As they scratched this itch, they found that the Murchison meteorite ferried a most unusual cargo. As old as the earth itself, it had wandered through outer space for eons, yet it contained several of the amino acid building blocks of proteins, as well as purines and pyrimidines, which are important DNA building blocks. Later work used twenty-first-century spectroscopy to show that it harbored more than ten thousand different kinds of organic molecules, although many of them in exceedingly small quantities.15
The Murchison meteorite, it is important to know, is not a freak of nature. Similar meteorites have landed on earth, and countless other rocks ferry organic cargoes through the heavens.16 Fortunately, we no longer need to wait until another one of them drops. Because molecules in the universe absorb or emit radiation that reveals their structure, the hypersensitive ears of radio telescopes can distinguish hundreds of different organic molecules whose voices whisper to us in multitudes from clouds of interstellar gas. Actually, they shout, since three-quarters of the molecules in these interstellar clouds are organic, and include key constituents of life, such as the amino acid glycine.17 Incidentally, the single most abundant three-atom molecule in interstellar clouds is water, another blow to the notion that we and our planet are oh so very special.
Life’s simpler building blocks are so prevalent in the universe that molecules from space may have seeded life on earth itself. Meteorites and comets, especially those that bombarded the early earth, discharged ten times more water than currently fills all of the earth’s oceans, and a thousand times more gases than its present-day atmosphere.18 What is more, they also delivered the rich buffet of organic molecules we find in interstellar space, and in a staggering number of servings. At least ten trillion tons of organic carbon, and perhaps a hundred times as much, have entered our atmosphere from outer space.19 That is at least ten times more than all the carbon that circulates in living cells today. Especially important is the dust trailing behind comets that pass our orbit. Unlike large meteorites whose white-hot temperatures destroy some of their organic cargo during their explosive landing, cometary dust merely blankets the earth in an invisible but unceasing rain of life’s seeds.20 Perhaps we really are made of stardust.
We may never know whether most of life’s molecules were created in outer space or on earth. But regardless, these observations contain some simple and important lessons. The first is that life’s molecules emerge spontaneously in the right environment. The second is that this environment need not be, like Darwin’s warm pond, a nearby and very special place in the universe. It could be light-years away or as ubiquitous as interstellar gas.
The third is a lesson about innovation—I already mentioned it—that is still valid today: Innovation revolves around new molecules and the reactions that create them. To understand innovability, we need to understand the origins of these molecules.
The molecules of life are not yet life itself, any more than a pile of bricks and lumber is a mansion. At a minimum, life needs a metabolism, a network of chemical reactions that harvests energy and combines chemical elements into life’s molecular building blocks. Life also needs the ability to make more of itself—to replicate—and pass its accomplishments on to future generations as heritable traits. Without offspring that resemble their parents, Darwinian evolution would be unworkable, natural selection impossible.
This doesn’t mean that metabolism and replication must have appeared simultaneously. Even today they do not always occur together. Viruses replicate but have no metabolism of their own, hijacking instead the metabolic machinery of their host cells. But true life requires both metabolism and replication, and that requirement raises the earliest chicken-and-egg question ever: Which came first?
Perhaps seduced by the ethereal beauty of the DNA double helix, mainstream science held for years that replication came first. But to explain its origins is a tall order, because modern replication is extremely complex. What’s more, the nucleotide letters of DNA do not replicate themselves. Rather, they only carry information. They are transcribed into RNA, which is translated into proteins (figure 1), and these proteins are tasked with just about everything else, including transcription and replication. Because no one protein masters all the necessary skills, dozens of proteins share these tasks, and each protein has a precisely specified amino acid sequence. This sophisticated division of labor leads to another chicken-and-egg problem, this one about whether proteins or nucleic acids—the collective term for DNA and RNA—appeared first. To have both emerge simultaneously would be asking too much—we would be back to odds-defying spontaneous creation. But if the first life consisted of a single replicator, this molecular Adam (or Eve) would have to be spectacularly versatile, able to both carry information and copy itself.
In their 1953 discovery of the double helix, Watson and Crick had already recognized that the key to replication was the pairing of complementary DNA bases—G with C and A with T—which glues the two strands of the double helix together. In their words, it “immediately suggests a possible copying mechanism for the genetic material.”21 That very mechanism excludes proteins as the first replicators, because their amino acid parts cannot transmit information in this way. They lack the simple complementarity that allows two DNA strands to build the twisted ladder of the double helix.
So proteins are poor replicators. But nucleic acids seemed to be just as poor at everything else. Could they do what proteins excel at? Could they catalyze their own replication? Could they catalyze anything at all? The very purpose and structure of DNA made that seem unlikely. DNA’s primary task is to do as little as possible except to store information, inertly, faithfully, generation after generation after generation.22 For more than half a century after the discovery of enzymes, most scientists thought that only proteins, not nucleic acids, could catalyze chemical reactions.
