- Byron Rogers
Mitochondrial DNA haplotypes and mating outcomes
A few years ago I did a lot of work on mitochondrial DNA. This is the DNA that is inherited matrilineally - that is - the female passes it on to her sons and daughters but only the daughters then pass it on to the next generation. Setting aside the physiological importance of mitochondrial DNA (mtDNA), it is also important from a historical perspective as it allows us to examine an individual and group them by their mtDNA haplotype, as this haplotype remains relatively unchanged for thousands of years and traces back through to a single source.
Back in 2012 an academic group out of Italy examine a vast array of equine breeds and catalogued their mitochondrial DNA, using the genetic changes to group them all they found a total of 18 major haplogroups, which were labeled in alphabetical order (from A to R), each defined by a specific genetic change. The deepest node, corresponding to the Ancestral Mare from which all modern horse mtDNAs derive, was dated at about 153 thousand years ago. For each major haplogroup, there are sub-groups known as haplotypes which are mares that had specific mutations that have occurred in more recent history. You can read the paper here - Mitochondrial genomes from modern horses reveal the major haplogroups that underwent domestication
As far as Thoroughbreds are concerned, the 'traditional' way to look at female families is the Bruce Lowe numbers. Lowe had traced back the pedigrees of the complete list of winners of the oldest English classics, the St. Leger Stakes, Epsom Derby Stakes and Epsom Oaks, grouping them by direct lines of tail female descent, from dam to granddam and on back until the family was no longer traceable in the General Stud Book. He then tallied the number of classic winners produced by each family and listed them in declining order. The family with the most classic winners, the one descending from Tregonwell's Natural Barb Mare, was designated Family #1, the Burton Barb Mare second, designated Family #2, and so on. Added at a later time were families that could not be traced back to a known female in the General Stud Book, but were native to particular regions so were designated as such:
Families 1 - 50 Bruce Lowe's original numbered English families, with Goos additions, traceable to the earliest volumes of the General Stud book
Families 51 - 74 Can be additionally traced to General Stud Book mares
Families Ar 1 - Ar 2 Native to Argentina
Families P 1 - P 2 Native to Poland
Families A 1 - A 48 Native to America
Families C 1 - C 76 Native to Australia and known as "Colonial families"
Families B 1 - B 26 Designated as "British Half-Bred"
It was inevitable that once I began mtDNA typing, it would prove that some errors have crept into matrilineal records spanning more than three centuries during which the Thoroughbred spread far and wide from its country of origin. It is actually quite remarkable, and an enduring testament to the earliest compilers of the General Stud Book, the extent to which so many of the Lowe-numbered families have actually remained somewhat conserved to the same mtDNA type since the closing of the stud book.
In all, I have sequenced the mtDNA of over 4,000 Thoroughbreds from varying Bruce Lowe families, and from samples in the US, Europe, Japan, Australia, New Zealand and South Africa. I have found some surprises. The most interesting of this is two samples that we had that on pedigree records both traced back through different female families to the mare Suntor (NZ), a 1961 mare by Summertime. This is noted as the Colonial #20 family that made its way into New Zealand in the 1850's. What made it noteworthy is that its mitochondrial haplogroup was G1, a haplogroup that has, to my knowledge, never previously been found in the Thoroughbred breed. What this means is that the distant female ancestor of this family is more likely of Middle Eastern origin than European and this haplotype has survived in a breed that it is not normally found in.
The above is a radial tree visualization of the mtDNA haplotypes and haplogroups of the different breeds of horses, along with some pictures of important horses that hail from each of the mtDNA haplogroups. You can see more about the interaction between what we believe on paper that a female family traces to and what is the actual genetic reality via mtDNA. You can see more about how this all works here - http://www.bloodlines.net/TB/PedigreeMatters/Fams/EquineGeneticGenealogy.htm
There are many occasions where we believe that one 'family' is different, but in fact they are all from the same female family. An example here is the haplotype of L3a1b. There are a number of female families that on paper are noted as distinct, but in reality these all trace to the same mare:
L3a1b haplotype/Lowe family
Branch 6-e - Error at or after Fenella (GB 1869), at or before Gondolette (GB 1902).
Branch 8, 8-a, 8-b, 8-c, 8-d, 8-e, 8-f, 8-h, 8-i, 8-j, 8-k - all trace to a Mare (GB 1695c) by Byerley Turk. It should be noted that a branch of 8 has an error in it as does a branch of 8b so this family is not perfectly conserved.
Branch 16, 16-a, 16-b, 16-c, 16-d, 16-e, 16-f, 16-g, 16-h - all trace back to a Mare (GB 1823) by Don Juan. Again there are further errors in the 16 branch (Furze Chat branch has multiple errors within this branch) and the 16-c family has an error, primarily through the popular branch of Denise's Joy which is actually the I haplogroup.
Setting the errors that I have noted above aside for one minute, generally speaking these branches of female families are in fact all one female family. Now, if we knew all this data when we were planning matings, don't you think you would do your matings a little differently???
So what does this all mean and how does it relate to matings? Apropos all of the above, I just got the following email from Alan Porter of Pedigree Consultants.....
I have just been writing up Accelerate for Lane's End. They wanted to focus on the Smartaire inbreeding for him (he's a grandson of Smart Strike and tail-female to Smartaire). What is interesting is that Smartaire is Bruce Lowe family A13, which on the one sample I know you have tested is haplotype M1a. The only other family that I can see is M1a is Lowe family #23 (conserved except the Domino error at Judith (1843) or before Lizzie G. (1867)).
Looking at the female line from Smartaire, her daughter, Smart Darlin, is by Alydar. The dam of Alydar, Sweet Tooth is by a stallion from the #23 family (On-and-On) out of mare by a stallion from the #23 family (Ponder). Alydar is a Native Dancer grandson, and Native Dancer is out of a mare by Discovery, who is also from the #23 family (as was the sire of his second dam).
Darlin Lindy, the third dam of Accelerate, is by Cox’s Ridge, and is 4x5 to Miz Clementine and her sister Two Lea (Miz Clementine is dam of Best Turn, sire of Cox’s Ridge; Two Lea, the dam of On-And-On, broodmare sire of Alydar). We might note that Wagon Limit, the grade one winner out of Darlin Lindy is by Conquistador CIelo, a Raise a Native/Native Dancer line horse (Discovery again), who goes back to a mare by Tim Tam (out of Two Lea), so has two crosses of Two Lea and one of Miz Clementine.
When we look at Accelerate, we note that his grandsire is Smart Strike. Not only is he a Raise a Native/Native Dancer out of a mare by Smarten (dam Smartaire), but he is also from the #23 family, so it seems that there is an awful lot of build up of the nuclear DNA elements/pedigree names that work well with the M1a haplotype.
I spoke to John (Eberth) about this and he said we’ll find some nuclear DNA combines well with more than one mtDNA haplotype. For some reason the #10 family, Queen Mary, line stallions such as Deputy Minister and Exclusive Native work well in association (both in Accelerate, and of course the Smart Strike/Deputy Minister cross) so that is something to consider. I know it's not direct inheritance, but interesting example of if x works with y...
There is definitely something that goes on with stallions and mitochondrial female families, even if it’s obscured by other inputs, as I see too many examples when I’m using the six generation catalog style pedigrees where there are several good horses from the one mitochondrial haplotype, by same sire or sire line, and where they are sufficiently distant in catalog terms that one wouldn’t have influenced the other.
Something to think about,