DNA polymerases build DNA from dNTPs and enable replication, whereas RNA polymerases make RNA from NTPs, permitting transcription. Though the subunit structures of DNA and RNA polymerases appear different, there are several single-subunit RNA polymerases in mitochondria and T-odd bacteriophages, which are thought to have evolved from an ancestral DNA polymerase. Now the question that arises is what are the changes that can make a DNA polymerase into an RNA polymerase; should be a difficult change to make given that in the presence of a vast excess of NTPs over dNTPs, DNA polymerases have to powerfully discriminate between the two substrates.
Though several mutations have been described so far, which allow a DNA polymerase to bind to NTPs and produce short stretches of RNA (typically <10nt) at which point the mutant polymerase simply stalls, the answer to the question above has remained elusive.
In a recent paper published in PNAS, Cozens and colleagues from the MRC-LMB and the NIH demonstrate that two targeted mutations in the replicative polymerase from Thermococcus gorgonarius can make an RNA polymerase out of it. The determining mutation is located at about 25 Angstroms from the active site in what is called a ‘thumb’ subdomain. The combination of the two mutations makes the DNA polymerase synthesise RNA upto 1.7kb long, and use various unnatural nucleotide substrates as well. The mutation hardly affects any of the enzyme kinetic parameters associated with its natural dNTP substrate. However, for NTPs (the RNA substrate), the mutation pair increases the Vmax 4-fold and decreases the Km (thus increasing substrate affinity) by three orders of magnitude. These effects might be explained by steric effects and charge differences in the thumb subdomain between the two forms of the polymerase.
The residue at the determining position in the thumb subdomain, though conserved in related organisms, varies in distant organisms. On the basis of the high-temperature adaptation of Thermococcus, the authors state that this residue (referred to as a ‘steric gate’)
may therefore be a specific adaptation to prevent the deleterious consequences of NTP mis-incorporation or replication of genomic lesions at high temperature. Conversely,…, in analogy to the diverse nature of steric-gate residues, the second gate may therefore be elaborated in different ways in the context of different thumb domain structures.