Introduction And History Of Merle

There have been many questions regarding Merle since Dr. LA Clark first idenfied the mutation in 2006. Since Merle affects so many breeds I am sharing this excerpt from my book in order to make this important information more accessible for everyone in all breeds.

MERLE - a SINE insertion (Short Interspersed Element) on the SILV gene.

Think of a piece of foreign DNA (viral-like) that has "inserted" itself into the chromosomes and has changed the number of base pairs by adding a piece of DNA. This SINE insertion lives off the host much like a parasite and can change the way cells function.

Both SINEs and LINEs (Long Interspersed Element) are common in canines and all mammals, including ourselves. They can be a form of evolution, sometimes destructive and sometimes a useful incorporation into the genome. Most SINE insertions occur in areas of the genome where they cause no damage or change to the host. However, some occasionally insert in regions where they can disrupt and alter expression.

There is some thought that it is these SINE and LINE insertions in canines that have helped to "create" (with our outside influence) our different dog breeds.

In the case of the Merle mutation, a SINE insertion impairs the ability of cells to produce normal pigment - a defective or faulty design.

Another example of a SINE insertion within the canine genome is the mutation for Tan Trim on the A Locus as previously mentioned. SINE insertions can also cause diseases like Narcolepsy in Dobermans and may be responsible for some cancers.

The Merle SINE mutation consists of 3 parts - a head, a body and a tail (poly-A tail). The tail contains a long string of repeating base pairs. Think of the poly-A tail as a section of a lariat-like thread of DNA that "catches and ties" into the dog's DNA, the head then attaches itself - "bites down" - so the whole invader can work its way into the DNA.
This is not very scientific language, but gives a good visual of the process.

The original Merle test only identified the body of the Merle mutation and the assumption was made that any reasonable length of tail produced some kind of Merle pattern; the longer the tail then the more Merle. No distinction was originally made between the tail length (base pairs) and Merle phenotype.

When Merle testing was first available any length of Merle was reported as a full "M" Merle. Numerous Catahoula owners had their dogs tested and were receiving results that their solid looking dog was in fact "M/M".

How could a solid dog be a double Merle!?

The Merle test results were just not making sense at all and were considered by many people to be "flawed."

Leo - Tested as "M/M" through IDEXX: Tested as "Mc/Mc" through Biofocus
Leo - Tested as "M/M" through IDEXX: Tested as "Mc/Mc" through Biofocus

We now understand that these dogs were most likely "Mc/Mc" - or even possibly "Ma/Mc" - having 2 cryptic Merle mutations - a length of Merle so shortened or truncated that it can no longer cause a change to the coat pattern.
Eventually the test results were given using "Mc", indicating that the mutation was not a full "M" but a shortened one and genotype results seemed to make more sense with the phenotype of the dog. 

As the years have passed more Catahoula owners, especially in Europe, were having their Catahoulas tested through a lab in Germany, Biofocus.

During this time Dr. Helena Synková, was the breed advisor for the Catahoula Club EU in the Czech Republic, as well as being a Catahoula owner and breeder herself. By observation alone of litters produced by dogs that appeared at first glance to be non-Merle, Helena suspected that there was more to Merle then just "M" and "Mc". These non-Merle, solid looking dogs had a certain quality to their coat, some with even very faint Merle patterning that the eye could see, while maybe not visible in photos. These dogs were Merle testing as "Mc/m" or "M/m", but did not show the typical Merle expression and did not breed as the tests concluded.

Helena approached Biofocus with this information and working closely with Dr. Beatrix Böckmann, they started testing dogs whose phenotype did not match their genotype. By paying special attention to the length of the base pairs (the length of the poly-A tail), the findings concluded that there was also a third length of Merle - a length longer than cryptic Merle "Mc", but shorter than a true full Merle "M".

Biofocus started reporting this length on lab results but continued to identify this shortened Merle as "M" - a full Merle. Helena noted that there was a very clear phenotype when this shortened Merle was inherited with either "M" or "m".

