⚗️ biochemistry

Ancestors of Arylmalonate Decarboxylase show increased Activity, Stability and Stereoselectivity

By employing ancestral sequence reconstruction, researchers engineered arylmalonate decarboxylase ancestors that exhibit significantly enhanced thermal stability, catalytic activity, and stereoselectivity compared to the extant enzyme, overcoming key limitations in the production of chiral carboxylic acids.

Original authors: van der Pol, E., Gerstenberger, J., Georgiadou, X., Hoffka, G., Krammer, L.-M., Schliep, K., Schuer, C., Shina Caroline Lynn Kamerlin, S. C. L., Kara, S., Kourist, R.

Published 2026-08-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: van der Pol, E., Gerstenberger, J., Georgiadou, X., Hoffka, G., Krammer, L.-M., Schliep, K., Schuer, C., Shina Caroline Lynn Kamerlin, S. C. L., Kara, S., Kourist, R.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

In the microscopic world of biology, life relies on tiny molecular machines called enzymes to speed up chemical reactions that would otherwise take forever. These proteins are the workhorses of nature, building everything from the DNA in our cells to the medicines we take. However, when scientists try to use these natural enzymes in factories to produce drugs or materials, they often face a frustrating problem: the enzymes are too fragile. They fall apart when the temperature rises slightly or when they are used for too long, making industrial processes expensive and inefficient. Furthermore, many natural enzymes are not picky enough; they might create a mixture of two mirror-image versions of a molecule, when only one version is safe or effective for human use. To solve this, researchers are turning to a method that looks backward in time to move forward. By using computer models to reconstruct the genetic blueprints of ancient, long-extinct enzymes, they can test these "ancestral" proteins in the lab. Often, these ancient versions are tougher and more versatile than their modern descendants, offering a fresh source of robust tools for modern industry.

A team of researchers set out to improve a specific enzyme known as arylmalonate decarboxylase, or AMDase. This enzyme is valuable because it can strip away a carbon dioxide group from certain molecules to create chiral carboxylic acids, which are essential building blocks for non-steroidal anti-inflammatory drugs like ibuprofen and naproxen. The natural version of this enzyme, found in a bacterium called Bordetella bronchiseptica, works well in the lab but has significant drawbacks for large-scale use. It is unstable, losing half its activity in just over an hour at moderate temperatures, and it struggles with certain types of raw materials, particularly those with small, bulky groups attached to them. It also sometimes fails to produce the purest form of the desired drug molecule. To fix these issues, the scientists did not try to tweak the modern enzyme piece by piece. Instead, they built a family tree of the enzyme's history and used it to predict what the ancestors looked like millions of years ago. They synthesized the genes for ten of these ancient ancestors and tested them to see if they possessed the durability and precision that the modern version lacked.

The results were striking. The researchers found that several of these ancient enzymes were far superior to the modern one. One particular ancestor, which the team named N131, stood out as a champion of stability. While the modern enzyme falls apart quickly when heated, N131 remained active for hundreds of hours under the same conditions. In fact, at a temperature of 45 degrees Celsius, the modern enzyme lasted only about an hour and a half, whereas N131 survived for over 380 hours. This massive improvement in longevity meant that N131 could produce hundreds of times more product before wearing out. The team also discovered that N131 was faster at its job, converting raw materials into the final product more efficiently than the modern enzyme. When they tested the ancient enzymes on a preparative scale, producing a full gram of the chemical, N131 completed the task in just 31 minutes, significantly outpacing the modern version.

To understand why N131 was so much tougher, the researchers ran detailed computer simulations of the protein's structure. They found that the ancient enzyme had acquired a coat of negatively charged chemical groups on its surface that the modern version lacked. These groups acted like a protective shield, holding a tight layer of water molecules around the protein. This hydration shell prevented the enzyme from clumping together and falling apart when heated. It was a natural solution to a problem that modern engineers often struggle to solve. Beyond stability, the ancient enzymes also showed improved precision. One ancestor, named N31, produced a drug building block with an optical purity of 99.7 percent, meaning almost every molecule created was the correct mirror image. This level of purity is crucial for pharmaceuticals, as the wrong mirror image can be ineffective or even harmful.

The study also explored whether these ancient enzymes could handle raw materials that the modern version rejected. The modern enzyme cannot process certain molecules with ethyl groups, but the ancestor N131 could convert them completely. The researchers also demonstrated that they could flip the enzyme's preference, making it produce the opposite mirror image of the drug molecule by swapping just two specific amino acids in the active site. This flexibility showed that the ancient enzymes were not just stronger versions of the modern one, but also more adaptable. By combining the natural evolution of the past with modern engineering, the team created a biocatalyst that is faster, longer-lasting, and more precise. This work suggests that looking into the deep evolutionary history of enzymes can reveal hidden solutions to the stability and selectivity problems that currently limit the production of life-saving medicines.

