A Blog by

How a quarter of the cow genome came from snakes

Genomes are often described as recipe books for living things. If that’s the case, many of them badly need an editor. For example, around half of the human genome is made up of bits of DNA that have copied themselves and jumped around, creating vast tracts of repetitive sequences. The same is true for the cow genome, where one particular piece of DNA, known as BovB, has run amok. It’s there in its thousands. Around a quarter of a cow’s DNA is made of BovB sequences or their descendants.

BovB isn’t restricted to cows. If you look for it in other animals, as Ali Morton Walsh from the University of Adelaide did, you’ll find it in elephants, horses, and platypuses. It lurks among the DNA of skinks and geckos, pythons and seasnakes. It’s there in purple sea urchin, the silkworm and the zebrafish.

The obvious interpretation is that BovB was present in the ancestor of all of these animals, and stayed in their genomes as they diversified. If that’s the case, then closely related species should have more similar versions of BovB. The cow version should be very similar to that in sheep, slightly less similar to those in elephants and platypuses, and much less similar to those in snakes and lizards.

But not so. If you draw BovB’s family tree, it looks like you’ve entered a bizarre parallel universe where cows are more closely related to snakes than to elephants, and where one gecko is more closely related to horses than to other lizards.

This is because BovB isn’t neatly passed down from parent to offspring, as most pieces of animal DNA are. This jumping gene not only hops around genomes, but between them.

This type of “horizontal gene transfer” (HGT) is an everyday event for bacteria, which can quickly pick up important abilities from each other by swapping DNA. Such trades are supposedly much rarer among more complex living things, but every passing year brings new examples of HGT among animals. For example, in 2008, Cedric Feschotte (now at the University of Utah) discovered a group of sequences that have jumped between several mammals, an anole lizard, and a frog. He called them Space Invaders.

The Space Invaders belong to a group of jumping genes called DNA transposons. They jump around by cutting themselves out of their surrounding DNA, and pasting themselves in somewhere new. They’re also relatively rare—they make up just 2 to 3 percent of our genome. BovB belongs to a different class of jumping genes called retrotransposons. They move through a copy-and-paste system rather than a cut-and-paste one, so that every jump produces in a new copy of the gene. For that reason, they spread like wildfire.

BovB was first discovered in the genomes of cows and other cud-chewing mammals in the 1980s, and seemed to be a signature of that group. Then, in the 1990s, Dusan Kordis and Franc Gubensek detected an extremely similar version of BovB amid the genes of the horned viper. It looked like this piece of DNA had jumped between species. Now, with complete genomes of the cow and other animals at hand, Walsh has more fully mapped BovB’s voyage through the animal kingdom.

A family tree of BovB sequences

Early on during its travels, BovB split into two major lineages. One group has made its way between a single lizard – the Lord Howe Island gecko –and the horse, the egg-laying platypus and echidna from Australia, and African mammals like the elephant and hyrax. The other group started off in lizards and snakes, and jumped from there into ruminants like cows and sheep, and marsupials like possums and wallabies.

To make sure that their results were real, the team took care to avoid contaminating samples from one animal with the DNA of another. Several different laboratories were involved, and no single one of them was responsible for collecting or processing all the samples. And if contaminated samples were producing phantom sightings of BovB, there’s no reason why those sightings would cluster in particular groups, like lizards or marsupials.

To the team, the best explanation for these bizarre patterns is that BovB jumped between species, and it must have done so at least 9 times during its history—far more than the one or two jumps that other scientists had envisaged.

How did it manage? Walsh found a huge clue when she discovered BovB in the genomes of two tick species, both of which suck the blood of lizards and snakes. Other related ticks bite mammals too, so it’s possible that by biting their way through the animal kingdom, these bloodsuckers inoculated fresh branches of the tree of life with jumping genes.