In 2011, she wrote an article for several dog magazines in Europe and for the Catahoula Club EU in which she gave this shortened Merle length the working name of "Atypical Merle" - "Ma".

Helena's translated article can be read on the following page.

PDF file of Helena's article printed in "Svět Psů" Magazine - Jan, 2012

In May 2015, Biofocus officially recognized this shortened Merle as an allele on the M Locus and adopted the working name that Helena had assigned to it in her article - Atypical Merle - "Ma".

Since that time, 8 more breeds have tested as having the "Ma" allele - Dachshunds, Great Danes, Chihuahuas, Pugs, French Bulldogs, Koolies, Australian Shepherds and Border Collies. I have no doubt that there will be more breeds added to this list as more testing is done. Every Merle breed has the potential for M to shorten to Ma.

Merle testing by Biofocus gave us 4 separate alleles on the M Locus:

  • Full Merle "M" - a length of 459 - 462 bp
  • Atypical Merle "Ma" - a length of 443 - 449 bp
  • Cryptic Merle "Mc" - a length of 400 - 425 bp - at this point the tail has been so shortened that Merle can no longer express.
  • Non-Merle - wild type "m"- no SINE insertion length of 199 - 200 bp

Note - The M, Ma and Mc lengths given are a combination of "m" length 200 + the length of the SINE insertion.

Example "M" = length of "m" 200 + SINE length of 259 - 262 = 459 - 462.

With all 4 alleles identified on the M Locus, breeding results are finally making sense.

This is a prime example of how breeders of all breeds, who are on the frontlines of recognizing the colors and patterns from parent to offspring, have been instrumental in helping labs and researchers to develop new testing.

Hope - Tested as "Mc/Mc" through Animal Genetics: Tested through Biofocus as "Ma/Mc"
Hope - Tested as "Mc/Mc" through Animal Genetics: Tested through Biofocus as "Ma/Mc"

With the inclusion of the "Mc" allele, the differences in Hope's Merle test results from 2 different labs are not as noticeably "flawed" as the example of Leo above. 

However, if Hope was bred to an "M/m" male, she would be able to produce "M/Ma" - Patchwork offspring. This would be confusing to the breeder since she tested as "Mc/Mc" at a lab who has not yet included the "Ma" allele in their test results.

The Shortening and Lengthening of Merle

To complicate things further, the Merle SINE is unstable - the number of base pairs in the length of the Poly-A tail can shorten or lengthen from parent to offspring. In one study done in 2006, it was surmised that this can happen in 3 - 4% of cases. It is important to note that this study was done at a time when Merle testing was not yet available so these findings were based strictly on phenotype alone.

Now that Ma has been defined as its own allele on the M Locus, we may find this happens more than first thought.

Note - M, Ma and Mc cannot shorten to "m". This would require the full removal of the DNA, which cannot happen.

Through testing of Catahoulas, we have found several examples of shortening from parent to offspring.

Full Merle - M shortening to Atypical Merle - Ma from parent to offspring.

Full Merle - M shortening to Mc from parent to offspring.

It is more probable that the changes in the Poly-A tail length are minor as opposed to a major jump. We have also found examples of M not shortening completely all the way to Ma but somewhere between the two alleles due to this "minor" shortening.
In order to make it easier to identify these lengths that are between alleles I have used the symbol "+"
Ma+ for a length that is between M and Ma
Mc+ for a length that is between Ma and Mc
Examples of Merle shortening start on Page **.
Note - Every Merle dog of any breed has the potential for the Merle Poly-A tail to shorten and therefore the Ma allele.

Merle's Poly-A tail can lengthen as well. Through the testing of hundreds of Catahoulas in Europe and following of generations, it is normal to have a slight variance of 1 - 3 base pairs within an allele from parent to offspring but a complete lengthening to another allele has not yet been documented.

I have found two examples of the lengthening of Ma to Ma+.

