduced by 1/4, and a 2:1 ratio of harls to blacks would be born (i.e. Mhm x Mhm results in 1MhMh: 2Mhm: 1mm). Sponenberg, however, suggest the M locus demonstrates a high degree of autosomal mutation to account for merles born to harl x harl breedings. It is not anecdotally reported that historically harlequin litters are consistently smaller that other Great Dane litters. It is further difficult to reconcile such a theory with the regularly reported statistical ratios of 1/4 black, 1/4 harl, 1/4 merle and 1/4 white born to harl x harl breedings. Such a high rate of mutation would, indeed, be unusual in canine coat color genetics.

Bagala (1966) proposed a separate W locus to account for harlequin coloration. MmWw clears the mid-ground (grey) patches to produce harls; Mmww results in merles. WW is lethal in embryo, whether as MMWW, MmWW or mmWW. Assumably all other breeds beside the Great Dane lack the W locus or are ww. Again in harl x harl and harl x black breedings commonly practiced, a 1/4 reduction in litter size is to be expected. This theory, as does Little's, generally accounts for the commonly reported ratios seen in harlequin breedings and the regular appearance of harl and merle siblings. However, it would not account for reports of harlequins born to merle x merle breedings. Jane Chopson (1992) has thoroughly diagrammed the expected ratios for all possible harlequin family color combinations by this theory, and notes that although a conception rate for harls, under this theory, is generally 1/4, the percentage of potential lethals, merles and blacks conceived can be altered by breeding choices.

Other regularly observed phenomena occur in heterogeneous Mm animals which may have some bearing on the action of the genes involved:

· Dense Pigment Tends to Increase Over Time: "Butterfly" noses regularly "fill in" to become black or near black at maturity. An increase in the percentage of pigmented hairs is often observed, particularly over areas of pigmented skin.

· Ratio of Pigmented Skin to Hair: There is a tendency for a larger percentage of pigmented skin than pigmented hair to be present, as with the "halo" effect of areas around patches where white hairs overlie pigmented skin. Ventral areas often will have "freckles" of pigmented skin also overlain with white hairs.

· Merle patches: If one observes closely the hairs that appear in merle patches, it may be noted that not all hair are dense (black) pigmented or non-pigmented (white). Varying reduced amounts of pigment are observable in the hairs of these "mouse-grey" patches, and the distribution of pigment in the hair shaft appears somewhat similar to Green's description of dun coloration in horses (Green, 1974).

A possible theoretical explanation for these phenomena is early completion of the action of the M allele and/or incomplete dominance of the M allele over the m allele in the heterozygote. In other words, the "merle gene" is a misnomer: MM produces a reduction in pigment with white dogs resulting. The heterozygote Mm demonstrates variation in density of pigment from white through greys to black due to incomplete dominance of M over m. The Mm dog is an "incompletely white" dog and the pattern action of the M allele results in the appearance of torn patches. As pigmentation in harls and merles is seen to alter over time, instability of coloration may be predicated for the heterozygote Mm dog, and may be the result of incomplete dominance at the M locus.

An alternate explanation for the appearance of harlequin and merle littermates could be predicated on modifier genes such as those seen at the S locus. If modifiers do occur, which act when the M allele is present, this could account for both the regular appearance of harl and merle siblings and for the quantitative variations of dense (black) and mid-range (grey) pigment areas not only seen in harlequins but also in merles. For an example, negative modifiers would act to clear mid-range pigment, and perhaps some dense pigment areas, to white. Dogs with a high ratio of negative to positive modifiers would appear as "clean" or "light" marked harlequins. More positive modifiers would lead to merle patches, and perhaps larger black patches on harls, and ultimately to merles. Most harls would be presumed to carry "a mix" of modifiers with a proportionally larger number of negative modifiers than their merle siblings. This "mix" would be enough to produce, if not always reproduce, harlequins. Occasional harl x harl breedings between two individuals with only negative modifiers would consistently produce harls without merle littermates. However, (genic) recombination in harl x harl breeding would generally result in harl and merle siblings, as well as a variation in the amount and distribution of spotting within the litter. Merle x merle or merle x black matings would rarely produce harls, but could if both individuals carry enough negative modifiers to combine to result in harlequin coloration. MM (whites) and mm (blacks) could carry any ratio of negative to positive modifiers. Assumably, the S locus acts independently of the M locus. The S locus restricts the pigmented areas on which the M allele and its modifiers could act. The M allele, then, in conjunction with the modifiers present, further reduces pigment; producing the range of observed harlequin coloration, as well as merles, "merlikins", etc.

