Late Archean molecular fossils from the Transvaal Supergroup record the antiquity of microbial diversity and aerobiosis

Abstract

Cores recovered during the Agouron Griqualand Drilling Project contain over 2500 m of well-preserved late Archean Transvaal Supergroup sediments, dating from ca. 2.67 to 2.46 Ga. Bitumen extracts of these strata were obtained using clean drilling, sampling and analysis protocols that avoided overprinting syngenetic molecular fossil signatures with contaminant hydrocarbons. Comparisons of biomarker contents in stratigraphically correlated intervals from diverse lithofacies in two boreholes separated by 24 km, as well as across a ∼2 Gyr unconformity, provide compelling support for their syngenetic nature. The suite of molecular fossils identified in the late Archean bitumens includes hopanes attributable to bacteria, potentially including cyanobacteria and methanotrophs, and steranes of eukaryotic origin. This molecular fossil record supports an origin in the Archean Eon of the three Domains of cellular life, as well as of oxygenic photosynthesis and the anabolic use of O₂.

Introduction

Widely accepted evidence for an active microbial biosphere during the Archean Eon (3.8–2.5 Ga) includes physically preserved objects, such as microfossils and stromatolites, and a range of chemical, isotopic and geologic signatures of biogeochemical processes (Schopf and Walter, 1983, Knoll, 2003, Schopf, 2006). Shales bearing abundant organic matter attest to vigorous primary production in marine ecosystems by the middle Archean, and biotic activity may have also played a role in deposition of the massive iron formations of the period (Cloud and Licari, 1968, Cloud, 1973). Although there is little doubt that life had established itself throughout much of the oceans no later than ca. 3.4 Ga (Allwood et al., 2006), there is scant information about what types of organisms were present in Archean marine environments, or what sorts of metabolic processes they relied on.

While there are numerous reports of microfossils in Archean sediments (Schopf, 2006), it is generally agreed that morphology cannot consistently document the phylogenetic affinities or physiological capabilities of Archean microbes. Several sets of criteria for judging the biogenicity of microstructures have been proposed (Schopf, 2006). Archean stromatolites are also controversial biogenic remains (Walter et al., 1980, Walter, 1983, Grotzinger and Rothman, 1996, Hofmann et al., 1999). Their occurrence and diverse morphologies tend to be associated with shallow water depositional settings (Allwood et al., 2006), and some authors attribute particular deposits to microbes capable of oxygenic (Buick, 1992, Altermann et al., 2006) or anoxygenic (Bosak et al., 2007) photosynthesis.

Chemical and isotopic traces of Archean life are widespread and may sometimes be directly associated with particular microfossils (House et al., 2000). Sulfur isotopic data provide indirect evidence for the early evolution of sulfate reduction (Shen et al., 2001, Shen and Buick, 2004). Carbon isotopes of total organic carbon in Archean rocks provide more general information about biogeochemical processes such as carbon assimilation, methanogenesis, methanotrophy and aerobiosis (Hayes, 1983, Eigenbrode and Freeman, 2006). This sedimentary organic matter is the direct geological legacy of microbial activity and, if it is of sufficiently low thermal grade, there is potential for a far more detailed evaluation of the microbiota present at the time of deposition using particular kinds of hydrocarbons, or biomarkers, preserved therein (Brocks and Summons, 2003).

Fossil biomarkers are chemically stable molecules that derive from the carbon skeletons of precursor lipids. Biomarkers become incorporated into sediments, either freely as bitumen or bound into macromolecular organic matter (kerogen), where they may be preserved for billions of years (Eglinton et al., 1964, Brocks et al., 1999, Brocks et al., 2003b). Where these compounds occur intact and uncontaminated, they represent a direct avenue for ancient organisms to leave identifiable traces of themselves in the fossil record. In contrast to bulk chemical and isotopic data, which only carry circumstantial evidence of the metabolic attributes of their sources, biomarkers can carry specific information about the identities and physiologies of organisms because they were, in their original state, functioning components of living cells. In their preserved state, they have chemical structures derived from the original biomolecules through reasonably well-known pathways of diagenetic alteration (Peters et al., 2005). Most paleobiologically informative biomarkers are structurally related to steroids, triterpenoids and photosynthetic pigments of various types (Ourisson and Albrecht, 1992, Brocks and Summons, 2003, Volkman, 2005).