Many scientists who work in this field have suggested that parasites, including worms, bugs, and viruses, could act as vehicles for hitchhiking genes. Indeed, in their very first paper on the BovB, Kordis and Gubensek said that ticks might be spreading the gene between animals. “It’s a scenario that remains hard to prove, but [Walsh’s] data are as close to a smoking gun as can be in the field,” says Feschotte.

A tick, by the CDC. Not quite the same species as the one that Walsh identified, but closely related.

Once it lands in a fresh genome, BovB’s ability to spread throughout its new home seems to vary greatly from one animal to another. Between 10 and 11 percent of the cow genome consists of BovB,and a further 14 to 15 percent are descendants of this jumping gene. BovB accounts for 15 percent of a sheep’s and 11 percent of an elephant’s, but horses, sea urchins and zebrafish only have a few copies each.  “Some genomes seem to have provided really favourable environments for BovB to take off, such as ruminants, but others have not, such as the horse,” says David Adelson, who led the study. “I don’t yet have a hypothesis to test to explore this.”

“BovB appears to have no function other than its own replication,” says Adelson. In one case, it seems to have been incorporated into a cow gene, but otherwise, these sequences don’t seem to do anything. Nonetheless, they’re often so common that they must have had some influence.

“The inescapable conclusion from this and a plethora of other recent studies is that horizontal DNA transfer has been a powerful engine of animal genome evolution, much like it is in bacteria,” says Feschotte. “The main difference being that while bacteria swap genes, animals swap transposons.” Adelson adds: “Despite public concern over the transfer of genetic material to create genetically modified organisms, it appears that Mother Nature has been quietly shuffling genomes for some time.”

Reference: Walsh, Kortschak, Gardner, Bertozzi & Adelson. 2012. Widespread horizontal transfer of retrotransposons. PNAS http://dx.doi.org/10.1073/pnas.1205856110

More on jumping genes:

21 thoughts on “How a quarter of the cow genome came from snakes

  1. It’s kind of neat to look at the genomic tree with regards to the Australian creatures that have picked up their own variants of BovB – South America and Australia were both part of Gondwana, which is why there are still marsupials found in South American (and North America in the form of opossums.) While most of the possums (and naturally wallabys/wallaroos) shown in the tree are Australian, monodelphis domestica is a south american one (so far as I know it doesn’t show up in Oz.)

    Additionally, the platypus/echidna variant shows as being closely related to the one in some African animals. Again, this matches with the history of Gondwana fairly well.

    The implication (to my untrained eye, anyways) is that the split between the versions found in monodelphis domestica (just noticed it shows up twice) and the versions found in other kangaroo/australian possums would have occurred after the split from Gondwana; the same would be true of the variant found in platypi (pus? pusses?) and the african animals.

    I love how things in one field of science can and often do reinforce findings in other fields. It’s so much fun, a giant interconnected web of awesome.

  2. “Despite public concern over the transfer of genetic material to create genetically modified organisms, it appears that Mother Nature has been quietly shuffling genomes for some time.”

    If Adelson is implying that this is a reason for the public not to be concerned, he is ignorant, confused, and/or intellectually dishonest. GM involves coding for specific and potent function, it is widespread over a very short time period, it bypasses natural selection that limits the spread of lethal pathogens, and it has significant legal consequences due to patent lawsuits and so on. Adelson’s comment is akin to a global warming “skeptic” talking about “natural cycles”.

  3. mk> “GM involves coding for specific and potent function” GM involves selecting for a single gene that people have been studying, the alternative is to just mash two organisms together and see what happens or alternatively bombard a plant with chemicals or radiation and see what random things happen, the offsprings of these sledge hammer lucky dip approaches can be sold as organic food, when the precise GM approach can not, there is nothing about GM that makes it more potent. “bypasses natural selection that limits the spread of lethal pathogens”, why would nature stop the spread of lethal pathogens, any pathogen that can successfully reproduce will survive. “significant legal consequences due to patent lawsuits” this is a problem with the legal system no GM and affect non GM crops ass well, companies will often produce crops that can not themselves produce offspring which stops farmers stopping farmers from using seeds from that years crop for next years.

