Phylogenetic conservation of freshwater lake habitat preference varies between abundant bacterioplankton phyla

Abstract

Despite their homogeneous appearance, aquatic systems harbour heterogeneous habitats resulting from nutrient gradients, suspended particulate matter and stratification. Recent reports suggest phylogenetically conserved habitat preferences among bacterioplankton, particularly for particle-associated (PA) and free-living (FL) habitats. Here, we show that independent of lake nutrient level and layer, PA and FL abundance-weighted bacterial community composition (BCC) differed and that inter-lake BCC varied more for PA than for FL fractions. In low-nutrient lakes, BCC differences between PA and FL fractions were larger than those between lake layers. The reverse was true for high-nutrient lakes. Nutrient level affected BCC more in hypolimnia than in epilimnia, likely due to hypolimnetic hypoxia in high-nutrient lakes. In line with previous reports, we observed within-phylum operational taxonomic unit (OTU) habitat preference conservation, although not for all phyla, including the phylum with the highest average relative abundance across all habitats (Bacteroidetes). Consistent phylum-level habitat preferences may indicate that the functional traits that underpin ecological adaptation of freshwater bacteria to lake habitats can be phylogenetically conserved, although the levels of conservation are phylum dependent. Resolving taxa preferences for freshwater habitats sets the stage for identification of traits that underpin habitat specialization and associated functional traits that influence differences in biogeochemical cycling across freshwater lake habitats.

Spatiotemporal distribution of bacterioplankton functional groups along a freshwater estuary to pelagic gradient in Lake Michigan

“…taxa carry out similar functions or may contribute differently to bacterially mediated processes remains to be determined. The compositions of PA bacterial communities are more variable across seasons and spatial gradients than FL communities, which is similar to previous reports in small inland lakes (Rösel and Grossart, 2012;Schmidt et al, 2016). Particles in aquatic environments are heterogeneous in nature by being composed of eukaryotic phytoplankton, small zooplankton, excretes of zooplankton, detritus, and other organic particulate matter including allochthonous materials from rivers (Anadón et al, 2002;Turner, 2015).…”

“…Not unexpectedly, seasonal changes in environmental conditions are a main driver of both species richness and differences in bacterial compositions, with more diverse and distinct communities, based on both species presence/absence and relative abundance of shared taxa, existing in the highly productive estuary compared to the Lake Michigan communities. Our study also highlights the distinct bacterial communities associated with particulate matter relative to free-living bacteria, which were also observed in previous studies in western Lake Erie (Mou et al, 2013) as well as smaller freshwater lake and marine systems (Allgaier and Grossart, 2006;Bižić-Ionescu et al, 2014;Rösel and Grossart, 2012;Schmidt et al, 2016). Particle-associated (PA) fraction bacterial populations are particularly interesting along the Muskegon transect as they are more diverse, highly distinct in phylogenetic and functional group composition, and more variable over time and space than free-living (FL) communities.…”

“…1C-F), which is another known factor shaping bacterial community composition (Hall et al, 2008;Kosten et al, 2012;Scheibner et al, 2014), and which our bioenv analysis identified among the subset of highest correlating environmental factors with community composition in both FL and PA fractions. In addition, nutrient levels differ between oligotrophic Lake Michigan and mesoto eutrophic Muskegon Lake, which can affect bacterial community composition and diversity (Jankowski et al, 2014;Lauro et al, 2009;Schmidt et al, 2016;Yannarell et al, 2003). The above factors, as well as increased mass effects from the Muskegon River and urban runoff, are likely driving factors for the BCC differences along this spatial gradient as suggested for the Milwaukee transect study .…”

“…Only bacterial sequences were retained (any reads classified as chloroplast, mitochondria, eukarya, archaea, and unknown were removed, ESM Table S1). For classification, we used a hybrid protocol using a freshwater-specific taxonomy (https://github.com/mcmahon-uw/ FWMFG) and the SILVA release 119 taxonomy (Quast et al, 2013) as previously described (Schmidt et al, 2016).…”

Lake Bacterial Assemblage Composition Is Sensitive to Biological Disturbance Caused by an Invasive Filter Feeder

“…As PA assemblages have been shown to significantly differ in composition from free-living (FL) assemblages, even at phylum- to class-level taxonomic resolution (30–34), we predicted that this selective removal would explain the observed assemblage composition shifts. Our data supported this hypothesis.…”