Humans and their best friend may share ERVs.[1]

My first thought was ‘is 10,000 years long enough?’


(By James Jun, source: Flickr)

Let me explain.

Firstly: ERVs are endogenous retroviruses.



Our genomes are DNA-based. Retroviruses are viruses with RNA-based genomes that, once they’ve infected a cell, create a DNA copy of their genome that they insert into the genome of their host. That’s us, in the case of human endogenous retroviruses, or HERVs.

Endo = within.

HERVs are found within our genomes. They have become part of us, they’re replicated and passed onto our children along with the rest of our DNA. To be passed on through generations, the virus had to have infected our germ cells sometime in the past.

Almost all ERVs no longer function as viruses because they are only remnants or are repressed.

Dogs have ERVs, too. The authors of the research I’m covering here have called them CfERVs: Canis familiaris ERVs. Canis are the dogs – wolves, coyotes, jackals included. familiaris is the one we know well. *Woof* :-)

The dog genome–from a female boxer, for what it’s worth–was examined for ERVs by a Swedish group of scientists lead by Göran Andersson. It turns out that dogs have about one-sixth of the number of ERVs than humans do, suggesting dogs are better at suppressing or eliminating them than we are.[2]

One group of CfERVs looked similar to a HERV, a human ERV (HERV-Fc). Could this have been shared between dogs and humans over the 10,000-odd years they’ve been hanging out together? The authors write in their abstract:

A novel group of gammaretrovirus-like CfERV with high similarity to HERV-Fc1 was found to have potential for active retrotransposition and possibly lateral transmissions between dog and human as a result of close interactions during at least 10.000 years.

10,000 years is a blink of time when you’re thinking about evolution. Active retroviruses evolve very fast, but once they become endogenous–become part of the host genome–they ought to change at the same mutation rate as the other genes of the host, which isn’t all that fast.

I have to admit I’m nonplussed by the reference to 10,000 years–an estimate of how long dogs have been domesticated–in the abstract and discussion of the paper, as the estimates of the age of the common ancestors of the dog and human HERV-Fc that Göran’s team report within the body of their paper are much older, millions of years old. I’m left not knowing quite what to make of it. The thought that horizontal transfer of ERVs between humans and dogs might have occurred during the period of the domestication of dogs is nice (if a bit cosy), but you’d think their evidence tells them that have to look further back in the past.[3]

Either way, there are suggestions that an ERV previously found only within primates has been seen in dogs, so–in a sense–we share some ERVs with our dogs.


ERV also has an article on this paper.

[1] Ugly, but PC. – man and their best friend.

[2] I’m leaving out a bunch of interesting results about where ERVs are in the dog genome. They’re particularly interesting if you’re interested in epigenetics and genome structure as I am, but they’re part of another story as you can see.

[3] One probably equally questionable speculative idea might be that the ERVs were transferred between the ancestors of primates and dogs up whilst eating, with primates perhaps being the more likely meal than the dogs’ ancestors.


Martí­nez Barrio, Á., Ekerljung, M., Jern, P., Benachenhou, F., Sperber, G., Bongcam-Rudloff, E., Blomberg, J., & Andersson, G. (2011). The First Sequenced Carnivore Genome Shows Complex Host-Endogenous Retrovirus Relationships PLoS ONE, 6 (5) DOI: 10.1371/journal.pone.0019832

Other articles on Code for life:

Autism — looking for parent-of-origin effects

Transcribing a gene, free poster

Finding platypus venom

Thoughts towards a human brain neural connection map

Coiling bacterial DNA