A jawbone discovered in a Tibetan cave turned out to be that of a Denisovan who had lived and died there about 160,000 years ago (see: Denisovan on top of the world; 6 May, 2019). That discovery owed nothing to ancient DNA, because the fossil proved to contain none that could be sequenced. But the dentine in one of two molar teeth embedded in the partial jaw did yield protein. The teeth are extremely large and have three roots, rather than the four more common in modern, non-Asian humans, as are Denisovan teeth from in the Siberian Denisova Cave. Fortunately, those teeth also yielded proteins. In an analogous way to the genomic sequencing of nucleotides (adenine, thymine, guanine and cytosine) in DNA, the sequence of amino acids from which proteins are built can also be analysed. Such a proteomic sequence can be compared with others in a similar manner to genetic sequences in DNA. The Tibetan and Siberian dentine proteins are statistically almost the same.

At present the most ancient human DNA that has been recovered – from an early Neanderthal in the Sima de los Huesos in Spain – is 430,000 years old (see: Mitochondrial DNA from 400 thousand year old humans; December 2013). Yet it is proving difficult to go beyond that time, even in the cool climates that slow down the degradation of DNA. The oldest known genome of any animal is that of mtDNA from a 560–780 thousand year old horse, a leg bone of which was extracted from permafrost in the Yukon Territory, Canada. The technologies on which sequencing of ancient DNA depends may advance, but, until then, tracing the human evolutionary journey back beyond Neanderthals and Denisovans seems dependent on proteomic approaches (Warren, M. 2019. Move over, DNA: ancient proteins are starting to reveal humanity’s history. Nature, v. 570, p. 433-436; DOI: 10.1038/d41586-019-01986-x). Are the earlier Homo heidelbergensis and H. erectus within reach?

It seems that they may be, as might even earlier hominins. The 1.8 Ma Dmanisi site in Georgia, now famous for fossils of the earliest humans known to have left Africa, also yielded an extinct rhinoceros (Stephanorhinus). Proteins have been extracted from it, which show that Stephanorhinus was closely related to the later woolly rhinoceros (Coelodonta antiquitatis). Collagen protein sequences from a 3.4 Ma camel preserved in the Arctic and even from a Tanzanian 3.8 Ma ostrich egg shell show the huge potential of ancient proteomics. Most exciting is that last example, not only because it extends the potential age range to that of Australopithecus afarensis but into tropical regions where DNA is at its most fragile. Matthew Warren points out potential difficulties, such as the limit of a few thousand amino acids in protein sequences compared with 3 million variants in DNA, and the fact that the most commonly found fossil proteins – collagens –  may have evolved very little. On the positive side, proteins have been detected in a 195 Ma old fossil dinosaur. But some earlier reports of intact diosaur proteins have been questioned recently (Saitta, E.T. et al. 2019. Cretaceous dinosaur bone contains recent organic material and provides an environment conducive to microbial communities. eLife, 8:e46205; DOI: 10.7554/eLife.46205)