Old gene, short new trick

Retrogene causes short legs in man’s best friends

We may never know how the zebra got its stripes, but we know how the wiener dog got its short legs. Height-challenged dog breeds — including dachshunds, corgis and basset hounds — have an extra copy of a normal gene to thank for their diminutive stature, new research shows.

LEG ENVY Dogs without the fgf4 retrogene have longer legs. Maggie Bartlett

SHORTY GENE Researchers traced the cause of the dachshund’s short, curved legs to a retrogene called fgf4. Ethan Hall

“It’s stunning to see a genetic modification like this,” developmental geneticist Douglas Mortlock of Vanderbilt University in Nashville, Tenn., says of the new study, published online July 16 in Science. “This is the gene that makes wiener dogs short-legged.”

As anyone who has been to a dog park can attest, man’s best friends come in a wide variety of body shapes and sizes. In this study, researchers focused on eight breeds among more than a dozen known to have a trait called chondrodysplasia — causing legs that are short relative to body size, curved and heavier-boned than normal.

Heidi Parker, of the National Human Genome Research Institute in Bethesda, Md., and her colleagues broadly compared the genomes of 95 short-legged dogs from the eight breeds with the genomes of 702 dogs from 64 breeds without the trait. Then, in a more detailed analysis, the researchers pinpointed an extra stretch of DNA on chromosome 18 in every dog from the eight short-legged breeds, but in none of 204 control dogs they examined.

This extra stretch of DNA turned out to have a sequence almost identical to another gene important for limb development, called FGF4. Located at the opposite end of chromosome 18 in dogs, the original FGF4 gene was duplicated at some point in the dog lineage, creating a new copy elsewhere called the fgf4 retrogene, the researchers say in their report.

In rare cases, messenger RNA — molecules made from DNA that carry information to cellular machinery, which then makes proteins—can get turned back into DNA. If this DNA then gets plopped back into the genome in a new neighborhood, and conditions are right for this genetic new kid on the block to become active, the extra DNA becomes a retrogene.

 “Finding a retrogene that is active and associated with certain traits is very rare,” says Parker. “We believe that they’re out there and that it happens, but they haven’t been identified. We may find more.”

Most pieces of DNA that hop around the genome are not functional, in part because they may land in inhospitable places in the genome. But this retrogene landed in a sweet spot that allowed it to change dog leg length. “This is very important because it’s the first time a clear morphological effect of a retrogene was detected,” says evolutionary geneticist Manyuan Long, of the University of Chicago.

Part of the reason this retrogene was found is that it is a dominant allele, meaning that only one copy is needed for the trait — in this case, chondrodysplasia — to appear. “It’s much easier to see the effects of really rare mutations if they are dominant,” says Mortlock, noting that the trait would be evident to dog breeders. “That’s why it was likely to be kept by humans.”

Intense breeding programs drive out a lot of genetic variation, but even dogs without fancy pedigrees can be affected. If a mutt has short legs, heavy bones and a little bit of curve in the legs, then it’s pretty likely that the dog has the retrogene, says Parker.

Parker and her colleagues also examined DNA from wolves, to see whether they could narrow the window of time during which the fgf4 retrogene originated. None of the wolves tested had the retrogene (which Parker says was not surprising; short-legged wolves would stand out). This gene duplication probably happened after dogs first became domesticated and before division of early dogs into modern breeds, putting the range, Parker says, anywhere between 300 and 15,000 years ago. 

Mortlock wonders whether other genes might also contribute to the short-leg phenotype. Further genetic experiments, like putting the retrogene into a mouse genome and seeing whether the resulting mice develop short limbs, would be revealing.

Still, the study’s strong, clear findings on leg length were exciting, Mortlock says.  “It’s a very satisfying story.” 

Laura Sanders is the neuroscience writer. She holds a Ph.D. in molecular biology from the University of Southern California.