Biases in Prokaryotic Community Amplicon Sequencing Affected by DNA Extraction Methods in Both Saline and Non-saline Soil

Abstract

High-throughput amplicon sequencing technology has been widely used in soil microbiome studies. Here, we estimated the bias of amplicon sequencing data affected by DNA extraction methods in a saline soil, and a non-saline normal soil was used as a control. Compared with the normal soil, several unique points were observed in the saline soil. The soil washing pretreatment can improve not only DNA quantity and quality but also microbial diversities in the saline soil; therefore, we recommend the soil washing pretreatment for saline soils especially hypersaline soils that cannot be achieved with detectable DNA amounts without the pretreatment. Also, evenness indices were more easily affected by DNA extraction methods than richness indices in the saline soil. Moreover, proportions of Gram-positive bacteria had significant positive correlations with the achieved microbial diversities within replicates of the saline soil. Though DNA extraction methods can bias the microbial diversity or community and relative abundances of some phyla/classes can vary by a factor of more than five, soil types were still the most important factor of the whole community. We confirmed good comparability in the whole community, but more attention should be paid when concentrating on an exact diversity value or the exact relative abundance of a certain taxon. Our study can provide references for the DNA extraction from saline and non-saline soils and comparing sequencing data across studies who may employ different DNA extraction methods.

Identifying Hidden Viable Bacterial Taxa in Tropical Forest Soils Using Amplicon Sequencing of Enrichment Cultures

“…This may be caused by the complexity of soil, which consists of several compounds (e.g. humic acid or salt) that can limit the recovery of microbes during DNA extraction and amplification [41,42]. Moreover, bias of PCR primers and low abundance of DNA could also be an issue [12,43].…”

Effects of DNA preservation solution and DNA extraction methods on microbial community profiling of soil

“…In general, Gram-positive bacteria are difficult to break open due to the thick multilayered and extensively crosslinked peptidoglycan layer in contrast to the thinner layer of peptidoglycan in Gram-negative cells (Mahalanabis et al 2009). Also, other studies revealed that the use of different DNA extraction kits may result in significantly different community profiles as a consequence of poor performance in DNA extraction from mainly Gram-positive bacteria (Wüst et al 2016;Zielińska et al 2017;Xie et al 2018). However, we also noted a better performance of the Zymo kit for the extraction of DNA from Gram-negative species from the phyla Planctomycetota and Chloroflexi.…”

“…DNA extraction kits developed by several molecular biology companies have a much better performance in the removal of humic acids and other types of PCR inhibitors, but their components are trade secrets (Zielińska et al 2017). It has been revealed that different types of DNA extraction kits show different community compositions and biodiversity indices (Martin-Laurent et al 2001;Wüst et al 2016;Zielińska et al 2017;Soliman et al 2017;Xie et al 2018). However, to our current understanding, the impact of the DNA extraction kit on single species and standardized microbial communities from soils has never been studied in detail.…”

Generating amplicon reads for microbial community assessment with next‐generation sequencing

“…There is no universal method that can be successfully applied to each kind of sample, but bead beating is far superior to chemical/enzymatic lysis in terms of microbial diversity recovered. There is also a study suggesting that there is no need to supplement bead-beating kits with enzymes, at least in the case of pine litter (Gołe z biewski et al 2019), but a study on saline soils suggests such a need (Xie et al 2018), therefore, we think that enzymatic pretreatment might be beneficial in case of certain sample types. The possible bias resulting from incomplete lysis and extraction was recognized long ago (e.g.…”