Examples of Merle lengthening start on Page **

Cryptic Merle - Possible Lengthening

It is still unclear if Mc would be able to lengthen at all given how truncated the tail length and how stable the allele is.
As of this time, there has been no DNA documented evidence of Mc lengthening.

Dr. Clark's 2006 study Retrotransposon Insertion in SILV is Responsible for Merle Patterning of the Domestic Dog is often cited as an example of the possibility of Mc lengthening.
The following quotes are taken from this study -
"Although rare, a dog that does not present with the overt merle phenotype may possess the merle genotype and subsequently produce merle offspring. Such a dog is termed a cryptic merle. The mechanism for this phenomenon is unknown."
The use of the term "cryptic merle" here does not refer to what is now identified as Mc - Cryptic Merle, but to what we also refer to as "ghost or phantom merle" as discussed on page 53.

"Additionally, expansion of the oligo(dA)-rich tail in the germ line of a non-merle dog having the smaller insertion may result in merle offspring and may be the mechanism behind the cryptic merle phenotype. Although the data presented herein suggest that the oligo(dA)-rich tail length determines phenotype (i.e., merle, non-merle), they are not sufficient to determine the precise threshold for this phenomenon."

It is very important to note that this study was done purely on phenotype and not on DNA testing of Merle's genotype especially taking into account the different lengths of M, Ma and Mc, which was not available at that time.

In the past, this example of a breeding between two Ma/m dogs and the resulting Ma/Ma offspring could have been considered by phenotype alone to be the lengthening of Mc from one parent to M in several offspring. We understand now that this is not correct. As well most labs still report Ma as Mc, so again with a breeding like this example, even if both parents where tested at a lab other than Biofocus, this may be assumed to be Mc lengthening. Being able to accurately report on the base pair lengths of Merle's Poly-A tail and the addition of the Ma allele has shown us how resulting Merle patterned offspring can be produced by two seemingly "Non-Merle" parents. To prove the possibility of Mc lengthening, testing of both parents and any offspring would have to be done at a lab who can report the exact base pair numbers.

As of this time there has been no DNA documentation of Mc lengthening.

Note - As of the first publishing of  "Catahoula Coat Color Genetics - Painting The Canvas" in June, 2016, Dr. Clark at Clemson University had not yet herself established the difference between the length of M and Mc.

This example breeding between Kelvar and Kiva and their resulting offspring, Colt, is also an excellent example of what we used to assume regarding Merle and what accurate DNA testing can now confirm. Again, either Kevlar or Kiva might have been assumed to be a "hidden" or "ghost" Merle - having a full M allele and not having the pattern expressing on their coat.
Or either parent could have been assumed to be Mc - Cryptic Merle that lengthened to M in their offspring.

Neither of these cases is correct and DNA testing by a lab who can provide us with the exact number of base pairs, has shown us that both Kevlar and Kiva are Ma/m and their offspring Colt is Ma/Ma.

Note - Dr. Clark and her team from Clemson University released an abstract at the "9th International Conference on Canine and Feline Genetics and Genomics" in May 2017 - "From genotype to phenotype: length variations in the Merle SINE oligo(dT) yield four distinct coat patterns".

Dr. Clark has intentions of changing the allele name of "Atypical Merle" - Ma to "Dilute Merle" - Md due to the diluted coloring that Ma can sometimes give to the coat. In late 2016 when I sent Dr. Clark information regarding the new allele, I explained to her at length and sent her many examples of Ma dogs that do NOT have a diluted coat color. As well the chart above of Kevlar and Kiva both Ma/m and both a normal black coloring with no visible dilution displays this very well.
The naming of an allele falls to the person who discovers the correlation between phenotype and genotype, which in the case of M
a is Dr. Helena Synková.

Merle - The Saga Continues

Once the different lengths of the Merle alleles started to become known, a lab in the UK - Laboklin and partner lab Labogen, started testing for the different base pair lengths of M, Ma and Mc. They also included Mh - Harlequin Merle in this testing.
The Mh allele is theorized in Dr. Clark's abstract mentioned on the previous page - "From genotype to phenotype: length variations in the Merle SINE oligo(dT) yield four distinct coat patterns".

