A Ribonucleotide ↔ Phosphoramidate Reaction Network Optimized by Computer-Aided Design

Andreas Englert, Julian Vogel, Tim Bergner, Jessica Loske, Max von DeliusJournal of the American Chemical Society 2022 · DOI

TL;DR

This paper presents a novel, computer-optimized chemical reaction network that mimics life's out-of-equilibrium nature. By fueling the formation of phosphoramidates from natural ribonucleotides and then controlling their hydrolysis, the researchers created a dynamic system capable of transient self-assembly, offering insights into early life chemistry.

What Problem Does This Paper Address?

Life as we know it is fundamentally an out-of-equilibrium system, constantly dissipating energy to maintain its complex organization. While chemists have been increasingly successful in creating artificial "dissipative systems" that mimic this behavior, most non-enzymatic examples rely on specific chemical fuels and pathways. This work tackles the challenge of building a more versatile and prebiotically relevant dissipative reaction network. Specifically, the authors aim to:

  1. Develop a new dissipative reaction cycle: They focus on the formation of phosphoramidates from natural ribonucleotides (like AMP and GMP) and their subsequent hydrolysis. This is a departure from common approaches using carboxylic acid anhydrides.
  2. Optimize complex reaction parameters: Chemical reactions, especially those in a cycle, involve numerous interconnected parameters (temperature, pH, reactant concentrations, catalysts, time). Optimizing these manually is inefficient and often misses crucial interactions.
  3. Demonstrate transient self-assembly: The ultimate goal is to leverage this optimized reaction cycle to create temporary, organized structures (compartments) from simple, prebiotically plausible building blocks, mimicking early cellular structures.

Methods

The core of this research lies in the application of Design of Experiments (DoE), a statistical and computational approach to efficiently explore and optimize complex systems with multiple interacting variables.

  • Design of Experiments (DoE): Instead of the traditional "one factor at a time" (OFAT) approach, DoE allows researchers to systematically vary multiple parameters simultaneously. This helps identify not only the main effects of each parameter but also their interactions. The authors employed DoE to optimize both the phosphoramidate formation (forward reaction) and hydrolysis (backward reaction) steps.
  • Response Surface Methodology (RSM): Within the DoE framework, RSM was used to build predictive models of the reaction outcomes based on the experimental data. This allows for the prediction of optimal conditions and the understanding of how parameters influence the reaction.
  • High-Performance Liquid Chromatography (HPLC): This is the primary analytical technique used to quantify the amounts of reactants and products (e.g., Bn-GMP, AMP, C7-PA) throughout the experiments, providing the "response" data for the DoE.
  • Dynamic Light Scattering (DLS): Used to characterize the size and aggregation state of the self-assembled structures.
  • Confocal Laser Scanning Microscopy (CLSM): Employed to visualize the formation and structure of the transient compartments, often using fluorescent dyes.
  • Cryogenic Transmission Electron Microscopy (cryo-TEM): Used for high-resolution imaging of the self-assembled structures, providing detailed morphological information.

What's novel here is the systematic application of DoE to optimize and understand a dissipative reaction cycle itself, rather than just a single reaction.

Datasets

Not discussed in the paper. The experiments were conducted using synthesized or commercially available chemical reagents.

Key Insights and Ideas

  • The power of DoE for complex systems: The paper strongly advocates for DoE in systems chemistry. It highlights how DoE can uncover critical parameter interactions that OFAT methods would miss, leading to more efficient optimization and a deeper understanding of reaction networks.
  • 1-ethylimidazole (EtIm) as a key catalyst: The researchers discovered that EtIm dramatically influences the hydrolysis rate of phosphoramidates. This catalytic effect is reminiscent of histidine residues in natural enzymes (like HINT1), suggesting a potential prebiotically relevant catalytic mechanism.
  • pH is critical for phosphoramidate formation: The study revealed that pH is the most significant factor in the forward reaction, with acidic conditions hindering nucleophilicity and promoting intermediate hydrolysis.
  • Temperature and time interactions are characteristic of dissipative cycles: The observation that high yields can be achieved either with long reaction times at low temperatures or short times at high temperatures is a hallmark of dissipative systems where backward reactions become dominant at higher temperatures.
  • "Sweet spot" for efficient cycles: By combining the DoE models for both forward and backward reactions, the authors identified a specific set of conditions (a "sweet spot") that allows for efficient phosphoramidate formation while ensuring rapid hydrolysis, thus closing the dissipative cycle.
  • Prebiotically relevant self-assembly: The system successfully demonstrated transient self-assembly of compartments using natural ribonucleotides and simple amines, suggesting a plausible pathway for the formation of early cellular structures.

