The iron–sulfur world hypothesis is a group of ideas about how life began and how it changed early in Earth's history. These ideas were introduced in a series of articles between 1988 and 1992 by Günter Wächtershäuser, a Munich patent lawyer who studied chemistry. He was supported in publishing his ideas by philosopher Karl R. Popper. The hypothesis suggests that early life may have developed on the surface of iron sulfide minerals, which is why it is called the iron–sulfur world hypothesis. The idea was formed by working backward from current biochemical knowledge (biochemistry that still exists today) and through chemical experiments.
Origin of life
Wächtershäuser suggests that the first living organism, called the "pioneer organism," formed in a volcanic hydrothermal flow under high pressure and high temperature (100 °C). This organism had a structure made of minerals with catalytic transition metal centers, primarily iron and nickel, but also possibly cobalt, manganese, tungsten, and zinc. These metal centers helped create organic compounds from inorganic gases like carbon monoxide, carbon dioxide, hydrogen cyanide, and hydrogen sulfide. These organic compounds attached to the mineral base as ligands, staying connected based on how strongly they bonded to the metals. This process created an autocatalytic "surface metabolism," where the organic products accelerated the metal centers, and the metabolism became self-sustaining through a cycle similar to the reductive citric acid cycle. As the metabolism grew, it produced more complex organic compounds, pathways, and catalytic centers.
The water–gas shift reaction (CO + H₂O → CO₂ + H₂) happens in volcanic fluids with or without catalysts. Ferrous sulfide (FeS, troilite) and hydrogen sulfide (H₂S) act as reducing agents, forming disulfide bonds (S–S) and pyrite (FeS₂) under mild volcanic conditions. This process has been debated. Nitrogen fixation using N₂ and pyrite formation has been demonstrated, with ammonia forming from nitrate using FeS/H₂S as a reductant. Methylmercaptan (CH₃-SH) and carbon oxysulfide (COS) form from CO₂ and FeS/H₂S, or from CO and H₂ in the presence of NiS.
When carbon monoxide (CO), hydrogen sulfide (H₂S), and methanethiol (CH₃SH) react with nickel sulfide and iron sulfide, they produce methyl thioester of acetic acid (CH₃-CO-SCH₃) and thioacetic acid (CH₃-CO-SH), which are simple versions of acetyl-CoA. These compounds serve as starting materials for energy-coupling reactions, such as forming (phospho)anhydride compounds. However, Huber and Wächtershäuser found low acetate yields (0.5%) from methanethiol (8 mM) and CO (350 mM), which are much higher than concentrations measured in natural hydrothermal vent fluids.
Nickel hydroxide reacting with hydrogen cyanide (HCN), along with ferrous hydroxide, hydrogen sulfide, or methyl mercaptan, forms nickel cyanide. This reacts with CO to produce α-hydroxy and α-amino acids, such as glycolate/glycine, lactate/alanine, glycerate/serine, and pyruvic acid. Pyruvic acid also forms from CO, H₂O, and FeS under high pressure and temperature with nonyl mercaptan. Reacting pyruvic acid or other α-keto acids with ammonia in the presence of ferrous hydroxide or FeS/H₂S produces α-amino acids like alanine. When α-amino acids react with COS or CO/H₂S in water, they form peptides, which later break down into dipeptides, tripeptides, and other compounds through N-terminal hydantoin or urea moieties.
A proposed mechanism for CO₂ reduction on FeS involves mackinawite (FeS) converting to pyrite (FeS₂) when reacting with H₂S up to 300 °C, but this requires an oxidant. Without an oxidant, FeS reacts with H₂S to form pyrrhotite. Farid et al. found that FeS can reduce CO₂ to CO above 300 °C, with FeS surfaces oxidizing and reacting with H₂S to form pyrite. In the Drobner experiment, CO and H₂O likely react to produce H₂.
Early evolution
Early evolution begins with the start of life and ends with the last universal common ancestor (LUCA). According to the iron–sulfur world theory, this period includes the development of cellular organization (cellularization), the genetic machinery, and the enzymatic processes of metabolism.
Cellularization happens in several steps. It may have started with the formation of simple lipids, such as fatty acids or isoprenoids, in the surface metabolism. These lipids collect on or inside a mineral base. This process helps lipids stick to the mineral’s surface, which encourages chemical reactions that form molecules instead of breaking them apart by reducing water and proton activity.
Next, lipid membranes form. These membranes are still attached to the mineral base and create a partially enclosed space, with the mineral and membrane working together. As lipids evolve further, they form self-supporting membranes and closed cells. The earliest closed cells are called pre-cells (sensu Kandler) because they allow frequent sharing of genetic material, such as through cell fusions. According to Woese, this frequent exchange of genetic material explains the shared origin of all life and the rapid early evolution. Nick Lane and coauthors note that non-enzymatic versions of glycolysis, the pentose phosphate pathway, and gluconeogenesis have been identified. Amino acids can also be made from α-keto acids through reductive amination and transamination reactions. Long-chain fatty acids can form through hydrothermal Fischer-Tropsch-type synthesis, a process similar to fatty acid elongation. Recent research suggests nucleobases might form using metal ions as catalysts, following the same pathways seen in modern life.
