Laboratory contamination over time during low-biomass sample analysis

Abstract

17Bacteria are not only ubiquitous on earth but can also be incredibly diverse 18 within clean laboratories and reagents. The presence of both living and dead bacteria 19 in laboratory environments and reagents is especially problematic when examining 20 samples with low endogenous content (e.g. skin swabs, tissue biopsies, ice, water, 21 degraded forensic samples, or ancient material), where contaminants can outnumber 22 endogenous microorganisms within samples. The contribution of contaminants within 23 high-throughput studies remains poorly understood because of the relatively low 24 number of contaminant surveys. Here, we examined 144 negative control samples 25 (extraction blank and no-template amplification controls) collected in both typical 26 molecular laboratories and an ultraclean ancient DNA laboratory over five years to 27 characterize long-term contaminant diversity. We additionally compared the 28 contaminant content within a homemade silica-based extraction method, commonly 29 used to analyse low-endogenous samples, with a widely used commercial DNA 30 extraction kit. The contaminant taxonomic profile of the ultraclean ancient DNA 31 laboratory was unique compared to the modern molecular biology laboratories, and 32 changed over time according to researchers, month, and season. The commercial kit 33 contained higher microbial diversity and several human-associated taxa in comparison 34to the homemade silica extraction protocol. We recommend a minimum of two 35 strategies to reduce the impacts of laboratory contaminants within low-biomass 36 metagenomic studies: 1) extraction blank controls should be included and sequenced 37 with every batch of extractions and 2) the contributions of laboratory contamination 38 should be assessed and reported in each high-throughput metagenomic study. 39 3 Main Text: 40In the new era of culture-independent microbiome research, targeted amplicon 41 or 'metabarcoding' approaches are now routinely used to amplify DNA from 42 microbial species across the tree of life. However, these methods lack the ability to 43 select for either specific species or to exclude contaminants [1]. Although these 44 techniques have provided invaluable insight into otherwise cryptic microbial 45 communities, the increased sensitivity and lack of target specificity leaves microbiota 46 studies particularly susceptible to the effects of contamination. Such effects are 47 widespread, as several recent studies have indicated that contaminant microbial DNA 48 can be routinely isolated from laboratory reagents and surfaces [2][3][4] and that this 49 signal has significantly impacted the interpretation and characterization of microbiota 50 in high-throughput sequencing studies. For example, Salter et al. recently 51 demonstrated that bacterial DNA present in laboratory reagents is present in both 52 quality-filtered 16S ribosomal RNA (rRNA) gene and shotgun metagenomic datasets 53 and significantly impacts the interpretation of results [3]. Multiple microbial 54

Hyperdiverse archaea near life limits at the polyextreme geothermal Dallol area

“…By contrast, amplicons derived from Dallol ponds, Black and Yellow lakes but also first-PCR 'negative'-controls were dominated by bacterial sequences. Most of them were related 25 to well-known kit and laboratory contaminants (17,18), other were human-related bacteria likely introduced during intensive afar and tourist daily visits to the site; a few archaeal sequences might result from aerosol cross-contamination despite extensive laboratory precautions (see Supplementary Methods). After removal of contaminant sequences (grey bars, Fig.…”

High‐resolution biomonitoring of plant pathogens and plant species using metabarcoding of pollen pellet contents collected from a honey bee hive

“…Another limitation with this metabarcoding study is that there were no blank samples included in the metagenomics workflow, which prevented us to discard commonly found contaminants from the extraction kits and other laboratory contaminants prior to sequencing (Weyrich et al, 2019). Given the high sensitivity aspect of HTS, there is also a risk of cross-contamination with positive controls used for PCR, in which case renders the interpretation of "true" versus "false" signals ambiguous.…”

Contaminations in (meta)genome data: An open issue for the scientific community

“…Multiple microbial contaminants, belonging to the bacterial genera Acinetobacter, Alcaligenes, Bacillus, Bradyrhizobium, Herbaspirillum, Legionella, Leifsonia, Mesorhizobium, Methylobacterium, Microbacterium, Novosphingobium, Pseudomonas, Ralstonia, Sphingomonas, Stenotrophomonas, and Xanthomonas have been identified within the published 1,000 (human) genomes. [16][17][18] Significant bacterial contaminations in existing human-derived sequence databases can be introduced by reagents, environment, handlers, or machines during the collection of the sample, the extraction of nucleic acid and the preparation of libraries. 7,[9][10][11] In 2014, it has been demonstrated that DNA extraction kits, commonly used in laboratories, present contaminating microbial DNA, which impacts both PCR-based 16S rRNA gene surveys and shotgun metagenomics, influencing the interpretation of the results obtained from samples containing a low microbial biomass.…”

“…7,[9][10][11] In 2014, it has been demonstrated that DNA extraction kits, commonly used in laboratories, present contaminating microbial DNA, which impacts both PCR-based 16S rRNA gene surveys and shotgun metagenomics, influencing the interpretation of the results obtained from samples containing a low microbial biomass. 7,18 Indeed, in the case of samples containing a low amount of microbial biomass, the starting material may be effectively swamped by the contaminating DNA, generating misleading results.…”