Metatranscriptomic Evidence of Chemolithoautotrophy in the Rifle (CO) Subsurface Relevant to C, S, N, and Fe Cycling

Abstract:

Although there is a limited understanding of the chemolithoautotrophic activity of aquifer microorganisms, such subsurface microbial activity could greatly influence the cycling of elements such as C, S, N, and Fe. Here, we present transcriptional (RNA-Seq) evidence of the emergence of such chemolithoautotrophic activities in groundwater filter samples from a 2-month experiment in which up to 1.5 mM nitrate (a native electron acceptor) was injected into a perennially suboxic/anoxic aquifer (Rifle, CO) containing a large reservoir of reduced Fe- and S-containing compounds. Illumina sequence data from rRNA-subtracted cDNA libraries was assembled and mapped to phylogenetically binned Rifle metagenome data.

Indicative of the activity of Fe(II)-oxidizing bacteria, many high-abundance transcripts mapped to the Gallionellaceae family, whose known members are chemolithoautotrophic bacteria that catalyze Fe(II) oxidation. For example, included among the most abundant transcripts were a cold-shock protein and an acyl carrier protein with 96-98% protein sequence identity to Gallionella capsiferriformans and a nitrite reductase (nirS) gene likely belonging to a Sideroxydans relative. The apparent activity of Gallionellaceae members is consistent with 16S rRNA iTag analyses of these samples, which indicated that Gallionella-related taxa accounted for up to ~50% of these communities. Evidence of sulfide oxidation also was apparent in these samples. For example, highly expressed subunits of APS reductase were very similar to those of the obligately chemolithoautotrophic S- and Fe(II)-oxidizing Thiobacillus denitrificans in terms of sequence identity (98-99%) and synteny of the mapped scaffold. Also highly expressed were a ß-Proteobacterial Form II RubisCO gene and a hydrazine oxidoreductase gene (93% identity to the planctomycete KSU-1), the latter strongly indicative of anaerobic ammonia oxidation (anammox) activity, which has seldom been reported in aquifer environments. Such gene-level data on CO₂ fixation and Fe(II), sulfide, and ammonium oxidation in the Rifle subsurface will contribute to genome-enabled modeling efforts aimed at developing a predictive understanding of biogeochemical processes at the site as part of LBNL's Sustainable Systems Scientific Focus Area (SFA) 2.0.