For a biomarker extracted from a rock to be considered a molecular fossil, we must be able to assess its syngeneity: that is, to discern whether or not a particular molecule derives from the original input of organic matter to a sediment. There are two principal routes for non-syngenetic biomarker hydrocarbons to be introduced into sedimentary rocks. First, under certain conditions, hydrocarbons can be widely mobile in sedimentary basins, so bitumen (operationally defined as the solvent-extractable portion of the organic matter) in a particular rock can potentially include material that has migrated between hydraulically connected strata of very different ages. This phenomenon is central to the accumulation of massive bitumen deposits – for example, oil reservoirs – in many petroleum systems. Second, human activity has suffused much of the surface environment with petroleum-derived hydrocarbons, rendering outcrop samples of bitumen-poor, thermally mature Precambrian rocks unsuitable for biomarker analysis. The heavy weathering experienced by most Archean terrains and their low bitumen contents mean that surface exposures are generally compromised. Sampling the subsurface by drilling affords long stretches of pristine stratigraphy, but necessitates contact of the core samples with drilling equipment and fluids. The trace quantities of biomarker molecules extractable from even the best-preserved Archean strata mean that attention to the possibility of even low-level contamination is essential to establishing a genuine molecular fossil record (Sherman et al., 2007).

To date there have been few detailed studies of biomarkers from Archean deposits. Two in recent years (Brocks, 2001, Eigenbrode, 2004) focused on biomarker analysis of resource-exploration cores drilled in the ca. 2.7–2.5 Ga Hamersley Basin, Western Australia. In both cases the syngeneity of the proposed Archean hydrocarbons was carefully assessed, although the approaches and methods differed. Both authors referred to the geological isolation of the basin, the structural integrity of the sediments studied and the absence of younger petroleum source rocks from the Hamersley Basin as valid reasons for discounting contamination from hydrocarbons migrated long distances from adjacent petroleum-prone basins. Brocks (2001) established that the identified Hamersley Basin hydrocarbons were associated with kerogenous shales, that they were at concentrations significantly above procedural blanks and that they showed maturity patterns, especially in respect to aromatic steroids, adamantanes and polyaromatic hydrocarbons, that were consistent with the prehnite-pumpellyite to lower greenschist metamorphic grade of the host rocks. He also showed that there was significant stratigraphic variation in biomarker compositions, that the biomarkers showed typical Precambrian patterns and that inappropriate compounds (e.g. plant terpanes) were not evident. Brocks et al. (2003a) concluded that the biomarkers from the Hamersley and Fortescue groups were ‘probably syngenetic with their Archean host rock’ although they could not absolutely rule out anthropogenic hydrocarbon contamination introduced during drilling, transport and storage of the cores.

Eigenbrode (2004) examined samples representing a wider suite of lithologies (and, therefore, paleoenvironments) and reached similar findings for those wells that were analyzed in common with Brocks (2001). Eigenbrode detected significant dispersion in the δ¹³C values of kerogen that was interpreted in the context of different paleoevironmental settings and a secular trend toward increasing apparent oxygenation of shallow water environments (Eigenbrode and Freeman, 2006). Further, it was shown that some biomarker ratios were strongly correlated to the δ¹³C values of associated kerogens or to dolomite abundance, supporting a syngenetic relationship with host sediments (Eigenbrode, 2004, Eigenbrode, 2008, Eigenbrode et al., 2008).

Another approach to molecular analysis of Precambrian organic matter is the study of hydrocarbons trapped in fluid inclusions. Hydrocarbon-bearing fluid inclusions in Proterozoic rocks from Australia (Dutkiewicz et al., 2003a, Dutkiewicz et al., 2003b, Dutkiewicz et al., 2004, Volk et al., 2003, George et al., 2008), Gabon (Dutkiewicz et al., 2007) and Canada (Dutkiewicz et al., 2006) dating as far back as >2.2 Ga have yielded suites of biomarker compounds including steroids and triterpenoids. Fluid inclusions present unique bitumen trapping conditions, including high fluid pressures and the absence of clay minerals, and the opportunity to assess the inclusion entrapment in the context of the alteration history of the host rock. Fluid inclusion analysis has provided insight into both the molecular fossil record of Precambrian life and the chemical behavior of biomarker hydrocarbons at high temperatures and pressures (George et al., 2008).

In this study, we examined the characteristics of organic matter in two cores (GKF01 and GKP01) drilled as part of the Agouron Griqualand Drilling Project with the express purpose of obtaining fresh, minimally contaminated late Archean sediments for sedimentological, geochemical and paleontological analyses (Beukes et al., 2004, Schröder et al., 2006, Sumner and Beukes, 2006). Importantly, these cores represent some of the first to be recovered from late Archean strata using protocols specifically designed to minimize potential for organic contamination throughout the drilling, handling and storage process. Given the extremely low quantities of extractable hydrocarbons in even the best-preserved Archean sediments, minimization of contamination is essential to avoid overprinting the indigenous organic signatures. A detailed discussion of these measures and the biomarker analysis methods used in this work is presented elsewhere (Sherman et al., 2007). Here we report the results of analysis of Archean biomarkers and correlations of their patterns across the same formations in two cores drilled 24 km apart.