In the past Harlequin Merle has been referred as "Herding Harlequin" as it often recognized in herding breeds such as the Border Collie and Australian Shepherd. This pattern typically has the same expression as H/h - Harlequin as is present in the Great Dane breed mentioned on page **, but these breeds all test as h/h - Non-Harlequin.
It was theorized that there could be another modifier of Merle causing this pattern, as these dogs routinely tested as M/m on the M Locus and S/S - Non-Piebald. These dogs were also not M/Ma - Patchworks, which also creates white within the Merle pattern. Something else was causing this Merle expression of areas deleted to white.

Several Catahoulas already tested through Biofocus were sent to Laboklin to ensure that their base pair numbers for M, Ma and Mc were correct. As well, Catahoulas with unexplained white patterns not related to M/Ma - Patchwork were tested. These dogs were confirmed as Mh - Harlequin Merle - a length of base pairs longer then M, causing areas of diluted Merle to delete to white.

Ava - Mh/Mc
Ava - Mh/Mc
Joy - Mh/Mc, S/S
Joy - Mh/Mc, S/S

It was exciting to have a test available that could report on all 5 alleles on the M Locus -
Mh, M, Ma, Mc and m - and finally explain dogs in all breeds with the "Herding Harlequin" pattern.

However Laboklin/Labogen has declined to share the base pair numbers for Mh - Harlequin Merle, as well as the number range they are using for the M, Ma and Mc alleles, citing "legal issues of confidentiality."

Their analyzer machine is different from the one that Biofocus uses (although still the same method of testing) and the base pair numbers slightly different and allowing for the Mh allele having higher base pairs then M. This will be explained in more detail in the following section.

Mh - Harlequin Merle

Vemodia - Laboratory of Molecular Veterinary Diagnostics

In March 2016, a representative from Vemodia Labs in the Czech Republic purchased the first release of "Catahoula Coat Color Genetics"and expressed great interest in developing their Merle test for the different alleles.
In early fall 2016, RNDr. Tereza Jančušková, Ph.D. and MUDr. Soňa Peková, PhD. contacted Dr. Helena Synková to arrange for the testing of 20 Catahoulas with already known genotypes.

Vemodia's data collecting method uses "Fluorescent Fragmentation Analysis" on an "ABI (Applied Biosystems) 3500 Capillary Sequencer". This genetic analyzer automatically counts the base pairs and produces a "chromatogram" - a "picture" showing the peaks of the base pairs and where they fall within the M Locus alleles.

Compared to Biofocus' analyzer, a Capillary Sequencer 310, the base pair numbers for the alleles are slightly smaller (tighter together on the number scale), resulting in higher base pairs numbers for each allele.The difference in numbers from both labs could be explained by several reasons -

Vemodia's analyzer is a newer and more advanced machine.
As well the kits of chemicals and enzymes used for the testing process are produced by many suppliers worldwide, each giving a possibly slighter difference.

How Merle Testing Is Done
DNA is isolated from samples - either blood, buccal swab or in some cases sperm. This sample is then processed using a technique called PCR - "Polymerase Chain Reaction" - this procedure amplifies (expands) a small segment of DNA into thousands or millions of copies with the identical sequence so that the sample can be studied in more detail. This is an extremely challenging process when Merle's long monotonous Poly-A tail is involved. This amplification is an artificial method; (unlike the DNA replication in living cells) thus the enzymes used for the in vitro (in a tube) amplification can make more errors then the in vivo system (the living cell) does. The conditions and quality of products used for this process must be of the highest quality to give optimal results.
A fluorescent dye is incorporated into one of the amplification primers which, in turn, "stains" the resulting PCR amplicon. This is referred to as Fluorescent "Labeling" or "Tagging". The obtained fluorescently labelled PCR products are then separated in a high-resolution polyacrylamide gel in a capillary sequencer, according to their size - the base pair numbers that can be read in the resulting chromatogram.

