Botanical microbiomes on the cheap: Inexpensive molecular fingerprinting methods to study plant-associated communities of bacteria and fungi.

David Johnston-Monje,Jessica Lopez Mejia

doi:10.1002/aps3.11334

Abstract

High‐throughput sequencing technologies have revolutionized the study of plant‐associated microbial populations, but they are relatively expensive. Molecular fingerprinting techniques are more affordable, yet yield considerably less information about the microbial community. Does this mean they are no longer useful for plant microbiome research? In this paper, we review the past 10 years of studies on plant‐associated microbiomes using molecular fingerprinting methodologies, including single‐strand conformation polymorphism (SSCP), denaturing gradient gel electrophoresis (DGGE), amplicon length heterogeneity PCR (LH‐PCR), ribosomal intergenic spacer analysis (RISA) and automated ribosomal intergenic spacer analysis (ARISA), and terminal restriction fragment length polymorphism (TRFLP). We also present data juxtaposing results from TRFLP methods with those generated using Illumina sequencing in the comparison of rhizobacterial populations of Brazilian maize and fungal surveys in Canadian tomato roots. In both cases, the TRFLP approach yielded the desired results at a level of resolution comparable to that of the MiSeq method, but at a fraction of the cost. Community fingerprinting methods (especially TRFLP) remain relevant for the identification of dominant microbes in a population, the observation of shifts in plant microbiome community diversity, and for screening samples before their use in more sensitive and expensive approaches.

Highlights

High-throughput sequencing technologies have revolutionized the study of plant-associated microbial populations, but they are relatively expensive
Molecular fingerprinting techniques are more affordable, yet yield considerably less information about the microbial community. Does this mean they are no longer useful for plant microbiome research? In this paper, we review the past 10 years of studies on plant-associated microbiomes using molecular fingerprinting methodologies, including single-strand conformation polymorphism (SSCP), denaturing gradient gel electrophoresis (DGGE), amplicon length heterogeneity PCR (LHPCR), ribosomal intergenic spacer analysis (RISA) and automated ribosomal intergenic spacer analysis (ARISA), and terminal restriction fragment length polymorphism (TRFLP)
We present data juxtaposing results from TRFLP methods with those generated using Illumina sequencing in the comparison of rhizobacterial populations of Brazilian maize and fungal surveys in Canadian tomato roots

Summary

Introduction

High-throughput sequencing technologies have revolutionized the study of plant-associated microbial populations, but they are relatively expensive. It should be noted that while no primer pair is either universal or unbiased, some are better than others, primer choice has significant implications on the abundances and richness of the sequences that are amplified (Pollock et al, 2018) These amplicons may be sequenced, or separated and analyzed, using various fingerprinting approaches. The earliest DNA-based technique to survey plant microbiomes was to clone and separate PCR products into plasmid-containing colonies of Escherichia coli and use Sanger sequencing on the resulting library to generate a list of all the microbial barcode DNA found in the environmental sample (Ward et al, 1990). The development of next-generation or high-throughput sequencing (HTS) technologies to analyze microbial amplicons from plant DNA samples has been a significant boon to the study of plant microbiomes (Knief, 2014). Scientists hoping to analyze just a few samples and who do not have their own HTS machine or sufficient resources to pay for multiple sequencing runs with a commercial service provider may find themselves wasting their money on excessive sequencing or sacrificing the quality of their experiment, for example, by reducing repetitions necessary for statistical confidence or analyzing incomplete/poor data sets when experimental errors arise (Prosser, 2010)

Objectives

Methods

Findings

Conclusion