And so the stuff of the first replicator remained mysterious. Until 1982, that is, when the chemists Thomas Cech and Sidney Altman transformed RNA from an ugly duckling into a white swan.23 RNA had been the stepchild of molecular biology, largely a messenger ferrying information from DNA to the ribosome, the hugely complex molecular machine that synthesizes proteins.24 But these two chemists jolted science with the discovery that RNA can catalyze chemical reactions all by itself.
The knowledge that RNA could perform the job of proteins catalyzed—pardon the pun—many other discoveries. Before long, biologists realized that RNA had an ancient history, even older than that of proteins and DNA, and that RNA had ruled over a sunken world of early life.25 Unlike the fabled Atlantis, however, this world has left many traces. One of them is that RNA remains a key molecule in today’s command centers of life. For example, in the ribosome machine, which contains dozens of proteins and also a few RNA molecules, the RNA—not the proteins—catalyzes the concatenation of amino acids into proteins, including those that make the ribosome’s own proteins.26
RNA may once have both carried life’s information and helped catalyze its own replication, but we are none the wiser how it got there. To find that first innovation, the origin of life itself, it would help if we could construct a simple molecule that replicates itself. This molecule would be an RNA replicase, an enzyme that catalyzes the replication of RNA.27
Some of today’s finest chemists are on a quest to find this single replicase. Their best efforts so far created a 189-letter-long RNA string with some talent as a copyist—it can’t actually copy itself, but only a shorter molecule template of about fourteen letters.28 This tells us that RNA-based replication might just work, if several obstacles can be surmounted. One of them comes from the very feature that makes nucleic acids replicable: base pairing. Complementary bases stick together, which means that a parent molecule and the complementary copy strung together by a replicase would anneal into a double-stranded RNA molecule like the familiar double-stranded DNA. To make another copy from them, the two strands would have to be separated, so their information can be read. But as soon as you—or a replic
ase—pulled them apart, their sticky bases would anneal again like two ribbons of Scotch tape. The very same feature that allows replication is also its worst enemy.
Another problem is that the first replicase would have needed to be unfathomably accurate, because a sloppy replicase would trigger an error catastrophe, a process discovered and baptized by the Nobel Prize–winning chemist Manfred Eigen.29
To understand error catastrophes, it is helpful to think of how medieval monks copied sacred texts, transcribing letter by letter by tedious letter. If one monk misread a letter, the error might be inherited in the copy. A second scribe might propagate the error, introduce errors of his own, and so on, until monk by monk, over generations and centuries, the text might slowly erode to a meaningless jumble of letters. An RNA replicase, whose one and only ability is encoded in a molecular text, faces a similar problem, but with an added twist: In an RNA world, the monk and the text he copies are one and the same. The replicase is a book that transcribes itself, and its errors erode not only the text itself but also its own ability to copy. Subsequent generations of monks become ever more error-prone.
Only a replicase that creates mostly error-free copies of itself can preserve the information in its letter sequence, which encodes the very ability to replicate itself. But if it is too sloppy, most of its copies will be inferior replicases—slower, perhaps, not as faithful—and will degrade over time into useless molecules from which the original information is erased. In 1971, four years after winning the Nobel, Manfred Eigen calculated the accuracy required to evade this error catastrophe. The longer a replicase is, he found, the more accurate it needs to be. As a rule of thumb, a replicase with fifty nucleotides would have to misread fewer than one in fifty nucleotides, a replicase with one hundred nucleotides would need to misread fewer than one in a hundred nucleotides, and so on.30 The current best candidate of 189 letters misreads several times as many.31 Even if it could replicate itself, it would gallop straight over the cliff of the error catastrophe.
Fortunately, life does much better than that today. The DNA-copying machinery of protein enzymes misreads fewer than one in a million letters.32 But its accuracy comes at a price: complexity. The machinery includes highly specialized proteins that proofread and correct the errors of other proteins, as if every text were copied by groups of monks looking over one another’s shoulders. These proteins are encoded by long genes, much longer than any primordial RNA replicase could have been. And to preserve information over time, the replication they supervise needs to be highly accurate. You may already see the next chicken-and-egg problem lurking here. It is also known as Eigen’s paradox: Faithful replication needs long and complex molecules, but long molecules require faithful replication. To this day, nature has not shown us an exit from this labyrinth, but as we shall see in chapter 6, a principle of innovability found in today’s life provides a clue.
The unwelcome stickiness of RNA and the catastrophic Eigen’s paradox are already daunting obstacles to the idea that replicators were life’s first innovations. But they are mere foothills to the Himalayas of a third problem: finding a sufficiently rich supply of raw materials—energy-rich molecules that also contain all the needed chemical elements, including carbon, nitrogen, and hydrogen. Examples include the energy-rich precursor nucleotides of DNA’s letters, of which modern replication proteins consume nearly a thousand every second when they copy DNA.33 And even if an early replicase were much slower and more inefficient—perhaps copying one letter per second, replicating itself in some three minutes—the demand for raw materials wouldn’t disappear.34 Since each copy would itself be a replicase, both the number of copies and the ability to make even more copies would increase in lockstep, which means that even a glacially slow pace of replication by modern standards would still result in an exponential population explosion and a huge demand on nucleotide raw materials. After the first six hours, this population would already have devoured 1 ton, after a day 2.5 tons, and after a week no fewer than 800,000 tons of nucleotide raw materials.
