RNA Replicase Ribozyme

RNA Replicase Ribozyme (or RNA-dependent RNA polymerase ribozyme, RdRp ribozyme) is a catalytic RNA molecule (ribozyme) that catalyzes the template-directed 5'→3' synthesis of complementary RNA strands from activated ribonucleoside 5'-triphosphates (NTPs) or trinucleotide 5'-triphosphates (pppNNN) without the assistance of any protein enzymes. In the theoretical framework of the RNA world hypothesis articulated by Carl Woese, Francis Crick, Leslie Orgel, and Walter Gilbert, an RNA replicase ribozyme is the primary macromolecular machine required to transition prebiotic chemical systems from slow, error-prone non-enzymatic RNA replication into a self-sustaining Darwinian system capable of hereditary replication, mutation, and natural selection inside early protocells.

Because no surviving natural RNA-dependent RNA polymerase ribozyme has been identified in modern cellular genomes—where genome replication was long ago taken over by high-processivity protein polymerases—biochemists have reconstructed and evolved candidate RNA replicase ribozymes in the laboratory using in vitro selection and directed evolution starting from synthetic random-sequence libraries. Between 1993 and 2024, lineages inaugurated by David P. Bartel, Jack W. Szostak, Philipp Holliger, and Gerald F. Joyce advanced from ligating a single phosphodiester bond to polymerizing >200-nucleotide RNA templates, amplifying functional ribozymes and tRNAs via PCR-like cycles, unwinding folded RNA secondary structures using triplet substrates, and demonstrating Darwinian evolution of a second ribozyme with heritable variation.

From the 1993 Class I RNA Ligase to Template-Directed Polymerization

All modern mononucleotide and trinucleotide polymerase ribozymes descend from the Class I RNA ligase ribozyme, isolated in 1993 by David P. Bartel and Jack W. Szostak from a pool of ~1.6 × 10¹⁵ random-sequence RNA molecules (220 nt randomized region) via SELEX. The Class I ligase catalyzes the formation of a canonical 3',5'-phosphodiester bond between the 3'-hydroxyl (3'-OH) group of an aligned substrate oligonucleotide and the α-phosphorus atom of a 5'-triphosphate (5'-pppG) at its own 5' terminus, releasing inorganic pyrophosphate (PP_i). Noticeably, this chemistry—nucleophilic attack of a primer 3'-OH on a 5'-triphosphate with displacement of PP_i—is chemically identical to the elongation step of all cellular DNA and RNA polymerases.

High-resolution X-ray crystal structures of the Class I ligase core solved by David M. Shechner and David P. Bartel (2009) revealed a tripod-like tertiary architecture built from three coaxially stacked helical stems joined by an A-minor triad and a solvent-excluded catalytic pocket. Inside the active site, a conserved cytidine (C47) and adjacent phosphate oxygens coordinate two catalytic Mg²⁺ ions that lower the pK_a of the primer 3'-OH nucleophile and stabilize the pentacoordinate phosphorane transition state and departing pyrophosphate leaving group, accelerating phosphodiester bond formation ~10⁷-fold over the uncatalyzed rate (k_cat > 100 min⁻¹).

The R18 Polymerase Ribozyme Breakthrough (2001)

While the original Class I ligase could polymerize up to 3–6 consecutive NTPs on a short tethered template (Ekland & Bartel, 1996), it could not copy arbitrary detached templates because it lacked a sequence-independent template–primer binding domain. In 2001, Wendy K. Johnston, Peter J. Unrau, Michael S. Lawrence, Margaret E. Glasner, and David P. Bartel appended a 76-nucleotide randomized accessory domain onto the Class I ligase core and performed 18 rounds of in vitro selection to isolate the R18 RNA polymerase ribozyme (189 nucleotides total).

