The limits of biomolecular palaeopathology: ancient DNA cannot be used to study venereal syphilis

To determine whether ancient DNA (aDNA) can be used to study the palaeopathology of venereal syphilis, we carried out a comprehensive analysis of the preservation of human and pathogen DNA in a set of 46 bones of various ages, most of which displayed osteological indications of the disease. Bones came from seven English cemetery sites that were in use during the 9th-19th centuries. Twelve of the 46 bones consistently yielded rnitochondrial DNA (mtDNA) sequences after replicate polymerase chain reactions (PCRs), and a further 13 bones yielded mtDNA sequences with less reproducibility. The sequence data enabled tentative mitochondrial haplogroups to be assigned to nine of the bones, and the identities and frequencies of these haplogroups were compatible with the geographical origins of the bones. Twenty-one bones consistently gave negative results with all mtDNA PCRs, indicating that at least these bones were not contaminated with modern human DNA, and those bones that gave positive results only yielded one sequence each, again suggesting that widespread modern contamination had not occurred. Mycobacterium tuberculosis sequences were obtained from seven bones, including three of five bones with tuberculous lesions. The cloned and direct sequences obtained from both the mtDNA and M. tuberculosis PCR products showed features typical of degraded aDNA. All of these results suggest that at least some of the 46 bones that we studied were suitable for aDNA analysis. All 46 bones were tested with nine different treponemal PCRs, each optimised to give a detection limit of <= 5 genomes. Although various bones gave PCR products of the expected size with one or more of these PCRs, sequencing showed that none of these products were authentic treponemal amplicons. Our failure to detect treponemal DNA in bones that were suitable for aDNA analysis, using highly sensitive PCRs, suggests that treponemal DNA is not preserved in human bone and that it is therefore not possible to use aDNA analysis to study venereal syphilis. Any past or future paper claiming detection of treponemal aDNA should therefore be accompanied by a detailed justification of the results. (c) 2005 Elsevier Ltd. All rights reserved.