If modifiers are involved in the production of harlequins, it would be theoretically possible to increase the percentage of harls over merles by observing the proportion of harls and merles born to particular breeding pairs, and even, perhaps, correlating this back to phenotype. However this may be of limited pragmatic help to harlequin breeders as breeding animals are, naturally, not selected based on coat color alone. Such a theory would also account for the varying proportions of harls and merles born in different harl x harl, harl x black, harl x merle and black x white breedings.

Many of our experienced breeders may have developed personal theories based on extensive observation. A complete understanding of harlequin and merle variants in the Great Dane would probably require a statistically large population survey with an accurate and detailed description of the phenotypes observed. It might further require experimental breedings of "off" color pairs such as merle x merle and merle x black to determine the exact effects and limits of the interacting alleles. This may not be pragmatically or ethically supportable to many breeders, however some valuable data might be collected from accidental, historical or uninformed breedings of this nature. It is hoped that further research, in conjunction with the accumulated wisdom of breeders, may someday provide a satisfactory explanation for the inheritance of the harlequin variant.

REFERENCES:
Bagala. 1966. (See Chopson, 1992)
Chopson, J. 1992. Inheritance of Great Dane Coat Color. GDCA Color Committee.
Burns, M. and Fraser, M.N. 1966. Genetics of the Dog: The basis of successful breeding. Edinburgh: Oliver & Boyd.
Little, C.C. 1957. The Inheritance of Coat Color in Dogs. New York: Howell Book House
Green, B.K. 1974. The Color of Horses. Flagstaff, AZ: Northland Press.
Klinkmann, G., Koniszewski, G. and Wegner, W. 1986. Light-microscope investigations on the retinae of dogs carrying the Merle factor. J. Vet. Med. A. 33:674-88.
Mitchell, A.L. 1935. Dominant dilution and other color factors in Collie dogs. J. Hered. 26: 424-30
Robinson, R. 1982. Genetics for dog breeders. Oxford: Pergamon Press.
Schaible, R.H. and Brumbaugh, J.A. 1976. Electron microscopy of pigment cells in variegated and nonvariegated piebald spotted dogs. Pigment Cell. 3: 191-220.
Sorsby, A. 1970. Ophthalmic Genetics. London: Butterworths.
Sorsby, A. and Davey, J.B. 1954. Ocular associations of dappling (or merling) in the coat color of dogs. 1. Clinical and genetical data. J. Gene. 52: 425-40.
Sponenberg, D.P. 1984. Germinal reversion of the merle allele in Australian shepherd dogs. J. Hered. 75:78.
Sponenberg, D.P. 1985. Inheritance of the harlequin color in Great Dane dogs. J. Hered. 76:224-5.
Sponenberg, D.P. and Lamoreux, M.L. 1985. Inheritance of tweed, a modification of merle, in Australian shepherd dogs. J. Hered. 76:303-4.
Willis, M.B. 1989. Genetics of the Dog. New York: Howell Book House.

GLOSSARY:

· Allele: An alternate form of a gene.

· Autosomal mutation: A mutation on a chromosome other than the sex chromosomes.

· Dominant: Relates to an allele of a gene which when present in a single dose, masks the presence of another. Usually depicted by uppercase letters.

· Genotype: The genetic structure of the animal.

· Heterozygote: An individual animal which carries two distinct alleles at a specific locus. A hybrid. Heterozygotes never "breed true", i.e. never consistently reproduce themselves.

· Homozygote: An individual animal which carries two of the same allele at a locus. Dominant homozygotes are "prepotent", i.e. always reproduce themselves (phenotypically). Recessive homozygotes consistently reproduce themselves when bred to other recessive homozygotes.

· Incomplete Dominance: A situation in which the heterozygote state is different from either homozygote.

· Locus (pl. Loci): Specific location of a gene on a chromosome.

· Phenotype: The physical expression of a trait.

· Recessive: Refers to an allele of a gene which must be present in duplicate in order to indicate its presence. In a single dose will be masked by a dominant allele. Usually denoted by a lower case letter; may be annotated in an allelic series by a superscript.

Return Home