One gait-keeper gene allows horses to move in unusual ways

By Ed Yong | August 29, 2012 1:00 pm

Icelandic horses can move in an odd way. All horses have three natural gaits: the standard walk; the two-beat trot, where diagonally opposite pairs of legs hit the ground together; and the four-beat gallop, where the four feet hit the ground in turn.  To those, Icelandic horses add the tölt. It has four beats, like the gallop, but a tölting horse always has at least one foot on the ground, while a galloping one is essentially flying for part of its stride. This constant contact makes for a smoother ride. It also looks… weird, like watching a horse power-walk straight into the uncanny valley.

The tölt is just one of several special ambling gaits that some horses can pull off, but others cannot. These abilities can be heritable, to about the same extent that height is in humans. Indeed, some horses like the Tennessee Walking horse have been bred to specialise in certain gaits.

Now, a team of Swedish, Icelandic and American scientists has shown that these special moves require a single change in a gene called DMRT3. It creates a protein used in neurons of a horse’s spine, those which coordinate the movements of its limbs. It’s a gait-keeper.

Leif Andersson, one of the leaders of the study, was completely surprised by the discovery. “There are hundreds of genes that contribute to variation in height in humans, and each has a tiny effect. Gait in horses sounds like it should be at least as complex a trait.” It’s not. By itself, DMRT3 explains much of the difference between horses that stick to the normal trinity of walk/trot/gallop, and those that adopt special gaits.

Lisa Andersson sequenced 30 four-gaited Icelandic horses that can tölt as well as walk, trot and gallop, and 40 horses that added a fifth gait—pace, where the legs on the same side of the body hit the ground together. She scoured their genomes for single mutations that were more common in the pacing animals, and found one. By sequencing the full genomes of two of the horses, she pinpointed the mutation to a gene called DMRT3.

Meanwhile, Klas Kullander had been searching for genes that are active in the spines of mice, and seem to control gaits. His group had independently identified DMRT3 as an important gene.  “When they heard about our horse discovery, they were really excited,” says Andersson.

DMRT3 creates a protein that controls the activity of other genes. In the five-gaited horses, both copies of DMRT3 carry a mutation that shortens the protein, creating a stunted finished product.

When the team sequenced 352 Icelandic horses and found that all the pacing animals have two copies of the shortening mutation. When they looked at other breeds, they found that the DMRT3 mutation is extremely common in those that show alternative gaits. Almost all Tennessee walking horses have it. So do all Peruvian pasos. So do all Kentucky mountain saddle horses. Even among the four-gaited Icelandic animals, which can tölt but not pace, two thirds of the animals carry at least one copy of the mutation.

By contrast, it’s absent in all horses that stick to the standard walk, trot and gallop. Thoroughbreds, Shetland ponies, wild Przewalski’s horses—all of them have unabridged DMRT3 proteins. The conclusion is stark: for a horse to move beyond its three natural gaits, it needs a stunted version of the DMRT3 protein.

What does DMRT3 do? Genetic manipulations held the answer, but since horses aren’t exactly easy to study in a lab, so three of Kullander’s team –Martin Larhammar, Fatima Memic and Hanna Wootz – turned to mice. They found that DMRT3 is active in a special class of ‘interneurons’ in the spines of developing mice, which coordinate the movements of the limbs. They connect to the flexor and extensor muscles that move the rodents’ legs. They also cross the spine to connect the right and left side of the body.

If mice lack any working copies of DMRT3 gene, these neural circuits don’t develop properly. The mice learn to compensate for their problems, so the adults seem outwardly normal, but newborn pups are different. “You see very disturbed signalling patterns in their spine,” says Andersson, “and there was a really strong defect in how they coordinate their hind legs.” In the adults, these problems came to the fore whenever they had to run. They walked normally, but the coordination between their legs broke down at high speeds.

