Research Article

14C-Dead Living Biomass: Evidence for Microbial Assimilation of Ancient Organic Carbon During Shale Weathering

See allHide authors and affiliations

Science  11 May 2001:
Vol. 292, Issue 5519, pp. 1127-1131
DOI: 10.1126/science.1058332


Prokaryotes have been cultured from a modern weathering profile developed on a ∼365-million-year-old black shale that use macromolecular shale organic matter as their sole organic carbon source. Using natural-abundance carbon-14 analysis of membrane lipids, we show that 74 to 94% of lipid carbon in these cultures derives from assimilation of carbon-14–free organic carbon from the shale. These results reveal that microorganisms enriched from shale weathering profiles are able to use a macromolecular and putatively refractory pool of ancient organic matter. This activity may facilitate the oxidation of sedimentary organic matter to inorganic carbon when sedimentary rocks are exposed by erosion. Thus, microorganisms may play a more active role in the geochemical carbon cycle than previously recognized, with profound implications for controls on the abundance of oxygen and carbon dioxide in Earth's atmosphere over geologic time.

The total global inventory of organic carbon (OC) contained in sedimentary rocks is estimated at roughly 1020 mol C (1), more than is present in other surface reservoirs, such as oceans, soils, and biomass. In most rocks, >95% of total organic matter (OM) is nonhydrolyzable, solvent-insoluble material termed kerogen. Kerogen composition is typically dominated by long chains of polymethylenic carbon, linked by carbon, sulfur, and oxygen bridging atoms and aromatic ring centers. As such, kerogen represents a natural form of refractory organic matter, comparable to synthetic polyalkylenes in composition and reactivity. However, given ample time, kerogen is susceptible to oxidation on Earth's surface, as revealed by OM loss (2–4) and CO2 generation (5) upon exposure of shales to oxidizing conditions.

A wide variety of subsurface microbial communities have been recognized that use forms of ancient sedimentary OC. Most of these are anaerobes, because subsurface environments are largely anoxic. These include sulfate reducers living in deep subsurface shales and sandstones (6, 7), anaerobes living in oil-field brines and oil pipelines (8), and hydrocarbon-degrading microorganisms living in petroleum-contaminated aquifers (9, 10). To date, none of these anaerobic communities has been shown capable of degrading kerogen. However, there is evidence that kerogen loss and degradation is geologically rapid in well-oxygenated environments. Studies reveal sharp oxidation fronts in turbidites (11, 12) and shales (2) after exposure to oxygen, by contrast with long periods of OM preservation under anoxic conditions in the subsurface.

14C provides a useful tracer of carbon sources in many surficial environments. Atmospheric CO2 is enriched in14C, which decays to 14N with a half-life of 5730 years. Thus, any organic material greater than ∼60,000 years in age, or any biomolecule assembled from an ancient carbon source, contains essentially no 14C. This discrepancy between modern and ancient carbon provides evidence for kerogen oxidation in glacial tills (5). Accelerator mass spectrometry of extremely small masses of carbon allows measurement of the14C content of individual organic compounds. Such compound-specific radiocarbon analyses demonstrate addition of ancient hydrocarbons to modern soils (13), reburial of ancient OM in modern marine sediments (14), and chemoautotrophic fixation of 14C-depleted deep-water CO2 (aq) by Archaea in the modern ocean (15).

Phospholipids are integral components of the cell membrane lipid bilayers. Because phospholipids are rapidly degraded after cell death, these compounds are not found in ancient sedimentary rocks and instead indicate the presence of living organisms. Bacterial and eukaryotic cell membranes contain a wide diversity of ester-linked phospholipid fatty acids (PLFAs). Whereas the straight-chain C16:0 (palmitic acid) and C18:0 (stearic acid) PLFAs (16) are ubiquitous among bacteria and eukaryotes, odd-numbered, methyl-branched, and monounsaturated PLFAs are commonly assigned a purely bacterial origin (17, 18). PLFAs are biosynthesized from common precursor biomolecules (e.g., acetyl–coenzyme A) and thus can be used for compound-specific14C analysis to trace carbon sources and pathways in microbial communities.

Here, we used natural-abundance 13C and 14C analysis of PLFAs to establish that microorganisms enriched from a black shale weathering profile can assimilate shale OM into cellular carbon. The New Albany Shale is a Late Devonian black shale [5 to 15% total organic carbon (TOC)] exposed in Illinois, Indiana, and Kentucky, USA. An exposure near Clay City, Kentucky, has been well characterized by previous studies of black shale weathering. At this site, a recent roadcut has exposed a weathering profile, revealing the complete transition from subsurface, unweathered, high-OC shale to heavily weathered, low-OC regolith at the surface of the profile (Fig. 1).

