Analysis of ancient human DNA and primer contamination: one step backward one step forward

The analysis of DNA from archaeological human remains is plagued by a unique set of methodological problems concerning contamination with modern exogenous DNA. Through an original approach, we propose complementary methods to identify all potential sources of contamination and complete guidelines for the validation of ancient human sequences. The study presented was conducted on non-European human samples (Polynesian and Amerindian) which were collected with all precautions during excavation. This permitted us to distinguish without ambiguity authentic and contaminant sequences. The samples' origins and histories were perfectly known, allowing us to trace all potential contamination sources and to determine the efficiency of precautions followed during all steps of the study. The data obtained confirm that precautions taken during sampling effectively prevent contamination. However, we demonstrate that human contamination can also be introduced during genetic analyses even if all precautions are strictly followed. Indeed, numerous human contaminations were detected in template-PCR products and negative controls, resulting in a striking diversity of contaminant mitochondrial DNA sequences. We argue that this contamination partly derives from the primers. To our knowledge, no previous experiment has been performed to investigate primers as a possible source of human contamination despite the fact that this specific type of contamination poses a real problem in terms of validating ancient human DNA studies. Finally, we confirm that the detection of contaminants in negative controls is clearly related to the number of PCR cycles used. This study enhances our understanding of contamination processes and confirms that, in reality, an absolutely contamination-free situation cannot be obtained. As a consequence, we propose improvements to the guidelines usually followed in the field in order to take the highly probable contamination of PCR reagents, including primers, into account.