Agriculture Reference
In-Depth Information
While 16S rRNA genes provide a taxonomic description of microbial communi-
ties, sequencing of DNA from microbial communities without initial amplification
(also known as “shotgun” sequencing) offers insight into the metabolic potential
of communities. The collection of sequences from shotgun libraries constitutes a
metagenome. Advances have been made in the application of metagenomic methods
in many environments (Dinsdale et al. 2008, Antonopoulos et al. 2009, Dethlefsen
and Relman 2011), but microbial communities in soil have received considerably
less attention. This is due in part to the remarkable diversity of microbes in soil and
in part to the difficulty of defining and sampling microbial communities in such a
heterogeneous structural matrix.
Heterogeneity in the physical structure of soil, including aggregates of recal-
citrant organic matter, minerals, and microbes, complicates the task of collecting
representative samples of the soil environment. Initial characterization of KBS
LTER soils suggested that the microbial communities changed dramatically over
relatively short distances; major variations in total microbial biomass, respiration,
and the number of cultured bacteria were detected over distances as small as 20
cm (Robertson et al. 1997). The importance of spatial heterogeneity to microbial
communities at KBS LTER was reinforced through analysis of 16S rRNA genes of
Burkholderia
isolates from the rhizospheres of nearby corn plants in the Biologically
Based cropping system of the Main Cropping System Experiment (MCSE; Table
6.1; Robertson and Hamilton 2015, Chapter 1 in this volume). This study revealed
dramatic differences in community composition and abundance between bacte-
rial communities on individual plants (Ramette et al. 2005), highlighting the chal-
lenge presented by spatial variability in even a single field intensively managed
for row-crop production. Variability in bacterial community structure was even
detected among soil aggregates in KBS LTER soils. These studies were carried out
before high-throughput sequencing became available. Sequence differences among
16S genes were identified indirectly by detecting specific sites for cleavage by
restriction enzymes in a process known as Terminal Restriction Fragment Length
Polymorphism (tRFLP; Fig. 6.1). As many as half of the differences between
tRFLP profiles of 16S rDNA genes from soil could be explained by interaggregate
variation, even when particles were less than 2 mm in diameter (Blackwood et al.
2006). This means microbial communities vary significantly from particle to parti-
cle—even in the same soil sample. This is a potential source of sampling bias when
comparing microbial distributions across field sites or systems, but fortunately, spa-
tial variability in microbial communities can be addressed with sufficient sample
size and replication in sampling. For example, RFLP profiles of the genes encoding
the final enzymatic step in denitrification (
nosZ
) were identical when triplicate 3-g
samples were analyzed from either unmanaged successional communities or agri-
cultural systems at KBS LTER (Stres et al. 2004).
In addition to spatial variability as a potential source of uncertainty, bias can also
be introduced in certain steps of processing samples for molecular surveys. Several
of these technical challenges have been addressed by research conducted with soils
from KBS LTER. Differences in cell lysis and therefore DNA extraction efficiency
can distort molecular surveys, revealing only a fraction of the tremendous diversity
present in soil communities (Feinstein et al. 2009). One strategy for reducing the
