You are here

Sequencing Neanderthal DNA

Challenges in Extracting Ancient DNA

Working with ancient DNA is very challenging, both in terms of finding sufficient material to work with after decomposition has occurred, and in terms of eliminating modern human contamination. Distinguishing between modern human and ancient genetic material is particularly difficult when the ancient DNA comes from close relatives of modern humans.

Organisms decompose after death. Water, oxygen and microbes break down DNA. Within 100,000 years, all DNA is destroyed. Ancient DNA tends to be found in small quantities. The DNA that is extracted is generally fragmentary and damaged. Some damage results in changes to the DNA sequence. Cytosine can change to uracil, which is read by copying enzymes as thymine, resulting in a C to T transition. Changes from G to A also occur. DNA errors are very common at the ends of molecules.

Contamination by modern DNA is a particularly difficult problem to solve. Labs and chemicals may be contaminated by the DNA of the people working in them, while many fossils have been handled by researchers for years. Contamination is difficult to detect because Neanderthals and humans share much of their genetic material, making some DNA sequences indistinguishable. Researchers have developed ways to analyze the results of ancient DNA sequencing efforts to determine whether contamination is likely and how much has occurred. Analysis of the results and efforts to keep labs and specimens free of modern DNA is very important as some researchers believe that the early studies of Neanderthal DNA included modern contaminants.


Sequencing the Complete Neanderthal Mitochondrial Genome

After successfully sequencing large amounts of DNA and devising strategies to deal with potential contamination, a team led by Svante Pääbo from the Max Planck Institute, reported the first complete mtDNA sequence for a Neanderthal (Green et al. 2008). The 0.3 gram sample was taken from a 38,000 year old Neanderthal from Vindija Cave, Croatia. Complete Neanderthal mtDNA sequences give researchers more information about the relationship between modern humans and Neanderthals, as well as information about Neanderthal population size.

The complete mtDNA sequence shows that Neanderthals were outside the range of modern human mtDNA variation. Researchers compared the mtDNA sequence with that of modern humans. They compared sequence changes that resulted in nonsynonymous amino acid changes with synonymous changes. They found a larger number of nonsynonymous changes in the Neanderthal lineage, possibly implying that Neanderthals had a small population size with weaker purifying selection (Green et al. 2008).

Map showing range of Neanderthals prior to the entry of Homo sapiens into Europe

Later, Svante Pääbo’s lab sequenced the entire mitochondrial genome of five Neanderthals (Briggs et al. 2009). Sequences came from two individuals from the Neander Valley in Germany, Mezmaiskaya Cave in Russia, El Sidrón Cave in Spain and Vindija Cave in Croatia. Though the Neanderthal sample comes from a wide geographic area, the Neanderthal mtDNA sequences were not particularly genetically diverse. The most divergent Neanderthal sequence came from the Mezmaiskaya Cave Neanderthal from Russia, which the oldest and eastern-most specimen. To look at whether age or geographic location contributed to genetic differences, the team sequenced part of the DNA of another Mezmaiskaya Cave Neanderthal that dated to 41,000 years ago. This more recent specimen grouped with the other Neanderthals, possibly showing that age was the cause of the sequence differences (Briggs et al. 2009). Other studies show the existence of eastern, western and southern groups of Neanderthals (Fabre et al. 2009).

On average, Neanderthal mtDNA genomes differ from each other by 20.4 bases and are only 1/3 as diverse as modern humans (Briggs et al. 2009).  The low diversity might signal a small population size, possibly due to the incursions of modern humans into their range (Briggs et al. 2009).

Neanderthal Fossilized Skull from La Ferrassie, France


Sequencing the Neanderthal Nuclear Genome

Recently, there have been efforts to sequence Neanderthal nuclear genes. Two studies, one by Svante Pääbo’s team and one by Edward Rubin, have sequenced large amount of Neanderthal nuclear DNA using different methods. Their results were announced in 2006. Given their success in sequencing some nuclear DNA, both labs launched projects to sequence the entire Neanderthal genome. Nuclear genomic sequences from Neanderthals show differences between modern humans and Neanderthals, and illustrate aspects of Neanderthal biology.  


One Million Base Pairs of the Neanderthal Sequence

Svante Pääbo’s team from the Max Planck Institute for Evolutionary Anthropology in Germany announced the sequencing of one million base pairs of nuclear DNA of a Neanderthal specimen in 2006 (Green et al. 2006). After a long search for specimens with a sufficient amount of undamaged DNA to sequence and for the ones with the least evidence of contamination, they focused on Vindija 80, a Neanderthal discovered in Croatia in 1980 that is approximately 38,000 years old.