  4. mk> it seems like you’re the one who is ignorant. Genetic modification is strictly controlled to provide the best results for everyone, such as the golden rice project to stop the millions of deaths caused by vitamin A deficiency. Its fear mongering nuts like yourself that stop progress in its tracks. Also, do you want to give some evidence to back up your claims? Oh wait, they’re baseless. get informed.

  5. Folks, impassioned debate is fine. Personal abuse is not. Conduct yourselves accordingly. Please don’t make me have to moderate this thread.

  6. Really interesting! Though I don’t understand how the jumping gene, brought by the tick in the bloodstream of the cow, can manage to get into a gamete, its only way to be passed on to the next generations…

  7. I think it is fascinating that Nature has been engaging in genetic modification in an unregulated and random fashion since life emerged on Earth. Without it evolution doesn’t happen. Please don’t misconstrue this fascination of mine or the underlying facts for a desire on my part to see unregulated GMOs in the wild. However many people solely or principally reject GMOs on grounds that they are not “natural”, and given the facts I would hope they might question that conviction. My 2 cents.

  8. @Cmdr Awesome: The split between the Afrotheria/Monotreme/Equid BovB and the Reptile/Tick/Ruminant BovB certainly got us thinking about potential phylogeographic explanations revolving around the breakup of Gondwana, however I think we need more data, particularly from birds and ticks or mites that might travel on migratory birds before we can come to any firm conclusions about what might have happened.

    @Raphael: Yes, this is the $64 question. I suspect the involvement of an arbovirus that might allow BovB to hitchhike and infect germ line cells in a new host. But I am no virologist and this is a really difficult thing experimentally test.

  9. So essentially, the next time I’m bitten by a blood-sucking insect, I should bear in mind that in addition to sucking out my life-blood and giving me malaria, it may ALSO be fundamentally altering my DNA to incorporate a bacterial symbiote it picked up off a squirrel or something.

    Brb buying bug-spray.

  10. Well, this is certainly a very interesting PROPOSED case of horizontal transfer and the results are very interesting.

    However I dispute the statement you make “The only other explanation for these bizarre patterns is that BovB jumped between species, and it must have done so at least 9 times during its history—far more than the one or two jumps that other scientists had envisaged.”

    This is without a doubt not the only other explanation. I can think of many things worthy of testing here as other possibilities including ancestral polymorphisms, convergent evolution, random noise (if you analyze 1000s of genes in organisms and a few show unusual patterns in phylogenetic trees is seems reasonable to conclude that this could just be statistical noise and not signal of any evolutionary event), ancient duplication of transposable element families and misidentification of orthology/ paralogy, bad sequence alignments, inaccurate phylogenetic tree inference, and many more.

    I am NOT saying the authors theories are wrong. But this is a somewhat extraordinary claim. And I think it requires one to test ALL reasonable theories to explain the pattern, not just one or two.

  11. Ah. GMO’s are mentioned
    For both sides of my 2 cents here we go. The proponents of GMO’s can prove their point if they open a restaurant that only serves GMO food, “better” food that has protein with three times as much essential amino acids, no toxins, no pesticides, high omega-3’s and full of golden vitamins, no carcinogens, etc, and flavor.. This is probably possible and this is the real potential of genetic engineering. Lots of traits can be inserted in tandem. In a short time the restaurant would be a big hit.
    It is high time the engineers put up.
    The restaurant could be named GMO for genetic engineered only.

  12. @David Adelson – Definitely more data would be needed. I’d not really thought closely about tick infestations carried on migratory birds but given the enormous flight range some of them have, I can see that being a viable transition vector.

    From a layman’s point of view (namely mine) the whole situation dovetails so nicely w/continental drift, that I just got a bit carried away with that idea.

