Microaerobic steroid biosynthesis and the molecular fossil record of Archean life

Abstract

The power of molecular oxygen to drive many crucial biogeochemical processes, from cellular respiration to rock weathering, makes reconstructing the history of its production and accumulation a first-order question for understanding Earth's evolution. Among the various geochemical proxies for the presence of O(2) in the environment, molecular fossils offer a unique record of O(2) where it was first produced and consumed by biology: in sunlit aquatic habitats. As steroid biosynthesis requires molecular oxygen, fossil steranes have been used to draw inferences about aerobiosis in the early Precambrian. However, better quantitative constraints on the O(2) requirement of this biochemistry would clarify the implications of these molecular fossils for environmental conditions at the time of their production. Here we demonstrate that steroid biosynthesis is a microaerobic process, enabled by dissolved O(2) concentrations in the nanomolar range. We present evidence that microaerobic marine environments (where steroid biosynthesis was possible) could have been widespread and persistent for long periods of time prior to the earliest geologic and isotopic evidence for atmospheric O(2). In the late Archean, molecular oxygen likely cycled as a biogenic trace gas, much as compounds such as dimethylsulfide do today.

Fig. 1.

A summary of the steroid biosynthetic pathway in yeast, showing the flow of carbon from glucose to ergosterol. The first oxygen-requiring step is the epoxidation of squalene; subsequently, O₂ is also required for the removal of three methyl groups from the steroid ring system and the introduction of two unsaturations in order to produce the functional membrane lipid. In these experiments, cells can either synthesize steroids de novo (and hence become labeled with ¹³C) or take up unlabeled ergosterol from the medium. Ultimately, steroid lipids may be incorporated into sedimentary organic matter and, after diagenetic alteration, observed as sterane molecular fossils.

Fig. 2.

Mass spectra of unlabeled (A) and ¹³C-labeled (B) trimethylsilyl derivatives of ergosterol. Arrows indicate mass shifts of selected fragment ions resulting from incorporation of ¹³C from labeled glucose, which produces multiisotopologue ion clusters shifted to masses higher than the unlabeled compound. Ionization was by electron impact at 70 eV; fragment ion assignments are after Brooks et al. (41)

Fig. 3.

Mass spectra of lipids from microaerobic growth experiments. Compounds are indicated (top); dissolved O₂ concentrations in the various growth conditions are indicated (Left). Spectra are labeled by detection of unlabeled (natural isotopic abundance; 98.9% ¹²C) and/or isotopically labeled (multiple-¹³C) versions of each lipid in each condition; ND, not detected. The anaerobic condition was in equilibrium with an atmosphere of < 0.5 ppmv O₂, i.e., < 0.7 nM dissolved O₂. Data are representative of at least two independent biological replicates for each culture condition.

Fig. 4.

The modeled average dissolved oxygen concentration in oxygenated regions of the Archean surface ocean, as a function of the sea-to-air O₂ flux and the proportion of the Earth’s surface contributing to that flux. The larger the outgassing flux, and the smaller the fraction of the earth’s surface it comes from, the more oxygenated surface waters in that region must be. Yellow contours indicate the dissolved oxygen concentrations of the steroid biosynthesis experiments (7 nM, 0.6 μM, and 6.5 μM) and the detection limit of the oxygen microelectrode (0.1 μM). Red bars indicate the range of O₂ fluxes in the Archean atmospheric evolution models of Pavlov et al. (21) (PBK01) and Zahnle et al. (22) (ZCC06).