The same is true for horses, and explains why the DMRT3 mutations are almost non-existent in the wild. Carriers find it hard to transition from trots and paces to full-blown gallops. They lack the coordination necessary to pull off the fastest gait, and predators would easily have removed them from the gene pool.

Humans were kinder, and saw a different sort of potential. Andersson imagines that early humans noticed that some horses could move in unique ways, and selected them for breeding, perhaps because they offered a smoother ride or were more versatile at intermediate speeds. Certainly, these animals also do very well in harness-racing, where trotting horses are disqualified if they break into a gallop. In our stables and tracks, an otherwise debilitating mutation has found a comfortable home.

Reference: Andersson, Larhammar, Memic, Wootz, Schwochow, Rubin, Patra, Arnason, Wellbring, Hjalm, Imsland, Petersen, McCue, Mickelson, Cothran, Ahituv, Roepstorff, Mikko, Vallstedt, Lindgren, Andersson & Kullander. Mutations in DMRT3 affect locomotion in horses and spinal circuit function in mice. Nature

Image by Paula Jautunen

More on the genetics of domestication:


Comments (21)

  1. Zach Miller

    Okay, Ed, “gait-keeper” is a new low.

  2. Interestingly, the same ‘gait-keeper’ pun is used on Nature’s print cover this week. Checking back, the headline on the press release that reporters all received was ‘Gait expectations’. Which doesn’t make much sense to me, but must have helped put the punning notion into people’s heads.

    [EY: “You wanna know my secret? I’m always punning.” – Me, riffing off Banner]

  3. Gretchen Icenogle

    We just returned from Iceland, where we had a chance to experience the smooth tölt of the Icelandic horse – it’s easy to see (or rather to feel, since it looks so funny) why the mutation has persisted in domestication.

    A couple of places where the article’s logic appears slightly frayed: at the beginning of the article, you say that all horses share the same three natural gaits, to which some add the tölt and/or the pace. At the article’s end, however, you suggest that horses with the DMRT3 mutation are incapable of a full-blown gallop. Similarly, you note that two-thirds of the Icelandic horses studied who could tölt possessed at least one copy of the mutated gene, from which we can infer that one-third did not – this calls into question the starkness of your conclusion that, in order for a horse to move beyond its three natural gaits, it needs the stunted protein.

    Is the gait open, shut or ajar?

  4. Super cool post! Watching these horses reminds me of that trick some people seem unable to pull off: ‘can you pat the top of your head with one hand while you rub your tummy with the other?’

  5. bonnie

    I would suggest looking at the research of Dr. James Rooney. Rooney mentions fossilized hoof prints of ancient horses showing those horses were gaiting. Many of the mustangs and horse that were developed from the mustang breeds also showed gait. Those horses can gallop quite well.

    On a personal note, I have many horses (including racing off the track harness horses that raced as pacers) that were quite lateral in gait who could also gallop which would not have prevented their survival in the wild. I have also bred a very lateral TWH stallion who passed on his gait to trotting and gaited horses but he could gallop very well. This stallion did very well in endurance as some other gaited horses that can gallop and gait that I have ridden.

  6. T

    Yeah, I noticed the same thing as Gretchen – if this section is true:

    “When they looked at other breeds, they found that the DMRT3 mutation is extremely common in those that show alternative gaits. Almost all Tennessee walking horses have it. So do all Peruvian pasos. So do all Kentucky mountain saddle horses. Even among the four-gaited Icelandic animals, which can tölt but not pace, two thirds of the animals carry at least one copy of the mutation.”

    Then this one doesn’t necessarily follow:

    “…for a horse to move beyond its three natural gaits, it needs a stunted version of the DMRT3 protein.”

    Where are the remaining horses who don’t have the DMRT3 gene getting the stunted DRMT3 protein?

