An important question we could ask at this point is: what if organisms aren’t relying on the game of Nucleotide Shuffle to evolve? Could nature be playing a more advanced game somehow?
Bacterial geneticist James A. Shapiro, discoverer of transposable elements in bacteria, recognized that classic evolutionary mechanisms such as mutations and natural selection weren’t sufficient to explain the complexity of life. He proposed that cells somehow shape their own genome, perhaps intelligently – a process he called “natural genetic engineering.” 1
This is an intriguing idea, and in many ways I think it’s superior to the game of Nucleotide Shuffle. It would be more like a game of “Sequence Shuffle,” where whole sequences of nucleotides are shuffled about, perhaps under the “intelligent” control of the cell.
If this were true, I suppose it could account for the serendipity wands of duplication, translocation and magical steps. These would no longer be based on incredible luck, but perhaps on a form of intelligent design, although the intelligence wouldn’t be at the divine level, but at the cellular one.
Cells do have the ability to manipulate their own genomes to a certain extent. As I discussed earlier, transposons are sequences of genetic material that can be “cut” and “pasted” in combination with transposase enzymes and specific markers in the genome, or “copied” from DNA into RNA and then “pasted” back into DNA somewhere else in the genome by reverse transcriptase machinery.
Genes can be transferred between bacteria, and sometimes between other organisms, in a process called horizontal gene transfer. In eukaryotes, genes can often be spliced into pieces, and those pieces can be shuffled, to produce many different variations, a process known as alternative splicing. DNA can be repaired by adding nucleotides to the genome.
These and other mechanisms imply that cells at least have the potential to genetically engineer themselves. Could natural genetic engineering be the explanation for the variety of life we see today? Did the cells themselves engineer life as we know it?
There are three main reasons why I think not. First of all, the functions needed to manipulate the genome require sophisticated machinery, such as the ability to “cut,” “copy” and “paste” sequences. How did the cell acquire these?
If they evolved in the classical evolutionary manner, by mutations and natural selection, then there is no particular reason why mutations and natural selection couldn’t produce everything else, without the need for the cell to intelligently engineer anything.
Another interesting question we could ask here is: if cells do their own genetic engineering, why didn’t they develop the machinery to directly create new proteins or sequences?
“Cut and paste” or “copy and paste” functions are fine if you want to re-use information that already exists. But if a cell wanted to create part of a gene or other genetic sequence from scratch, it would be useful to have machinery equivalent to a keyboard, which allows someone to type directly into a word processor.
For example, if I wanted the sentence THE CAT SAT ON THE MAT to appear here, I can just type this sentence directly into a word processor using my keyboard, rather than having to search for, copy and then paste the individual words THE, CAT, SAT, ON, THE and MAT from elsewhere in the document. If all I could do was cut, copy and paste, and I hadn’t used the word CAT before, I’d have to copy and paste the individual letters C, A and T. I’d also have to paste the word THE twice, and put them in just the right place for the sentence to have the desired meaning.
But despite supposedly evolving for several billion years, and despite having cut, copy and paste facilities, the cell doesn’t appear to have evolved a mechanism to directly write up genetic inventions. In all that time, why hasn’t it evolved the cellular equivalent of a keyboard? Maybe it’s because the cell wasn’t the author of those genetic sequences after all, and it had no need for this facility, or the ability to create it in the first place.
Either way, the lack of a keyboard implies cells can’t do the kind of genetic engineering that could eventually produce frogs and princesses.
A second reason why I don’t think natural genetic engineering is the answer, is that each potential tool in the cell’s engineering toolkit has important limitations.
For example, the ability of the cell to cut, copy and paste sequences doesn’t mean these features can be used at will. This is a common mistake made by evolutionary theorists. They take an idea that has been demonstrated to be true in a limited or highly specific context, and generalize the idea as being true in a much wider context that is often unproven. We could call this the “fallacy of generalization.”
Consider the concept of “translocation.” Just because certain sequences such as transposons have the ability to translocate across the genome through highly regulated mechanisms, it doesn’t follow that the cell can translocate any sequence anywhere at will. This would be a fallacy.