Figure 1

Field site near Clay City, Kentucky, USA, where a weathering profile developed on New Albany Shale has been exposed by a recent roadcut. Unweathered shale (left) grades laterally into heavily weathered regolith (right). Evidence for weathering includes loss of color, increased fissility and friability, loss of organic carbon, and loss of pyrite.

Sampling and microbial cultures. A core into the Clay City weathering profile was obtained in January 2000 and sectioned on site (19) (Table 1). Samples for PLFA analysis were removed from six depths for PLFA characterization (20). Small masses were removed from the interior of the core at several depths and transferred to sterile glass culture tubes containing OC-free growth medium for enrichment culturing (21). Given the potentially wide diversity of metabolic types in the weathering profile samples, the goal of culturing was to enrich specifically for populations of microorganisms able to use shale OM as their carbon source, namely, aerobic heterotrophs that can assimilate kerogen. After 10 days, a small volume from each of these first-generation enrichments was added to fresh medium that contained shale substrate (22). Multiple subcultures over several months in fresh medium and substrate ensured that subsequent generations of cultures were accessing only substrate OM, and not OM entrained in the inoculum. The abundance and morphology of cells (Fig. 2) were monitored by examining stained samples by epifluorescence microscopy (23). A large-volume culture was initiated to generate sufficient biomass for PLFA14C analysis, inoculated with material enriched from the 50- to 60-cm core depth after six dilutions (24). After 90 days' growth, the large-volume culture was harvested for PLFA characterization (25). The large sample size afforded sufficient quantities of four classes of PLFA, which were submitted for14C analysis at the National Ocean Science Accelerator Mass Spectrometry Facility, Woods Hole Oceanographic Institution, after further sample preparation (26).

Figure 2

Photomicrographs of stained microorganisms in association with shale particles. (A) Core sample from 50- to 60-cm depth, stained with DAPI. OM fragments fluoresce yellow-orange; stained cells fluoresce blue. Cells appear associated with clusters of mineral fragments and OM. (B) Large-volume enrichment culture after 90 days' growth, stained with acridine orange. Stained cells fluoresce green; OM fragments fluoresce red-orange. Cells are gathered along the surfaces of OM fragments. Most cells are prokaryotic rods and cocci, although several eukaryote filaments are visible. Bars, 25 μm.

Table 1

Corg, Δ14C, PLFA, C16:0, and calculated biomass for core samples and enrichment culture. ND, not determined.

View this table:

PLFA distributions in kerogen-degradation environments.Forty individual PLFAs were identified in core samples (Fig. 3). In all core samples, C16:0, C18:0, C18:1, and C18:2 are the most abundant PLFAs. The presence of methyl-branched, monounsaturated, and cyclopropyl PLFAs revealed substantial bacterial contributions to the total PLFA pool. The highest concentration and greatest diversity of PLFAs were measured in the surface samples. Abundance of total PLFAs decreased downcore from 226 nmol/g at the surface to 12 nmol/g in the deepest sample, accompanied by a shift toward relatively greater contributions from branched, monounsaturated, and cyclopropyl PLFAs. Only seven PLFAs were identified in the large-volume culture, but total PLFA concentrations in culture were similar to those in deeper sections of the weathering profile core.

Figure 3

Concentrations of total PLFA and compound classes for selected depths (in cm) within weathering profile core and large-volume culture. In the core surface sample, linear PLFAs were predominant. In all other samples, unsaturated and cyclopropyl PLFAs were predominant.

PLFAs isolated from microorganisms living on black shales provide a window into the diversity and metabolic activity of subsurface microbial communities. The abundance of total PLFA in the core examined in this study is lower than PLFA concentrations measured in typical forest soils (27) or marine sediments (28), but greater than that measured in anoxic subsurface environments (7–10). This suggests that even in environments where O2 is available, kerogen degradation supports only a limited microbial population compared with environments where fresh terrestrial or marine OM is consumed. PLFAs in this weathering profile core also provided some taxonomic information. PLFA distributions suggest a community dominated by Bacteria with some Eukarya. Certain PLFAs detected in the core have been correlated with the presence of sulfate-reducing bacteria and other groups, e.g., 10Me16:0, which is a marker for Desulfobacter spp. (29), and cyclopropyl C17:0 and C19:0, which indicate microaerophilic Gram-negative bacteria (30). Whole-cell hybridizations with domain-specific probes confirm the presence of Bacteria and Eukarya (31); however, more-refined phylogenetic characterization of these samples is not possible with PLFA and, instead, will require DNA extracted from these samples.

