[Originally published in October 2012 and republished here following Labour MP Trevor Mallard’s desire to see the Moa resurrected]
Moa birds disappeared from New Zealand following the arrival of human settlers in the 13th century, but their fossils now provide us with a valuable clues about long-term DNA survival and how DNA decays over thousands of years.
In a paper published today in Proceedings of the Royal Society B, we show that fragments of DNA from these extinct, flightless birds can be used to estimate how DNA decays over time. The findings have implications in choosing where to look for DNA-containing fossils and in forensic casework involving bones.
Sadly, the findings won’t satisfy those people wanting us to talk about dinosaur DNA. When telling someone our day-job involves working with ancient DNA (aDNA), more often than not, the response is: “Ahhhh, you mean like Jurassic Park?”.
Then we explain how resurrecting the dinosaurs is not possible and that the early papers describing isolation of dinosaur DNA were likely based on a combination of contaminating human and chicken DNA.
In other words, when the early aDNA researchers cranked up the DNA “photocopier” they copied DNA from themselves, or what they had for lunch, rather than dinosaurs. A number of high-profile (never retracted) publications citing dinosaur-era DNA, coupled with Michael Crichton’s famous novel, gave aDNA researchers a serious headache for a while.
How far back?
So let’s return to the conversation about our day-job. On explaining that dinosaur DNA is a pipe dream, the next question tends to be: “How far back in time can you go, then?”.
That’s not an easy question to answer. Questions of DNA survival and rate of decay have been central for a long time to scientists engaged in aDNA research, and also in forensic cases where old and degraded DNA is encountered.
But as yet, it has proven extremely difficult to predict DNA survival, and there has been little progress towards understanding the rate of DNA degradation.
It was shown in the early 1970s that free DNA molecules in a solution would fragment and “disappear” over time, but this was measured in a controlled laboratory setting under artificial conditions and therefore didn’t resemble the conditions in fossil bone.
Researchers have since attempted to predict the decay rates of DNA and establish theoretical survival times, but never very convincingly, largely because there has been a lack of empirical data on this topic.
Why has it been so difficult to assess? Well, mainly because fossils with good DNA preservation are rare. This means large comparative studies on this topic are rare too.
Moreover, many studies that have addressed this question have included samples from a range of different burial environments which vary in soil conditions, temperature, and so on. All this variation results in a high degree of variability in DNA preservation that can mask any true signal of DNA decay over time.
Our study was different in that we had access to a very large number of moa fossil bones from sites within 5km of each other and which importantly were all subject to similar environmental conditions.
With funding from the Marsden Fund of the Royal Society of New Zealand we were able to uniformly drill small holes in hundreds of moa bones and radiocarbon-date them.
The upshot is that the samples, for the first time, gave us a very solid framework in which to investigate the question of DNA decay.
With this unusual setting we could show that DNA does indeed degrade at a certain rate and it therefore makes sense to talk about a half-life of DNA. That is, the length of time required before half the original (starting amount) of DNA is left.
In moa, for the targeted DNA fragment, we were able to estimate a half-life of 521 years and propose a rough, general model that estimates DNA decay at certain temperatures.
In a perfect world, this would mean one could predict the amount of DNA remaining if the age of a bone is known, or use it as a tool to assess whether it is worthwhile drilling into a valuable fossil collections in the search for aDNA. But our research also shows that things are, unfortunately, not that simple.
Despite demonstrating the overall relationship between age and DNA preservation, it is evident that time alone can still only explain about 40% of the variation in DNA content in these old moa fossils.
So although all these fossils were preserved under the same temperature and roughly similar environmental conditions it is clear other factors need to be considered if we want to try and reliably predict DNA decay. In that sense, our study is just the first step on the path to better predictive models.
Beating the survival record
Our study shows that DNA in bone seems to decay at a rate that is almost 400 times slower than previously measured for DNA in solution.
Combined with the fact modern DNA sequencing technology allows us to target very short fragments of DNA, it seems likely future research will identify DNA from permafrozen bone that is considerably older than the current DNA survival record of about half a million years from Greenlandic ice cores.
Under the best possible scenario, discovering a million-year-old fossil DNA sequence seems like a very real possibility. While this is a long way short of 65-million-year-old dinosaur DNA, it would still represent a considerable achievement.
To illustrate the extreme improbability of isolating authentic DNA fragments from 65-million-year-old dinosaur bones, our model predicts, under extremely favourable conditions, all “letters” in the DNA code (your genome has 3 billion of them) within bone would be broken after 6.8 million years.
In other words, the last break to separate the final two bases in the DNA code occurs after a maximum of 6.8 million years – and this is indeed a highly optimistic estimate.
In addition, we know that much longer DNA fragments are required in order to be sequenced and meaningful, and such fragments would be gone a long time before the 6.8 million year mark.
So dinosaur DNA is still science fiction … but then again, we all know now that birds are actually (theropod) dinosaurs, so in that sense it is easily argued our entire study is based on DNA from extinct dinosaurs.
Michael Bunce receives funding from the Australian Research Council and The Royal Society of New Zealand (Marsden Fund).
Morten Erik Allentoft receives funding from the Royal Society of NZ (Marsden Fund).