Sediment Metagenomes as Time Capsules of Lake Microbiomes

Abstract

The reconstruction of ecological time series from lake sediment archives can retrace the environmental impact of human activities. Molecular genetic approaches in paleolimnology have provided unprecedented access to DNA time series, which record evidence of the microbial ecologies that underlaid historical lake ecosystems. Such studies often rely on single-gene surveys, and consequently, the full diversity of preserved microorganisms remains unexplored. In this study, we probed the diversity archived in contemporary and preindustrial sediments by comparative shotgun metagenomic analysis of surface water and sediment samples from three eastern Canadian lakes. In a strategy that was aimed at disentangling historical DNA from the indigenous sediment background, microbial preservation signals were captured by mapping sequence similarities between sediment metagenome reads and reference surface water metagenome assemblies. We detected preserved Cyanobacteria, diverse bacterioplankton, microeukaryotes, and viruses in sediment metagenomes. Among the preserved microorganisms were important groups never before reported in paleolimnological reconstructions, including bacteriophages (Caudovirales) and ubiquitous freshwater Betaproteobacteria (Polynucleobacter and Limnohabitans). In contrast, ultramicroscopic Actinobacteria ("Candidatus Nanopelagicales") and Alphaproteobacteria (Pelagibacterales) were apparently not well preserved in sediment metagenomes even though they were numerically dominant in surface water metagenomes. Overall, our study explored a novel application of whole-metagenome shotgun sequencing for discovering the DNA remains of a broad diversity of microorganisms preserved in lake sediments. The recovery of diverse microbial time series supports the taxonomic expansion of microbiome reconstructions and the development of novel microbial paleoindicators.IMPORTANCE Lakes are critical freshwater resources under mounting pressure from climate change and other anthropogenic stressors. The reconstruction of ecological time series from sediment archives with paleolimnological techniques has been shown to be an effective means of understanding how humans are modifying lake ecosystems over extended timescales. In this study, we combined shotgun DNA sequencing with a novel comparative analysis of surface water and sediment metagenomes to expose the diversity of microorganisms preserved in lake sediments. The detection of DNA from a broad diversity of preserved microbes serves to more fully reconstruct historical microbiomes and describe preimpact lake conditions.

Keywords: DNA preservation; bacterioplankton; metagenomics; paleogenomics; paleolimnology; shotgun sequencing; viruses.

FIG 1

Map of Canada with the starred locations of study lakes: Lac Paula, Eightmile Lake, and Grand lac Touradi. The map was created in QGIS v. 3.10 (77) with the Canada Atlas Lambert projection and shapefile data sourced for the hydrographic features (78) and political boundaries of Canada (79). Spatial coordinates, morphological characteristics, and physicochemical profiles of study lakes are summarized in Table 1.

FIG 2

(a) As illustrated in this experimental design schematic, the recruitment of unassembled metagenome reads (abbreviated subscript R) to metagenome assemblies (subscript A) tracks the sequence similarities between focal and reference metagenomes. Captured metagenomes are abbreviated SW_A → TS_R (top sediment reads mapped to the surface water assembly), SW_A → BS_R (bottom sediment reads mapped to the surface water assembly), and TS_A → BS_R (bottom sediment reads mapped to the top sediment assembly). Arrows point in the direction of DNA preservation, from the overlying source to the underlying sediment archive. (b) Metagenome capture was performed with a 90% sequence identity threshold derived from the delineation of closely (solid curve, ≥90%) and distantly (dashed curve, <90%) related sequences. Reads mapped to scaffolds containing ribosomal or transfer RNA genes are included (exceptionally) in this figure, with negligible impact on the results presented (see Table S1 in the supplemental material).

FIG 3

Relative coverage of prevalent phyla in the free and captured metagenomes of three lakes. Represented in this heat map are taxa with ≥10% relative coverage in at least one metagenome.

FIG 4

(a) Relative coverage of prevalent taxonomic orders in the free and captured metagenomes of three lakes. Represented in this heat map are taxa with ≥5% relative coverage in at least one metagenome. (b) PCA of order-rank taxonomic composition data of free and captured metagenomes. The predominant taxon contributions to the first two PC axes are shown. Groups of taxonomically similar metagenomes are highlighted by grey-shaded ellipses.

FIG 5

Gene counts of selected preserved Cyanobacteria, bacterioplankton, microeukaryotes, and viruses in free surface water and sediment metagenomes in three lakes.