First off, let me say that the book Seven Clues to the Origin of Life addresses your question very directly, and I think is written on a level any layman can understand. (Plus, it raised the question of Irreducible Complexity long before Behe- and it was written by an evolutionist!)
Here’s my abiogenesis FAQ (to be precise, a rough draft.) It talks in some detail about how the genetic code originated in the RNA world. If you are completely unfamiliar with DNA, proteins, and the genetic code, you’ll want to read the first few sections of my molecular genetics FAQ:
http://psyche11.home.mindspring.com/ben/WritingIndex.htm
ABIOGENESIS MINI-FAQ
Creationists are fond of attacking the idea of abiogenesis. No matter how many times you tell them that abiogenesis is a separate issue from evolution, they throw it in the face of evolutionists, usually with some sneers about “if you can explain that, there’s a Nobel Prize waiting for you.” Frequently you hear them say that the question of the origin of life (and in particular the origin of the genetic code) is so difficult that mainstream scientific journals contain a “thundering silence” on the issue. Of course, this is all nonsense. Aside from the fact that the creationists are using a classic God-of-the-Gaps argument, there’s plenty of fascinating work being done on a variety of issues relating to abiogenesis, and I’ve written this FAQ to present some of this work to laymen.
This FAQ is not meant to be comprehensive. There are a large number of issues relating to abiogenesis, and I’m simply not familiar with them all. For example, many scientists are working on the question of what chemical conditions are necessary to produce the basic building blocks of life. In this FAQ I’m not going to answer the chemical questions, and instead I will try to address some of the informational questions. Moreover, I am only going to address the informational questions in broad outline, without detailing some of the debates over the precise details (if you’re curious, you can read the papers in the bibliography.) The real questions that creationists are demanding answers to are these: how did those building blocks organize themselves into the first living organism? If the genetic code is irreducibly complex, how could it evolve? I don’t expect my presentation of the answers to these questions to convince any creationists, but that’s not the point. All I want to do is to show that the questions aren’t so impossible as creationists would have us believe, and I hope to teach you some interesting stuff along the way.
The RNA world
In modern organisms, DNA stores genetic information which directs the synthesis of protein machines, which carry out the work of the cell by catalyzing chemical reactions (for example, digesting food, copying DNA, and so forth.) As first blush, this system is irreducibly complex: the DNA can’t do anything without proteins, and the cell doesn’t know how to make proteins without the information in DNA. How could such a system evolve?
In 1986 Walter Gilbert suggested that the answer might lie with RNA. RNA is a DNA-like molecule that is heavily involved in the steps by which the information in DNA is used to make proteins. Gilbert suggested that at an early stage in the history of life, the machinery of life was entirely made of RNA, which served both to store information (like DNA) and to do work (like proteins.) Later, the RNA lifeforms evolved the ability to use DNA for information storage and proteins for catalysis. This idea was vindicated when Thomas Cech and Sidney Altman discovered that RNA molecules can, in fact, catalyze reactions just like proteins can. For this work, they won the 1989 Nobel Prize in Chemistry. (See? The creationists were right- there really are Nobel prizes available for scientists who work on abiogenesis!)
RNA is, as I said, used heavily in protein synthesis. First, the information in the DNA is copied to RNA. This “messenger RNA” is then sent to the RNA-rich ribosome, which assembles amino acids into the protein whose sequence is encoded in the messenger RNA. The ribosome grabs onto the amino acids by RNA handles called “transfer RNA.” (For a fuller explanation, see my molecular genetics FAQ xxx.) The parts of the ribosome that are directly involved in the chemical reactions that link the amino acids together are made of RNA, and the RNA forms a “catalytic triad” that mimics the triad found in digestive enzymes that catalyze similar reactions. The proteins in the ribosome have been compared to “mortar” that holds the RNA “bricks” together: the RNA does the real work, and the proteins just make the RNA more stable. In fact, the ribosome retains much of its ability to synthesise proteins even if you strip away all the ribosomal proteins, leaving behind pure RNA. The fact that the protein-synthesizing machinery is so heavily built of RNA lends support to the idea that RNA-based lifeforms gradually gained the ability to manipulate amino acids and link them together to make proteins. Moreover, other “molecular fossils” of the RNA world can be found in our biochemistry. For example, our cells store energy in the form of ATP, which is one of the components of RNA. Other biomolecules have a “handle” of ribose, another component of RNA, which they use to interact with proteins.
