How Proteins Evolved

Fifty years ago molecular biologists began to uncover the inner workings of the cell and one of their profound discoveries was that genetic information, stored in the double helix DNA molecule, was translated according to a code to produce a string of amino acids which, after being hitched to each other like train cars, folded up to produce a protein that did something useful in the cell. Interestingly, a given protein’s amino acid sequence was found to have some degree of flexibility. Hemoglobin proteins, for instance, across different species revealed quite a few changes to the sequence while still functioning as a hemoglobin.

Two discoveries that later followed seemed to bode well for evolution. First, protein sequences fit the evolutionary pattern when compared across different species. For instance, a given protein such as hemoglobin has a very similar sequence of amino acids when compared between two similar species, and a much more different sequence when compared between distant species. Second, the degree of flexibility of a given protein’s amino acid sequence was substantial. Hemoglobin sequences in distant species had practically no resemblance. This suggested that a relatively small amount of information was actually needed to code for a given protein. Proteins have hundreds of amino acids, but perhaps only tens of them are important in determining the structure and function of the folded protein structure. This would make evolution’s task of constructing proteins less daunting.

These discoveries are a bit technical and so not always easy to convey to non scientific audiences. But evolutionists have tried, for these discoveries definitely gave them added confidence their theory was right. What evolutionists have been less vocal about are the discoveries that have been made more recently. We now know that for both these discoveries, the story is more complicated than it originally appeared.

The evolutionary pattern

First, the observation that protein sequences fit the evolutionary pattern is not particularly surprising. If we discovered a new bone or tissue, would we not generally expect that it would be more similar between species that otherwise are similar, and more different in more distant species? After all, it functions within a context. So it is no shocker that cellular components such as proteins generally follow the pattern.

But this is not always the case. This pattern that evolution predicts is sometimes violated. Biology reveals some very big differences between otherwise similar species, and some striking similarities between otherwise distant species. This is true for visible features as well as at the molecular level. The evolutionary pattern has now been falsified many times over. Nonetheless, the textbooks continue to proclaim the good news that proteins fit the predicted pattern, and curiously omit any mention of the more recent failures.

The flexibility of a protein sequence

Second, we now understand that the degree of flexibility in a protein sequence is far less than once imagined. One way to think of this is with an analogy to English. Consider these two sentences:

Evolutionary theory has produced a vast number of falsified predictions.

and,

Most of the expectations generated by evolution have turned out false.

These two sentences share the the same meaning yet if you align and compare them letter-by-letter, you would find most of the letters are different. They seem to be unrelated at that level. Could it be that there are only a few key letters that determine the entire meaning of the sentence?

Of course not. We all know that while very different text messages can share the same meaning, this does not imply that most of the letters carry no effect and can be mutated without little or no effect. In fact, even for such short messages there is a large number of possible letter sequences. Each letter has 26 possibilities and in the above example there are about 60 letters. This means there about 10^85 (a one followed by 85 zeros) different letter sequences that are theoretically possible. Of that astronomical set of possibilities, only a relatively tiny set of selections is grammatically correct. The set of selections that makes any sense is even smaller, and the set that carries the same meaning as those above is tiny.

Likewise, it is now clear that while a given protein such as hemoglobin may come in very different sequences, only few changes to that sequence can be sustained. There are many ways to code for the hemoglobin machine, and they may appear to be very different, but that does not mean there is a relatively large number of sequences that can produce a hemoglobin. Here is how one recent paper explained the new findings:

The accepted paradigm that proteins can tolerate nearly any amino acid substitution has been replaced by the view that the deleterious effects of mutations, and especially their tendency to undermine the thermodynamic and kinetic stability of protein, is a major constraint on protein evolvability—the ability of proteins to acquire changes in sequence and function.

Another recent paper explained that only a few percent of a protein’s amino acids can tolerate change at any given point in time.

