The wooly mammoth (Mammuthus primigenius) is an iconic animal, like the saber tooth tiger or dire wolf, from a time in human history when our position at the top of the global food chain was decidedly not assured (and being something's prey was not limited to just other humans). Perhaps this is a reason that resurrection of mammoths using Jurassic Park-like technology has some currency and appeal (but see How to Clone a Mammoth for reasons why this may not be such a good idea). Perhaps paradoxically, the mammoth arose in Africa 5 million years ago and like its (very) distant Homo sapiens relatives migrated to colonize the Northern Hemisphere, with most of the evolutionary adaptations to new habitats (and speciation that accompanied them) rising in the Pleistocene Era (from about 2.6 million to 12 years ago). From these speciation events arose the Columbian mammoth in North America about 1.5 million years ago and the "classic" wooly mammoth in northeastern Siberia about 700,000 years ago.

Recently an international team* elucidated some genetic relationships between mammoths using DNA extracted from specimens from the Early and Middle Pleistocene, in a paper entitled "Million-year-old DNA sheds light on the genomic history of mammoths," in the journal Nature. These results, termed "deep-time paleogenomics" by the authors, were obtained from three specimens, two of which were more than 1 million years old. The specimens represented two distinct lineages present in Eastern Siberia in early Pleistocene, one of which resulted in wooly mammoth populations that survived to become contemporaneous with humans. The third lineage gave rise to first mammoths in North America (Mammuthus columbi), resulting from a hybrid between these two lineages in Siberia in the Middle Pleistocene. And perhaps not surprisingly, these researchers found most of the adaptation to cold seen in these animals were present in the samples from one million years ago.

As reported by these researchers, DNA was extracted from the three specimens using now established techniques for highly degraded DNA samples and libraries prepared. The libraries were sequenced and these sequences compared to genomic DNA sequences from the African savannah elephant (Loxodonta africana) and mitochondrial DNA from the Asian elephant (Elephas maximus). Perhaps not unexpectedly, the specimen DNA was "considerably more fragmented and had higher levels of cytosine deamination than DNA from permafrost-preserved samples dating to the Late Pleistocene subepoch," which required modifications to the sequence analysis protocols. Nevertheless, these researchers reported that they obtained complete mitochondrial DNA genomes from the three samples and 49 million (from the oldest sample, named Krestovka), 884 million (the Adycha sample) and 3,671 million base pairs (the Chukochya sample) of nuclear genomic data from the three samples (with the highest DNA recovery coming from the most recent specimen, which from tooth morphology they thought to be an early wooly mammoth specimen).

The ages of these specimens were then determined using the mitochondrial DNA sequences, ranging from 1.65 Ma (2.08–1.25 Ma; Krestovka), 1.34 Ma (1.69–1.06 Ma; Adycha) and 0.87 Ma (1.07–0.68 Ma; Chukochya). Genomic DNA was used to verify the two specimens having the highest amount of gDNA sequence obtained, against estimates of the highest number of changes from the latest Africa savannah element ancestor, which yielded 1.28 Ma (1.64–0.92 Ma) and 0.62 Ma (1.00–0.24 Ma), respectively. These results were similar to the dates derived from the generally more intact mitochondrial DNA, but the authors cautioned that "this analysis is based on low-coverage data and the confidence intervals are wide." Nevertheless, the authors also note that "[t]he DNA-based age estimates for the [the two genomic DNA] specimens are consistent with the geological age inferences that were independently derived from biostratigraphy and palaeomagnetism."

The genetic analyses indicated that the samples were divergent from known mammoth species from the Late Pleistocene, and were consistent with the younger two specimens (Adycha and Chukochya) being ancestral to the wooly mammoths, while the oldest specimen (Krestovka) apparently diverged prior to the species split between the wooly mammoth and the Columbian (North American) mammoths. The oldest specimen was estimated by these authors to have split from the later mammoth lineages between 2.66 and 1.78 million years ago based on both mitochondrial and genomic DNA data and had fewer derived alleles that differed from the Asian elephant than the younger two (indicating a closer evolutionary relationship). Due to the relative closeness in time for the eldest and one of the younger specimens, these results are consistent with these two lineages together inhabiting eastern Siberia in the Early Pleistocene, with the earliest having diverged from all later mammoth species before mammoths appeared in North America.

