Human evolution used to be the preserve of two groups of academics: the ones who liked fossils and the ones who liked stone tools. Both regarded the other as peculiar for being obsessed with the wrong part of a massive jigsaw puzzle. Then in 1987 the geneticists arrived and they’ve been making things much more untidy ever since…

As recently as 40,000 years ago there were at least four species of hominin living in Eurasia. While we know of three from fossils and archaeological material. The fourth, however, is known almost purely from ancient DNA. ‘Species X’, or more commonly the Denisovans, are helping to re-write our understanding of human evolution during the later Pleistocene. But who were they and were did they go?

Denisova cave, image by Nerika via Wikimedia Commons

The Denisovans are a species of hominin that occupy a very peculiar place in the human evolutionary story. They are the most poorly understood species in relation to fossils (one tiny finger bone, a toe bone and two teeth!) but they also have the most complete genome of any extinct hominin species. Compared with the fossil record of the Neanderthals, which runs to hundreds of individuals the Denisovan fossils are not even a blip on the radar, but when we factor in the data that comes from ancient DNA we can start to see the power of palaeogenomics as an analytical tool.

Denisova is a cave site located in the Altai Mountains in Siberia near the border with China and Mongolia. Both Homo sapiens and the Neanderthals used the site at least as far back as 125,000 years ago.  Excavations in 2008 uncovered a small finger bone which was initially thought to belong to either a Homo sapiens or Neanderthal (often it can be difficult to distinguish between the two species when only very small or fragmentary fossils are found).  Radiocarbon dating of associated archaeological materials suggested an age of around 40,000 years.

The story would have likely ended there had it not been for the amazing developments in the study of ancient DNA and the pioneering work carried out on the Neanderthal Genome by Svante Pääbo and his team at the Max Planck Institute in the field of palaeogenomics (the recovery of ancient DNA from fossils and reconstruction of the genomes of extinct species). The environmental conditions of the cave site (the average annual temperature is 0^oC) meant that conditions for DNA preservation were excellent. The finger bone was analyzed for ancient DNA by the lab of Svante Pääbo. The team was able to initially recover mitochondrial DNA and their results of their analysis turned our thinking about human evolution upside down.

We don’t know very much about what the Denisovans looked like.  The few scanty fossils that have been recovered suggest a robust and well-built hominin that was well adapted to its cold and relatively inhospitable environment. The finger bone is very wide and much thicker than would be found in Homo sapiens. What is especially interesting to note is that the bone comes from a female – which suggests that they were a lot more robust than us but broadly similar to the Neanderthals. But when we look at the teeth they are completely different to both the Neanderthals and ourselves.

One of the extremely rare physical evidence of their existence: a tooth from Denisova cave. Image from Reichs et al. 2010

Yet the archaeological material hints at a much more complicated picture. Stone tools made by Homo sapiens and a toe bone from a Neanderthal reveal this site that was used by three different hominin species and hints at the possibility of being used at the same time by at least two of them. Intriguingly analysis of Denisovan DNA shows genes that are found in modern humans that code for some of the phenotypic traits commonly associated with modern Melanesian populations (e.g. dark skin, brown hair and eyes) but care must be taken when dealing with such superficial phenotypic traits.

DNA analysis suggests that there was a complex pattern of interbreeding between the three species (Homo sapiens, Neanderthals and the Denisovans). The study carried out by Prüfer et al (2014) also identified a 4^th source of DNA, suggesting that there was (at least) one more completely unknown hominin species – which means that Eurasia during the Pleistocene was a rather exciting place to be.

Gene Flow from an Unknown Ancient Population (Image from Birney & Pritchard, 2014).

When we examine the Homo sapiens genome it shows evidence of mating with both the Neanderthals and the Denisovans. The picture is much more complex than at first glace though. Up to 4% of non-African human DNA is from Neanderthals (but from populations found at the join between the Middle East and Eurasia and different from Neanderthal DNA found at Densiova). This means that while we interbred with Neanderthals we only did so after some groups had left Africa.

The Denisovan picture is even more complicated. Research carried out by Rasmussen et al, (2011) compared the Denisovan genome with six modern human samples from the !Kung (South Africa), Nigeria, France, Papua New Guinea, Bourgainville Islands (Solomon Islands archipelago) and Han (China).  The results showed that between 4-6% of the DNA of the genome of the Melanesians (PNG & B. Islands) came from the Denisova species.  Further research has shown that Denisovan DNA is also found in Australian Aboriginal populations.  While Prüfer’s 2013 study showed that mainland Asian and Native Americans both have approximately 0.2% Denisovan DNA. What this tells us is that the Denisovans must have once occupied a large area of Asia and come into contact (very personal contact!) with incoming groups of Homo sapiens.

What is especially exciting about the Denisovan hominin is the relationship it appears to have had with Neanderthals from the same area. There seems to have been at least 0.5% gene flow from the Neanderthals into the Denisovan genome. Furthermore there are a greater number of derived alleles shared by the Denisovan hominin and Denisovan Neanderthals than with Neanderthals from other parts of Europe which hints at a much more complicated pattern of inter-species local relationships than we had previously thought. Basically they kept it local rather than going in for long distance relationships.