Let’s look at two more potential tools of natural genetic engineering, and their limitations. Alternative splicing is a feature of eukaryotic cells where exons from a gene can be pieced together differently, to make varying proteins. In theory, this could provide the cell with an opportunity to experiment.
The process seems to be regulated in a number of ways, involving proteins that act as silencers or enhancers, genetic and epigenetic markers, growth factors in the cell cycle, and the structure of the chromosomes.
The development of an organism also seems to play a major role in splicing regulation. When development is over, and if the cell isn’t in the process of dividing, presumably it can focus on the business of being alive, and can perhaps respond and react to signals from its environment or other cells. However, it’s not clear that the cell has any truly creative capacity to use the splicing equipment for experimentation.
Let’s consider the DNA repair mechanism known as non-homologous end-joining, where broken ends of the DNA double helix can be repaired by adding DNA that isn’t necessarily the same as the original.
From a natural genetic engineering point of view, this is seen as a potential toolkit for the cell, because it might be able to use this new information to build new things. However, it’s not clear how the cell could know what to do with the new information, so in classical evolutionary thinking, this is simply treated as a mutational process. Either way, the repair mechanism is a feature meant for survival rather than for creative design.
A third reason why I don’t think natural genetic engineering explains life as we know it, is that prokaryotic and eukaryotic cells both have their own unique limitations.
Prokaryotes are single-celled organisms such as bacteria. They can do lots of remarkable things such as swap genes, hunt for food together, and sometimes form what looks like a multi-cellular organism. They use ion channels to communicate with one another, which enables a kind of memory. They can regulate their own genes as a response to cell population density, in a process called “quorum sensing.”
Bacterial colonies also have their own “pangenomes” or “supergenomes,” which we could think of as a large library of genes, out of which an individual bacterium owns a smaller subset. The size of the supergenome has been estimated to be up to ten times bigger than an individual bacterial genome.2
Indeed, bacteria seem to defy the classic evolutionary mechanisms. They lose and gain genes at an order of magnitude faster than they duplicate them, so any supposed evolution doesn’t happen by small changes, but by big leaps as a result of swapping genes. It’s almost as if they were designed to rapidly adapt to their environment. They are, in effect, nature’s little helpers.
We know that cells have built-in mechanisms such as error correcting, to limit mutations. Intriguingly, in situations where they are starving, bacteria switch off error correcting to a certain extent, meaning any genetic sequences they copy in this state are more prone to mutation. This is considered by natural genetic engineering proponents to be another potential toolkit for the cell. Perhaps by allowing more mutations, the colony as a whole can engage in a natural search for any protein variations that might allow at least some of them to survive. On the other hand, when the colony is starving, they would probably lack the resources needed to function normally, so perhaps bacteria have no choice but to shut down this form of error correction.
Either way, if they were capable of engineering their own genomes, surely they could make the proteins needed to survive. But if they have to rely on a higher mutation rate to search for genetic sequences that might keep them alive, this suggests any genetic engineering they can do is limited.
Eukaryotes, by contrast, are usually multi-celled organisms. Eukaryotic cells are much more sophisticated than prokaryotic cells. They usually have a nucleus, which acts as a data management center, where millions or billions of bits of data are stored and processed. The main part of the cell, the cytoplasm, contains mitochondria that provide the power needed to sustain more sophisticated machinery and processes.
Animals, which are eukaryotes, go through a process of development. For humans, this means being a sperm and egg first, then an embryo. The blueprint for this must be stored in the genome. As we discussed earlier, LINE1 transposons are critical for development.
The evidence we have about cells in multi-cellular organisms indicates that they act in a highly regulated manner. They follow blueprints, algorithms and sequences. This is how a human sperm and egg can grow into a human adult with an extraordinary degree of consistency. Development is strictly controlled, otherwise we would see more people with ten heads and twelve legs, in disturbing natural genetic engineering experiments.
There is room for variety, through built-in systems like sexual recombination, gene expression, and epigenetic markers not directly coded for in the DNA blueprint. Cells and organisms can adapt and change to a certain extent, but they do it in highly regulated ways, based on feedback from the environment or other cells. However, the ability of an individual cell to create innovations is limited by being part of an organism.
