Self-replicating RNA may lack the fidelity needed to originate life
The hypothesis of an ‘RNA world’ as the font of all life on Earth has been with us now for more than 30 years, the term having been coined by the biologist Wally Gilbert in 1986. You could be forgiven for thinking that it pretty much solves the conundrum of how the replication of DNA could have avoided a chicken-and-egg impasse: DNA replication requires protein enzymes, but proteins must be encoded in DNA. The intermediary RNA breaks that cycle of dependence because it can both encode genetic information and act catalytically like enzymes. Catalytic RNAs, known as ribozymes, play several roles in cells.
It’s an alluring picture – catalytic RNAs appear by chance on the early Earth as molecular replicators that gradually evolve into complex molecules capable of encoding proteins, metabolic systems and ultimately DNA. But it’s almost certainly wrong. For even an RNA-based replication process needs energy: it can’t shelve metabolism until later. And although relatively simple self-copying ribozymes have been made, they typically work only if provided with just the right oligonucleotide components to work on. What’s more, sustained cycles of replication and proliferation require special conditions to ensure that RNA templates can be separated from copies made on them.
It’s an alluring picture – catalytic RNAs appear by chance on the early Earth as molecular replicators that gradually evolve into complex molecules capable of encoding proteins, metabolic systems and ultimately DNA. But it’s almost certainly wrong. For even an RNA-based replication process needs energy: it can’t shelve metabolism until later. And although relatively simple self-copying ribozymes have been made,1 they typically work only if provided with just the right oligonucleotide components to work on. What’s more, sustained cycles of replication and proliferation require special conditions to ensure that RNA templates can be separated from copies made on them.
In the soup
Perhaps the biggest problem is that self-replicating ribozymes are highly complex molecules that seem very unlikely to have randomly polymerised in a prebiotic soup. And the argument that they might have been delivered by molecular evolution merely puts the cart before the horse. The problem is all the harder once you acknowledge what a complex mess of chemicals any plausible prebiotic soup would have been. It’s nigh impossible to see how anything lifelike could come from it without mechanisms for both concentrating and segregating prebiotic molecules – to give RNA-making ribozymes any hope of copying themselves rather than just churning out junk, for example.
In short, once you look at it closely, the RNA world raises as many questions as it answers. Even one of its chief advocates, Gerald Joyce of the Scripps Research Institute in California, suggested recently that it might be necessary to consider that the RNA world was preceded by ‘some other replicating, evolving molecule’ such as peptide-nucleic acid hybrids.2 That, of course, may simply defer some of the problems rather than solving them.
Joyce and his coworkers have now found a ribozyme that holds the potential to copy heritable ‘pre-genetic’ information into RNAs considerably more complex and structured than any seen before.3 This molecule was found through a process of in vitro evolution, the criterion for selection in each round being the ability to keep adding individual nucleotides to a ‘primer’ RNA segment to make long strands – that is, to act as an RNA polymerase.
In short, once you look at it closely, the RNA world raises as many questions as it answers. Even one of its chief advocates, Gerald Joyce of the Scripps Research Institute in California, suggested recently that it might be necessary to consider that the RNA world was preceded by ‘some other replicating, evolving molecule’ such as peptide-nucleic acid hybrids. That, of course, may simply defer some of the problems rather than solving them.
Joyce and his coworkers have now found a ribozyme that holds the potential to copy heritable ‘pre-genetic’ information into RNAs considerably more complex and structured than any seen before.1 This molecule was found through a process of in vitro evolution, the criterion for selection in each round being the ability to keep adding individual nucleotides to a ‘primer’ RNA segment to make long strands – that is, to act as an RNA polymerase.
Joyce and colleagues started with a known ribozyme called the class I ligase (which links nucleotide fragments together). After 14 rounds of evolution, their best performing ribozyme was capable of joining nucleotides into three separate RNAs that would then spontaneously assemble (after purification) into the original ligase. In other words, it was capable of synthesising the RNA molecule from which it was derived – to make its own ancestor. Joyce and colleagues say this is ‘the most complex functional RNA that has been synthesized by a ribozyme’ from single nucleotides, and believe that, with further in vitro evolution, they might obtain a ribozyme of similar complexity that is genuinely able to make itself.
Copy errors
Does this help us to understand how an RNA world could have arisen? On the contrary, it points to another problem. The best RNA polymerase the researchers obtained this way had a roughly 8% chance of inserting any nucleotide wrongly, and any such error increased the chance that the full chain encoded by the molecule would not be replicated. What’s more, making the original class I ligase was even more error-prone and inefficient – there was a 17% chance of an error on each nucleotide addition, plus a small chance of a spurious extra nucleotide being added at each position.
These errors would be critical to the prospects of molecular evolution, since there is a threshold error rate above which a replicating molecule loses any Darwinian advantage over the rest of the population – in other words, evolution depends on good enough replication. Fidelity of copying could thus be a problem, hitherto insufficiently recognised, for the appearance of a self-sustaining, evolving RNA-based system: that is, for an RNA world.
Maybe this obstacle could have been overcome in time. But my hunch is that any prebiotic molecule will have been too inefficient, inaccurate, dilute and noise-ridden to have cleared the hurdle. Rather, we’ll need to look for ways in which noisy, heterogeneous and perhaps compartmentalised molecular collectives could have bootstrapped their way to life. And that, after all, makes complete sense when you recognise that this is precisely what cells still are.
Reference
1 K F Tjhung et al., Proc. Natl Acad. Sci. USA, 2020, 117, 2906 (DOI: 10.1073/pnas.1914282117)
References
1 N Paul and G F Joyce, Proc. Natl Acad. Sci. USA, 2002, 99, 12733 (DOI: 10.1073/pnas.202471099)
2 M P Robertson and G F Joyce, Cold Spring Harbor Perspect. Biol., 2012, 4, a003608 (DOI: 10.1101/cshperspect.a003608)
3 K F Tjhung et al., Proc. Natl Acad. Sci. USA, 2020, 117, 2906 (DOI: 10.1073/pnas.1914282117)
1 Reader's comment