Key Results

  • Optimized Dissipative Cycle: The researchers successfully established a fuel-driven (EDC) reaction cycle for phosphoramidate formation and hydrolysis.
  • Catalytic Hydrolysis: EtIm was shown to be a crucial catalyst for phosphoramidate hydrolysis, enabling the backward reaction under mild conditions.
  • Predictive Models: DoE models accurately predicted reaction outcomes, validated by experimental confirmation runs with good agreement (e.g., predicted Bn-GMP hydrolysis below 15% was experimentally confirmed).
  • Transient Self-Assembly: The optimized cycle was used to generate transient self-assembled structures (vesicular or colloidal) observed via DLS and CLSM, which disappeared upon hydrolysis of the phosphoramidate.
  • Multiple Cycles Demonstrated: The dissipative reaction cycle could be refueled and run for at least four cycles, demonstrating sustained out-of-equilibrium behavior.

Limitations and Future Work

  • Fuel limitations: The authors acknowledge that EDC, while a useful model reagent, has questionable prebiotic plausibility. Future work could explore more naturally occurring fuels.
  • Side product accumulation: Pyrophosphates (like GppG and AppA) were observed to accumulate over multiple cycles. This "kinetic sink" can limit the long-term sustainability of the system. Future research could focus on catalysts that can hydrolyze these pyrophosphates.
  • Temperature dependence of self-assembly: While the reaction cycle was optimized at 55 °C, self-assembly was only observed at lower temperatures (25 °C). This temperature mismatch is interesting and could be explored further in the context of thermophoresis and early life environments.
  • Broader scope of amines: While the approach was shown to work with both benzylamine and a simple alkylamine (heptylamine), further exploration of different amine structures and their impact on reaction kinetics and self-assembly is warranted.
  • Prebiotic relevance: While the building blocks are prebiotically plausible, the overall system is still a laboratory construct. Further work is needed to bridge the gap between this system and actual prebiotic chemistry.

Summary updated

Abstract

A growing number of out-of-equilibrium systems have been created and investigated in chemical laboratories over the past decade. One way to achieve this is to create a reaction cycle, in which the forward reaction is driven by a chemical fuel and the backward reaction follows a different pathway. Such dissipative reaction networks are still relatively rare, however, and most non-enzymatic examples are based on the carbodiimide-driven generation of carboxylic acid anhydrides. In this work, we describe a dissipative reaction network that comprises the chemically fueled formation of phosphoramidates from natural ribonucleotides (e.g., GMP or AMP) and phosphoramidate hydrolysis as a mild backward reaction. Because the individual reactions are subject to a multitude of interconnected parameters, the software-assisted tool "Design of Experiments" (DoE) was a great asset for optimizing and understanding the network. One notable insight was the stark effect of the nucleophilic catalyst 1-ethylimidazole (EtIm) on the hydrolysis rate, which is reminiscent of the action of the histidine group in phosphoramidase enzymes (e.g., HINT1). We were also able to use the reaction cycle to generate transient self-assemblies, which were characterized by dynamic light scattering (DLS), confocal microscopy (CLSM), and cryogenic transmission electron microscopy (cryo-TEM). Because these compartments are based on prebiotically plausible building blocks, our findings may have relevance for origin-of-life scenarios.
BibTeX
@article{englert2022ribonucleotide,
  author = {Andreas Englert and Julian Vogel and Tim Bergner and Jessica Loske and Max von Delius},
  title = {{A Ribonucleotide ↔ Phosphoramidate Reaction Network Optimized by Computer-Aided Design}},
  journal = {Journal of the American Chemical Society},
  year = {2022},
  doi = {10.1021/jacs.2c05861},
}
Figure