Every time a cell divides, its entire genome, three billion base pairs in humans, must be duplicated with near-perfect fidelity. That process begins with an enzyme called DNA primase, which lays down short RNA primers that DNA polymerases need to start copying. How primase itself gets started has been a long-standing puzzle: it must select exactly the right two nucleotides to form the first dinucleotide bond, without the benefit of a pre-existing strand to extend from.
A study published July 3 in Nature Communications by researchers at ETH Zurich and RWTH Aachen reveals the answer in atomic detail using nuclear magnetic resonance spectroscopy. The mechanism involves an unexpected asymmetry, a “second-first” ordering in which the second nucleotide drives template recognition while the first waits unpaired.
The work was led by Pengzhi Wu and Frederic Allain at ETH Zurich, studying a minimal, thermostable primase from the pRN1 plasmid of the archaeon Sulfolobus islandicus. Because this primase is small, single-subunit, and robust at high temperatures, it is an ideal system for structural dissection.
The two-step assembly
DNA primase binds a DNA template and two initiating nucleotides simultaneously in its ancillary domain. The key finding is that only the second of these two nucleotides initially base-pairs with the template. The first nucleotide remains unpaired at this stage, held in place by protein interactions.
This unpaired first nucleotide triggers a process called template-base flipping, the template strand rotates its opposing base out of the normal helical stack and into a binding pocket, analogous to mechanisms seen in DNA repair enzymes. The flipped base engages a linker interaction that induces a closed conformational change in the enzyme.
In the closed state, the second nucleotide moves from the initiation site to the elongation site, and the first nucleotide finally forms a proper base pair in the initiation site. Both nucleotides are now positioned for the catalytic formation of the initial phosphodiester bond.
“It’s a built-in proofreading checkpoint before committing to catalysis,” the mechanism suggests: by requiring the second nucleotide to correctly base-pair before the first one locks into place, the primase avoids initiating with a mismatched first base.
Conservation and implications
The pRN1 primase is an archaeal enzyme, but the mechanism is thought to be conserved across domains of life, meaning the same “second-first” assembly likely operates in human primases as well. If confirmed, the finding would solve a fundamental puzzle in replication biology: how primases achieve fidelity in the initiation step, where classical proofreading (3′-5′ exonuclease activity) is unavailable because there is no nascent strand to check.
The structural details also open avenues for designing inhibitors that block primer synthesis in pathogens, viruses, bacteria, and parasites, without targeting host primases. In synthetic biology, the minimal pRN1 system provides a template for engineering artificial primases with tunable initiation specificity.