Biogeochemical Regimes in Shallow Aquifers Reflect the Metabolic Coupling of the Elements Nitrogen, Sulfur, and Carbon

Abstract

Near-surface groundwaters are prone to receive (in)organic matter input from their recharge areas and are known to harbor autotrophic microbial communities linked to nitrogen and sulfur metabolism. Here, we use multi-omic profiling to gain holistic insights into the turnover of inorganic nitrogen compounds, carbon fixation processes, and organic matter processing in groundwater. We sampled microbial biomass from two superimposed aquifers via monitoring wells that follow groundwater flow from its recharge area through differences in hydrogeochemical settings and land use. Functional profiling revealed that groundwater microbiomes are mainly driven by nitrogen (nitrification, denitrification, and ammonium oxidation [anammox]) and to a lesser extent sulfur cycling (sulfur oxidation and sulfate reduction), depending on local hydrochemical differences. Surprisingly, the differentiation potential of the groundwater microbiome surpasses that of hydrochemistry for individual monitoring wells. Being dominated by a few phyla (Bacteroidetes, Proteobacteria, Planctomycetes, and Thaumarchaeota), the taxonomic profiling of groundwater metagenomes and metatranscriptomes revealed pronounced differences between merely present microbiome members and those actively participating in community gene expression and biogeochemical cycling. Unexpectedly, we observed a constitutive expression of carbohydrate-active enzymes encoded by different microbiome members, along with the groundwater flow path. The turnover of organic carbon apparently complements for lithoautotrophic carbon assimilation pathways mainly used by the groundwater microbiome depending on the availability of oxygen and inorganic electron donors, like ammonium.IMPORTANCE Groundwater is a key resource for drinking water production and irrigation. The interplay between geological setting, hydrochemistry, carbon storage, and groundwater microbiome ecosystem functioning is crucial for our understanding of these important ecosystem services. We targeted the encoded and expressed metabolic potential of groundwater microbiomes along an aquifer transect that diversifies in terms of hydrochemistry and land use. Our results showed that the groundwater microbiome has a higher spatial differentiation potential than does hydrochemistry.

Keywords: groundwater; metagenomics; metatranscriptomics; microbiome.

FIG 1

(A) Characterization of the Hainich CZE groundwater transect. Sites/wells that have not been sampled are marked with an x. The availability of metagenome/metatranscriptome data sets is indicated by (filled) squares. Important characteristics are plotted as lollipop charts based on data for sampling campaign PNK 66 extracted from Kohlhepp et al. (13). (B to D) Principal-component analyses were carried out to obtain hydrochemical data (B), taxonomic profiles (C), and functional metagenome profiles (D). Sub-data sets relating to sampling campaigns PNK66 and PNK69 were extracted from available hydrochemical data (13). The corresponding analysis for metagenome taxonomic profiles was done on the phylum level. Taxonomic profiles were based on the method introduced by Menzel et al. (65). Functional metagenome profiles used for principal-component analysis were deduced from KEGG-based annotations. The color code for hydrochemical clusters is the same across all subpanels. Cluster nomenclature is based on Kohlhepp et al. (13). HTL, Hainich transect lower aquifer assemblage (including wells H1-3, H1-4, H2-1, H3-1, H4-1, and H5-1); HTU, Hainich transect upper aquifer assemblage (wells H3-2, H4-2, H4-3, H5-2, and H5-3).

FIG 2

(A and B) Taxonomic profiles on the metagenome (A) and metatranscriptome (B) levels. When available, replicate data sets were plotted individually. (C) Relative abundance ratios are shown for selected taxa, based on data from PNK66.

FIG 3

Functional assignment of metagenome data sets (A) and taxonomic profiling (B) of function specific sub-data sets. Functional (Func.) profiles are based on assignments to KEGG (69) KO identifiers (K numbers) for functional categories of interest. Percentages show the proportion of functionally assigned sequences to any KEGG category. Balloon sizes refer to relative abundances. Sets of representative K numbers/gene functions were selected based on metabolic pathways of interest. Respective pathways are given below. Commonly accepted gene abbreviations are given if available. K15020_acrC, K15019_3hpd, and K14469_3hps refer to acryloyl-coenzyme A reductase, 3-hydroxypropionyl-coenzyme A dehydratase, and 3-hydroxypropionyl-CoA synthetase genes, respectively. For these three genes, there no commonly accepted gene abbreviations available yet. Sequences assigned to functions of interest were extracted from the total data set and subjected to taxonomic profiling on the family level using DIAMOND (69) against NCBI RefSeq (68). Waffle charts reflect relative (Rel.) abundances, within respective sub-data sets based on taxonomically assigned sequences. The color code of the waffle charts refers to selected family-level affiliations.

FIG 4

Functional assignment of metatranscriptome data sets (A) and taxonomic profiling (B) of function-specific sub-data sets. Details about functional assignment, balloon sizes, and color code of the waffle charts are the same as those for Fig. 3. PNK66, sampling campaign 66; PNK69, sampling campaign 69.

FIG 5

Functional (A) and taxonomic (B) assignment of metagenome/metatranscriptome data sets to selected CAZyme functions. Selections of relevant CAZyme functions were chosen for cellulose, chitin, xylan, and pectin turnover. Additional functions were selected based on a potential involvement in hemicellulose breakdown. Functions are enumerated and indicated at the bottom. % CAZymes refers to the proportion of sequences that were assigned to any CAZyme. Balloon sizes refer to relative abundances. Taxonomic profiles were determined for cellulose, chitin, and xylan turnover on the metagenome/metatranscriptome level. Heatmaps display relative abundances within the corresponding subset. PNK66, sampling campaign 66; PNK69, sampling campaign 69.

FIG 6

Dominating microbial community functions along the Hainich CZE aquifer assemblages. The visualization represents a summary of the findings presented in Fig. 2 to 4. Values in parentheses refer to summed relative abundances of gene functions linked to the respective processes (e.g., nitrification) on the metagenome and metatranscriptome levels. WLP, Wood-Ljungdahl pathway; CBB, Calvin-Benson-Bassham; rTCA, reductive tricarboxylic acid; 3H4H, 3-hydroxypropionate/4-hydroxybutyrate; [CH₂O]_n, biomass; AOA, ammonia-oxidizing archaea.