Robert Frost’s poem Fire and Ice ponders which of these two will eventually cause life on Earth to cease. Conversely, scientists have long been interested in whether life on this planet originated in hot or cold conditions. Did life arise in a hot volcanic environment1,2 or, as Charles Darwin suggested in a letter (see go.nature.com/2q8w3n5), in “some warm little pond”? Might we need to invoke a global setting that includes ice3 or would an ocean-floor setting suffice1–5? Writing in Nature, Ménez et al.6 report an analysis of rocky material below the ocean floor at sites called serpentinizing hydrothermal vents, which are hot springs that discharge alkaline, gas-rich water. The authors provide evidence that a chemical reaction occurs there that could have set the stage for life to begin — the generation of an amino acid by a process that is not biologically mediated.
A well-characterized form of hydrothermal vent on the ocean floor, called black-smoker chimneys, occurs in a volcanic, magma-rich setting. These vents emit acidic fluids into the ocean that have high concentrations of gases and also of metals that are in a chemically reduced form.
However, in 2000, the serendipitous discovery of a mid-Atlantic ocean-floor site called Lost City revealed another type of hydrothermal-vent environment7. The Lost City vents were found to arise by a process termed serpentinization — a chemical interaction between water and a type of rock called peridotite that contains minerals enriched in magnesium, iron and silica. Serpentinization generates alkaline conditions that aid the formation of majestic carbonate-rich towers (Fig. 1). Serpentinization also produces hydrogen and a variety of organic molecules, including formate, acetate and pyruvate, which might be important in supporting the microbial life at Lost City and could also have been used in the biochemical steps that led to life on Earth8,9. Studies of Lost City have provided a treasure trove of scientific discoveries that have greatly altered our understanding of hydrothermal vents and the geological history of the early Earth10. They have also led to testable hypotheses about possible environmental settings for the origin of life.
In laboratory experiments that simulate conditions in magma-hosted hydrothermal vents, amino acids can be synthesized by chemical reactions11 that do not require biological input. However, whether such abiotic generation of amino acids can occur in serpentinizing hydrothermal vents was unknown. Amino acids have been detected12 in fluids emanating from Lost City, but their source was undetermined.
To investigate further, Ménez and colleagues analysed material from rock samples retrieved by drilling around 170 metres below the ocean floor at Lost City. The authors present extensive evidence for the presence of the nitrogen-containing amino acid tryptophan in a context in which it was unlikely to have been produced by a biologically mediated process. They report data obtained from three high-resolution techniques that are consistent with the presence of tryptophan. The researchers also found other organic compounds that might be intermediates in the synthesis of tryptophan. Ménez et al. propose that this synthesis could be the result of Friedel–Crafts reactions because they found the molecule indole, which is an intermediate organic compound in the synthesis of tryptophan by this type of reaction. Their case for the abiotic synthesis of tryptophan is strengthened by the absence of other amino acids that would be present if a biological source was there, such as microbial contamination of the rock sample.
To extend Ménez and co-workers’ report of the abiotic synthesis of tryptophan, future studies at Lost City should try to collect adequate volumes of fluid to determine a structural property, called chirality, of the tryptophan present. Molecules can exist in two mirror-image chiral forms. Synthesis of a molecule by a non-biological process generally results in equal proportions of these two forms, whereas biologically synthesized amino acids are usually made in predominantly one form or the other.
The authors’ work also sheds light on the long-standing mystery of what mechanism reduces nitrogen molecules (N2) to ammonia under hydrothermal-vent conditions. In most of the cases in which ammonia has been detected in hydrothermal-vent environments, it was found to originate from buried sediment sources of organic material and not from abiotic synthesis in the hydrothermal vents13. Ménez and colleagues propose that saponite, an iron-containing clay mineral that they detected, and which is reported14 to be a catalyst for the synthesis of organic compounds and the reduction of N2 to ammonia, might be involved in tryptophan synthesis. Abiotic generation of a source of ammonia, together with saponite’s proposed catalysis of heterocyclic-amine molecules such as tryptophan, also raises the possibility of abiotic synthesis of other heterocyclic amines called pyrimidines and purines, which are components of the nucleic acids DNA and RNA. Moreover, saponite has the potential to promote the formation of organic polymers15,16.
Beyond the potential for the synthesis and accumulation of organic compounds that were probably important in the origin of life, serpentinization has two other characteristics that have intriguing implications regarding the origin of life and the establishment of habitable conditions5,8. One characteristic is that serpentinization produces heat. The gradient of temperature can reach more than 200 °C at the site of the serpentinization reaction4. This, in turn, promotes hydration and therefore expansion of rocks, which is the other intriguing characteristic of serpentinization. However, if part of the ocean floor ‘sinks’ (subducts) into Earth’s interior as tectonic plates move, the greater heat and pressure encountered on its descent into the deep subsurface region would reverse such serpentinization, and the water released during this reversal could help to give rise to volcanoes on the ocean floor14. This, in turn, might help to recycle key elements that support life.
The geological record at the time of life’s origin, 3.5 billion to 4.4 billion years ago, is enriched in iron- and magnesium-containing minerals (characteristic of the rocks that form Earth’s mafic crust), and in other elements that could have been extracted from rock in a process mediated by high-temperature water, strongly pointing to hydrothermal activity at that time. Yet, during the first billion years of Earth’s history, the heat from the mantle was too great for plate tectonics to occur17,18. Consequently, heat would have been lost from Earth’s interior mainly through volcanoes on the ocean floor. Earth’s crust would have been rich in silicate minerals and iron17,18, allowing high rates of serpentinization and producing high concentrations of hydrogen and organic compounds.
Extensive circulation of seawater through volcanic rock during this time might have resulted in heat, fluid and gas ascending from the depths to create convective cells — a phenomenon characterized by currents due to density differences in the liquids or gases present. In the volcanic-rock environment, this could have led to associated gradients of temperature, pressure, chemical composition and wet–dry cycles (hydration–dehydration cycles known to promote chemical reactions that include the polymerization of organic compounds). Regardless of how life originated and in what environmental setting it was first established, serpentinization probably had an important role in facilitating the availability of organic chemicals required for life.
Understanding serpentinization at Lost City has wider scientific implications. Saturn’s icy moon Enceladus has many of the chemical properties known to support life19,20 that are seen in serpentinizing environments such as those of Lost City. Whether or not Enceladus, or indeed other icy moons such as Jupiter’s Europa, could or did support life, they nevertheless could provide insight into geochemical processes that might lead to life. Such geochemical analysis seems to support the hypothesis that hydrothermal systems might have had an essential role in the origin of life. A more far-reaching implication of the work by Ménez and colleagues, and of others investigating hydrothermal vents, is that efforts to understand the characteristics of these settings might aid efforts to search for life beyond Earth. A planetary body with evidence of geophysical properties, including plate tectonics and hydrothermal systems, might have a higher probability of acquiring and supporting carbon-based life than planetary bodies lacking such geophysical properties. If true, then targeting such planets might also increase our probability of finding such life.