Microbial Community Structure of Submerged Aquatic Vegetation in the Potomac River
Submerged aquatic vegetation (SAV) are plants that are rooted in sediment and fully submerged most of the time, and have many adaptations for coping with varied salinity and osmotic conditions. We focus here on one aspect of SAV - their microbiome - which was studied in the Potomac River along a salinity gradient as the river empties into the Chesapeake Bay. The goal was to find a link between the microbial communities on different SAV species and the changing salinity across the river.
One of the four successfully sampled sites was very different from the rest in terms of microbial community and water/sediment chemistry, clustering separately from the other sites on PCoA plots. Methylotenera, Planctomyces, Rhodobacter, and Providencia are commonly found amongst most SAV species across all sites, and sulfur oxidizing bacteria were present in high relative abundance in the roots of Potamogeton perfoliatus at one site.
Site location, which had distinct water and sediment chemistries, was a main driver of the microbial community structure. Host species of SAV and sample types (leaves or roots) also have different microbial communities. Due to the small sample size in this study, it is difficult to draw robust conclusions about the impact of salinity on microbial community structure. Therefore, future efforts will sample more thoroughly along the Potomac river, as well as along the length of the James River, which provides a nearby, parallel salinity gradient.
Salinity is an important factor in determining the distribution of aquatic organisms. Nevertheless, there are some species adapted to a wide range of osmotic conditions, including fish (Whitehead 2011), insects (Pallarés 2015), and plants (Garrote-Moreno 2014). Submerged aquatic vegetation (SAV) are plants that are rooted in the sediment and remain fully submerged most of the time. They are a polyphyletic assemblage restricted to shallow waters by light penetration. They posses several, convergent adaptations to life underwater, including the loss of a waxy cuticle, air cells that allow leaves to float, and specialized epidermal osmoregulartory cells (Les 1997). The presence of SAV is known to influence the structure, density, and metabolic activity of sediment microbial communities (Regier 2012)(Zhao 2013)(Menon 2013)(Meng 2015), and there is some evidence that SAV species may have an effect on rhizosphere microbial community structure (Gagnon 2007)(Meng 2013). Whether the microbial communities found in association with SAV (i.e., the SAV microbiome) play any role in plant adaptation to a fully aquatic lifestyle, and in particular, in plant adaptation to high salinity, is unknown.
The Chesapeake Bay is the largest estuary in the United States, fed by over 100 rivers and flowing into the Atlantic Ocean (Boesch 2001). One of these, the Potomac River has a salinity gradient that ranges from 0 to 12.5 ppt, and SAV in the Potomac River exhibit variation in their ability to thrive across this salinity gradient. For example, Ruppia maritima is known to survive in water ranging from 0 ppt up to 70 ppt (Kantrud 1991) (for reference, ocean salinity averages 30-35 ppt). Other species are more restricted, like Vallisneria americana, which grows poorly at a salinity of 8 ppt, and does not tolerate 18 ppt (Boustany 2009). The James River also flows into the Chesapeake Bay, and offers the opportunity to study a second, parallel salinity gradient. In this study, we aimed to characterize the microbial communities associated with the SAV species found along a salinity gradient in both the Potomac and James Rivers. To do this, we joined a field trip planned by our colleague, Andrew Whitehead, to sample killifish at 6 sites along each river. Unfortunately, while killifish were abundant at all sites, SAVs proved to be more elusive, and in fact were never found at the sites along the James River. Therefore, we report here our findings for a total of 5 species of SAV, collected from 4 sites along the Potomac River.
Triplicate leaf and root samples were taken of an individual of each distinct species of SAV found at each site along the Potomac River. Twelve sites were visited in total – 6 along the Potomac River, and 6 along the James River. Any observed SAV species were sampled. Plants were drawn and distinguishing characteristics, like leaf morphology, were noted. A single plant of each species at a particular site was uprooted with gloved hands and swirled in source river water to dislodge sediment. Alcohol-sterilized surgical scissors were used to cut approximately one inch of leaf tissue from the center of the length of the leaf. Alcohol-sterilized surgical scissors were also used to cut 5-10 roots. These were stored at room temperature in 500 µl of Zymo Xpedition Lysis/Stabilization Solution (Zymo Research, Irvine, CA) in a 2 mL sterile tube until DNA extraction.
DNA was extracted from the samples upon return to the University of California, Davis. DNA extraction was done using the Mo Bio PowerSoil DNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA) with minor adjustments to the provided protocol. These adjustments included the following: heating the sample at 65ºC for 5 minutes after adding C1 solution to dissolve a precipitate that forms from C1 solution and Zymo Xpedition Lysis/Stabilization Solution; bead beating for 1 minute after heating with C1 solution; eluting in 50 µl of nuclease-free water instead of 100 µL of C6 solution. DNA concentration was determined using a Qubit fluorometer (Thermo Fischer Scientific, Carlsbad, CA).
We use the bacterial/archaeal primers 515F/806R (Caporaso 2012). There are 25 each of forward and reverse primer barcodes that can be combined to make 625 total unique barcode combinations. The forward and reverse primers are at 10µM each and are mixed in equal parts to create the 10µM combinations.
Samples were cleaned with Agencourt AMPure XP magnetic beads and adapter (Beckman Coulter, Indianapolis, IN) and quantified for amplicons with Qubit fluorometer (Thermo Fischer Scientific, Carlsbad, CA). Samples were pooled to 1 ng of DNA per sample, and sent in for sequencing on an Illumina MiSeq (Illumina, San Diego, CA) at the UC Davis Genome Center Sequencing Core.
Sequences were demultiplexed using a custom in-house script (https://github.com/gjospin/scripts/blob/master/Demul_trim_prep.pl), which automates quality assessment and trimming. Fasta files were then analyzed using a standard QIIME (Caporaso 2010) workflow for analyzing microbial community data (http://www.wernerlab.org/teaching/qiime/overview). Figures were produced in R version 3.2.2 using the phyloseq (McMURDIE 2011) package. We used unweighted UniFrac (Lozupone 2005) to produce the figures for our data, and QIIME to calculate statistics on alpha and beta diversity data (p-values less than 0.05 were considered significant).