A preview of this full-text is provided by Springer Nature.
Content available from Nature Microbiology
This content is subject to copyright. Terms and conditions apply.
Articles
https://doi.org/10.1038/s41564-019-0391-z
1Water Studies Centre, School of Chemistry, Monash University, Melbourne, Victoria, Australia. 2School of Earth, Atmosphere & Environment, Monash
University, Melbourne, Victoria, Australia. 3School of Biological Sciences, Monash University, Melbourne, Victoria, Australia. 4Australian Centre for
Ecogenomics, School of Chemistry and Molecular Biosciences, The University of Queensland, St Lucia, Queensland, Australia. 5School of Biological Sciences,
University of Auckland, Auckland, New Zealand. 6Institute for Marine and Atmospheric Research Utrecht, Utrecht University, Utrecht, The Netherlands.
7These authors contributed equally: Adam J. Kessler, Ya-Jou Chen, David W. Waite. *e-mail: perran.cook@monash.edu; chris.greening@monash.edu
At least half of the continental margin is covered by permeable
sediments1. Defined as sands and gravels with permeabilities
exceeding 1 × 10−12 m2, these sediments are highly dynamic
across space and time, especially in contrast to cohesive sediments
(that is, muds and silts)2. Pore-water advection and physical disrup-
tors (for example, tidal flows and groundwater discharge) drive con-
tinual exchange of dissolved particles, solutes and microorganisms
between these sediments and the water column3–5. These sediments
therefore shift between being oxic and anoxic over short distances
and timescales, and rarely become as stratified in their redox chem-
istry as cohesive sediments2,6–8. It was long assumed that sands
are less biogeochemically active than muds, given they harbour
low levels of organic carbon. However, more recent studies have
revealed that sands are highly active: most organic carbon produced
by photoautotrophic and chemolithoautotrophic microorgan-
isms is immediately mineralized by heterotrophs2,9. Cultivation-
independent surveys have consistently shown that sands harbour
phylogenetically and functionally diverse communities of prokary-
otes and microbial eukaryotes10–15. Permeable sediments are now
recognized as key systems for regulating global biogeochemical
cycling and supporting oceanic primary production2,16.
The biogeochemical processes and microbial communities that
mediate carbon cycling are likely to differ between coastal perme-
able and cohesive sediments. In coastal muds, carbon mineraliza-
tion pathways are highly stratified, with oxygen availability and
redox potential decreasing with sediment depth. Within anoxic
zones, organic carbon is primarily mineralized by obligately fer-
mentative bacteria (for example, Clostridiales). The dominant end
products, namely organic acids and molecular hydrogen (H2), are
oxidized by respiratory bacteria through redox cascades controlled
by the concentration and energy yield of available oxidants17,18. H2
is generally maintained at low steady-state concentrations (<2 nM)
through tight coupling of fermentative H2 producers (hydrogeno-
gens) and respiratory H2 consumers (hydrogenotrophs), notably
sulfate reducers (for example, Desulfobacterales)17,19,20. Anoxic
sands also maintain high mineralization rates, but the processes and
communities responsible are unresolved7,21. Denitrification and sul-
fate reduction occur in such sediments, but measured rates for these
processes vary and are often too low to account for carbon min-
eralization rates21–30. Other processes, namely iron reduction and
methanogenesis, occur at low rates despite the availability of elec-
tron acceptors21. Hence, unlike cohesive sediments, anoxic miner-
alization processes in permeable sediments may not be principally
governed by the availability of electron acceptors.
We have recently produced evidence that hydrogenogenic fer-
mentation may be the dominant carbon mineralization pathway
in anoxic permeable sediments. We used flow-through reactor
(FTR) experiments to simulate shifts from oxic to anoxic condi-
tions. Following the transition to anoxia, carbon mineralization
rates were sustained and H2 concomitantly accumulated to high
levels (>1 µM). In contrast, low rates of respiration of sulfate,
nitrate, nitrite and ferrous iron were observed21. Other recent
Bacterial fermentation and respiration processes
are uncoupled in anoxic permeable sediments
AdamJ.Kessler 1,2,7, Ya-JouChen3,7, DavidW.Waite4,5,7, TessHutchinson1, SharlynnKoh1,
M.ElenaPopa 6, JohnBeardall 3, PhilipHugenholtz 4, PerranL.M.Cook 1* and ChrisGreening 3*
Permeable (sandy) sediments cover half of the continental margin and are major regulators of oceanic carbon cycling. The
microbial communities within these highly dynamic sediments frequently shift between oxic and anoxic states, and hence are
less stratified than those in cohesive (muddy) sediments. A major question is, therefore, how these communities maintain
metabolism during oxic–anoxic transitions. Here, we show that molecular hydrogen (H2) accumulates in silicate sand sedi-
ments due to decoupling of bacterial fermentation and respiration processes following anoxia. In situ measurements show
that H2 is 250-fold supersaturated in the water column overlying these sediments and has an isotopic composition consistent
with fermentative production. Genome-resolved shotgun metagenomic profiling suggests that the sands harbour diverse and
specialized microbial communities with a high abundance of [NiFe]-hydrogenase genes. Hydrogenase profiles predict that H2 is
primarily produced by facultatively fermentative bacteria, including the dominant gammaproteobacterial family Woeseiaceae,
and can be consumed by aerobic respiratory bacteria. Flow-through reactor and slurry experiments consistently demonstrate
that H2 is rapidly produced by fermentation following anoxia, immediately consumed by aerobic respiration following reaera-
tion and consumed by sulfate reduction only during prolonged anoxia. Hydrogenotrophic sulfur, nitrate and nitrite reducers
were also detected, although contrary to previous hypotheses there was limited capacity for microalgal fermentation. In com-
bination, these experiments confirm that fermentation dominates anoxic carbon mineralization in these permeable sediments
and, in contrast to the case in cohesive sediments, is largely uncoupled from anaerobic respiration. Frequent changes in oxygen
availability in these sediments may have selected for metabolically flexible bacteria while excluding strict anaerobes.
NATURE MICROBIOLOGY | VOL 4 | JUNE 2019 | 1014–1023 | www.nature.com/naturemicrobiology
1014
The Nature trademark is a registered trademark of Springer Nature Limited.