The building blocks from which DNA is made are known as deoxynucleoside-5′-triphosphates (dNTPs), which are made up of the familiar four DNA bases – Adenine, Guanine, Cytosine, and Thymine – plus a deoxyribose sugar and three phosphate groups. dNTPs are produced by the enzyme ribonucleotide reductase (RNR), and the enzymes that replicate DNA must obtain these molecules from the solution around them.
Imagine you have a tower of randomly coloured lego bricks, which you are trying to construct an identical copy of. The problem is that the materials you are using are not orderly arrayed on the floor but are instead being thrown at you randomly and must be caught before use. An efficient way to work in this hellish environment might be to catch a given brick, see if it is the colour you need next, drop it immediately if it isn’t and catch a new one. You’d build up statistical rules of thumb about the bricks you were getting pelted with: determining say that only about one in four of the bricks you caught with your right had would be blue, giving you time for your left hand to move the tower into position and to check to see if you got the previous brick in the right place. Maybe yellow bricks come up slightly more often than blue bricks, meaning that when you are waiting for one you don’t have time to check your last block and have to do that instead on a red turn. It’s not unachievable, though maybe a superhuman robot might be a better builder than you or I.
Now, imagine the ratio of colours suddenly changes without warning. Now there are twice as many blue bricks, and half as many yellow bricks. You’re going to make mistakes if you can’t adjust, and it turns out that DNA replicating enzymes have the same problem. A recent paper in PNAS reports the results of mutations of the RNR enzyme that produce different ratios of dNTP concentration. Turns out relatively moderate changes – twice as many of one base, half of another – end up increasing the mutation rate of E. coli by a thousand times, which is more than they expected. They conclude that the mechanisms that naturally check and repair mistakes in the new DNA strands get saturated, producing an “error catastrophe” which “can lead to poor growth or even death.”
Brian Thomas writes Delicate Balance in DNA Production. The most important paragraph for us to look at is this one:
This research clearly implies that cells not only need DNA copying, error-checking, and repair enzymes but also require molecular detectors that monitor nucleotide levels, mechanisms that communicate those levels to the dNTP manufacturers, and a process to govern production rates accordingly. In other words, this discovery adds one more critical component to the already long list of parts that must all simultaneously be in place for any cell to function.
He says that this result means that there must be a mechanism in the cell that measure the level of dNTPs, another that passes this information to the manufacturer (RNR), and another that makes RNR “govern production rates accordingly.” Of course, this couldn’t have happened spontaneously, so abiogenesis is impossible, evolution didn’t happen, and the Earth is young.
I must have remarked in the past on how Thomas’ mechanistic steps don’t seem to mesh with reality. I sometimes bring up for a variety of reasons Adrian Thompson’s experiment with evolutionary electronics. He had a board filled with simple electronic “cells,” and the puzzle was to connect them in such a way that the circuit could distinguish between two different input frequencies, and output certain voltages accordingly. Constructing this circuit in the natural human way, like Thomas would describe it, would involve building some kind of clock to compare the frequencies to, along with “mechanisms to monitor” input frequencies and “mechanisms that communicate” the result to the mechanism that produced the output voltage.
But Thompson didn’t build the circuit himself, he used natural selection to craft it for him. And it did – but without building a recognisable clock, and with less components than would be necessary via the our method above. Evolution doesn’t have to make a system via the same process that a human would design it, it just has to make something that works.
Like the electronics, our enzymes also don’t work in a the manner that Thomas describes. Instead, the RNRs that produce dNTPs also interact with them. dNTPs bind to the enzyme, causing its shape to change and preventing it making more. The higher the concentration of dNTPs the more likely this is to occur, and the less that will be produced. This maintains a comfortable equilibrium, in a much simpler and easier step than Thomas’ design.
And it’s evolvable too – the mutations of RNR have the result of modifying this equilibrium, meaning that it can change and evolve. It’s also possible that the DNA-checking enzymes could work with different ratios, perhaps less well, if they weren’t adapted so closely to the one that RNR presently provides.
But Thomas concludes:
So many precisely interacting parts do not just come together by chance. No wonder Job wrote, “Who among all these does not know that the hand of the LORD has done this?” As in other cases where a precise balance of supplied materials is required, DNA replication reflects purposeful creation through a wise Designer.
Look: a biblical quotemine. The “this” in Job 12:19 refers to the horrible things that God has done to Job – the whole book is an exercise in answering the question of “why do bad things happen to good people?” – and he wants to know why. But sure, pervert your own sacred scriptures to underline a flawed argument. I’m not going to stop you.