The paper in Nature Ecology & Evolution is here: rdcu.be/wnNz or http://go.nature.com/2wtN76z

All life is continuously building up and breaking down its constituent chemical building blocks, an energy consuming process called metabolism. How did metabolism get its start? The answer might lie with a hurricane. A hurricane’s building blocks are just air and water, and yet when these building blocks are exposed to heat in just the right way, they self-organize into a very active (and dangerous) system with a characteristic dynamic pattern. 

The dynamic pattern of a hurricane is just as fundamental to its nature as the building blocks that make it up. So it is with metabolism. Metabolism is a self-organized system expressed through chemistry that takes on very characteristic and dynamic chemical patterns as it builds up and breaks down its molecules, dissipating energy in its environment. Just as we think of a hurricane’s dynamic patterns and its building blocks (air and water) as being equally fundamental to its nature, so too should metabolism’s dynamic chemical patterns and chemical building blocks be viewed as equally fundamental.

In our opinion, however, much experimental chemical research on the origins of life is overly focused on the synthesis of chemical building blocks through de novo synthetic pathways without sufficiently appreciating life’s larger biochemical self-organization. An explanation for the origin of life must account for life's dynamic patterns, and the most parsimonious proposal is that metabolism recapitulates prebiotic chemistry. Life ultimately builds all of its molecules from carbon dioxide, yet it is surprisingly lacking in innovation in this respect. 

Despite nearly 4 billion years of evolution, autotrophic organisms use only six pathways to build their molecules from CO₂. Two of these pathways – the reductive acetyl CoA pathway (also known as the Wood-Ljungdahl pathway) and the rTCA cycle (also known as the reverse Krebs cycle) - are thought to be ancestral, with just five molecules within them serving as the universal chemical precursors for all of biochemistry. These two pathways were likely once fused together, in whole or in part. How and why did these pathways get their start? Are they sophisticated products of evolution or are they deeply rooted in chemical determinism – a product of prebiotic self-organization – a chemical hurricane?

To answer this question experimentally, a small team in my lab has embarked on a systematic search to find simple, non-enzymatic chemical or mineral catalysts and reagents that can promote the reactions of core anabolism, particularly the acetyl CoA pathway and the rTCA cycle. After finding as many ways as possible to promote each reaction, we can then start to triangulate mutually compatible conditions where many reactions can occur in sequence. The more of "core anabolism" that we can achieve under a single set of purely chemical conditions, the more likely it is to have constituted early prebiotic chemistry rather than a later product of chemical and biological evolution. 

To start, could we find simple catalysts or reagents that could promote the reduction, dehydration and hydration reactions of the reverse Krebs cycle? The question was nearly wide open - only a couple of non-enzymatic reduction reactions of the cycle were known. Armed with in-house know-how on catalyst screening and after a considerable amount of time finding a suitable high throughput analytical method, we started systematically running experiments to find as many ways as possible to do each reaction. The sheer number of experiments was massive - even by limiting the catalysts screened to simple metal ions and the reducing agents to a small selection, we still had many variables to cover: starting substrate, temperature, pH, catalyst and reducing agent. 

After over 4,000 experiments, we found that iron (a reducing agent) in the presence of Zn²⁺ and Cr³⁺ ions was able to promote all six of the reduction, dehydration and hydration reactions of the rTCA cycle, as well as amino acid synthesis, under a single set of conditions and in sequence. Reducing conditions and metal ions are about as simple as it gets. Our results tell us that at least half of the chemistry of the rTCA cycle is so simple that it could have constituted prebiotic chemistry.

It is not our intent to claim that Zn²⁺ and Cr³⁺ were necessarily the catalysts for this pathway at its origin. We only screened metal ions, but there are hundreds of minerals and clays, some of which might be able to do this chemistry just as well or better. 

Our next big task is to try to find reagents or catalysts for the CO₂-fixation steps of the AcCoA pathway and the rTCA cycle, which are arguably the most challenging chemical reactions in the ancestral network. If we can do that - and we optimistically believe it can be done -  it may be time to start thinking of some biological anabolic pathways as self-organized prebiotic chemical paths of least resistance rather than as sophisticated inventions of evolution. Giants in the origins of life field have raised sensible objections to this hypothesis, but similar ideas for catabolic pathways have recently found experimental support for the oxidative TCA cycle and glycolysis. How much of core metabolism might be thought of in this way? How big is the "chemical hurricane"? Stay tuned!