PLFA abundance in the enrichment cultures provides an estimate of microbial carbon assimilation. After 90 days' growth, the 40-g culture supported a microbial population containing 200 μg C as PLFA. Given that dry-cell biomass is ∼10% phospholipid, this approximates to 2 mg of total biomass carbon. The 40-g shale substrate contained 1.6 g TOC. Assuming a steady rate of carbon assimilation, this corresponds to 20 μg C/day in a population of ∼108cells/ml, given well-mixed conditions and ample substrate and growth medium. However, these organisms were probably actively respiring CO2 during the course of the incubation, so the rate of kerogen degradation is probably much greater.

PLFA carbon isotopic composition. Stable carbon isotope (δ13C) values for PLFAs isolated from the core samples indicated δ13C variability among the PLFAs (Fig. 4). Measured δ13C values for kerogen (−29.5‰) and the C16:0 and C18:0PLFAs (−28 to −30‰) were similar, whereas the C18:1 and C18:2 were 3 to 5‰ depleted relative to kerogen. The C16:0 and C18:1 PLFAs had consistent δ13C values throughout the core, whereas the C18:2 and C18:0 revealed depth-related trends in isotopic composition. For depths where those compounds are present, the unsaturated and methyl-branched PLFAs were between 3 to 6‰ depleted relative to kerogen. Leaf litter and humus collected from the core top measured –25.7‰, which is ∼4‰ enriched relative to shale kerogen. Isotopic compositions for PLFAs from the culture share similarities with PLFAs from the core. The C18:0and C16:0 PLFAs are most enriched, and the unsaturated and cyclopropyl PLFAs are more depleted. Overall, however, these PLFAs are enriched in 13C relative to the kerogen substrate provided in the culture.

Figure 4

Stable carbon isotopic composition of PLFAs from the three deepest sections of weathering profile core, reported in ‰ on the δ13C scale.

The 13C depletion in PLFAs isolated from the core relative to kerogen can be understood as the mixed contribution from heterotrophy of kerogen and chemoautotrophy of weathering profile CO2. Because the soil pore spaces are in limited exchange with the atmosphere, respiration results in accumulation of CO2 derived from organic matter. Thus, the CO2available for autotrophy is 13C-depleted, and the resulting lipids will reflect this depletion.

The Δ14C value of the kerogen substrate used for the culture measured –990‰, or essentially no 14C. Four PLFA compound classes were isolated from the culture for 14C analysis: (i) C16:0, (ii) C18:0, (iii) C18:1 + C18:2, and (iv) cycC17:0 + cycC19:0. The measured Δ14C for these four PLFA classes ranged from –711‰ to –922‰ (Table 2). These Δ14C values were used in a linear mixing model to estimate the relative contribution of kerogen (−990‰) and modern carbon (+100‰). The estimated fraction of carbon in these PLFAs derived from kerogen ranged from 74 to 94%. The remainder of the carbon probably derived from chemoautotrophic fixation of atmospheric CO2 dissolved in the low-pH growth medium.

Table 2

14C and 13C analysis of PLFA compound classes isolated from enrichment culture grown on New Albany Shale, and calculated fraction of PLFA carbon derived from ancient kerogen.

View this table:

Carbon pathways in kerogen-degrading microorganisms. Taken together, the 13C and 14C data on PLFAs from the large-volume culture provide interpretation of carbon pathways in this microbial community. In culture, there are two carbon sources of distinct 13C isotopic composition: kerogen (−29.5‰) and dissolved CO2, which is itself a mixture of atmospheric CO2 (−8‰), dissolved carbonate (∼0‰), and CO2 derived from degraded kerogen (−29.5‰). PLFA Δ14C values require that these organisms obtain 74 to 94% carbon from a 14C-depleted source. This excludes atmospheric CO2 fixation as the dominant carbon pathway in this community. Other pathways include (i) kerogen assimilation by heterotrophs, (ii) remineralization followed by fixation of kerogen-derived CO2, or (iii) dissolution of shale carbonate followed by fixation of carbonate-derived CO2(substrate measured <0.1% Cinorganic). Lipids of most autotrophic organisms are significantly (20 to 30‰) depleted in 13C relative to their CO2 source, whereas lipids of heterotrophic organisms measure generally ±4‰ relative to their carbon source. Culture PLFAs ranged from –24 to –27‰. These values are possible only under scenarios (i) and (iii) above; fixation of CO2 derived from oxidized kerogen would be associated with substantial isotope depletion. Scenario (iii) is less likely than (i), because the growth medium was well mixed and in free exchange with the atmosphere, and during the 90 days of the experiment trended toward isotopic equilibrium with atmospheric CO2. Subsequent culturing experiments have further excluded scenario (iii). Cultures grown with HCl-digested shale and demineralized kerogen as substrate have revealed that microorganisms enriched from the weathering profile are able to grow in the absence of carbonate and on pure OM without pyrite (a likely electron donor for chemoautotrophy) (32). Cell morphologies, generation times, and stationary-phase populations in these incubations were comparable to growth on shale substrate. Growth experiments with polished New Albany Shale blocks (33) as substrate have also revealed a strong affinity for microbial colonization along select, OM-rich microlaminae within the rock (Fig. 5). This association reveals that population growth is enhanced along surfaces rich in OM, suggesting that the organisms are actively using kerogen and not CO2 released into solution by oxidation. The explanation consistent with 13C and 14C analysis and the various culturing experiments is that these organisms are heterotrophs that use kerogen.

