Wednesday, September 12, 2012

The Pace Gene

Icelandic horses are famously either four-gaited or five-gaited. They can walk, trot, and canter (or gallop), like all horses can. But Icelandics can also tolt (a kind of running walk) and pace, which is a racing gait. No other breed of horse can coordinate the movement of its legs in so many different ways, I discovered when researching my book A Good Horse Has No Color in the late 1990s. Recently I learned exactly why:

In August, a group of Swedish, Icelandic, and American scientists published an important scientific paper in the journal Nature identifying the genetic mutation that causes Icelandic horses to tolt and pace. See

Led by geneticist Leif Andersson of the Swedish University of Agricultural Sciences and Uppsala University, the group first compared the genes of 30 four-gaited Icelandic horses to 40 five-gaited Icelandic horses. They found “a highly significant association between the ability to pace and a single nucleotide polymorphism” or SNP (pronounced “snip”) mutation on one chromosome in the horse genome.

They confirmed that association by testing an additional 352 Icelandic horses (both four-gaited and five-gaited), along with 808 horses of other breeds.

The SNP mutation occurred in a gene named DMRT3. Investigating this region in mice, the researchers discovered that it controls leg movement. The gene is expressed in the spinal cord, producing neurons that affect length of stride, swing time or flexion, and both front-back and left-right coordination.

To pace, an Icelandic horse must possess two copies of the mutant DMRT3 gene. To tolt, a horse needs one copy. 

You can watch an Icelandic horse tolt on YouTube, courtesy of my friend Stan Hirson:

To understand what the Nature paper meant, I needed to review some terminology. Fortunately, I had learned a lot of that when coauthoring the book Mendel in the Kitchen: A Scientist’s View of Genetically Modified Foods with Penn State geneticist Nina Fedoroff in 2004.

The genome, I recalled, was the horse’s complete set of genes, packed into the chromosomes inside each of its cells. Genes are made of DNA, a long molecule that forms a double helix, looking like a twisted ladder. DNA itself is made up of four small molecules, the nucleotides adenine (known as A), cytosine (C),  thymine (T), and guanine (G).

In the 1960s, scientists learned how to read the DNA code. Each sequence of three letters, each codon, such as GAG or GAA or AAA, stands for an amino acid (except for three codons, which instead mean “stop”). When the code is read, the cell’s machinery links the specified amino acids together into a chain, which is then folded up to become a functioning protein. Each protein made by the body has its corresponding DNA code—a gene.

There’s more than genes in a genome, if by gene we mean the DNA sequence that codes for a protein. Many DNA sequences don’t code for proteins. Instead, they control when and where in the body certain proteins are made. Still other sequences are thought to be useless junk, though scientists keep finding uses for more and more of this junk, as the recent ENCODE study shows. See:

An Icelandic horse showing trot.

But the mutation that allows Icelandic horses to tolt and pace is a relatively easy one to understand. A SNP mutation, or single nucleotide polymorphism, means that one nucleotide (one A, C, T, or G) in the DNA sequence of the DMRT3 gene was wrong. The three-letter codon that this one nucleotide was part of should have coded for an amino acid. Instead, it said “stop.” The mutant form of the gene found in Icelandic horses is significantly shorter than the wild type found in three-gaited horses.

In fact, all gaited horse breeds tested, including the Icelandic horse, the Kentucky Mountain Saddle horse, the Missouri Fox Trotter, the Paso Fino, the Peruvian Paso, the Rocky Mountain horse, the Standardbred, and the Tennessee Walking horse, had the short, mutant type. All three-gaited breeds—Arabians, Gotland ponies, North-Swedish draft horses, Przewalski’s horses, Shetland ponies, Swedish Ardennes, Swedish Warmbloods, and Thoroughbreds—had the long, wild type.

Now the terms “wild type” and “mutant” are value-laden. Geneticists used to use “wild type” to refer to the typical form found in nature. But it’s now accepted that genes can come in many different forms in a single species without any one of them being more typical than another.

In this study, the researchers seem to have assumed that, since all horses can walk, trot, and canter/gallop, these gaits are more typical. The fact that the Przewalskis horsethe closest wild relative of the domesticated horsehas only three gaits supports this conclusion. The researchers therefore assigned the label “wild type” to the long version of the gene, the one that produces three-gaitedness.

But the reverse could be true. And even if the Icelandic horse’s shorter DMRT3 gene is not the wild type, the mutation must have happened very early in the history of horse evolution: When anthropologist Mary Leakey uncovered the tracks of three 3.5-million-year-old equids in East Africa, she found their footfalls to match those of a tolting Icelandic horse traveling at ten miles an hour.

The same horse showing canter.

Horse historians, especially those (like me) within the Icelandic horse community, have long argued that gaitedness was bred out of horses since medieval times, not bred in.

As I wrote in A Good Horse Has No Color, the first Icelandic horses were bred for strength, speed, color, and a smooth, untiring traveling gait. Few of Iceland’s bridle paths, though ancient and well-worn, are smooth and even. Interrupted by birch brush and roots, they cross farm fields heaved and hummocked by frost, pocked and pitted with bogs. They angle up and around precipitous slopes, wind along rocky rivers, or pick their way through lava wastes. Yet even today, when riding fast and far is only a pleasure, not a necessity, Icelandic horses routinely cover thirty miles a day. On one trek organized in 1994, one thousand kilometers—six hundred miles—were ridden in two weeks, an average of over forty miles each day.

