These are the chronicles of two Shetland Sheepdogs and their adventures in rally-o, obedience, flyball, agility, tracking and therapy dog work.
Also including information on raw feeding, canine epilepsy, positive training, and lots and lots of Sheltie hair!

Thursday, October 9, 2008

Shetland Sheepdog Coat Color Genetics

Shelties come in a wide variety of colors, but the genetics behind it all is really quite simple!

My intent is to gear this article towards the average Sheltie owner. I will attempt to explain things so that even those that are not top of the class in genetics will be able to follow along. All information present is scientifically based and deals with the actual genes behind the traits and are 100% up to date with the current scientific research. There are many coat color sites out there that are still following the work of a researcher named C. C. Little. He published a book in 1957 that attempted to detail canine coat color genetics. His data was derived from breeding experiements, upon which he attempted to postulate loci and gene inheritance. He got some things correct, but other things were incorrect. The nomenclature that we use now in coat color genetics is based on the theories set out by Little, though since his time there have been changes to his original postulations as knowledge of gene structure, inheritance, and interaction have been revealed with current genetic research possibilites.

Basic Overview of Genetics
Shelties: It's All About Agouti

Color Varieties:
Color Headed White

Useful References

When possible, I have included photographs to demonstrate the various phenotypes. Thank you to everyone that volunteered photos of their beautiful Shelties! I have not been able to obtain photos of bi-blue Shelties, though there is one photo of a bi-blue CHW Sheltie in the Color Headed White section.

Basic Overview of Genetics

A primary understanding of genetics is essential to be able to understand coat color inheritance in Shelties. Each individual has a full set of genomic material, called the genome. That genome is present in fragments called chromosomes, and there are two copies of each chromosome. Dogs have 38 pairs of autosomes, the non-sex chromosomes, and one pair of sex chromosomes. During fertilization, the dam gives one copy of each chromosome and the sire gives another copy of each chromosome. Each chromosome is made up of many many genes, all linked together. So because every individual has two copies of each chromosome, they also have two copies of every gene.

A gene is essentially a string of small elements called nucleotides, or base pairs. There are 4 types of base pairs, generally designated as A, C, T, and G. These 4 base pairs can be arranged in an endless number of patterns or sequences to form genes. During evolution, sometimes the pattern gets messed up a little bit, a base pair might be switched with a different base pair, a base pair might be deleted from a gene, or an extra one inserted. Or whole chunks of a gene may be deleted, duplicated, or inserted. These changes, called mutations, can alter the way that the gene works.

Let's take an example:
Pretend that this is the sequence for a gene, we will call Gene W.

If a mutation event occurs at the 4th base pair, we might end up with:

Not a big change. But it might be enough to alter the way that the gene works, changing the physical outcome, or the phenotype. This mutation would be called a 4T>A SNP (single nucleotide polymorphism). Meaning that the 4th base pair was changed from a T to an A. Each variant, if it produces a change in phenotype, is called an allele. Let's also pretend, for example's sake, that the first sequence with the T allele allows for a functional gene. But when the A allele is present, the gene is no longer functional. Therefore, a dog that had two copies of the A allele (A/A) would not have a Gene W that functioned at all. A dog with one copy of the A allele and one copy of the T allele (T/A) would have half of the copies of Gene W functional, and the other half non-functional. A dog with both copies of the gene with the T allele (T/T) would have all copies of that gene as functional.

Depending on what the mutation does to the gene, the particular trait (phenotype) produced may be expressed preferentially over the trait produced by the allele without the mutation. If only one copy of the allele is required to produce the altered phenotype, then the allele is said to be dominant. If two copies of the mutant allele are required to produce the altered phenotype, then the allele is said to be recessive.

When a gene is found to influence coat color, it is given a locus name. These locus names are modeled after the work of C. C. Little (1957), usually designated with a capital letter of the alphabet. Each mutation in that gene that causes a different phenotype is then given an allele name, usually a combination of letters including the letter of the locus name. The loci known to affect coat color in dogs include:
A (Agouti)
B (Brown)
D (Dilute)
E (Extension)
M (Merle)
S (Spotting)

The genes for all of those loci have been discovered and characterized. Meaning that we know what the gene is, and what mutations give rise to the various alleles.

There are other loci that are postulated based on observed phenotypes. But since coat color genetics is a relatively young science, not all genes have been discovered yet. The postulated loci include:
C (Chinchilla)
G (Progressive Greying)
I (Intense)
R (Roan)
T (Ticking)

Within each loci, there can be a number of alleles. Sometimes, a locus only has one mutation, in which case it has two alleles, the wildtype (usually designated with a capital letter) and the mutant form (usually designated with a lower case letter). However, sometimes there is more than one possible mutation within a gene, creating more than two possible alleles. In this case, there is generally a "dominance hierarchy" within the locus. This means that each allele is progressively more dominant, with the most dominant allele being expressed preferentially over all the others, and the least dominant allele needing two copies in order to be expressed.