We are a long way from having hard numbers, but even conservative estimates of the number of protein sequences that are viable, for a given type of protein, are tiny. Any given residue in a protein may not be required to have a specific type of amino acid in order to form a hemoglobin, but in the context of the remainder of the sequence there is a tight requirement. And the number of such contexts is relatively small.

Years ago creationists argued that evolution was improbable because a protein sequence is a long shot. A typical globin sequence has about 140 amino acids. This means the sequence represents a one in 10^182 chance (a one followed by 182 zeros).

It was an imprecise argument, for there is far more than just one sequence that can do the job. But in this case, being off by orders of magnitude hardly matters. A viable globin sequence may be a mere one in 10^150 chance. Who knows, perhaps it is even a one in 10^120 chance. The fact is we do not know, but today’s science is telling us that a viable globin sequence makes finding a needle in a haystack seem easy. And the hemoglobin protein is a relatively small one. Many proteins are several times longer.

How do proteins evolve?

Contrary to early notions that protein sequences were extremely flexible, science is now telling us the opposite. This indication that viable protein sequences occupy a tiny sliver of sequence space suggests that they are difficult to evolve.

If you ask an evolutionist how a protein evolved, you will likely hear the standard answer: via gene duplication and subsequent divergence. In other words, the protein arose from a different type of protein that was pre existing. The gene for that protein duplicated, and then mutated until landing on a new protein that was helpful. And of course this story must have repeated itself thousands of times to create the many different proteins in biology.

It is an unlikely, just-so, story, for viable protein sequences are hard to find. If the different types of proteins each have their own tiny slivers of sequence space as science is suggesting, then gene duplication and divergence, alone, doesn’t stand a chance.

What would be needed are long trails of intermediate, functional, proteins connecting the different types of proteins. These proteins would not only need to be functional, their particular function would have to be useful at the time.

And why would the known proteins just happen to be fortuitously connected by these trails? Science gives us no reason to think such a lucky circumstance is built into the protein world. So either there are no such trails, which means evolution has a problem, or there are such trails which means someone has monkeyed with the fundamentals of protein chemistry.

And we have not yet even addressed the problem of how the first proteins evolved. Remember the evolutionist’s standard explanation for how proteins evolved is by gene duplication and subsequent divergence. But that requires the pre existence of other types of proteins. In other words, the question of how proteins evolve in the first place has been swept under the rug.

The problem of how evolution could create a new type of protein from an existing protein, via gene duplication and subsequent divergence, as difficult as it is, pales in comparison to how evolution was supposed to have created new proteins from scratch. Evolutionists speak of an initial world where RNA molecules do the work of proteins. But even this heroic story doesn’t magically make the problem of protein evolution go away. Whether there were RNA precursors or not, there is a substantial difficulty in explaining how the first proteins could have evolved.

Stepping stone: ATP binding

A few years ago an experiment showed that randomly constructed short proteins have a one in 10^12 shot (a million million) of having function. The function, in this case, was the binding of ATP, the cell’s unit of energy. While that is an interesting experiment, the results do little to help evolution. Most obviously, that function alone is quite minor. Yes, proteins bind to ATP or other chemicals, but that binding is in a complex, tight coordination with other functions. Comparing ATP binding with the incredible feats of hemoglobin, for example, is like comparing a tricycle with a jet airplane.

And even the one in 10^12 shot, though it pales in comparison to the odds of constructing a more useful protein machine, is no small barrier. If that is what is required to even achieve simple ATP binding, then evolution would need to be incessantly running unsuccessful trials. The machinery to construct, use and benefit from a potential protein product would have to be in place, while failure after failure results. Evolution would make Thomas Edison appear lazy, running millions of trials after millions of trials before finding even the tiniest of function. Why would this machinery be in place? Why would it continue to construct and test? Why would evolution maintain such an incompetent test bench?

Evolutionary answers to such questions are like all their stories. It did it because it did it. And if it seems unlikely, then remember there are many planets revolving about many stars, in many galaxies. And beyond that there are many universes. And in any case, a different sort of life—or no life at all—could have evolved. This is what evolution has done to science. Religion drives science and it matters.