Microbial communities have been detected in anoxic fractures and rock at up to 4.4 km depth in Precambrian crystalline bedrock, but they are known almost exclusively from marker-gene surveys and low-quality metagenomes, i.e., from genetic potential rather than measured activity. The unresolved problem is to determine what actually sustains and structures these deepest continental communities: whether chemolithoautotrophic primary producers are present and active or their apparent absence is a sampling artifact (ultra-small cells passing 0.2 μm filters, ultra-low biomass, contamination), what in situ metabolic rates (nitrate and sulfate reduction, methanogenesis) and geochemical energy sources (H2, CH4, geogenic organic carbon) support them, and what environmental factors drive the observed depth stratification. Later literature (2020–2026) provides global compositional syntheses, improved contamination methodology, and in-situ rate measurements at other deep sites, but none of this resolves these questions at ultradeep crystalline-bedrock sites themselves.

Background

The deep biosphere is estimated to hold on the order of 15% of Earth's total biomass, yet the deepest reaches of the deep continental biosphere — microbial life in rock fractures and pore fluids hundreds of meters to kilometers below the land surface — are among the least sampled habitats on the planet. A reference point for this problem is the study of Purkamo and colleagues (2020), which characterized microbial communities from two sites in the Fennoscandian Shield (the ancient Precambrian crystalline bedrock of northern Europe): fracture fluids from the Pyhäsalmi mine at 2.4 km depth, and crushed rock from geothermal boreholes at Otaniemi, Espoo, Finland, down to 4.4 km depth — among the deepest microbial samples ever analyzed. The communities were profiled with amplicon sequencing of 16S rRNA genes (the phylogenetic marker used to identify bacteria and archaea; operational taxonomic units, OTUs, are sequence clusters treated as approximate taxa) and of the fungal ITS region (an internal spacer sequence used equivalently for fungi), with functional inference from metagenomics (sequencing of total environmental DNA) and from FAPROTAX, a database that maps taxa to metabolisms but only for cultured, characterized species. Copy numbers of functional marker genes for sulfate reduction (dsrB), methanogenesis (mcrA), and nitrate reduction (narG) were quantified by qPCR (quantitative PCR, an amplification-based counting method for specific genes in a sample).

The study found diverse but unusual communities: chemoheterotrophs — organisms that gain both carbon and energy from pre-formed organic compounds — dominated (e.g., Pseudomonas at Otaniemi), while chemoautotrophs (chemolithoautotrophs), which build biomass from inorganic CO2 and gain energy from reactions between inorganic compounds such as H2 and sulfate, were conspicuously scarce. This was surprising because chemoautotrophs are generally considered the primary producers of the deep biosphere, in the absence of sunlight. The result, however, carries large methodological caveats that the authors themselves emphasized: biomass was so low that autotrophs could have been missed; some deep-subsurface microbes are ultra-small cells that pass through the 0.2 μm filters used to collect biomass; reagent and drilling-fluid contaminants (e.g., Pseudomonas, Streptococcus) could not be fully excluded; the metagenomic assembly was too fragmented (average contig length 408 bp, N50 398 bp — meaning half the assembled fragments were under 398 base pairs) to recover genomes; and FAPROTAX could assign functions to only 11–20% of the detected diversity, since most deep-subsurface lineages have no cultured representative. Consequently, only genetic potential, not metabolic activity or rates, was inferred, and the geochemical energy sources (H2, CH4, organic carbon) that could fuel the community were not directly tied to it. Only two locations were sampled, and the study did not determine which environmental factors drive the distinct community compositions it observed at the different Otaniemi depths. The authors also suggested, as an astrobiological implication, that because the number of autotrophs supporting deep-subsurface communities is sometimes much lower than that of heterotrophs, it might be more reasonable to aim life-detection missions to Mars and icy moons towards chemoorganotrophs rather than autotrophs.