Shelties: It's All About Agouti

While there are numerous loci that affect coat color in dogs, not all mutations at each loci are present in all breeds of dogs. With Shelties, their coat color is primarily caused by mutations at the A or Agouti locus, and the M or Merle locus. The S or Spotting locus also comes into play in some cases. The A locus is encoded by the Agouti Signaling Protein gene and the M locus is encoded by the PMEL17 gene.

The Agouti locus has 4 possible alleles.
ay produces fawn or sable
aw produces wildtype banded hairs
at produces black and tan coloration (think Doberman Pinscher or Rottweiler)
a produces recessive black.

The dominance heirarchy is ay > aw > at > a. Meaning that ay (fawn/sable) is the most dominant allele. aw is next, followed by at and finally a. In Shelties, only the ay, at, and a alleles are present. aw is not seen in Shelties.

Color Varieties in Shelties


Sable is caused by the ay allele at the A locus. Since it is the most dominant allele at the A locus, it can be present as ay/ay, ay/at, or ay/a. ay/at is said to be "tri-factored", as the dog "carries" tricolor. Meaning that it will pass on the tricolor (at) allele to approximately half of its offspring. ay/a is said to be "bi-factored", as the dog carries the bi-black allele.

ay allows for expression of red/yellow pigment (phaeomelanin) as well as black/brown pigment (eumelanin). The pattern that is shown in with the ay allele is a "tipped" hair. This means that each hair has a red base with a black tip. The size of the black tip can vary greatly, from barely noticeable to significant amounts of black. There is a common theory that tri-factored or bi-factored sable dogs will have larger black tips to to the hairs than pure for sable (homozygous ay) dogs. While this theory is popular among breeders, there is no scientific evidence yet to date that shows that this is completely true.

The ay allele allows for great variation among dogs, and also presents a changing phenotype as an individual dog ages. Since each individual hair is comprised of a black tip portion and a red base portion, as the hair matures and grows in length and texture, the proportion of black to red can change. So a pup may look very dark when born, but mature to a light red color, or vice versa. The mechanism that controls this variation has not yet been determined.

Click on the photos for larger images.

From left to right: Layla owned by Chris; Guiness owned by Dianna; Callie owned by D. Drass

From left to right: Rockstar owned by
Lauri Veneri
; Gio owned by Dayna; Karma owned by Bob Burns

From left to right: Brandy owned by Lauri Veneri; Krystal owned by Bob Burns


Tri-color is produced by the at allele. It is actually "black and tan", like that seen in Rottweilers, Doberman Pinschers, etc. The addition of the white spotting of the Sheltie makes it three, or tri, colors. Hence the term tri-color. The at allele is the 3rd dominant in the Agouti hierarchy, but since Shelties do not have the aw allele (the second dominant allele), it is the second dominant phenotype in the breed. As such, a tri-color Sheltie can either be at/at (homozygous) or at/a. at/a dogs carry the bi-black allele, so will pass on the allele for bi-black to approximately half of their pups.

Often there is confusion between tri-colored Shelties and Shelties that are a very dark sable, often termed Mahogany sable. People assume that because there are technically three colors in a Mahogany sable, that it counts as a tri-color. It does not. The clue is in the individual hair. A tri-colored dog has SOLID colored hairs, either red or black. A sable dog has both red and black on each individual hair. So if you look at the hair of a Mahogany sable dog, it will have a large black tip and a red base. If you look at the individual hairs of a black and tan dog, they will only have one color on them, either red or black.

The pattern that black and tan follows is always the same. A black body, with red on the lower legs, cheeks, "eye brows", and under the tail. With dogs like Shelties that have white spotting, sometimes the red areas are covered by white spots, so are not visible.

Click on the photos for larger images.

From left to right: Romeo owned by Dayna; Wheezer owned by Bob Burns; Ch Karayshel Holly Berry ADC; Telly owned by Lauri Veneri


The final allele of Agouti, the a allele or recessive black, results in the bi-black color pattern. Since the a allele is the lowest allele on the Agouti hierarchy, a bi-black dog must be a/a. There is no Agouti allele that is recessive to a.

The recessive black allele is not present in all breeds. It is primarily present in herding breeds, like Shelties, Australian Shepherds, Border Collies, German Shepherd Dogs, etc. It produces a uniform black color over the whole body of the dog. In Shelties, this is topped off with white spotting, producing a two (bi) color pattern.

Click on image to view larger.

From left to right: Morgan; Frodo


The gene that produces the merle phenotype is PMEL17. It is a gene that encodes for a protein that assists in "pushing" pigment granules into the hair shaft. When the protein is not functional, the pigment granules are not packed into the hair shaft at the same concentration as in normal hairs. To the human eye, this makes the hair appear dilute in color.

