Sterile paper points as a bacterial DNA-contamination source in microbiome profiles of clinical samples

Objectives High throughput sequencing of bacterial DNA from clinical samples provides untargeted, open-ended information on the entire microbial community. The downside of this approach is the vulnerability to DNA contamination from other sources than the clinical sample. Here we describe contamination from sterile paper points (PPs) used in microbial sample collection. Methods Peri-implant samples from 48 individuals were collected using sterile PPs. Control samples contained only PPs or DNA extraction blank controls. 16S rRNA gene libraries were sequenced using 454 pyrosequencing. 16S rRNA gene copy numbers were measured by quantitative PCR. Results Nearly half of the sequencing reads belonged to two OTUs classified as Enterococcus (25% of reads) or Exiguobacterium (21%), which are not typical oral microorganisms. Of 87 peri-implant samples, only 10 samples (11%) contained neither of the two OTUs. The relative abundance of both unusual OTUs correlated with each other (p<0.001; r=0.828, Spearman correlation). The control samples showed that 2 of 4 (50%) of the sterile unused PPs contained bacterial DNA equivalent to 1.2×103 and 1.1×104 cells respectively, which was within the range of DNA in the clinical samples (average 1.8×107, SD 4.8×107, min 4.4×102, max 2.8×108). The microbial profile from these PPs was dominated (>83% of reads) by the two unusual OTUs. Conclusions Sterile PPs can contain contaminating bacterial DNA. The use of PPs as a sampling tool for microbial profiling of clinical samples by open-ended techniques such as sequencing or DGGE should be avoided. Clinical significance Clinicians working with PPs as sampling tools for bacterial DNA should consider using an alternative sampling tool, because sterile unused PPs can be a considerable source of foreign bacterial DNA. We recommend sterile curettes for collecting clinical samples for open-ended techniques, such as sequencing or DGGE.