They estimated that 7.9% of the changes in human DNA compared with that of the chimpanzee occurred after the split with Neanderthals. They dated the split between the ancestors of modern humans and Neanderthals to 465,000 to 569,000 years ago. They also found that the effective population size of the Neanderthals was small. Their success in sequencing this amount of DNA indicated that a large-scale project to sequence the Neanderthal genome is possible.


Rubin's Neanderthal Nuclear DNA

Edward Rubin’s team from the Lawrence Berkeley National Laboratory in California also sequenced Neanderthal nuclear DNA (Noonan et al. 2006).  They sequenced about 65,000 base pairs from the 38,000 year old Vindija, Croatia specimen. The technique used here produces a copy of the Neanderthal sequence that can be retained forever, reducing the need for repeated destructive sampling. The DNA is then cloned in bacteria.

The average split time between the Neanderthal and modern human populations was around 370,000 years ago.  They used the sequence to look at the possibility of interbreeding between Neanderthals and moderns. Admixture would be seen as derived alleles that are found in Neanderthals and in low frequencies among modern humans. They did not detect this in their sample. A simulation to test the Neanderthal contribution to the human genome found a 0% chance of Neanderthal input with a 0% to 20% confidence range. With this data, the authors cannot definitively rule out admixture (Noonan et al. 2006).

Some aspects of the two sets of nuclear DNA do not fit together, possibly because of contamination and sequencing errors, especially in the Green et al. (2006) study (Wall and Kim 2007).  This has led the researchers to develop new methods of detecting and preventing contamination to ensure that only ancient DNA is being sequenced.


A Draft Sequence of the Neanderthal Genome

In 2010, Svante Pääbo’s lab announced a draft sequence of the Neanderthal genome (Green et al. 2010).  This new study has produced evidence consistent with interbreeding between Neanderthals and anatomically modern Homo sapiens and points to aspects of the human genome that may have changed since the split between humans and Neanderthals.

DNA was extracted from three Neanderthal bones from Vindija Cave, Croatia. By comparing sequences from their mtDNA and their nuclear DNA, scientists determined that the three bones came from different individuals, although two of them might be related on their mother’s side. The researchers used several methods to ensure that the DNA they were sequencing was derived from the Neanderthal specimens rather than from contamination by modern humans in the lab.

The Neanderthal sequence was compared to those of five modern humans from France, China, Papua New Guinea, as well as Africans from the San and Yoruba groups. Tests indicated that Neanderthals shared more derived alleles with non-African modern humans than with African modern humans. They compared parts of the Neanderthal genome with pairs of modern humans. While the European and Asian pairs had similar amounts of derived material compared with the Neanderthal, Neanderthals had more similarities with non-African humans than with Africans. The simplest explanation for these results is gene flow from Neanderthals into modern humans. Gene flow could also have occurred from modern humans into Neanderthals. Interbreeding events between Neanderthals and modern humans might be obscured if the modern human population was large.

Neanderthals have contributed approximately 1% to 4% to the genomes of non-African modern humans. This evidence of interbreeding sheds light on how we think of the expansion of modern  humans out of Africa. It refutes the strictest scenario in which anatomically modern humans replaced archaic hominins completely without any interbreeding. However, even with some interbreeding between moderns and archaic hominins, most of our genome still derives from Africa.

The data also points to the time when interbreeding might have taken place. Since the Neanderthal DNA was equally related to that of the modern samples from France, China and Papua New Guinea, admixture between moderns and Neanderthals must have occurred before the Eurasian populations split off from each other. Remains of both modern humans and Neanderthals dating to around 100,000 years ago have been found in the Middle East. A few interbreeding events during this period could have produced the results found in this study.

The sequence of our close hominin relative also shows us how humans are unique. Researchers found 78 sequence differences that would have affected proteins in which Neanderthals had the ancestral state and modern humans had a newer, derived state. Five genes had more than one sequence change that affected the protein structure. These proteins include SPAG17, which is involved in the movement of sperm, PCD16, which may be involved in wound healing, TTF1, which is involved in ribosomal gene transcription, and RPTN, which is found in the skin, hair and sweat glands. Scientists do not know the function of the CAN15 protein, which was also one of the differences. Other changes may affect regulatory regions in the human sequence. Some changes are in regions that code for microRNA molecules that regulate protein manufacture.

The comparison also pointed out regions that might have been under positive selection in modern humans. Though some of the genomic areas that may have been positively selected for in modern humans may have coded for structural or regulatory regions, others may have been associated with energy metabolism, cognitive development and the morphology of the head and upper body.