Furthermore, if anything new is to be inherited by future generations, it must be made in the germline, the cells that pass on the genetic material to their offspring. This means a cell can’t really test the effects of an innovation on the organism as a whole, but can only pass on the change, relying on natural selection to determine whether it was beneficial or not, which isn’t really any different from classical evolutionary thinking.
Incidentally, one line of evidence used to support the idea of natural genetic engineering is the story, which I discussed in a previous chapter, that one day a cell engulfed or was invaded by another cell, and that over time the host was able to transform the cell invader into mitochondria, the cell batteries, in a process called “symbiogenesis” or “endosymbiosis.”
If the story is true, it suggests that cells can indeed engineer their own genomes. After all, they were supposedly able to build themselves cell batteries out of pieces of genetic flotsam and jetsam! However, I have already explained why I think this is an evolutionary fairy tale, widely believed because mitochondria use circular DNA like bacteria do.
The story highlights the philosophical differences between the more traditional evolutionary theorists, who believe that mutations and natural selection, with some duplications, translocations and magical steps thrown in, are enough to turn genetic flotsam and jetsam into a cell battery, and the natural genetic engineering proponents who see this as an obvious form of intelligent design on the part of the host cell.
But perhaps the story is simply fiction. Maybe mitochondria weren’t gradually pieced together from bits and pieces of another cell, but eukaryotic cells were equipped with mitochondria by some other means, such as outside design.
Whatever the case, when we ignore evolutionary storytelling and look at what we know cells are capable of – and we don’t know they are capable of engineering batteries out of genetic bits and pieces – we can say that cells have a number of built-in mechanisms and programs to respond to other cells and their environment.
They contain genes, blueprints, algorithms and gene regulatory networks. They use these to build proteins, and machines made out of those proteins, and more cells in their image. They can regulate how and when sequences are cut, copied or pasted. Genes can be switched on or off, or their expression can be varied according to gene regulatory networks or external feedback.
It is remarkable and truly extravagant how the single-celled pond-dwelling Oxytricha breaks up its genome into nearly a quarter of a million pieces, and then rearranges them into 16,000 chromosomes. It is clearly following a plan, and not doing this at random. Some of its genes can probably be pieced together in different ways, so Oxytricha might make a good case for natural genetic engineering.
However, I would suggest that if the organism is making any choices about which genes to pass on, it is based on feedback from its fellow pond dwellers or the environment. The organism is likely programmed to respond and adapt in certain ways, based on outside signals. In other words, it seems that any engineering Oxytricha do is already built into their programming. They are designed to be flexible and adapt to their environment.
In summary, cells are incredibly complex and exquisite things, and yet they act much more like automatons executing their programming, rather than engineers creatively designing things. The fact that cells are pretty predictable allows biologists to perform repeatable scientific experiments on them in the first place. This also allows human parents to plan for the birth of an actual child, rather than a randomly shaped blob with an uncertain number of mouths to feed.
If intelligence is involved in the design of life, I’d suggest the same intelligence was also behind the origin of life, including all the functions and features needed for living cells to operate, which would require a top-down design approach. I will present a more thorough case for this in the next chapter.
I think James Shapiro was right to question the ability of mutations and natural selection to explain the complex things that happen in living cells. He recognized that there seems to be a kind of intelligence in the way cells adapt.
However, I would say that if the smartest human minds struggle to put together even a self-replicating protocell, and can barely mimic the incredible nanotechnology found within a cell, this highlights the level of ingenuity required to design such things, and suggests that the cells themselves weren’t the inventors of the technology they use, and neither was the blind shuffling of nucleotides. In that case, it becomes time to start asking one of the Forbidden Questions, the questions scientists aren’t allowed to ask.
1 See James A. Shapiro’s book Evolution: A View From The 21st Century for a detailed explanation of the concept of natural genetic engineering; FT Press, 2011. 2 Eugene Koonin, “The Turbulent Network Dynamics Of Microbial Evolution And The Statistical Tree Of Life”, Journal Of Molecular Evolution, 2015.