Shadows of Forgotten Ancestors

Those same proteins that surround the DNA (themselves manufactured, of course, on instructions from the DNA) are responsible for switching genes on and off, in part by uncovering and covering the DNA. At appointed times, the exposed ACGT information of the DNA makes copies of certain sequences and dispatches them as messages out of the nucleus into the rest of the cell; in response to the commands in these telegrams, new molecular machine tools, the enzymes, are manufactured. They in turn control all the metabolism of the cell and all its interactions with the outside world. As with the children’s game called “Telephone” in America and “Grandmother’s Whispers” in Britain—in which a message is whispered successively by each player into the ear of the next—the longer the sequence of relays, the more likely it is that the communication will be garbled.

It’s a little like a kingdom with the distant DNA, isolated and guarded in the nucleus, as the monarch. The chloroplasts and mitochondria play the role of proudly independent dukedoms whose continuing cooperation is essential to the well-being of the realm.* Everybody else, every other molecule or complex of molecules working for the cell, has as its sole obligation punctilious obedience to orders. Great care must be taken that no message is mislaid or misunderstood. Occasionally, decisions are delegated to other molecules by the DNA, but generally every machine in the cellular toolshop is on a short tether.

However, even to the rank-and-file molecular workers in the cell, the monarch often seems half-witted and his decrees garbled and meaningless. As we’ve mentioned, most DNA of humans and other eukaryotes is genetic nonsense which the START and STOP instructions—like prudent assistants to a mad president—duly ignore. Immense reams of nonsense are in effect thoughtfully preceded by the notice “DRIVEL AHEAD. PLEASE IGNORE,” and followed by the message “END OF DRIVEL.” Sometimes the DNA goes into a stuttering frenzy in which the same ravings are repeated over and over. In the kangaroo rat of the American Southwest, for example, the sequence AAG is repeated 2.4 billion times, one after the other; TTAGGG, 2.2 billion times; and ACACAGCGGG, 1.2 billion times. Fully half of all the genetic instructions in the kangaroo rat are these three stutters.⁴ Whether repetition plays another role—maybe some internecine struggle for control by different gene complexes inside the DNA—is unknown. But superposed on precision replication and repair, and the meticulous preservation of DNA sequences from ages past, there is an element in the life of the eukaryotic cell that seems a little like farce.⁵

Some 2 billion years ago, several different hereditary lines of bacteria seem to have begun stuttering—making full copies of parts of their hereditary instructions over and over again; this redundant information then gradually specialized, and, excruciatingly slowly, nonsense evolved into sense.⁶ Similar repetitions arose early in the eukaryotes. Over long periods of time, these redundant, repetitive sequences undergo their own mutations, and sooner or later there will be, by chance, rare short passages among them that begin to make sense, that are useful and adaptive. The process is much easier than the classic imaginary experiment of the monkeys poking at typewriter keys long enough that eventually the complete works of William Shakespeare emerge. Here, even the introduction of a very short new sequence—representing only a punctuation mark, say—may be able to increase the survival chances of the organism in a changing environment. And here, unlike the monkeys at their typewriters, the sieve of natural selection is working. Those sequences that are slightly more adaptive (to continue the metaphor, we might say those sequences that correspond even slightly to Shakespearean prose—“TO BE OR,” immersed in gibberish, would be a start) are preferentially replicated. Out of randomly changing nonsense, the accidental bits of sense are preserved and copied in large numbers. Eventually, a great deal of sense emerges. The secret is remembering what works. Just such a drawing forth of meaning from random sequences of nucleotides must have happened in the very earliest nucleic acids, around the time of the origin of life.

An illuminating computer experiment analogous to the evolution of a short DNA sequence was performed by the biologist Richard Dawkins. He starts with a random sequence of twenty-eight English-language letters (spaces are counted as letters):

WDLTMNLT DTJBKWIRZREZLMQCO P.