But perhaps the single most tantalising genetic clue found in the Denisovan aDNA is a section of genes that come from an unkown hominin that diverged long before the Homo sapiens/Neanderthal/Denisovan split – opening up the possibility that it might be Homo erectus DNA.

So what does it all mean? Human Evolution is not a science that stands still and rapid changes are the norm. New fossil discoveries frequently sweep away old ideas and make us change our thinking. Ancient DNA analysis is opening up a whole new way of thinking about our evolution and how groups we call different species (based on fossils) interacted with each other on a day-to-day basis. The only safe prediction is that the Denisovan hominin won’t be the only new discovery that aDNA will find.

Postscript – A quick guide to ancient DNA

DNA can be recovered from a range of organic materials – bone, teeth, hair and soft tissues but the key to successful recovery is the level of preservation. As a general rule of thumb DNA in bones and teeth is much more likely to be recoverable when the conditions are ‘high and cold’ – contrast this with the warm and wet conditions found in Indonesia which have created the problems finding ancient DNA in the Homo floresiensis (aka Hobbit) remains.  Soft tissues and hair do sometimes survive in archaeological contexts but this is often in very arid and hot environments or in the case of bog bodies when the DNA has been completely denatured by the chemical processes that preserve the body.  Although there have been a handful of remains recovered from permafrost they are extremely rare.  If preservation conditions are optimal DNA can be recovered from approx. 0.25g of dry bone – but it should be remembered that in the field ‘optimal’ conditions are rather like hen’s teeth and contamination and degradation of sample quality is all too common.  Even breathing on bone during excavation could cause the introduction of modern DNA – mixing up the picture produced by later analysis.

Ancient DNA (aDNA) in human evolution is typically used to reconstruct population history. This can be done within a species – such as how Homo sapiens left Africa and dispersed around the world. Or between species such as when last shared a common ancestor with the Neanderthals (Homo heidelbergensis c. 500,000 years ago).  Two forms of ancient DNA are most commonly used: Mitochondrial DNA (mDNA) and Nuclear DNA (nDNA) and both sources have their benefits and problems for the researcher.

mDNA is maternally inherited and can be used to reconstruct the female lineage (the Y Chromosome can be used in a similar way to reconstruct the male lineage). mDNA is not subject to recombination during reproduction and is not acted on by natural selection meaning that mutations are built up over large timespans and can be used to reconstruct the maternal relationship. mDNA is also extremely prevalent in our body – each cell has hundreds of copies of our mDNA compared to only two copies of nuclear DNA per cell. It is also relatively small compared to our nDNA – just under 17,000 base pairs in a closed circular genome that gives greater preservation after death.

nDNA by comparison has billions of base pairs, is contained in a open ended double helix strand and is modified by meiosis. Because of the structure of nDNA it is relatively unstable after death, which can create problems when trying to recover samples from fossils. As a general rule the best conditions for DNA preservation in fossils are ‘high and cold’ which is what makes the Denisova site so well suited for aDNA recovery. Techniques for the analysis of aDNA are improving at a very rapid pace that opens up the possibility that older and older material might be recovered in the future.

Written by Simon Underdown (@sunderdown)

Edited by Ross Barnett (@DeepFriedDNA)

Further Reading:

Birney, E., & Pritchard, J. K. (2014), ‘Archaic humans: Four makes a party’, Nature, 505(7481), pp,32–34. [Abstract only]

Cann, R. L, Stoneking, M, & Wilson, A. (1987), ‘Mitochondrial DNA and human evolution’, Nature. 325(6099). pp.31-36. [Abstract only]

Cooper, A & Stringer, C. B, (2013), ‘Did the Denisovans cross the Wallace line?’ Science. 342(6156). pp.321-323. [Abstract only]

Gibbons, A. (2012), ‘Ancient DNA. A crystal-clear view of an extinct girl’s genome’, Science. 337(6098). pp.1028-1029. [Full article]

Houldcroft, C. & Underdown, S. (2015), ‘Neanderthal Genomics Suggests a Pleistocene Time Frame for the First Epidemiologic Transition’. http://biorxiv.org/content/early/2015/03/31/017343

Krause, J, et al. (2010), ‘The complete mitochondrial DNA genome of an unknown hominin from southern Siberia’, Nature. 464(7290). pp.894-897. [Abstract only]

Pääbo, S. 2014. Neanderthal Man: In Search of Lost Genomes. London, Basic Civitas Books. [Book]

Pennisi, E, (2013), ‘More genomes from Denisova Cave show mixing of early human groups’, Science. 340(6134). pp.799. [Abstract only]

Prüfer, K, et al. (2013), ‘The complete genome sequence of a Neanderthal from the Altai Mountains’, Nature, 505(7481). pp.43-49. [Abstract only]

Rasmussen, et al, (2011), ‘An Aboriginal Australian genome reveals separate human dispersals into Asia’, Science. 334(6052). pp.94-98. [Abstract only]

Reich D,  et al. (2010) ‘Genetic history of an archaic hominin group from Denisova Cave in Siberia’, Nature. 468(7327). pp.1053-1060.[Full Text]

Reich, D, et al. (2011), ‘Denisova Admixture and the first modern human dispersals into Southeast Asia and Oceania’, The American Journal of Human Genetics. 89. pp.516-528. [Abstract only]