Bengal cats are prized for their appearance; the exotically marbled and spotted coats of these domestic pets make them look like small, sleek jungle cats. But the origin of those coats — assumed to come from the genes of Asian leopard cats that were bred with house cats — turns out to be less exotic.
Stanford Medicine researchers, in collaboration with Bengal cat breeders, have discovered that the Bengal cats' iridescent sheen and leopard-like patterns can be traced to domestic cat genes that were aggressively selected for after the cats were bred with wild cats.
"Most of the DNA changes that underlie the unique appearance of the Bengal cat breed have always been present in domestic cats," said Gregory Barsh, MD, PhD, an emeritus professor of genetics. "It was really the power of breeding that brought them out."
For a study published online March 25 in Current Biology, Barsh and his colleagues analyzed genes collected from nearly 1,000 Bengal cats over the course of 15 years. Barsh is the senior author of the paper.
The results shed light not only on the Bengal cat's coat but also help answer broader questions about how appearance is encoded in genetics and how different genes work together to yield colors, patterns and physical features.
Wild origins
Barsh and his colleagues, including senior scientist Christopher Kaelin, PhD, use cats and other animals to study the genetics of physical features. In previous studies, they identified genes responsible for the color coat variation in tabby cats and for the unique markings on the Abyssinian cat.
"The big-picture question is how genetic variation leads to variation in appearance," Barsh said.
"This is a question that has all kinds of implications for different species, but we think that cats offer an especially tractable way to study it."
From the 1960s through the 1980s, breeders, led by biologist Jean Mills, crossed the wild Asian leopard cat species Prionailurus bengalensis with domestic cats to create a new, visually striking cat breed. Over many generations, the cats with the desired physical characteristics and temperaments were progressively selected and bred. By 1986, the Bengal cat was recognized as its own new breed by the International Cat Association.
Barsh and Kaelin saw Bengals — with their recent genetic origin and unique appearance — as a particularly interesting way to study how genetic variation causes diversity in form, color and pattern. In 2008, they began reaching out to cat breeders, attending cat shows, and collecting cheek swabs and photographs of Bengal cats.
Genetic surprises
The Stanford Medicine team suspected that Bengal cats might give them an accessible way to probe the genetics of wild cat colors and patterns that had evolved naturally. But after sequencing 947 Bengal cat genomes, they found something surprising: There were no parts of the wild Asian leopard cat genomes that were found in all Bengal cats.
"Nearly every Bengal cat breeder and owner has this idea that the distinctive look of the domestic Bengal cat must have come from leopard cats," Barsh said. "Our work suggests that's not the case."
Instead, the genetic signatures suggested that the unique appearance of Bengals was a result of variations in genes that had already been present in domestic cats.
The team found something similar when they looked specifically at "glitter": About 60% of all Bengal cats have particularly soft, iridescent fur that glitters like gold in the sunlight. A mutation in the gene Fgfr2, they showed, is responsible for glitter and comes not from leopard cats but from domestic cats. Glitter and the underlying Fgfr2 mutation are nearly specific to Bengal cats. Interestingly, the mutation reduces the activity of the protein encoded by Fgfr2, rather than rendering it inactive as many mutations do. This sheds light on how variations in genes can cause subtle changes in appearance, the researchers said.
Finally, Barsh and Kaelin's group analyzed the genetics of "charcoal" Bengals, a rare subset of the breed with darker coloring. They uncovered a leopard cat gene linked to the charcoal color, but only when it was combined with domestic cat genome. The leopard cat gene, known as Asip, essentially doesn't work as well when it's mixed with the domestic genes — a phenomenon known as genomic incompatibility. So, in leopard cats, Asip doesn't cause charcoal coloring, but the same gene in domestic cats does.
"Hybridization between different species can happen naturally and is responsible for the small amount of Neandertal DNA found in many human genomes," Barsh explained. "But the wild leopard cat and the domestic cat are more different from each other than humans are from chimpanzees, and it's remarkable to see how DNA from these distantly related species can exist and work together in a popular companion animal."
A boost for biology and breeders
A better understanding of the genetic origins of Bengal cat traits is already helping Bengal breeders fine-tune the way they breed animals to create new colors and patterns. Over the past 15 years, Barsh and Kaelin have worked closely with Bengal cat organizations and given talks at cat shows. They often return ancestry and genetic data to owners to help guide their breeding.
"Breeders are extremely interested in our data," Kaelin said. "They not only want to contribute their cats' DNA but they also want to be involved and help analyze data and hear about our results. It's been a great collaboration and a true example of citizen science."
The researchers say there are lessons to be learned in just how powerful artificial selection can be, as the Bengal cat coats could probably have been selected for without the help of the Asian leopard cat.
"People have this idea that we have to get access to these distantly related animals to breed beautiful individuals and designer animals," Barsh said. "But it turns out all the diversity was already there waiting in the domestic cat genome."
Scientists from HudsonAlpha Institute of Biotechnology, Gencove Inc., University of Bern, and Texas A&M University were also authors of the paper.
Funding for this research was provided by the HudsonAlpha Institute for Biotechnology and the National Institutes of Health (grant AR082708).