A 3′ terminal sugar-specific polymerase for terminator-free single-nucleotide extension of RNA oligonucleotides

Technical Summary: A 3′ Terminal Sugar-Specific Polymerase for Terminator-Free Single-Nucleotide Extension of RNA Oligonucleotides

Problem Statement
Current methods for enzymatic oligonucleotide synthesis rely heavily on single-nucleotide extension (SNE) strategies that utilize terminator-modified substrates (e.g., 3′-O-allyl or 3′-ONH₂ groups). These chemical modifications necessitate complex nucleotide synthesis, require additional deprotection steps that expose growing chains to reactive species (reducing fidelity), and limit the attainable length of synthesized products. Furthermore, the synthesis of these terminator-modified feedstocks often lacks viable biosynthetic alternatives, undermining the environmental sustainability and high-fidelity potential of enzymatic synthesis. There is a critical need for a fully enzymatic, terminator-free SNE strategy that avoids cumulative deprotection damage and complex chemical feedstock synthesis.

Methodology
The authors investigated E. coli poly(A) polymerase (EcPAP), a template-independent class II nucleotidyltransferase, to determine if it possesses intrinsic sugar specificity that could be exploited for controlled extension. The study employed a multi-faceted approach:

1. Biochemical Screening: EcPAP activity was tested against RNA primers with varying 3′-terminal sugars (ribose vs. deoxyribose) and internal deoxynucleotides to map sugar discrimination.
2. Kinetic Analysis: Steady-state kinetic parameters ($k_{cat}$, $K_m$, and specificity factors) were derived for wild-type EcPAP using Mg²⁺, Mn²⁺, and Co²⁺ cofactors to quantify the discrimination between ribo- and deoxy-terminated primers.
3. Structural Biology: Cryo-electron microscopy (cryo-EM) single-particle analysis was performed to resolve structures of EcPAP in three states: apo, RNA-bound, and a substrate-engaged ternary complex (EcPAP:(rA)₂₄-dA:rATP). This included the use of additives (arginine/glutamate, sodium sulfate, AavLEA1) to prevent aggregation and preferred orientation.
4. Computational Modeling: Molecular dynamics (MD) simulations were conducted to analyze sugar pucker conformations (2′-endo vs. 3′-endo) and the stability of the stalled ternary complex.
5. Protein Engineering: Rational mutagenesis was guided by structural insights to reprogram the enzyme's nucleotide-binding pocket. Specific mutations (R150K, R197M/L) were introduced to alter sugar and nucleobase specificity, enabling the incorporation of all four canonical deoxynucleotides (dNTPs) while maintaining the "self-termination" mechanism.

Key Results

• Primer Terminal Sugar Specificity (PTSS): Wild-type EcPAP exhibits a strict preference for 3′-ribonucleotide-terminated primers. A single 3′-deoxynucleotide at the primer terminus completely blocks further extension, with a specificity factor ($f_{PTSS}$) ranging from $10^2$ to $10^4$ depending on the metal cofactor. This discrimination is catalytic (affecting $k_{cat}$) rather than binding-based ($K_m$), as the enzyme binds both primer types with comparable affinity.
• Structural Mechanism of Inhibition: Cryo-EM structures reveal that when EcPAP binds a primer with a 3′-deoxynucleotide (e.g., dA), it undergoes a substrate-induced conformational switch. The N-terminal head domain rotates ~10°, and a previously disordered $\beta$-turn (R86-D137) becomes ordered. This creates a "closed" but catalytically inactive dead-end complex. The 3′-OH of the terminal deoxynucleotide is sequestered by residue E108 and positioned ~4.4–5.4 Å away from the $\alpha$-phosphate of the incoming nucleotide, preventing nucleotidyl transfer. MD simulations confirm the terminal deoxynucleotide adopts a catalytically inert 2′-endo sugar pucker, unlike the 3′-endo pucker required for catalysis.
• Engineering dMIL: The authors engineered an EcPAP variant, EcPAPR150K/R197M, which overcomes the enzyme's natural preference for rATP and adenine.
  • The R150K mutation relaxes the strict 2′-OH requirement, allowing the incorporation of dATP and dCTP.
  • The R197M mutation (replacing a positively charged arginine with methionine) removes steric and electrostatic clashes with guanine, enabling the incorporation of dGTP and dTTP via van der Waals interactions rather than specific hydrogen bonding.
  • Optimization of reaction conditions (ionic strength, additives like spermidine, PEG, and trimethylglycine) suppressed "leakage" (N+2 products), ensuring that extension stops precisely after a single nucleotide addition.
• Demonstration of dMIL: The engineered enzyme successfully performs deoxy-mononucleotide inhibited ligation (dMIL). It quantitatively extends various RNA primers with any of the four canonical dNTPs to produce a single, monodisperse +1 nt extension product (RNA-3′-dN) without further elongation.

Significance and Claims
The paper claims to present the first fully enzymatic, terminator-free strategy for single-nucleotide extension. By leveraging the intrinsic PTSS of EcPAP and reprogramming its substrate specificity, the authors have established a "self-terminating" mechanism that replaces the need for chemical blocking groups.

• Novel Mechanism: This work identifies the first structural evidence of a substrate-induced open-to-closed conformational transition in a class II nucleotidyltransferase, distinguishing EcPAP from related CCA-adding enzymes which rely on substrate translocation rather than head-domain movement.
• Enzymatic Synthesis: The study establishes EcPAP as a viable platform for next-generation enzymatic oligonucleotide synthesis. The dMIL strategy enables the addition of a 3′-terminal deoxynucleotide to RNA primers with high fidelity and no chemical deprotection steps.
• Future Outlook: While the current work demonstrates dMIL (RNA primer + dNTP), the authors note that a complete terminator-free synthesis cycle would require a complementary "ribo-mononucleotide inhibited ligation" (rMIL) strategy (DNA primer + rNTP), which remains to be discovered or engineered. Nevertheless, dMIL immediately expands the toolkit for site-specific RNA labeling and editing.

The authors emphasize that this approach addresses the core limitations of current methods—cumulative deprotection damage and complex feedstock synthesis—by utilizing a purely enzymatic route with single-nucleotide precision.