Figure 5

(A and B) Photomicrographs of microbial colonies attached to a polished surface of New Albany Shale. Polished blocks were colonized for 96 hours, rinsed in growth medium, stained with acridine orange, and viewed by epifluorescence microscopy. Colonies did not grow radially, but instead followed fine, dark laminae within the shale that are rich in OM. Colonies appear to have grown around large mineral grains, indicating selective attachment to the shale surface. Bars, 25 μm.

The relative patterns of PLFA δ13C values between the core and culture samples are similar. C16:0 and C18:0 PLFAs are enriched in 13C relative to cyclopropyl and to unsaturated and branched PLFAs, consistent with a previous study on bacterial PLFA isotopic composition variability (34). The overall enrichment in13C in culture samples relative to the core can be understood as a difference in the δ13C of CO2available to chemoautotrophic organisms in these two environments.

14C measurements of PLFAs from the weathering profile were not initially sought because resulting data are likely to be more ambiguous than the simple culture experiment owing to the need to account for multiple carbon sources. These include kerogen, atmospheric CO2, soil OM, respired kerogen and soil OM, and carbonates.

Implications for global carbon and oxygen cycling.The presence of microorganisms capable of using kerogen may have significant implications for global-scale cycling of carbon and oxygen. The concentration of oxygen in Earth's atmosphere is controlled by a balance between the burial of OM in sediments and the subsequent weathering of ancient sedimentary rocks on the continents (35, 36). Photosynthesis results in fixation of CO2—chemical reduction of CO2 to carbohydrates—in the process releasing O2 to the atmosphere and ocean surface waters. Although most biomass carbon is respired or otherwise remineralized to CO2 within months after fixation, a small fraction of biomass becomes buried in sediments. Organic carbon buried in sediments represents a net addition of O2 to the atmosphere, understood as total oxygen produced by photosynthesis less the oxygen consumed by respiration. Eventually, sediments become lithified to sedimentary rocks, and processes of uplift and exhumation result in exposures of ancient sedimentary rocks on Earth's continents. Weathering and erosion expose chemically reduced organic carbon in sedimentary rocks to the O2-rich environment of the atmosphere. At sites of weathering of ancient sedimentary rocks, oxidation and loss of OM in sedimentary rocks, as a result of chemical weathering, have been shown. This oxidation consumes O2 by mechanism(s) that, on a global scale, to a large degree balance the net oxygen release associated with OM burial. This study shows that one of these mechanisms is biological assimilation of kerogen. Remarkably, this near balance between O2 production and consumption has been maintained for at least the past ∼560 million years. Small imbalances between global rates of OM burial in sediments and OM oxidation during weathering have resulted in modest changes in atmospheric O2 during Earth's past (37,38). However, controls on OM oxidation during weathering remain poorly understood, and thus understanding of the evolution of Earth's atmosphere remains limited.

Relict, ancient OM has been detected in modern sediments (14), suggesting that OM oxidation is not 100% efficient in all environments during weathering. The presence of this residual OM after weathering implies that quite commonly, oxidative weathering of ancient OM occurs more slowly than rates of erosion and mass transport. This suggests a kinetic barrier to OM oxidation in shale weathering profiles in spite of abundant available oxygen.14C-depleted bacterial phospholipids demonstrate that kerogen is or becomes to some extent bioavailable and able to support growth of some microorganisms during weathering. Thus, biological processes may play a previously unrecognized pivotal role in controlling the rate of oxidation of OM during weathering of sedimentary rocks. The limited taxonomic and metabolic information obtained from PLFA analysis indicates that this community contains a consortium of heterotrophs and autotrophs, and aerobes and anaerobes. Thus, it is unclear what effect variations in O2concentration would play on the rate of kerogen degradation by this community. Further study is required before any possible feedback between atmospheric oxygen concentration and rates of microbial kerogen degradation can be established. Biological activity has long been acknowledged as a key component of the geochemical carbon cycle, owing to the role that organisms play in OM production and subsequent degradation during sediment burial and diagenesis. Discovery of14C-depleted living organisms growing on ancient shale OM indicates that remineralization of ancient organic matter during weathering of sedimentary rocks may likewise be controlled by microbial activity.

  • * To whom correspondence should be addressed. E-mail: spetsch{at}


Stay Connected to Science

Navigate This Article