by Jean Hofve, DVM

Purina was the first to trumpet the results of an letter in the journal Nature reporting an in-depth study of the genetics of dogs vs. wolves. Their findings suggest that dogs have adapted their DNA to be able to digest and assimilate a high-starch diet. Purina is all about how wonderful corn is for dogs; they (and other big pet food makers) must be ecstatic about these findings.

However, Purina’s version isn’t the quite the same as what the researchers actually found.Before the oh-boy-let’s-feed-more-grains movement gets too much impetus from this study (which is not even in print yet!), I thought I would share my thoughts about it (but with the caveat that I am far from up-to-date on genetics and statistics!).
First, let’s define a couple of terms:
  • Gene: a sequence (area or region) of DNA that determines a particular characteristic, such as gender or coat color. 
  • Allele: a different version of a gene. An organism inherits two alleles of every gene, one from each parent. 
  • Genome: the entirety of hereditary information (DNA) of an organism. The dog’s genome has been sequenced (decoded); it contains about 19,000 genes. 
  • Genotype: the genetic makeup underlying a trait, cell, or individual. 
  • Phenotype: the physical manifestation of an organism’s DNA. This includes things like size, appearance, and heritable diseases. Phenotype is determined by genotype, but because an organism inherits DNA from both parents, the offspring’s phenotype will exhibit a mixture of the parents’ characteristics.

This seems to be a sound study using appropriate methods and analysis. For the first analysis, the researchers used pooled blood samples from 60 dogs of many breeds and 12 wolves from Europe and North America. They also examined the genotypes of 72 dogs of 38 breeds, and 19 wolves; as well as another sample of 507 dogs and 17 wolves. (It is not clear how much overlap there was among these groups.) With these numbers, the statistical power of the analyses should be good and the results reliable.

The authors also provide massive amounts of supplementary information describing all the technical data from the DNA analyses, as well as notes on the dog breeds used and original of the wolves. (One of the authors confirmed that DNA and serum samples came from wild wolves, whereas pancreas samples were obtained from captive wolves in zoos. This is important because there are known biochemical differences between wild and captive wolves.)

In simple terms, the researchers found that there were more copies of 36 different genes in the dog than in the wolf, suggesting that these genes were “selected for” during domestication. Ten of those genes involved digestion, and three specifically involved starch digestion.
  • Dogs had 4-30 copies of the gene for pancreatic B-amylase, wolves only 2. This suggests that dogs born with additional copies were more successful in the domestication process, and their genes were more likely to be passed to their offspring.
  • The maltase pathway had one or two nucleic acid substitutions in two genes.
  • A major glucose uptake gene had one nucleic acid substitution that was present in all dogs in the sub-sample, but no wolves.

Some DNA regions were “fixed” in the dog; these areas are present in every dog, and are no longer subject to change within the dog population. 

In the study, mention of the increase in copies of starch-related genes was last on the authors’ list, though they did several more tests to pinpoint the changes and thus discussed it more extensively. Alterations in the nervous system (122 affected genes) and reproductive system (19 affected genes) were also found.

These changes were similar to those taking place in the Siberian fur farm fox study, in which silver foxes were selected for tameness and selectively bred. Over many generations, tame foxes had significant alterations in behavior and reproduction, as well as changes in physical appearance.  Forty genes differed between the tame and other farm foxes; but the difference between farm foxes and their wild cousins involved some 2,700 genes.

Limitations of the study

There is broad agreement today that dogs are descended from wolves, period. No jackals, no coyotes, just wolves. Dog and wolf DNA are 99.8% identical, despite the major differences in looks and temperament.

Dog breeders, of course, try to minimize genetic diversity by breeding to a set standard of size, color, and other traits. However, it’s hard to selectively breed for one trait without inadvertently dragging other nearby genes along. Compounding the situation, many breeds suffer from the “Founder Effect” that occurs when the new breed is established with only a few individuals; or from a “bottleneck” that wipes out all but a few individuals of that breed. The result is a loss of genetic diversity, and amplification of  traits from the original or remaining breeding stock–whether good or bad. These issues are at the root of why there are so many heritable diseases in today’s dogs, such as hip dysplasia and heart defects. 

There are four breed groups based on clusters of alleles:

  • dogs of Asian and African origin:
  • herding dogs and sight hounds
  • modern hunting dogs such as terriers, hounds and retrievers
  • large mastiff-type dogs.

In addition, 14 breeds have been identified by DNA analysis as “ancient” or “basal.” All other breeds are thought to derive from them:

  • Afghan Hound
  • Akita
  • Alaskan Malamute
  • Basenji
  • Canaan
  • Chow Chow
  • Dingo
  • Finnish Spitz
  • New Guinea singing dog
  • Saluki
  • Samoyed
  • Shar-Pei
  • Shiba Inu
  • Siberian Husky 
While a variety of dog breeds were represented in this study, they were all from Europe or North America. They included Samoyeds and Shar Peis (along with many other breeds), but no other basal breeds. The genetic variation between breeds is about 27.5%; in contrast, the biggest difference between human populations is just  5.4%.

While all dogs evolved from wolves, they are thought to have descended from several geographically distinct wolf populations. Since there is so much genetic variation between dogs of different breeds, and much of it is distinguishable among the four major groups, we just don’t know if the three genes described in this study are altered in all dogs.

The Big Question

Even if we accept the idea that expression of these these three genes is increased in all dogs, the big question remains: is this a good thing or a bad thing for dogs?

The increased ability to digest starch undoubtedly helped early dogs out-compete wolves for survival in the early days of human agriculture and settlement. It helped them survive as humans dispersed across the globe, and as humans’ diet became more and more plant-based. Across the ages, meat became a luxury, much more accessible to the rich than the peasant. Most humans have subsisted primarily on diets based on wheat, corn, or rice for thousands of years. And their dogs have clearly adapted to the changes.

But there is no denying that dogs today, particularly purebreds, are subject to hundreds of heritable diseases and disorders that wolves are not. In humans’ quest to create dogs designed for particular functions or looks, the starch-related genes obviously came along for the ride; but what might have they brought with them?

Just because dogs can utilize starch better than wolves, does that mean that a high-starch diet is better for them than an ancestral wolf-type carnivorous diet? That, too, is unanswered in this study. Nevertheless, this is what pet food companies will be claiming.

(And if you’re wondering about cats…I think they’re safe, for now. Modern research on feline metabolism has tended to confirm their stubborn status as obligate carnivores! Sorry, Purina!)
Axelsson E, Ratnakumar A, Arendt ML, et al. The genomic signature of dog domestication reveals adaptation to a starch-rich diet. Nature. 495:360–364  doi:10.1038/nature11837 (http://www.nature.com/nature/journal/vaop/ncurrent/full/nature11837.html)