R18 bound an external primer–template duplex in trans through sequence-independent minor-groove sugar-backbone contacts and catalyzed the template-directed incorporation of up to 14 consecutive ribonucleoside triphosphates (ATP, GTP, CTP, UTP) with an average Watson–Crick base-pairing fidelity of 96.7% per nucleotide. However, R18 was still far short of self-replication: it required a 189-nucleotide sequence of its own, exhibited weak processivity (K_m for primer–template in the micromolar to millimolar range), and dissociated before completing two helical turns.

Sequence-Specific Tethering and Eutectic Water-Ice Phases (tC19Z and tC9Y)

To overcome the weak binding affinity of R18 for its primer–template duplex, Philipp Holliger, Aniela Wochner, and James Attwater (2011) evolved the tC19Z polymerase ribozyme, which incorporated a 5' single-stranded tail that base-paired to a complementary tag on the template (functioning as a processivity clamp) along with core mutations that enhanced sequence generality. tC19Z polymerized 95 consecutive nucleotides on a favorable cytosine-rich repeat template and synthesized a functional 24-nucleotide cleavage ribozyme from an RNA template.

Holliger and colleagues next discovered that freezing polymerase ribozyme reactions into eutectic water-ice phases at −7 °C to −15 °C enhances polymerization yield and lifetime. As pure water crystallizes into hexagonal ice crystals, the RNA enzyme, template, NTPs, and Mg²⁺ ions are concentrated up to 200-fold inside microscopic liquid brine channels between the ice crystals, while the low temperature lowers the rate of Mg²⁺-catalyzed 2'-OH backbone hydrolysis by more than an order of magnitude. Using compartmentalized bead-tagging evolution in eutectic ice, Attwater, Wochner, and Holliger (2013) isolated the tC9Y ribozyme, which synthesized RNA strands up to 206 nucleotides in length—exceeding the 202-nucleotide length of the polymerase ribozyme itself for the first time.

Multi-Cycle RNA Amplification (RiPCR) and Darwinian Evolution (24-3, 38-6, and 52-2)

Despite copying >200 nucleotides on unstructured C-rich templates, tC19Z and tC9Y stalled whenever they encountered folded RNA secondary structures (such as G–C stems or hairpins) required by functional ribozymes. Using microfluidic in vitro compartmentalization and kinetic selection for completion of structured templates, David P. Horning, Katrina F. Tjhung, Nikolaos Papastavrou, and Gerald F. Joyce (2016–2024) evolved the 24-3, 38-6, and 52-2 polymerase ribozymes.

These variants copy structured templates directly from NTPs and perform Ribozyme-Catalyzed Polymerase Chain Reaction (RiPCR)—entering an exponential growth phase (exp(kt)) that amplifies both the (+) and (−) strands of functional RNA aptamers, hammerhead ribozymes, and complete 76-nucleotide cloverleaf yeast phenylalanyl-tRNAs across thermal cycles. In 2024, Papastavrou, Horning, and Joyce used the evolved 52-2 polymerase ribozyme to propagate a population of hammerhead ribozymes across multiple rounds of (+) and (−) strand replication, observing the emergence and selective enrichment of new adaptive hammerhead sequence variants—the first experimental demonstration of ribozyme-catalyzed Darwinian evolution of a functional genetic molecule.

The t5⁺¹ Triplet Polymerase Ribozyme (pppNNN)

A persistent biophysical barrier for mononucleotide polymerase ribozymes is the strand-invasion / secondary-structure problem: single-stranded RNA templates naturally fold into stable intramolecular hairpins, and once a complementary strand is synthesized, the resulting long double-stranded RNA duplex has a melting temperature (T_m) exceeding 80–90 °C. In 2018, James Attwater, Philipp Holliger, and coworkers evolved the t5⁺¹ triplet polymerase ribozyme—a heterodimeric ribozyme (comprising a 172-nt catalytic subunit paired with a 135-nt non-catalytic RNA cofactor functioning like a processivity clamp) that uses trinucleotide 5'-triphosphates (pppNNN, a pool of all 64 possible RNA triplets) rather than mononucleotides (NTPs) as substrates.

