Deficiencies and challenges in the study of ancient tuberculosis DNA

We explore the standards of research and reporting needed to justify the destructive analysis of archaeological human bone for biomolecular studies of ancient tuberculosis (TB). Acceptable standards in osteological interpretation have been met in some biomolecular papers, but there are also cases where insufficient care has been taken in distinguishing between pathognomonic lesions and those that are 'consistent with' a diagnosis of TB. Some biomolecular studies have failed to recognize that archaeological bones might be contaminated with environmental mycobacteria whose DNA could give rise to false positives in polymerase chain reactions directed at members of the Mycobacterium tuberculosis complex. The difficulties of applying spoligotyping to ancient DNA have also been underestimated and conclusions drawn from such analyses are often weakly supported. Assumptions that mycobacterial DNA preserves better than human DNA, and that contamination with modern DNA is less of a problem, has led in some cases to a laxity in research standards with insufficient attention paid to the need to authenticate ancient DNA results. We illustrate our concerns by reference to a recent paper reporting biomolecular detection of ancient TB DNA in skeletons from the eastern Mediterranean Neolithic settlement of Atlit-Yam. We are unconvinced that the skeletal evidence presented in this paper gives sufficient indication of TB to warrant destructive analysis, and we are concerned that during the biomolecular part of the project inadequate attention was paid to the possibility that results might be due to laboratory cross-contamination or to amplification of environmental mycobacterial DNA present in the bones.