Outgrowing God
Page 15
Around the unfinished puzzle, floating in the liquid, are thousands of jigsaw pieces. These might be sodium and chloride ions floating in water. Every time one of the floaters bumps into a crystal, it finds the correctly shaped ‘hole’ and slots itself in. So that’s another way of picturing how a crystal grows at the margins. Now we’re going to use the jigsaw metaphor to talk about what goes on in living creatures. In particular we’re going to look at enzymes. We’ll see what enzymes are in a moment.
Remember the picture in Chapter 7 of the chemical reactions going on in a cell: that enormously complicated spaghetti-junction of arrows and blobs? You might wonder how all those different chemical reactions can go on in the same tiny space, inside the same cell, without interfering with each other and getting muddled up. Suppose you went into a chemistry lab, grabbed all the bottles off the shelves and tipped them, all at once, into a great vat. You’d get a horrible mess – and maybe set off a lot of horrible reactions, even explosions. Yet somehow, in the cells of a living creature, lots of chemicals manage to stay separate without interfering with each other. Why don’t they all react with one another? It’s as though each one was in a separate bottle. But they aren’t. How does it work?
Part of the answer is that the interior of the cell is not a single vat. It is filled with a complicated system of membranes, and these can act rather like the glass walls of test tubes. But that’s not the whole story. There’s something more interesting going on. And this is where enzymes come in. Enzymes are catalysts. A catalyst is a substance which speeds up a chemical reaction without actually being changed itself. It’s a kind of fast-working miniature lab assistant. Catalysts can sometimes make a chemical reaction go millions of times faster, and enzymes are especially good at this. All those chemicals, muddled up together, don’t react with one another unless there’s a catalyst present: and it has to be a particular catalyst for each reaction. Particular reactions are turned on only when they are needed, by adding the right enzyme. You might think of an enzyme as a switch, which can be on or off, almost like an electric switch. Only when a particular enzyme is present in a cell is its one particular chemical reaction switched on. Even better, enzymes can ‘switch on’ other enzymes. You can see how elegant control systems could be built up with switches switching on (or off) other switches.
We know, at least in outline, how enzymes work. This is where the jigsaw puzzle idea comes in. Think of all the hundreds of molecules whizzing about in the cell as jigsaw puzzle pieces. Molecule X needs to find molecule Y in order to join up with it and make XY. The X/Y marriage is just one of the hundreds of vitally important chemical reactions in Chapter 7’s ‘spaghetti’ diagram. There’s a chance that an X will happen to bump into a Y. There’s a smaller chance that they’ll happen to bump at just the right angle to slot in and combine together. That happens so seldom that the rate at which XY is formed is extremely slow – so slow that if left to chance it would almost never happen. (This reminds me of my very first school report, when I was seven: ‘Dawkins has only three speeds, slow, very slow and stop.’) But there’s an enzyme whose particular job is to speed up the rate at which Xs combine with Ys. And in the case of many enzymes, ‘speed up’ is an understatement. Again, the process works using the jigsaw principle.
An enzyme molecule is a great big complicated lump, with bulges and crevices all over its surface. When I say ‘great big’, it’s only big by molecular standards. By the standards we’re used to in our everyday lives it is tiny, too small to be seen by a light microscope. Let’s take the particular case of the enzyme that speeds up our ‘XY’ chemical reaction. Among the crevices in its surface is an X-shaped hole which just happens to be right next to a Y-shaped hole. This is why it is a good ‘lab assistant’, specifically good at speeding up the X/Y combination. An X molecule falls, jigsaw-style, into the X-shaped hole. A Y-shaped molecule falls, jigsaw-style, into the Y-shaped hole. And, since the two holes are next door to each other in exactly the right way, the X and the Y find themselves snuggled together at exactly the right angle to combine with each other. The newly formed XY combination then pops out and floats away into the cell, leaving the two precisely shaped holes free to do the same thing with another X and another Y. So the enzyme molecule can be seen not just as a lab assistant but as a kind of factory machine, churning out XY molecules, using a steady stream of Xs and Ys as raw material. And, in that cell and in other cells elsewhere in the body, there are other enzymes, each one perfectly shaped – that is, with the right ‘crevices’ or ‘dents’ in the surface – to speed up other chemical reactions. I must stress that my language of ‘crevice’ and ‘shape’ constitutes a great oversimplification, but I’ll stick with it because it’s helpful for the purposes of this chapter. ‘Shape’ can mean not just physical shape but chemical affinity.
