How Did Life Begin on Earth? What Science Can and Cannot Say

The Earth formed about 4.54 billion years ago and was uninhabitable for its first stretch: molten, bombarded, and struck by a Mars-sized body that made the Moon. Liquid water oceans existed by around 4.4 billion years ago, judging by the chemistry of the oldest surviving crystals. The oldest widely accepted fossils are microbial mats called stromatolites from Australia, about 3.5 billion years old, and there are chemical traces in rocks from Greenland and Canada that some researchers argue push life back to 3.8 billion years or more.

That leaves a window of a few hundred million years between the first oceans and the first clear life, and possibly much less. Whatever happened, it did not take long on a geological scale, which suggests that it was not an outrageous fluke.

In 1953 Stanley Miller sealed water, methane, ammonia and hydrogen in a flask, sparked it with electricity to imitate lightning, and found amino acids, the components of proteins, in the water a week later. His atmosphere was probably wrong, the early Earth's was richer in carbon dioxide and nitrogen, but repeats with more realistic mixtures still produce amino acids, and so do meteorites: the Murchison meteorite that fell on Australia in 1969 contained dozens of them, along with sugars and the bases used in DNA and RNA. The components of life form readily wherever there is carbon, energy and water.

The difficulty is not making the parts. It is getting them to assemble into something that copies itself, and that is where the field divides.

Living cells face a chicken-and-egg problem: DNA stores the instructions but needs proteins to copy it, and proteins are made from DNA's instructions. The most widely held answer is that neither came first. RNA, DNA's single-stranded cousin, can both carry information and, as Thomas Cech and Sidney Altman showed in the 1980s, act as a catalyst that speeds up chemical reactions. A world of self-copying RNA molecules could have preceded both DNA and protein.

Support has grown steadily. Laboratory evolution has produced RNA molecules that copy other RNA, though not yet themselves completely. John Sutherland's group, then at the University of Manchester, showed in 2009 a plausible route from simple chemicals, hydrogen cyanide and its relatives, to RNA's building blocks under conditions resembling a drying pond struck by ultraviolet light. And the ribosome, the machine in every living cell that builds proteins, turns out to be made mostly of RNA doing the actual work, a fossil of the RNA world still running inside us.

Two settings compete. Hydrothermal vents on the ocean floor, particularly the alkaline vents discovered in 2000 at a site called the Lost City, produce warm, mineral-rich fluid flowing through porous rock towers, with natural chemical gradients across thin mineral walls that resemble the gradients living cells use to make energy. Nick Lane and others argue that such a vent supplies the energy, the confinement and the catalysts that a drying pond does not.

The pond camp replies that the key chemistry, especially making RNA and joining its units into chains, needs cycles of wetting and drying, concentrated solutions and ultraviolet light, none of which exist at the bottom of the sea. Volcanic hot springs on land, like those in Yellowstone or Iceland today, give all three. Charles Darwin guessed at a warm little pond in a letter of 1871. It is still a live option, and the honest position is that both settings have chemistry in their favour and neither has been made to work end to end.

Whatever the route, it converged. All life on Earth shares the same genetic code, the same set of twenty amino acids in proteins, the same handedness of its molecules and the same core metabolism, which means everything alive descends from one population of early cells, called the last universal common ancestor. Comparing the genes shared across all three domains of life suggests that ancestor already lived in a hot, mineral-rich, oxygen-free environment and used hydrogen and carbon dioxide for energy, which fits the vent picture, though the argument is far from settled. What is established, what is supported and what is open:

• •Established: life existed by 3.5 billion years ago, within a few hundred million years of the oceans
• •Established: amino acids, sugars and nucleobases form easily on planets and in space
• •Well supported: an RNA-based stage came before DNA and proteins
• •Contested: whether it happened at deep-sea vents or in ponds on land
• •Open: how the first self-copying system arose and became a cell

Every living cell is wrapped in a membrane, and the first cells needed one too, to keep their chemistry together and to separate one copying system from another. Modern membranes are made of complicated molecules built by enzymes, but simpler relatives called fatty acids can do a similar job. In water, above a certain concentration, fatty acids gather on their own into hollow spheres called vesicles, much as soap does, with no instructions needed. Fatty acids have been found in meteorites and can form in the kinds of chemistry thought to occur at vents and hot springs.

Experiments in Jack Szostak's laboratory at Harvard showed how much such simple bags can do. Fatty acid vesicles can grow by absorbing more fatty acid from their surroundings, divide into smaller vesicles when gently stirred, and let small building blocks such as nucleotides pass through their walls while holding on to larger chains of RNA. Certain clays speed up both the formation of vesicles and the joining of RNA units, bringing the pieces together. None of this is a living cell, but it shows that a container able to grow and divide needs no evolved machinery at all.

For most of history people assumed that simple life arose from non-living matter all the time: maggots from rotting meat, mice from stored grain. Francesco Redi undermined the maggot case in 1668 by covering jars of meat with gauze, which kept flies off and kept the meat free of maggots. Louis Pasteur settled the microbial version around 1860 with flasks whose long necks bent downwards in an S shape. Boiled broth inside stayed clear because dust and microbes from the air settled in the bend, yet it clouded soon after the neck was snapped off.

Pasteur's result, that life today comes only from life, is sometimes presented as a disproof of any natural origin of life. It is not, because the two questions concern different worlds. Pasteur's broth sat in an oxygen-rich atmosphere crowded with organisms, while the origin of life took place in the young Earth's very different chemistry, over spans of time no laboratory can match. Darwin saw the point in the same 1871 letter where he imagined his warm little pond: today, he wrote, any such newly formed matter would be instantly devoured or absorbed, which would not have happened before living creatures existed. A new start would now simply be eaten.

Many of life's molecules come in two mirror-image forms, like a pair of gloves, built from the same atoms but impossible to superimpose. Living things are strikingly one-sided about them. The amino acids in proteins are almost all the left-handed form, and the sugars in DNA and RNA are right-handed. The choice matters, because a chain built from a random mixture of both forms folds badly and copies poorly. Yet ordinary chemistry, including Miller's spark experiment, produces the two forms in equal amounts. Somewhere on the way to life, one hand had to win, and explaining how is one of the field's lasting puzzles.

Several clues point towards answers without settling the question. Careful analysis of amino acids in the Murchison meteorite found a small excess of left-handed forms in some of them, suggesting that processes in space, perhaps involving polarised light in star-forming regions, could tip the balance slightly. Chemists have also found reactions that amplify a tiny imbalance: in one, discovered by the Japanese chemist Kenso Soai in the 1990s, the product helps to make more of itself, so a slight excess of one form snowballs into almost pure product. Whether nature used such amplification, and whether the starting bias was chemical or simply chance, is still argued over.

Life on Earth appeared quickly once there was water, its building blocks form easily, and an RNA world very probably preceded the DNA-and-protein system every cell now uses. Whether it started at a deep-sea vent or in a pond on land, and how the first molecules began to copy themselves, remain unsolved, which is the difference between a mystery and a research programme.