  7. Lori M.

    The article is interesting. I did notice the discrepancies as well. There are gaited mustangs, though they are relative newcomers when compared to the ancient breeds, and come from mixed heritage. Anybody who says a horse is at a disadvantage to have the smooth gaits and that he can’t run because of it, has never seen my Missouri Fox Trotter mare explode into a gallop across my pasture. She’s fast. She’s not a Thouroughbred, but she’s fast and agile! She also traces to old thoroughbreds. So I take issue with this mutation being called a genetic weakness. Gaited horses have been around a long time, and I’m betting if they do enough looking, they will find it in the ancient types as well. I think this discovery is a beginning to understanding the genetics of gait, but I’m willing to bet the expression of it is a whole lot more complicated than that. By the way, there are several easy gaits besides the tolt and many gaited horses can do more than the signature gaits of their breeds. Also nothing was mentioned of the 3 beat canter as one of the standard gaits. Even the Thouroughbred was created by using some gaited stocks and crossing with the Barbs, Turks, and Arabs. I am hoping a friend of mine might read this and reply as well. She has done a lot of research on the origins of gaited horses. And lastly, I love my gaited horses. They can do the standard gaits and also have a bunch of nice to ride extra gears!

  8. Rasem Brsiq

    But, you see:

    “The mice learn to compensate for their problems, so the adults seem outwardly normal, but newborn pups are different”.

    So the problem, I understand, is in living long enough to compensate.

  9. This is really interesting. Thanks, Ed. And for what it’s worth, I’d’ve been very disappointed had the gait-keeper pun not made an appearance.

  10. Lori M.

    From birth, or at least once able to stand and move, gaited breed foals will show gait, and also from birth they can zip around galloping. They are no more disabled than any breed of foal. All newborn foals need to sort out there legs in the first hours of life.

  11. Dawn

    I’ve raised Tennessee Walkers for years. All foals do a running walk, along with just about everything else. They certainly don’t seem to be disabled in any way. Nor do their parents.

    Hipparion hoof prints found by Mary Leaky display the distinctive running walk pattern, indicating it was not maladaptive some 3 million years ago.

    Nor are horses the only animals who do the running walk: elephants do it as well. It certainly hasn’t been detrimental to them.

  12. Alison

    Interesting article and interesting research but I very much disagree with the conclusions.
    I believe also that gaited horses have been about since ancient times and were much more common than they are now, maybe in fact they were the norm and it is the animals without the extra gaits which are the mutation.

    Horses which used trot as their first choice of gait were selectively bred when horses were used in great numbers for carriage driving. Because of that simple reason the gaited horses became more of a speciality market and the gaits have survived best in places such as Iceland where, due to the terrain, the horses were ridden rather than driven.

    I have seen some gaited Shetlands too, despite what it says in the article, so it would be interesting to see what proportion of that breed still carry the gene.

  13. Phyllis Lewis

    Down through the ages there have been smooth gaited horses. These were once the norm until working ability of a horse for pulling became more critical as oxen fell out of favor and carriage horses became a status symbol. The higher the head and the trot, the flashier the movement and looks, the more standing you had. The average gaited horse was ridden, not driven then. There are records of knights keeping both their war horses and a palfrey, a gaited traveling horse of lighter weight.

    Modern gaited horse breeds are known for their versatility in usage. They do it all, farm work, showing, jumping, trail, roping, barrel racing, you name it, they can do it all and be the most comfortable ride you have ever had.

  14. Katy

    Do all of the different “alternative-gait” breeds have the same shortening mutation in DMRT3? I was struck by the fact that while all these breeds report the same mutation in the same gene that apparently bestows their extra-gaitedness, their extra gaits (e.g., tölt, pace, running walk, rack) do not all look the same. I wonder what kinds of influences dictate what specific extra gait an animal with this mutation will develop? Could it be other aspects of their physiology?

  15. As a breeder of gaited Morgans, I enjoy this article but agree with the other gaited horse owners here that mine have been able to gallop from day one. Most of mine can do the extra gaits and trot also although I have two mares that do not trot, only walk, gait ( tolt type gait called rack in America and also foxtrot and stepping pace- lateral and diagonal versions of smooth gaits), canter and gallop. so that part of the theory I dismiss. It is easier on horses to trot in deep mud or sand so I think they started breeding for more trot when roads are better developed and the horses are pulling wagons and carriages instead of only being ridden like in very wild country.