Because triplets bind to single-stranded templates much more tightly than single nucleotides and cooperate in string-like binding along the template ahead of the active site, the triplet polymerase invades and unwinds stable hairpin secondary structures without thermal cycling. Moreover, t5⁺¹ can initiate synthesis without a primer and can polymerize templates in non-canonical 3'→5' as well as 5'→3' directions, enabling it to synthesize all five fragments of its own catalytic (+) and complementary (−) strands directly from triplet substrates.

Cross-Chiral D- and L-RNA Polymerase Ribozymes

In 2014, Jonathan T. Sczepanski and Gerald F. Joyce demonstrated a second solution to the template-folding problem: a cross-chiral RNA polymerase ribozyme (83 nt) that catalyzes the polymerization of opposite-handedness (enantiomeric) RNA. Because D-RNA (containing natural β-D-ribose) and L-RNA (containing mirror-image β-L-ribose) cannot form contiguous Watson–Crick base pairs with one another, a D-RNA cross-chiral polymerase recognizes its L-RNA template and L-NTP substrates purely through tertiary structural contacts rather than base pairing. Consequently, the D-ribozyme never hybridizes parasitically with its L-RNA template, allowing a D-ribozyme to synthesize its own mirror-image L-ribozyme counterpart from oligonucleotide building blocks.

Why is an RNA replicase ribozyme central to the RNA World hypothesis?

For a population of prebiotic RNA molecules to undergo Darwinian evolution by natural selection, sequence information must be copied from parent templates to progeny strands faster than the RNA backbones degrade by hydrolysis, and with an error rate below Manfred Eigen's quasispecies error threshold (μ < ln σ / L). While non-enzymatic RNA replication can copy short oligoribonucleotides (10–25 nt), an evolved RNA replicase ribozyme is required to copy longer, structured ribozymes (50–200 nt) with sufficient speed and fidelity to sustain cellular life before protein polymerases evolved.

How does freezing an RNA polymerase ribozyme reaction in ice accelerate RNA synthesis?

When an aqueous reaction containing an RNA polymerase ribozyme, RNA templates, NTPs, and Mg²⁺ ions is cooled to −7 °C to −15 °C, pure water freezes into hexagonal ice crystals while excluding solutes into a connected network of microscopic liquid eutectic brine channels. This eutectic concentration increases local reactant and ribozyme concentrations up to 200-fold—promoting template binding and ribozyme dimerization—while the sub-zero temperature suppresses Mg²⁺-catalyzed 2'-OH hydrolysis of the RNA backbone, extending the productive catalytic lifetime of the ribozyme from hours to weeks.

What is the advantage of using trinucleotide triphosphates (triplets) instead of single NTPs?

Single-stranded RNA templates rapidly fold back on themselves into stable G–C stems and hairpins that block mononucleotide polymerase ribozymes from advancing. In the t5⁺¹ triplet polymerase ribozyme (Attwater et al., 2018), trinucleotide triphosphates (pppNNN) bind cooperatively along the template via three base pairs at a time, providing enough binding free energy to melt and invade secondary structures and copy folded catalytic RNAs without requiring high-temperature strand denaturation.

1. Bartel, D. P., & Szostak, J. W. (1993). "Isolation of new ribozymes from a large pool of random sequences." Science, 261(5127), 1411–1418. https://doi.org/10.1126/science.7690155
2. Johnston, W. K., Unrau, P. J., Lawrence, M. S., Glasner, M. E., & Bartel, D. P. (2001). "RNA-catalyzed RNA polymerization: Accurate and general RNA-templated primer extension." Science, 292(5520), 1319–1325. https://doi.org/10.1126/science.1060786
3. Attwater, J., Raguram, A., Griehl, A. S., & Holliger, P. (2018). "Ribozyme-catalysed RNA synthesis using triplet building blocks." eLife, 7, e35255. https://doi.org/10.7554/eLife.35255
4. Papastavrou, N., Horning, D. P., & Joyce, G. F. (2024). "RNA-catalyzed evolution of catalytic RNA." Proceedings of the National Academy of Sciences, 121(11), e2321592121. https://doi.org/10.1073/pnas.2321592121