A comparison of the ability of PLFA and 16S rRNA gene metabarcoding to resolve soil community change and predict ecosystem functions. (February 2018)

• Record Type: Journal Article
• Authors: Orwin, K.H.; Dickie, I.A.; Holdaway, R.; Wood, J.R.
• Is Part Of: Soil biology and biochemistry. Volume 117(2018)
• Page Start: 27
• Page End: 35
• Publication Date: 2018-02
• Subjects: PLFA -- 16S rRNA gene metabarcoding -- Land use -- Carbon cycling -- Nutrient cycling -- Stability; Soil biochemistry -- Periodicals; Soil biology -- Periodicals; Sols -- Biochimie -- Périodiques; Sols -- Biologie -- Périodiques; Sols -- Microbiologie -- Périodiques; Bodembiologie; Biochemie; 631.46
• DOI: 10.1016/j.soilbio.2017.10.036
• Languages: English
• ISSNs: 0038-0717

Abstract

Abstract: Soil bacterial community structure has traditionally been measured using phospholipid fatty acid (PLFA) profiling. However, with the development of high-throughput sequencing technologies and metabarcoding techniques, more studies are now using 16S rRNA gene metabarcoding to measure bacterial community structure. Metabarcoding provides a much greater level of detail than PLFA profiling does, but it remains unclear whether or not the two techniques have a similar ability to answer many of the common questions asked by ecologists. We test the relative ability of the two techniques to quantify differences in bacterial community structure among five land uses (natural and planted forest, unimproved and improved grasslands, and vineyards), and to predict ecosystem functions. We also test whether PLFA- and metabarcoding-based metrics indicative of microbial growth strategies are correlated to each other. We show that both techniques showed broadly similar patterns of bacterial community composition change with land use and a remarkably similar ability to predict a wide range of ecosystem functions (carbon and nutrient cycling, and responses to drought). However, they were also complementary, as each showed different strengths in discriminating land uses and predicting ecosystem functions. PLFA metrics (i.e. the gram-positive:gram-negative ratio and fungal:bacterial ratio) were strongly correlated with the equivalent 16S rRNA gene metabarcoding metrics (i.e. the gram-positive:gram-negative and oligotrophic:copiotrophic ratios), although PLFA metrics were less well correlated with the Proteobacteria:Acidobacteria ratio. For many ecological questions the two techniques thus give broadly comparable results, providing confidence in the ability of both techniques to quantify meaningful changes in bacterial communities. Highlights: We test the comparability of PLFA and 16S rRNA gene metabarcoding results. Both methods distinguished some bacterial communities across a land-use gradient. Equivalent metrics calculated using each method were correlated with each other. Both methods were similarly able to predict a wide range of ecosystem functions. Although results were broadly comparable, the methods showed different strengths.