The inheritance pattern for merle is called co-dominant. This means that, in order to express the merle phenotype, the dog must have one dominant allele and one recessive allele, M/m. When a dog is m/m, it is solid colored. For example, your average tri-color, sable, or bi-black Sheltie. When the dog is M/m, there is merle on top of the base color. For instance, a tri-color dog becomes blue merle, a sable dog becomes sable merle, and a bi-black dog becomes bi-blue. When a dog is M/M, with two dominant alleles, they are termed Double Merle. Double Merle dogs often have other developmental problems, including but not limited to vision impairment and deafness. As such, no responsible breeder would ever breed a merle (M/m) dog to a merle (M/m) dog, as 25% of the resulting pups will be Double Merle.

Since merle is a co-dominant trait, it means that half of the PMEL17 proteins are functional and half are not. How those functioning PMEL17 proteins are dispersed across the body is completely random. As such, you get the very sporadic and variable patterns associated with the merle phenotype. Alternating patches of solid pigmented and dilute pigmented hairs appear randomly across the body.

Click on images to view larger.

Blue Merle

From left to right: Miley owned by Dianna; Karosel After Party; Windy owned by Bob Burns

Sable Merle

From left to right: MACH2 Aslan RAE (Photo by Sirius Photography); Cassbar's Eat My Dust FM, HIC,TN-O,NJC owned by Dale and Pat Burroughs


From left to right: Andy Blue Anderson owned by Robin Anderson; Amberlyn's Magical Merlin FMX,NAC,OJC,NGC,CGC,Therapy Dog owned by Dale and Pat Burroughs

Color Headed White

Color Headed White (CHW) is a spotting phenotype that can be applied on top of any of the other phenotypes already discussed. When thinking of spotting in dogs, remember that there are only white spots. There are no such things as "black spots" or "red spots", etc. Spots are always white. Picture a solid black dog, if you take a large white bedsheet, cut it into strips, and toss it over the dog's body, you will end up with a black and white dog. The white spots cover the solid color underneath. If the white spot were not there, then that area of the body would be the color dictated by its base phenotype, in the case of Shelties dictated by the A locus.

In the past, spotting phenotypes have been associated with one locus, the S locus. Within that S locus, the postulated dominance hierarchy flowed S > si > sp > sw. With S being solid colored, si being Irish spotted (like Shelties), sp being piebald spotted, and sw being extreme white. Current research raises a question that indicates that the postulated hierarchy may not be completely accurate. Though the true answer has yet to be determined.

In Shelties, you primarily have two spotting patterns to be aware of. The majority of Shelties exhibit the Irish spotting pattern. With a white collar, bib, paws, and tail tip, with or without a white blaze on the face. The CHW phenotype is an example of piebald spotting. Where the majority of the body is white, usually with one or two pigmented areas on the body, and a pigmented head. The pigment "underneath" the white spot can be any of the regular color phenotypes discussed previously. CHW is recessive to Irish spotting, so a dog must have two sp alleles (sp/sp) to exhibit the CHW phenotype.

Click in image to view larger.

From left to right: Smoke (bi-blue CHW) owned by Julie; Deepfork's Battle of Jericho RE, OA, NAJ, CD (Sable Merle CHW) (Photo by Mickey Rabanek); Oreo (Tri-color CHW)

Useful References

Berryere, T. G., Kerns, J. A., Barsh, G. S., Schmutz, S. M. 2005. Association of an Agouti allele with fawn or sable coat color in domestic dogs. Mammalian Genome. 16:262-72.

Kerns, J. A., Newton, J., Berryere, T. G., Rubin, E. M., Cheng, J.F., Schmutz, S.M., Barsh, G.S. 2004. Characterization of the dog Agouti gene and a nonagoutimutation in German Shepherd Dogs. Mammalian Genome. 15:798-808

Karlsson, E. K., Baranowska, I., Wade, C. M., Salmon Hillbertz, N. H., Zody, M. C., Anderson, N., Biagi, T. M., Patterson, N., Pielberg, G. R., Kulbokas, E. J. 3rd, Comstock, K. E., Keller, E. T., Mesirov, J. P., von Euler, H., Kämpe, O., Hedhammar, A., Lander, E. S., Andersson, G., Andersson, L., Lindblad-Toh, K. 2007. Efficient mapping of mendelian traits in dogs through genome-wide association.
Nature Genetics. 39:1321-8.

Schmutz, S. M., Berryere, T. G. 2007. Genes affecting coat colour and pattern in domestic dogs: a review. Animal Genetics. 38:539-49.


the Corgi Girls said...

You're sucha nerd, I love you! You're awesome, to the gal who explained 'cryptic' merle to me! YOU ROCK!!!

M & I

Bree/Reilly said...

Hey......Oreo looks just like an 'old me'!!!!! I wonder why it is that the majority of us color headed whites all seem to look the same, our patches are in the same place, they even look the same shape and size.

Reilly - anti word verification unit

Copyright & Link Restriction

Please do not use any text or images from this blog without expressed permission of the blog owner. If you are a organization, group, community, commercial site, are acting on behalf of a commercial site (sponsored by them), or are trying to build an audience for any site with the intent of financial gain, YOU MUST obtain permission from the blog owner BEFORE reproducing or copying any material from this site, or embedding this blogs feed into your own site.