  13. I was left with two questions I wonder if you might briefly answer:

    1. What is the particular function of BovB that its usefulness is limited to a single species (i.e. snakes or cows)? Is it inert in any case, or is there a cross-species function?

    2. What is the line of division between repetition that represents pointless or harmful aggregation, vs. useful duplication (e.g. backup)?

  14. Forgive me – I have no training in biology. But I have a couple of questions.

    First – when you say that “between 10 and 11 percent of the cow genome consists of BovB,and a further 14 to 15 percent are descendants of this jumping gene,” does this mean that the stated percentages represent the percentage of DNA coding represented by these genes in each member of a population, does the percentage represent a sort of weighted average over the DNA information represented throughout the individuals of a species, or is there another meaning to these percentages? If the jumping gene information is present in substantially all members of a given species it would seem to have had to jump into a common ancestor at a time determinable based on analysis of the variation of this material throughout the population. Has this been done?

    Is there a way to determine whether a particular portion of DNA actually has no effect on an organism or is this simply an educated guess? Is it possible to decide if a portion of coding is inert by detecting a different rate of change through mutation (i.e., more mutations may remain within useless info as any change that is not actively harmful would seem to be preserved over time as opposed to changes to consequential genes where overtly harmful changes would be eliminated as well as changes which reduced efficacy or eliminated efficacy)? Thank you – the article is fascinating. If these questions aren’t a waste of your time I have more.

  15. Very interesting work here. Makes me think of a question I never asked in my biology class:
    If C.elegans behavior can be affected by eating E.coli bacteria expressing a silencing DNA structure (RNAi), should I conclude that MY DNA profile – or transcriptome in worst case – potentially can be changed by eating food. Any, normal, regular, daily food?

  16. Not many people know that cows (and other farm animals) eat meat. Living on a farm as a child I saw cows eat mice , rats and snakes. It is not common but it does happen.
    A quick search on the web I found this story from 1910 in Australia


    MORTLAKE, Wednesday-An unusual sight was witnessed at Framlingham on Tuesday by two men engaged in erecting fences they killed a snake in the morning and hung it on the fence Some few hours afterwards they saw a cow chewing a snake, some eighteen inches of which was hanging out of the animal’s mouth.

  17. The same ticks biting both cows and snakes, as well as cows occasionally eating snakes, are relatively small amounts of biological material transfered, and yet it has such a dramatic DNA impact. Then imagine how much DNA must be transfered between different cells in the same body! Goodbye, Weismann’s barrier, and welcome, soma-to-germline gene transfer.

  18. I found this article to be Fascinating! Have any advancements been made since?
    Have any correlations between plant or fungus matter been found between snakes/cows?

  19. I’m with Mr. Eisen, it’s a good study and a valid working theory, but we should be open to other possibilities. There doesn’t seem to be a good explanation for the jump between land animals and “purple sea urchin, the silkworm and the zebrafish.” Too long we’ve taken the simplistic view about many things that “The obvious interpretation is that [X] was present in the ancestor of all of these animals, and stayed in their genomes as they diversified.” Besides HGT, we know of many factors that affect genomes that were unimagined when it was first discovered that the was a DNA code that determined our biological heredity. Now we are learning that there are epigenetic factors that are very important as well. A lot of DNA that was consigned to the “Junk” drawer have been found out to have purpose, in a way that we hadn’t thought DNA acted. We’ve also learned that viruses can make themselves part of our genes (ERVs). I have a feeling some things, like this case, might not involve transfers of any sort, but simply be “random” changes which are more likely to occur with certain DNA code sequences, unrelated to what they code for but purely a matter of the base sequences and the copying / checking / repairing processes. Maybe if we test more things and find more organisms having variations of the sequence and different variations, we’ll have to look for some other explanation. Then again, once a theory gets started, it can be “adjusted” to fit new data a long way…

Leave a Reply

Your email address will not be published. Required fields are marked *