Is “Horse Sense” Overrated? Nonsense Mutation Allows Novel Gaits

1 September 2012 by Anne-Marie Hodge, posted in Uncategorized

By Dagur Brynjólfsson (CC-BY-SA-2.0, via Wikimedia Commons)

Most horses are capable of three basic gaits: walk, trot, and canter/gallop.  Some specialized breeds, however, are capable of two additional patterns of locomotion that are beyond the reach of their equine brethren. These displays have impressed human observers so much that "gaitedness" has been selected for in some popoulations, creating highly specialized breeds such as at the Icelandic horse, Paso fino, and others. Pacing and harness competitions (in which horses perform gallop-paced trotting), although less well known than traditional races such as the Kentucky Derby, are significant sports in the equine world.

Icelandic horses in particular are known for their ability to “tölt,” a smooth gait that uses the same step pattern as a walk, but at very high speeds.  It is analogous to a the infamous "granny speed walk", yet far more elegant.  Some horses can also perform the “flying pace,” in which the legs are coordinated laterally, with both feet on the same side of the body (for example, right forefoot and right hindfoot) striking the ground in concert.  Not all extra-gaited horses can perform both the tölt and the flying pace, but those that do display a full five-gait repertoire are highly coveted.

By Dagur Brynjólfsson (CC-BY-SA-2.0, via Wikimedia Commons)

Anyone lucky enough to witness this (turns out you are lucky, see the video below) is almost inevitably struck by the fact that something unique is on display.  This begs a question: how do trainers teach horses to perform such unnatural-looking movements?  Also, why does this ability show up in a smattering of other breeds from vastly different parts of the globe?

New evidence suggests that training may actually be the least important component (although surely not insignificant). It had already been established that gaitedness shows a high degree of heritability (Albersdóttir et al. 2011), and a team of researchers from around the world sought to expand upon this by determining what specific genes were at play. Their results were reported in the 30 August 2012 edition of Nature.

After conducting whole-genome sequencing of four- and five-gaited Icelandic horses, the researchers narrowed the source of the ability down to DMRT3, part of the DMRT family of transcription factors.  In some horses, this gene has a misplaced stop codon, a condition known as a “nonsense mutation,” which disrupts protein synthesis and results in an alternate form of the final gene product (Figure 1g).  All of the five-gaited horses in the study were homozygous for this DMRT3 mutation.* Perhaps "horse sense" is overrated after all.


From Andersson et al. 2012


The study also included horses of other gaited breeds: the Kentucky mountain saddle horse, Missouri fox trotter, Paso fino, Peruvian fino, Rocky mountain horse, and Tennessee walking horse, three varieties of Standardbred, and the French trotter. All of these breeds showed an even more extensive DMRT3 mutation rates than the Icelandic horse, with allele frequencies of the mutation ranging from 0.95-1.00.  In contrast, the non-gaited horses that were tested showed 0.00 mutation allele frequencies across the board (Table 1).


From Andersson et al. 2012


It was noted that although being a DMRT3 mutant allows for novel gaits, it actually decreases performance value for galloping or trotting, indicating that it has probably been selected against in horse breeds that compete in dressage, jumping, traditional races, and other more well-known equine sports. In a slightly novel approach, the authors also tested to see whether genotype was associated with breeding value and prize money earned by each horse, and found that DMRT3 mutants scored significantly higher on both counts (p < 0.0001 and p = 0.007, respectively).  Definitely challenges the derogatory connotations of the term "mutant," no?

So, it is clear that the DMRT3 mutation affects limb coordination in horses, allowing for the execution of an expanded repertoire of gaits. But how does this nonsense mutation actually achieve such a feat?

The authors sought to test the effects of DMRT3 mutations in mice, an easier animal to breed and "put through its paces," so to speak, in the lab.  They used Dmrt3-null mice, and performed an array of tests to evaluate their locomotory patterns and abilities. In addition, immunostaining techniques were used to trace the development of Dmrt3  during embryogenesis.

The results showed that Dmrt3 is expressed in the spinal cord, and plays a critical role in the function of the spinal circuitry controlling vertebrate stride and limb coordination. The paper asserts that in mice, “Dmrt3 is required for normal development of a coordinated locomotor network in the spinal cord.”  There is overwhelming reason to believe this applies to mammals and even vertebrates in broader sense, which establishes that the gene identified as being altered in gaited horses does indeed control gait and coordination. This ties in beautifully with the analysis of horse gaitedness, which indicates that DMRT3 has a critical regulatory role in the coordination of both left/right and fore-/hind leg coordination in equids.

This brings us to the take home-message of the study: a misplaced stop codon in horses results in an amazing broadening of abilities and that has spawned entire breeding industries and new equine sports.  An impressive achievement for a single point of nonsense.





Albersdottir, E., Eriksson, S., and Arnason, T. 2011. Genetic analysis of “breeding field test status’ in Icelandic horses. J. Anim. Breed. Genet. 128:124-132.
Andersson, L. S. et al. (2012). Mutations in DMRT3 affect locomotion in horses and spinal circuit function in mice. Nature, 488: 642-646 DOI: 10.1038/nature11399


* There was one exception out of nearly 350 Icelandic horses tested, which was later chalked up to a "phenotypic misclassification"--it wasn't a five-gaited horse after all.

Leave a Reply

+ two = 5