Metabolic intermediates like glucose, pyruvate, ribose 5-phosphate, and erythrose-4-phosphate are created naturally in the presence of Fe(II). Fructose 1,6-biphosphate, a key molecule in gluconeogenesis, accumulates in frozen solutions. Its formation is faster with lysine and glycine, suggesting the earliest enzymes were amino acids. Studies show that iron-sulfur clusters, such as 4Fe-4S, 2Fe-2S, and mononuclear iron, form naturally in low concentrations of cysteine and at alkaline pH. Methyl thioacetate, a precursor to acetyl-CoA, can be made under conditions similar to hydrothermal vents. Phosphorylating methyl thioacetate creates thioacetate, a simpler precursor to acetyl-CoA. Thioacetate can then form acetyl phosphate, which is a building block for ATP and can help create ATP from ribose and nucleosides. This suggests acetyl phosphate was likely made through thermophoresis and mixing of acidic seawater with alkaline hydrothermal fluid in tiny pores. It may have helped form nucleotides on mineral surfaces or in areas with less water. Thermophoresis at hydrothermal vent pores can concentrate polyribonucleotides, but how it promoted coding and metabolic reactions remains unclear.
In mathematical models, self-catalytic nucleotide synthesis is thought to help protocells grow by also helping fix CO₂. Strong nucleotide catalysis of fatty acids and amino acids might slow protocell growth, but weaker or moderate catalysis could help protocells divide and grow. In 2017, a simulation of a protocell near an alkaline hydrothermal vent showed that some hydrophobic amino acids bind to FeS nanocrystals, creating three effects: more catalytic surface area, FeS nanocrystals moving to the membrane, and a proton-motive site for fixing carbon, similar to the enzyme Ech. Maximum ATP synthesis would have occurred in freshwater with high water activity, but high Mg and Ca levels in modern oceans prevent ATP synthesis. Hadean oceans had much lower Mg and Ca concentrations, and alkaline hydrothermal vents typically have lower levels than the ocean. These conditions would have produced Fe, which could help form ATP from ADP. Mixing seawater with alkaline hydrothermal fluid might have helped cycle Fe and Fe. Research on prebiotic reactions, such as NAD reduction and phosphoryl transfer, supports the idea that life began near alkaline hydrothermal vents.
William Martin and Michael Russell suggest that the first cells may have formed inside alkaline hydrothermal vents at deep-sea spreading zones. These vents have tiny caverns lined with thin metal sulfide membranes. This model places LUCA within the physical structure of an alkaline hydrothermal vent, not as a free-living cell. The final step before free-living cells would be forming a lipid membrane, allowing organisms to leave the vent. The later development of lipid membranes guided by genetic peptides explains why archaea, bacteria, and eukaryotes have different types of membrane lipids. The vents they describe are chemically similar to warm off-ridge vents like Lost City, not the hotter black smoker vents.
In a world without life, temperature and chemical concentration differences near vents could help form organic molecules. Hotter areas near vents have more chemicals, while cooler areas have fewer. Molecules move from high-concentration areas to low-concentration areas, creating a flow that supports early metabolic processes, like making acetic acid and oxidizing it.
This process allowed many reactions now found in central metabolism to occur before cell membranes formed. Each vent compartment acts like a single cell. Communities of chemicals that are stable and adaptable would thrive, using up key chemicals in their area. Over time, these chemicals could be incorporated into cell membranes, increasing metabolic complexity inside cells while simplifying the environment outside. This might eventually lead to complex systems that can sustain themselves.
Russell adds that semi-permeable mackinawite (an iron sulfide mineral) and silicate membranes could naturally form under these conditions, linking chemical reactions in different areas through electrochemical processes.
Alternative environment
Six out of the 11 metabolic intermediates in the reverse Krebs cycle are supported by iron (Fe), zinc (Zn), and chromium (Cr) under acidic conditions. This suggests that early life may have developed in areas on Earth with high levels of metals and acidic water, such as hydrothermal fields. Acidic conditions appear to help protect RNA, a molecule essential for life. These hydrothermal fields likely experienced repeated freezing and thawing, along with changes in temperature, which could have encouraged chemical reactions that form sugars, nucleobase building blocks, and RNA molecules. Energy for processes like making ATP and converting iron into other forms might have come from sunlight, lightning, meteorite impacts, or volcanic activity.
Wet and dry cycles in these fields could have helped form RNA and proteins. When protocells (early cell-like structures) gathered in a wet, gel-like environment during these cycles, they might have shared chemical products with nearby protocells. This process resembles a simple form of horizontal gene transfer, where genetic material is exchanged between cells. Fatty acid membranes, which could form early cell boundaries, would be stabilized by certain polymers and magnesium (Mg), which is needed for RNA-related reactions. These processes might have occurred in shaded areas that shielded early life from harmful ultraviolet light. Long-chain alcohols and simple organic acids could also have formed through a chemical process called Fischer–Tropsch synthesis. Hydrothermal fields likely contained deposits of transition metals and concentrated elements like carbon, hydrogen, nitrogen, oxygen, phosphorus, and sulfur (CHNOPS). Energy from geothermal convection might have powered early processes such as proton movement, energy transfer, and transport of molecules across cell membranes. Scientists David Deamer and Bruce Damer note that these environments resemble Charles Darwin’s idea of a "warm little pond" where life began.
However, the hypothesis that life originated in subaerial (land-based) hydrothermal fields has issues. The chemical reactions proposed do not match known biochemical processes. These environments were rare and lacked protection from harmful meteorites or ultraviolet light. Clay minerals in these areas might have trapped organic molecules, preventing them from reacting. Pyrophosphate, a molecule involved in energy transfer, does not dissolve well in water and requires a specific chemical to become active. This hypothesis also does not explain how chemiosmosis (a process used by cells to generate energy) or the differences between the domains Archaea and Bacteria arose.