Once life has arrived, it quickly multiplies into an army of molecules that is voraciously hungry for a continuous supply of energy-rich materials. Like any military expedition, it would quickly collapse without such a supply chain. What is more, absent a steady food supply, Darwinian evolution and natural selection cannot work. Their power unfolds over many generations and thus needs replication—a lot of it. In addition, it does not help that replicases, like soldiers, are mortal.35 Over time, they decay through random collisions with other molecules. Starving, they would decay faster than they could make copies. Life’s campaign to conquer the planet would fizzle out like a wet match, moments after it had ignited.
The supply chains created by Miller’s experiments and by interstellar chemistry are not strong enough to sustain this army. Although they create molecules like amino acids that life craves as raw materials, they do not create enough of them to feed life continuously. Miller’s experiments take days to produce a few milligrams of organic molecules from about a kilogram of carbon.36 And while meteorites may help import megatons of organic carbon given enough time, the first replicators would starve without having food in enormous quantities, at the same time, and in the same place. Depending on meteorites to sustain life is like relying on a manure-carrying truck to crash into your backyard garden every few days.
All this leaves a gnawing suspicion that the replicator-first idea puts the cart before the horse. Seduced by the beauty of the double helix, its advocates have dreamed up a gleaming and sophisticated automobile factory—before reliable parts suppliers existed. The factory’s high-throughput assembly line is useless until wheels, axles, transmissions, and engines can be built in quantities. And if the supply’s trickle is so slow that only one car can be built every few years, then decay and eventual bankruptcy become inevitable. The obvious alternative is that before a single self-replicating molecule could emerge, the supply network had to be in place, a network of chemical reactions that could produce life’s raw materials.
In other words, life started not with a replicator, but with a metabolism.37
When the right molecules come close enough, the chemical reactions needed to produce energy and life’s building blocks proceed—eventually. But eventually can be a long time, a very long time. Some of life’s chemical reactions would take thousands of years to proceed without help. For this reason, metabolism needs catalysts, molecules whose main job is to speed up chemical reactions. Catalysts have a remarkable feature: Driven by heat—the incessant bouncing and vibrating of atoms and molecules—they can arrange other molecules such that their atomic parts come in contact and react, but they themselves stay above the fray and are not eaten up by the reaction. Catalysts are accelerants for the fire of metabolism. Their main job is to lower the activation energy for a particular chemical reaction, and accelerate it by several orders of magnitude. The catalysts of modern metabolism are protein enzymes, extremely efficient and sophisticated chemical agents, each one specific to one reaction, some of them accelerating their reactions by more than a trillionfold.38 Our bodies harbor thousands of these catalysts. A good thing, too. Were one of them to falter, we would die, snuffed out like an early replicator without supplies.
But 3.8 billion years ago, protein catalysts had yet to be invented. Darwin’s “warm little ponds” are poor sources of catalysts, which is one reason why many scientists became disenchanted with them.39 Another is that two molecules have to meet before they can react. Because molecules are jostled erratically through water by heat’s atomic vibrations, molecular encounters are chance events, and the chances are directly proportional to the number of molecules in a given volume of water: too few molecules, too few reactions. In other words, a metabolism can get going only if its molecules are concentrated. Dilute them in a bowl too large, and primordial life would end before it began. This is why chemists perform experiments in small test tubes and not in swimming pools. Washed out into the primordial ocean, newl
y created molecules would never be seen again.
Some think that tidal pools, a variant of Darwin’s warm ponds, might solve this last problem. At low tide, water in such a pool evaporates through heat, thus concentrating chemicals. At high tide, new water flooding in can stir up the broth. But earth’s violent youth casts doubt on this scenario as benign as a beach vacation. The moon orbited only a third as far away as it does today and tugged ferociously at the oceans, creating gigantic tides at least thirty times higher than today’s. What is more, the moon positively whirled around the earth (which itself rotated twice as fast), circling it at least every five days, and would have created these extreme tides every few hours, leaving little time to concentrate life’s ingredients.40
Evolutionary biology had been aching for better and smaller test tubes for decades, when an answer to its prayers arrived out of the blue—the deep blue. In 1977, the research submarine Alvin discovered an exotic menagerie on the Pacific seafloor near the Galápagos Islands, more than two thousand meters below the surface.41 Red-plumed mouthless tube worms more than two meters long, snails with feet and shells armored with iron minerals, and eyeless shrimp thrived down there, cushioned by lawns of never-before-seen microbes that do double duty as food. But even more bizarre than this community is how it survives: Its raw materials come straight from Mother Earth herself, through searingly hot fissures in the earth’s crust that overflow with nutrients, chemical energy, and the very catalysts that warm little ponds lack.
Arrival of the Fittest: Solving Evolution's Greatest Puzzle Page 5