LAUSR

Abstract The discovery in the 1980's that DNA could be extracted and sequenced from extinct animals opened-up a whole new area of research in paleobiology. The oldest authenticated sequence ever recovered is between 1.65 and 1.1 million years (My) old and extrapolation models on DNA degradation suggest that this is close to the temporal limit of DNA survival. However, recent data from cell nuclei in 30 to 80 My-old fossil cells from plants and dinosaurs show positive staining with standard DNA stains and fluorochromes. We heavily discuss and scrutinize the results of these studies and argue that the intracellular stainings seen in these Cenozoic and Cretaceous fossils are consistent with the presence of endogenous DNA and inconsistent with contamination. Properties of the stains suggest that the preserved molecules are made of at least the sugar-phosphate backbone of DNA, and in some instances they may be double stranded with preserved base pairs. Previous works on DNA damage also suggest that the material is crosslinked and filled with chemically modified nucleotides, which may explain why it is apparently not in a PCR-amplifiable nor in a sequenceable form. However, even though many questions remain, it cannot yet be ruled out that retrieving ancient sequences in fossils older than the Pleistocene will be possible in the future. Here, we summarize and reassess all current evidence and propose new methods and ideas on how to further understand DNA preservation in deep time. Notably, microscopy-based DNA sequencing may offer the most promising results. The main goal of this review is to show the need for new collaborations between the fields of Ancient DNA, Molecular Paleontology and Paleohistology to better seek the truth about DNA fossilization.