Regarding the Columbian (North American) mammoth, the genetic analyses also indicated that these mammoths shared a higher proportion of it ancestry from the earlier-diverging Krestovka species, which showed "excess derived allele-sharing between the Columbian mammoth and the Krestovka specimen." These statistics were further consistent with one migration of n admixture of mammoth populations of the species from the Krestovka specimen (38-43% ancestry) and the wooly mammoth (57-62% ancestry). Further analysis of later Columbian mammoth specimens indicated a second admixture with North American wooly mammoth species.

The researchers then assessed the genetic adaptations expected to be found in animals descendent from African elephants living in Siberia, notably cold-tolerance and open-habitat adaptations. These analyses were performed using protein-coding portions of the genetic information derived from these samples in comparison with known adaptations from Late Pleistocene wooly mammoths (as further compared with African and Asian elephants, who carried the "ancestral" alleles). As reported in the paper, "85.2% (782 out of 918) and 88.7% (2,578 out of 2,906) of the mammoth-specific protein-coding changes were already present in the genomes of Adycha . . . and Chukochya [specimens], respectively." These specimens showed no specific genetic changes arising in these populations in response to known changes in climate and mammoth morphology arising in the Middle Pleistocene. These researchers also interrogated the DNA samples for 91 changes known to have arisen in the mammoth lineage. These included genes thought to be involved in "hair growth, circadian rhythm, thermal sensation and white and brown fat deposits," the researchers finding "the vast majority of coding changes were present in both the Adycha (87%) and Chukochya (89%) genomes." In contrast, another gene ("TRPV3, which encodes a temperature-sensitive transient receptor channel that is potentially involved in thermal sensation and hair growth") possessed only 2 of 4 amino acid sequence changes found in Late Pleistocene wooly mammoths, which indicated to these authors that these genetic alterations "occurred over several hundreds of thousands of years, rather than during a single brief burst of adaptive evolution."

The authors conclude by a discussion of their results suggesting an admixture of an earlier-arising mammoth ancestor with wooly mammoths to produce the Columbian (North American) mammoth through a "hybrid speciation event" resulting in an almost 50:50 mix of ancestor species DNA. Mitochondrial DNA assessments suggested that the most recent common female ancestor of all Late Pleistocene Columbian mammoths lived about 420 thousand years ago, which these authors use as a minimum date for the admixture event. Because it is known that mammoths existed in North America 1.5 million years ago, these authors posit initial North American colonization by the Krestovka ancestral species. They conclude their report by stating:

The retrieval of DNA that is more than one million years old confirms previous theoretical predictions that the ancient genetic record can be extended beyond what has been previously shown. We anticipate that the additional recovery and analysis of Early and Middle Pleistocene genomes will further improve our understanding of the complex nature of evolutionary change and speciation. Our results highlight the value of perennially frozen environments for extending the temporal limits of DNA recovery, and hint at a future deep-time chapter of ancient DNA research in which specimens from high latitudes will have an important role.

* From Centre for Palaeogenetics, Stockholm, Sweden; Department of Bioinformatics and Genetics, Swedish Museum of Natural History, Stockholm, Sweden; Department of Cell and Molecular Biology, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Uppsala University, Uppsala, Sweden; Department of Zoology, Stockholm University, Stockholm, Sweden; Section for Computational and RNA Biology, Department of Biology, University of Copenhagen, Copenhagen, Denmark; The Francis Crick Institute, London, UK; Department of Biomolecular Engineering, University of California Santa Cruz, Santa Cruz, CA; Institute for Biochemistry and Biology, University of Potsdam, Potsdam, Germany; Department of Biological Sciences, Middle East Technical University, Ankara, Turkey; Department of Earth Sciences, Natural History Museum, London, UK; College of Plant Protection, China Agricultural University, Beijing, China; The Arctic University Museum of Norway, UiT – The Arctic University of Norway, Tromsø, Norway; Geological Institute, Russian Academy of Sciences, Moscow, Russia; Department of Ecology and Evolutionary Biology, University of California Santa Cruz, Santa Cruz, CA; Howard Hughes Medical Institute, University of California Santa Cruz, Santa Cruz, CA; and Department of Archaeology and Classical Studies, Stockholm University, Stockholm, Sweden.

Image of Woolly mammoth model Royal BC Museum in Victoria by Thomas Quine, from the Wikimedia Commons under the Creative Commons Attribution 2.0 Generic license.