Think of the M Locus as a sliding scale starting from Mc to Mh.
When comparing the base pair numbers from Biofocus vs Vemodia  there is not a very noticeable difference lower on the bottom end of the scale near Mc.
As the numbers proceed higher toward Ma, M and Mh the number results from each lab's analyzer become further separated with a more noticeable difference. With this more precise count result, Vemodia's analyzer and method allows for the identification of the higher number of base pairs of the Mh - Harlequin Merle allele, which Biofocus was not able to reliably detect.

Note - this does not change the accuracy of the Mc or Ma allele results from Biofocus.
Dogs tested with Mc and Ma alleles through Biofocus are still tested the same by Vemodia, although with slightly different base pair numbers reported. There is a distinct separation between Mc, Ma and M that Biofocus' analyzer has no problem detecting. As well, many dogs tested as M/Ma or M/Mc or M/m by Biofocus are still confirmed with these results by Vemodia - the result of M is still correct in most cases.The difference between Biofocus' and Vemodia's testing machines becomes most apparent when the length of M and Mh are distinguished. The number of base pairs between M and Mh are not as large as the gap between Mc and Ma and the gap between Ma and M. With the numbers between M and Mh being closer/tighter together, the more accurate ABI 3500 Capillary analyzer and Vemodia's testing method is able to pick up this difference.

The following diagram is an example showing the smaller/tighter increments of base pair numbers when comparing Biofocus and Vemodia's results.

On the lower end of the scale there is not a significant difference between the Mc alleles from either lab.
As the scale climbs higher, the differences in base pair numbers for each allele becomes more noticeable.
Also notice the large gap between Mc and Ma.
This large gap is the reason why it would be very difficult, if not impossible for Mc to lengthen to Ma as this would require a significant jump in base pair numbers.

As the scale proceeds past Ma the number of base pairs within each allele's parameter is less.
Looking at this scale of the M Locus you can see how in most cases the minor lengthening of base pairs from parent to offspring of 1 or 2 bp would not cause much difference in the expression of the dog's Merle pattern but M and Mh are very close together on the scale so a lengthening of only 1 or 2 base pairs could move the dog into the higher allele range and change phenotype noticeably. You will also notice on the scale that Vemodia has made a distinction for base pair lengths between Mc and Ma - a new allele of Mc+ has been added. And a distinction made for the base pair lengths between Ma and M - a new allele of Ma+ has been added. The use of the "+" sign for these "between" lengths was adapted from the definition I assigned to them in my book. Test results do not often fall between alleles but the distinction of these lengths does make a difference in the dog's phenotype and how the dog breeds.

Examples of Ma+ being passed through 3 generations will be noted starting on page ** -
"Test Results from Vemodia."

Merle testing by Vemodia has now given us 7 alleles on the M Locus

  • m - Non-Merle - Wild Type
  • Mc - Cryptic Merle         200 - 230 bp
  • Mc+                                  231 - 246 bp
  • Ma - Atypical Merle       247 - 254 bp
  • Ma+                                  255 - 264 bp
  • M - Merle                        265 - 268 bp
  • Mh - Harlequin Merle    269 - 280 bp

Vemodia has been a wonderful lab to work with, very helpful and forthcoming with all their testing information.
They were also receptive to input from Helena and I when it came to genotype and phenotype.
I would like to thank RNDr. Tereza Jančušková, Ph.D. and MUDr. Soňa Peková, PhD. for all their work.
Examples of dogs tested for their trial and the corresponding chromatograms results will be shown starting on page ***

Example of a Merle test result Chromatogram from Vemodia M/Ma - 266/251 bp

There have also been "Mosaic" results with a dog having 3 (or 4) alleles on the M Locus.
This dog tested as M/Ma/Mc. By parents this dog should have been M/Ma - either of these alleles has mutated/shortened early on in the embryo development and created some cells with the Mc allele.
"MOSAICISM" will be covered in detail starting on page "**"