There are hundreds of enzymes, each one shaped differently, each one shaped to speed up a different chemical reaction. But in most cells only one or a few of the available enzymes are present. Enzymes are the main (though not the only) answer to the riddle of why the chemical reactions don’t all happen at once, and don’t all interfere with each other.
Enzyme molecules, then, sound like magic. Just as a cheetah’s legs are beautifully shaped to run fast, enzymes are beautifully shaped to speed up particular chemical reactions. Just one special chemical reaction per enzyme. How do they get their beautiful shape? Are they carved into shape by a divine molecular sculptor? No. They come into being by a more complicated version of what growing crystals do. It’s self-assembly again.
Every protein molecule is a chain of smaller molecules called amino acids. There are lots of different kinds of amino acids, but only 20 of them are found in living things. They all have names, and I could write out the 20 names, but let’s not bother with the details. There are 20, and that’s all we need to know here. Each protein molecule is like a necklace with amino acids for beads (a necklace with the clasp unfastened, not a closed loop). Proteins differ from each other in the exact sequence of beads from which each is made, all taken from the repertoire of 20 kinds of amino acids: 20 kinds of beads.
You remember that salt crystals grow when jigsaw pieces floating in water recognize their ‘opposite numbers’ at the edge of the crystal and slot themselves in. Well, think of the beads in the protein necklace as a selection from 20 kinds of jigsaw pieces. And some of them slot into other jigsaw pieces somewhere along the same chain. The result of this self-jigsawing, happening in various places all along the chain, is that the chain folds into its special shape. Like a piece of string tying itself into a very particular knot.
Now, I described an enzyme molecule as a complicated lump with bulges and crevices. That doesn’t sound like a chain, does it? But it is. The thing is, any chain of amino acids has a tendency to fold itself into a particular three-dimensional shape. As I said, it’s a bit like tying itself into a knot. The ‘lump with knobs and crevices’ is the knotty shape into which the chain assembles itself. Links in the chain are attracted to other particular links in the chain and stick to them, jigsaw-wise. And these hook-ups help to ensure that every instance of a particular chain folds itself into the same shape with the same bulges and crevices.
Actually that’s not always quite true – and the exceptions are interesting. Some chains can tie themselves in one of two alternative knots. That can be extremely important, but I’m going to leave it aside here because this chapter is already complicated enough. For our purposes, we can think of each protein molecule as a chain of jigsaw pieces (amino acids) which folds itself into a very particular shape. The shape really matters, and it is determined by the particular sequence of amino acids and their tendency to slot, jigsaw-wise, into other amino acids in the same chain.
Here I can’t resist a little story, which may seem unconnected but throws an interesting light on this idea of jigsaw pieces slotting in. It’s about our sense of smell. Imagine the smell of a rose. Or of honey. Or onions. Apples. St
rawberries. Fish. A cigar. A stagnant marsh. Every smell is different, unmistakable: beautiful or horrible, smoky or fruity, fragrant or foul. How is it that molecules, borne on the air into our nose, give rise to this smell or that smell? The answer is jigsaws again. The lining of your nose has thousands of differently shaped molecular crevices, each one just waiting for a molecule of exactly the right shape to slot in. A molecule of, say, acetone (nail-varnish remover) fits snugly into an acetone-shaped crevice, just like in a jigsaw puzzle. The acetone-shaped crevice sends a message to the brain saying, ‘My kind of molecule has just slotted in.’ The brain ‘knows’ that this particular crevice is an acetone-shaped crevice, so the brain ‘thinks’: Aha, nail-varnish remover. The smell of a rose, or of a fine vintage wine, is made up of a complex mixture of jigsaw molecules, not just one as with acetone. But the point is the same: it’s the molecular jigsaw principle at work.