  16. Anthony

    I believe (and may be wrong) that the issue isn’t the inability of the animals to gallop, it’s that they have difficulty transitioning from a tolt/amble to a gallop. That is, an animal that trots can transition from stationary, walk, or trot to a gallop, but a gaited animal can (normally) only transition from stationary or walk. Under some escape circumstances this would be a disadvantage.

  17. Shari

    I have two, 5 gaited Icelandic horses, they walk, trot, Tolt, canter, Pace and they can both Gallop like a bat out of hell.
    I have to question, you saying, gaited horse’s can’t pace from gallop. That is how Pace racing goes,,, the Icelandic is in full gallop and drops down into Pace. Anyone from Iceland should know that.
    I have also seen them Pace and go right into a gallop smoothly.
    Also ridden pace, gallop to pace, canter to a pace, pace to gallop and so on….

    Nor are they in any pain or handicapped in any way, due to being gaited.
    So your research is a start but I think there is more to it that what you have found so far.

  18. Lori M.

    There has been some discussion of this article among gaited circles, including in groups on FB. Many point out earlier research on prehistoric horses being gaited. Here is an interesting article about the origins of gaited horses:

    It just doesn’t make sense that this mutation would be maladaptive if it has indeed been around as long as the research of prehistoric horses suggests. I would like to see if this same mutation, identified in the study, is found in donkeys because there are indeed gaited donkeys who could not have acquired the mutation passed from horses. It must be a very ancient mutation.

  19. TW

    For the dark side of showy horse gaits and the assholeness of humans just look up the practice of soring in the Tennessee Walkers and other training methods.
    Some gaits while they make for a smoother ride for the human come with a price of damaging stress on the horse’s back.

  20. SPS

    In addition, some horses that do not gait naturally can be trained to rack, etc — there have been quite a few purebred Arabians trained in 5 gaits.

    TW, this is a good point — however, it’s not the natural running walk that is the problem, it’s the human desire for an extreme far beyond natural that “inspires” the abusive training and soring. Natural TWHs are lovely horses and do not require mistreatment to deliver a wonderful smooth ride.

  21. “All horses have three natural gaits: the standard walk; the two-beat trot, where diagonally opposite pairs of legs hit the ground together; and the four-beat gallop….”

    The canter is a 3-beat gait with laterality:

    The Right-Lead Canter:
    (1st beat) left hind leg
    (2nd beat) right hind / left fore
    (3rd beat) right fore leg

    The Left-Lead Canter:
    (1st beat) right hind leg
    (2nd beat) left hind / right fore
    (3rd beat) left fore leg

    The gallop develops out of the canter and the 2nd beat is separated into two separate beats:

    The Right-Lead Gallop:
    (1st beat) left hind leg
    (2nd beat) right hind leg
    (3rd beat) left fore leg
    (4th beat) right fore leg

    The Left-Lead Gallop:
    (1st beat) right hind leg
    (2nd beat) left hind leg
    (3rd beat) right fore leg
    (4th beat) left fore leg

    In some equestrian circles, the expected speeds are expressed in meters per minute. A general rule of thumb for a 16 hand Warmblood or Thoroughbred type would be:

    – 150 mpm (“forward” walk)
    – 250 mpm (medium trot)
    – pace of 300 mpm (medium to fast trot)
    – pace of 325 mpm (slow canter)
    – pace of 350 mpm (canter)
    – pace of 375 mpm (fast canter)
    – 400 mpm (slow gallop)

    In 3-day eventing, the base cross-country pace is the gallop.

    Preliminary eventing speed = 520 mpm
    Intermediate speed = 550 mpm
    Advanced speed = 570 mpm


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.

See More

Collapse bottom bar