How did these first, RNA-based lifeforms come to be? Ultimately, all one needs for life to begin is a molecule of RNA that can replicate itself, or a small number of RNAs (say, three or four) that form a self-replicating system. Once that RNA starts replicating, it can mutate, which means that it can evolve into more complicated RNA-replicating systems which contain more and more different RNAs with specialized functions. There’s nothing particularly inconceivable about the idea that the initial, self-replicating RNA could come to be. For example, suppose that if you made an RNA at random, there’s a one in a billion chance that the RNA will be able to self-replicate. If the primordial ocean contains (just for the sake of argument) a trillion random RNA molecules, then a thousand of them will be able to self-replicate! Of course, in reality less than one in a billion RNA molecules will have that ability. On the other hand, the number of random RNA molecules available might be quite large: if you hold up a pin against the background of the night sky, the head of that pin blots out thousands of galaxies, each containing trillions of stars. (Mind you, my estimate of “thousands” is probably far too conservative.) How many planets are there on which the conditions are right for forming random RNA molecules? Scientists are currently trying to put more specific numbers on this argument. First, they are trying to find an RNA molecule that can replicate itself (RNA molecules have already been found which can replicate other RNA’s.) Once they do this, they will be able to determine which parts of the RNA are critical to its function, and thus they can calculate what percentage of random RNA molecules will have the same function. Secondly, scientists are trying to find out roughly how many planets have the right conditions for these processes to take place.
Amino acids enter the scene
How did this RNA world gain the ability to synthesise proteins? It is thought that the first interactions between RNA and amino acids came about when RNA enzymes (or “ribozymes”) evolved the ability to use amino acids as cofactors. Cofactors are molecules that proteins use to enhance their chemical abilities. For example, hemoglobin uses a heme cofactor to bind oxygen more efficiently than amino acids alone could. In an RNA world, the diverse chemical functionalities of amino acids would make amino acids attractive cofactors. Ribozymes thus evolved which had the ability to bind to amino acids and use them in chemical reactions. (Even today, some ribozymes still use amino acids as cofactors.)
However, the loops of RNA which are needed for a ribozyme to recognize and bind a particular amino acid are complicated, and it’s inefficient for each ribozyme to have to independently evolve such structures. On the other hand, it’s easy for one RNA loop to recognize another. The ribozyme lifeforms thus evolved a system by which some ribozymes would recognize particular amino acids and attach themselves as “handles” to individual amino acid molecules. Other ribozymes could then simply evolve a short stretch of sequence that would bind to the handle, and they would thereby be able to snag an amino acid molecule for use as a cofactor. These ribozyme handles ultimately evolved into the transfer RNAs which serve as handles for amino acids during protein synthesis.
(I should mention that there are a number of differing opinions on the details of this step. Some scientists follow the model I describe above, whereas others argue that instead of binding individual amino acids, the ribozyme handles bound chains composed of one type of amino acid repeated over and over. Some scientists also believe that the association of particular amino acids with particular handles- and therefore with particular codons in the modern genetic code- is entirely arbitrary, whereas others believe that the codon assignments are the result of a physical affinity between the particular amino acid and an RNA handle containing its anticodon. Currently, experiments are underway to determine which of these views is correct. Again, if you want more details, see the papers in the bibliography.)
Over time, ribozymes evolved which could use two or more amino acid cofactors for the same reaction. As time went on, the RNA parts of the ribozymes started to shrink as they waned in importance, while more and more cofactors were added. Meanwhile, ribozymes evolved which could link these amino acids into short chains, perhaps to enhance the stability of the enzyme. Ultimately, most ribozymes became nothing more than recognition sequences that could grab onto the appropriate RNA handles and bring together the right combination of amino acids for a job; these ribozymes became our modern messenger RNA. The handles became transfer RNA, while the ribozymes that linked the amino acids together became ribosomes.
