The expansion of early genomes presents something of a paradox. For a genome to expand, enzymes that prevent accumulation of errors are required, but a large genome is needed to encode these enzymes. A recent study indicates that this might not have been a serious problem after all — it predicts that primordial genomes were able to tolerate more errors than previously appreciated.

Kun and colleagues investigated the maximum genome size that could have been maintained before accurate replication evolved. This is determined by the rate of replication errors that can be tolerated before fitness is severely affected, as a larger genome will undergo more mutation, which can lead to loss of function.

The authors estimated the error threshold for ribozymes — RNA-based enzymes — as representatives of early RNA genomes. Ribozyme activity is not only affected by nucleotide sequence, but also by secondary structure. Combining this knowledge with results from previous mutagenesis experiments, the authors calculated the activity of many possible ribozyme sequences. Based on this, the maximum error rate that could be tolerated without loss of fitness was significantly higher than previous estimates that took only nucleotide sequence into account.

The error rate of the ribozymal replicators that copied the first genomes is unknown, so the authors used rates from RNA virus replicases as a rough estimate. Using the threshold for error tolerance calculated for ribozymes, an RNA-based protocell would support a genome of ∼7,000 nucleotides. This is enough to contain 100 genes — many more than was previously thought.

By helping to resolve a long-standing paradox Kun and colleagues have revealed how primordial genomes could have increased in complexity — a key step towards understanding how life as we know it evolved.