Back to the main story. We’ve seen that the sequence of amino acids in the ‘necklace’ is responsible – through ‘self-assembly jigsawing’ – for the lumpy crevicy shape of the protein ‘knot’. And we’ve seen that the crevices in turn are responsible for the protein’s particular role as an enzyme, speeding up – usually so much that it amounts to switching on – a particular chemical reaction. There are lots of chemical reactions that could be going on in a cell at any one time. The ingredients are all there, ready to go. All each one requires is the right enzyme. And there are lots of enzymes that could be there, but only one is. Or only a few. So which enzymes are present is utterly crucial. They determine what the cell does. What the cell is, indeed.
So now you must be asking yourself, what determines the sequence of amino acids in the necklace of any particular enzyme, and therefore the lumpy shape into which the chain folds itself? That’s obviously a hugely important question because so much else depends on it.
And the answer is: the genetic molecule, DNA. An answer whose importance is impossible to exaggerate. Which is why I’ve given it a paragraph to itself.
Like a protein molecule, DNA is a chain, a necklace of jigsaw pieces. But here the beads are not amino acids, they are chemical units called nucleotide bases. And there aren’t 20 different kinds, only four. Their names are shortened to A, T, C and G. T jigsaws only with A (and A only with T). C jigsaws only with G (and G only with C). A DNA molecule is an immensely long chain, much longer than a protein molecule. Unlike a protein chain, the DNA chain doesn’t tie itself into a ‘knot’. Instead, it stays as a long chain – actually two chains jigsawed together in an elegant spiral staircase. Each ‘step’ of the staircase is a jigsawed pair of bases, and there are only four kinds of step:
A–T
T–A
C–G
G–C
The sequence of bases carries information, in the same way (almost exactly the same way) as a computer disc. And the information is used in two completely different ways: the genetic way, and the embryological way.
The genetic way is just copying. By a rather complicated version of jigsawing, the entire staircase is copied. This happens when cells divide. The embryological way is amazing. The code letters are read in triplets – three at a time. There are 64 possible triplets of four (4 × 4 × 4 = 64), and each of those 64 triplets is ‘interpreted’ either as a punctuation mark or as one of the 20 amino acids that go into making protein chains. When I say ‘read’, there is, of course, nobody to do the reading. Once again, it’s all done automatically using the jigsaw principle. I’d love to go into the details, but that isn’t what this book is about. For our purposes, what matters is that the sequence of the four types of bases in a stretch of DNA, when read in threes, determines the sequence of the 20 types of amino acids in a protein chain. The sequence of amino acids in a protein chain then determines how that protein chain coils up into a ‘knot’. The shape of the ‘knot’ (its ‘crevices’ and other things) determines how it works as an enzyme, and therefore which particular chemical reaction it switches on in a cell. And the chemical reactions in a cell determine what sort of cell it is and how it behaves. Finally – and this is perhaps most wonderful of all – the behaviour of cells working together in an embryo determines how the embryo develops and turns into a baby. So it was ultimately our DNA that determined how each one of us developed from a single cell into a baby, and then grew into what we are now. This is the subject of the next chapter.
*I owe my understanding of snowflakes to Brian Cox’s beautiful book Forces of Nature (London, Collins, 2018).
A great scientist – and larger-than-life character – of the twentieth century, J. B. S. Haldane, was once giving a public lecture. Afterwards, a lady stood up and said something like this:
‘Professor Haldane, even given the billions of years that you say were available for evolution, I simply cannot believe it is possible to go from a single cell to a complicated human body, with its trillions of cells organized into bones and muscles and nerves, a heart that pumps without ceasing for decades, miles and miles of blood vessels and kidney tubules, and a brain capable of thinking and talking and feeling.’
Haldane gave a wonderful reply: ‘But madam, you did it yourself. And it only took you nine months.’