When Iceland closed its borders to the importation of horses sometime in the Middle Ages (no medieval source gives us an exact date), today’s fine distinctions between fast tolt, slow tolt, trot-tolt, pacey-tolt, traveling pace, and piggy pace were unknown. A good horse was comfortable to ride all day, which meant it possessed some mid-speed gait other than a jarring, jouncing trot.

In the rest of medieval Europe, such a horse was known as an ambler, palfrey, or jennet, and would be ridden by ladies or noblemen or knights not in armor. The word “ambler” comes from the early Roman ambulator, or walker, while a trotting horse in these same documents is called a cussator or cruciator, a “tormentor” or “crucifier.” “Palfrey” derives from the Greek parippi or the Celtic paraveredus, both meaning a horse used for long journeys. The jennet was originally a particular breed out of Spain, known for its easygoing slow pace.

When roads became widespread during the Renaissance and carriages were the noble way to travel, selective breeding for smooth-gaited riding horses declined. People bred for trotters, faster and with more stamina when under harness. Only in the roadless outposts of European culture were the smooth, lateral gaits like the tolt retained, producing, in the New World, such breeds as the Tennessee Walking Horse and the Peruvian Paso.

The same horse showing tolt.

The Icelandic horse experts I consulted in the 1990s, while writing A Good Horse Has No Color, assured me that the ability to tolt was genetic and that it depended on the pace. According to one, “It is part of the quality of the trait, how easy it is to breed it. Well-bred Icelandic horses should therefore be almost self-trained to tolt.” Said another, “Not all horses have pace worth showing, but it is considered good for the tolt if the horse has some pace. It is thought that without breeding for pace, the tolt would be lost.”

The Nature study proves they were right. If you breed a mare and a stallion who each have only one copy of the short DMRT3 gene, you can get a foal that has no copies of it: This foal will be three-gaited. That’s why this study is so important to Icelandic horse lovers. It tells us exactly how to preserve what we love best about our breed: the gaits.

For breeders who want to know in advance of training and evaluation if a foal is five-gaited, a simple DNA test is already available from Capilet Genetics: The company notes, however, that “the quality of the pace is a quantitative trait that is determined by many genes working together and the environment (training).”

In other words, the short DMRT3 gene is necessary, but not enough for true flying pace. The Swedish study found that 31 percent of the horses considered four-gaited were genetically five-gaited: They had two copies of the short DMRT3 gene. The classification was done by Thorvaldur Arnason of the Agricultural University of Iceland in Borgarnes, based on WorldFengur records. Their pace was simply not good enough for them to show it in competition.

The same caveat holds for the tolt. Without one copy of the gene, your horse cannot tolt. But genetics alone cannot explain how well your horse tolts.

The horse in the gait photos above is Parker fra Solheimum, a first-prize Icelandic stallion owned and ridden by Sigrun Brynjarsdottir. Visit them at http://usicelandics.comOn the US Icelandic Horse Congress’s website (, you learn more about Icelandic horses. Or take a virtual ride on my friend Stan Hirson’s video blogs, and Life with Horses ( Stan's beautiful video of an Icelandic horse tolting is also available there.

Join me again next Wednesday at for another adventure in Iceland or the medieval world.


  1. That tolt is amazing! It looks just like the Tennessee Walking Horse's gait, only without the appalling "soring" that some resort to, to achieve it.

    1. It is amazing--smooth and fast--and completely natural. My horses tolt in the field, without a rider or shoes.

    2. hi, I'm trying to understand if any other gaited horse breed can pace besides the Icelandic, or if a cross to an Icelandic can pace. The DNA research sounds to me like, it's a no,,,the offspring of an Icelandic cross cannot produce a horse that will pace. I would really like to know. thanks! nan

    3. Hi Nan, Cross-breeding is really discouraged with Icelandics, so I don't know if anyone would be able to answer that question. You could go here to learn more about how to test your horse for the "pace gene":

    4. Hi Nancy!
      There are two breeds, other than Icelandics that I'm aware of that can be five-gaited. They are the Standardbred and the Basuto Pony of Lesotho (Horse really same as the icey :) )

  2. Thank-you SO much Nancy, for explaining this in detail with respect to Icelandic horses. The shorter summaries I had read were confusing and misleading, and I did not want to struggle through the deeper genetic research. This seems like a significant breakthrough, and helps discourage simply breeding two amazing four-gaited tolters to get an amazing four-gaited tolter... At least without having the DNA test.

  3. I always thought that palfrey and jennet referred to donkey/horse or donkey/pony crosses. This is very interesting to hear.

  4. Sent me to the dictionary--thanks! A jennet or jenny can be a female donkey, but it's also used to refer to a smooth riding horse. Palfrey is only used in reference to horses, as far as I can learn.

    1. The word 'Jennet' is the English corruption of the Spanish 'Jinete' (=rider) and implies a gaited saddle horse. :)

  5. Hillary BartkowiakJanuary 18, 2013 at 6:50 PM

    Whoa! So Icelandic horses could only do that? That horse is really awesome doing tolt huh? What type of performance horses could do tolt?