Work since 2020 has deepened the context without closing these questions. Global comparative analyses of subsurface microbiomes (Ruff et al. 2024; González-Rosales et al. 2025) have shown that the terrestrial deep subsurface harbors a core set of recurring taxa and that the inferred contribution of autotrophy increases with depth across four continents — which sharpens, rather than answers, whether the apparent absence of chemoautotrophs at Otaniemi is real or artifactual. A 2024 continental scientific drilling campaign in central Sweden (Westmeijer et al. 2024) demonstrated that notorious kit and drilling-fluid contaminants still dominate sequence libraries from low-biomass crystalline rock even after stringent control-based removal, with autochthonous deep taxa below 0.1% of reads — confirming that contamination control at depth remains unsolved. In-situ observatories elsewhere (e.g., the 2.4 km Kidd Creek observatory in Canada, the Deep Mine Microbial Observatory in the USA) have begun to measure active growth and temporal dynamics, and reviews of the terrestrial deep subsurface and of deep geothermal reservoirs consistently identify microbial activity rates, energy sources, and ecosystem function as unresolved. No published study has returned to the 4.4 km Otaniemi horizon with contamination-controlled, activity-resolving methods.

Problem Statement

Determine what sustains and structures microbial life in anoxic crystalline bedrock at ultradeep depths (below ~2 km, exemplified by the 4.4 km Otaniemi and 2.4 km Pyhäsalmi sites in the Fennoscandian Shield), moving from genetic potential to measured function. A complete answer must establish, from contamination-controlled evidence that accounts for ultra-low biomass and for the possible loss of ultra-small cells through 0.2 μm filtration: (1) whether chemolithoautotrophic primary producers are present and metabolically active in these communities, or whether their apparent absence is an artifact of low biomass, 0.2 μm filtration bias, or contamination; (2) the actual in situ metabolic activities and rates of the communities — including nitrate reduction, sulfate reduction, and methanogenesis — rather than marker-gene inference, and which geochemical energy sources (H2, CH4, geogenic organic carbon) fuel them; and (3) the environmental factors (temperature, pressure, fluid chemistry, depth) that drive the observed differences in community composition between depths and sites. The components form one objective — a functional, mechanistic account of ultradeep crystalline-bedrock life. A complete answer is such an account, built from existing contamination-controlled evidence; where direct measurements of activity or rates at these sites are unavailable, it must specify which measurements would settle components (1) and (2). No new measurements are required.