The lady could have retorted, ‘Ah, but my nine months as a developing baby were orchestrated by the DNA my parents gave me. I didn’t have to start from scratch.’ That is, of course, true. And her parents got the DNA from their parents, who in turn got it from their parents and so on back through the generations. What was happening during all the billions of years of evolution was that the DNA instructions for how to make babies were being gradually built up. Built up – honed and improved – by natural selection. Those genes that were good at making babies got passed on, at the expense of the genes that weren’t. And the kind of babies that were made was changing, ever so gradually and slowly, over the millions of generations.
There’s a rather charming hymn, ‘All things bright and beautiful’. Perhaps you know it. It praises God for the detailed beauty of his creations, especially living creatures:
He made their glowing colours
He made their tiny wings.
But even if you believe God had something to do with creating animals, you’ll realize that he didn’t directly make glowing colours. Or wings, tiny or not. Wings and glowing colours, and all the other bits of a living body, develop anew, from a single cell by means of the processes of embryonic development. And embryonic development is supervised by DNA, via enzymes, which are made in the way we saw in the previous chapter. If God made glowing colours or fashioned tiny wings, he did it by manipulating the development of an embryo. Nowadays we know that means manipulating DNA (which then manipulates protein, and so on, in ways outlined in Chapter 9). And if – which is actually true – it is natural selection that (indirectly) paints those glowing colours, and fashions those tiny wings, natural selection too does it via DNA. DNA supervises the development of bodies, and DNA in turn is ‘supervised’ over many generations by natural selection. So, indirectly, natural selection ‘supervises’ the development of bodies.
You may have heard that DNA is a ‘blueprint’ for a body, but that’s deeply wrong. Houses and cars have blueprints. Babies don’t. The difference is entirely separate from the fact that cars and houses are designed whereas babies aren’t. Here’s the deeper difference. In a blueprint there’s a one-to-one ‘mapping’ between each bit of the house (or car) and each bit of the blueprint. Neighbouring bits of house correspond to neighbouring bits of blueprint. If the blueprints of a house have been lost, you can redraw them simply by taking meticulous measurements of the house and drawing out a scaled-down version on paper. I’ve just had that done for my house. A man came with a laser gun to measure every room, and it only took him a couple of hours to draw out a complete plan, good enough to build an exact replica of my house.
You can’t do that with a baby. There’s no one-to-one mapping between points on a D
NA ‘blueprint’ and points on a baby. In theory there could have been – it’s not a totally silly idea. The plans of my house, carefully reconstructed by measuring every room, could be digitized in a computer. A modern genetics laboratory is capable of turning any computer information into DNA code, and that could include the digitized plans of my house. You could put the DNA in a test-tube and send it to another genetics lab, in Japan for example, where they could read the DNA and print out a faithful copy of the drawings. An exact replica of my house could then be built in Japan. Maybe on some other planet something like that happens when parents transmit their genetic information to their children: the parental body is ‘scanned’ and turned into a blueprint, which is then digitized in DNA (or that planet’s equivalent of DNA). The digitized scan is then used to build a body of the next generation. But nothing remotely like that happens on our planet. And, between you and me, I suspect that it wouldn’t ever work, not on any planet. One reason for this (only one reason out of several) is that a scan of the parent’s body couldn’t help reproducing things like scars and broken legs. Each generation would accumulate the scars and broken limbs of all the ancestors.
Yes, DNA is a digital code, just like computer code. And yes, DNA transmits digital information from parents to children and so on down countless generations. But no, the information transmitted is not a blueprint. It is not in any sense a map of a baby. It’s not a scan of a parent’s body. A genetics lab can read it, but it couldn’t print out a baby. The only way to turn human DNA information into a baby is to put the DNA into a woman!
If DNA is not a blueprint of a baby, what is it? It’s a set of instructions for how to build a baby, and that’s a very different matter. It’s more like a recipe for making a cake. Or like a computer program whose instructions are obeyed in order: first do this, then do that, then if so-and-so is true do…otherwise do…and so on for thousands of instructions. A computer program is like a very long recipe, complicated by branch points. A recipe is like a very short program, with only a dozen or so instructions. And a recipe is not reversible, like the building of a car or a house is. You can’t take a cake and reconstruct the recipe by taking measurements. And you can’t reconstruct a computer program by watching what it does.