References

  1. Purkamo L., Kietäväinen R., Nuppunen-Puputti M., Bomberg M., Cousins C. (2020). Ultradeep Microbial Communities at 4.4 km within Crystalline Bedrock: Implications for Habitability in a Planetary Context. Life 10(1):2. DOI: 10.3390/life10010002. https://doi.org/10.3390/life10010002
  2. Nuppunen-Puputti M., Kietäväinen R., Purkamo L., Rajala P., Itävaara M., Kukkonen I., Bomberg M. (2020). Rock Surface Fungi in Deep Continental Biosphere—Exploration of Microbial Community Formation with Subsurface In Situ Biofilm Trap. Microorganisms 9(1):64. DOI: 10.3390/microorganisms9010064. https://doi.org/10.3390/microorganisms9010064
  3. Nuppunen-Puputti M., Kietäväinen R., Raulio M., Soro A., Purkamo L., Kukkonen I., Bomberg M. (2022). Epilithic Microbial Community Functionality in Deep Oligotrophic Continental Bedrock. Frontiers in Microbiology 13:826048. DOI: 10.3389/fmicb.2022.826048. https://doi.org/10.3389/fmicb.2022.826048
  4. Nuppunen-Puputti M., Kietäväinen R., Kukkonen I., Bomberg M. (2023). Implications of a short carbon pulse on biofilm formation on mica schist in microcosms with deep crystalline bedrock groundwater. Frontiers in Microbiology 14:1054084. DOI: 10.3389/fmicb.2023.1054084. https://doi.org/10.3389/fmicb.2023.1054084
  5. Rajala P., Nuppunen-Puputti M., Wheat C. G., Carpen L. (2022). Fluctuation in deep groundwater chemistry and microbial community and their impact on corrosion of stainless-steels. Science of The Total Environment 824:153965. DOI: 10.1016/j.scitotenv.2022.153965. https://doi.org/10.1016/j.scitotenv.2022.153965
  6. Kietäväinen R., Nyyssönen M., Nuppunen-Puputti M., Bomberg M. (2025). Naturally occurring volatile organic compounds in deep bedrock groundwater. Communications Earth & Environment 6:64. DOI: 10.1038/s43247-025-02053-2. https://doi.org/10.1038/s43247-025-02053-2
  7. Wilpiszeski R. L., Sherwood Lollar B., Warr O., House C. H. (2020). In Situ Growth of Halophilic Bacteria in Saline Fracture Fluids from 2.4 km below Surface in the Deep Canadian Shield. Life 10(12):307. DOI: 10.3390/life10120307. https://doi.org/10.3390/life10120307
  8. Meyer-Dombard D. R., Malas J. (2022). Advances in Defining Ecosystem Functions of the Terrestrial Subsurface Biosphere. Frontiers in Microbiology 13:891528. DOI: 10.3389/fmicb.2022.891528. https://doi.org/10.3389/fmicb.2022.891528
  9. Escudero C., Amils R. (2023). Hard rock dark biosphere and habitability. Frontiers in Astronomy and Space Sciences 10:1203845. DOI: 10.3389/fspas.2023.1203845. https://doi.org/10.3389/fspas.2023.1203845
  10. Bregnard D., Leins A., Cailleau G., Vieth-Hillebrand A., Eichinger F., Ianotta J., Hoffmann R., Uhde J., Bindschedler S., Regenspurg S., Junier P. (2023). Unveiling microbial diversity in deep geothermal fluids, from current knowledge and analogous environments. Geothermal Energy 11:28. DOI: 10.1186/s40517-023-00269-z. https://doi.org/10.1186/s40517-023-00269-z
  11. Ruff S. E., de Angelis I. H., Mullis M., Payet J. P., Magnabosco C., Lloyd K. G., Sheik C. S., Steen A. D., Shipunova A., Morozov A., Reese B. K., Bradley J. A., Lemonnier C., Schrenk M. O., Joye S. B., Huber J. A., Probst A. J., Morrison H. G., Sogin M. L., Ladau J., Colwell F. (2024). A global comparison of surface and subsurface microbiomes reveals large-scale biodiversity gradients, and a marine-terrestrial divide. Science Advances 10(51):eadq0645. DOI: 10.1126/sciadv.adq0645. https://doi.org/10.1126/sciadv.adq0645
  12. González-Rosales C., Rezaei Somee M., Buck M., Bertilsson S., Mehrshad M., Dopson M. (2025). A global deep terrestrial biosphere core microbiome. ISME Communications 5:ycaf176. DOI: 10.1093/ismeco/ycaf176. https://doi.org/10.1093/ismeco/ycaf176
  13. Westmeijer G., Escudero C., Bergin C., Turner S., Ståhle M., Mehrshad M., Leroy P., Buck M., López-Hernández P., Kallmeyer J., Amils R., Bertilsson S., Dopson M. (2024). Continental scientific drilling and microbiology: (extremely) low biomass in bedrock of central Sweden. Biogeosciences 21:591–604. DOI: 10.5194/bg-21-591-2024. https://doi.org/10.5194/bg-21-591-2024
  14. Osburn M. R., Casar C. P., Stevenson B. S., Selensky M., Kruger B., Momper L. M., Flynn T. M., Amend J. P. (2026). Microbial Ecology of the Heterogeneous Terrestrial Deep Biosphere Over 4 years in the Deep Mine Microbial Observatory (DeMMO). Journal of Geophysical Research: Biogeosciences 131:e2025JG009515. DOI: 10.1029/2025JG009515. https://doi.org/10.1029/2025JG009515
  15. Saha P. K., Sar P., Sarkar S., Mukherjee D., Kazy S. K. (2026). Deep subsurface rock-hosted chemolithotrophic bacterial communities exhibited differential CO2 assimilation and bioconversion potential under varying oxygen level. Scientific Reports 16:21352. DOI: 10.1038/s41598-026-51641-5. https://doi.org/10.1038/s41598-026-51641-5