Technical Summary: Ancestors of Arylmalonate Decarboxylase Show Increased Activity, Stability, and Stereoselectivity

Problem Statement

Bacterial arylmalonate decarboxylase (AMDase) from Bordetella bronchiseptica (BbAMDase) is a valuable biocatalyst for generating $\alpha$ -chiral carboxylic acids, including non-steroidal anti-inflammatory drugs (NSAIDs) and chiral building blocks. However, its broader industrial application is hindered by three primary limitations:

Low Stability: The enzyme exhibits poor operational and thermal stability, with a half-life ( $t_{1/2}$ ) of only 1.2 hours at 30 °C and significant activity loss at temperatures above 40 °C.
Limited Substrate Scope: The enzyme struggles with $\alpha$ -ethyl substituted malonates (due to steric constraints in the active site) and shows reduced activity toward smaller substituents.
Suboptimal Stereoselectivity: While generally selective, the enzyme shows reduced enantiomeric excess (ee) for substrates with sterically similar substituents (e.g., 2-methyl-2-vinylmalonic acid) and requires extensive engineering to invert enantiopreference from (R) to (S).

Traditional protein engineering via site-directed mutagenesis often compromises catalytic efficiency when introducing stabilizing or selectivity-enhancing mutations. The authors hypothesized that Ancestral Sequence Reconstruction (ASR) could provide a suitable sequence context to recover or enhance these properties.

Methodology

The study employed a multi-step approach combining phylogenetics, computational biology, and biochemical characterization:

Phylogenetic Analysis and ASR:
- A dataset of 222 AMDase sequences and 4 outgroup sequences (cis-trans maleate isomerase, MI) was aligned.
- A maximum likelihood (ML) phylogenetic tree was constructed using the WAG model.
- Ten ancestral nodes (N5, N10, N31, N87, N107, N109, N131, N151, N164, N200) were selected to cover different clades and hierarchical levels.
- Ancestral sequences were reconstructed using marginal reconstruction in phangorn, treating gaps as distinct states.
Biochemical Characterization:
- Ancestral genes were expressed in E. coli BL21 (DE3).
- Stability: Thermal stability was assessed via Differential Scanning Fluorimetry (DSF) to determine melting temperatures ( $T_m$ ). Operational stability (half-life) was measured at 30 °C and 45 °C.
- Activity: Specific activity was determined against a panel of malonic acids (1a–6a), including prochiral and $\alpha$ -ethyl substrates. Kinetic parameters ( $K_M$ , $k_{cat}$ ) were calculated for promising variants.
- Stereoselectivity: Enantiomeric excess (ee) was analyzed via chiral GC for (R)-selective ancestors and engineered (S)-selective variants (created by swapping catalytic cysteines G74C/C188G and incorporating hydrophobic pocket mutations).
Computational Analysis:
- Sequence Analysis: Tools like ProtParam, SCMTPP, and Protein-Sol were used to predict thermostability, solubility, and surface charge properties.
- Molecular Dynamics (MD): 10 $\times$ 0.5 $\mu$ s MD simulations were performed on wild-type BbAMDase and the most stable ancestor (N131) at 300 K and 360 K. Analysis focused on Solvent Accessible Surface Area (SASA), hydration shells, hydrogen bonds, and salt bridges.
- Mechanistic Study: Isotope-labeled 4a (pro-R-13C) was used to trace the stereochemical pathway of decarboxylation.
Preparative Scale Application:
- A gram-scale decarboxylation of 2-methyl-2-phenyl malonate (2a) was performed using N131 to evaluate space-time yield and total turnover numbers (TTN).

Key Results

1. Enhanced Stability and Activity

Thermal Stability: Ancestral N131 exhibited a $T_m$ of 62 °C (7 °C higher than wild-type). More remarkably, at 45 °C, N131 showed a 294-fold increase in half-life (381 hours) compared to BbAMDase (1.3 hours). Ancestor N5 also showed a 72-fold improvement.
Catalytic Efficiency: N131 displayed a 1.8-fold increase in specific activity toward 2-methyl-2-phenyl malonate (2a) compared to BbAMDase (25.1 U·mg⁻¹ vs. 14.3 U·mg⁻¹). Kinetic analysis revealed N131 has a significantly lower $K_M$ (0.3 mM vs. 25.5 mM) and higher $k_{cat}$ (33.9 s⁻¹ vs. 29 s⁻¹) for this substrate.
Substrate Scope: N131 achieved full conversion of the sterically demanding $\alpha$ -ethyl substrate (3a), whereas BbAMDase only achieved partial conversion (15%).

2. Improved Stereoselectivity

Natural Selectivity: Ancestor N31 produced (R)-4b (from 2-methyl-2-vinylmalonic acid) with 99.7% ee, a significant improvement over the wild-type's 96.8% ee.
Inverted Selectivity: Engineered (S)-selective variants of N5 (specifically N5 ICPLLG) achieved >99.8% ee for (S)-2b, surpassing the engineered BbAMDase ICPLLG (89.0% ee).
Mechanism: Isotope labeling confirmed that the high selectivity of N31 results from the suppression of the pro-S decarboxylation pathway, increasing the energy difference between transition states to 3.9 kcal·mol⁻¹ (compared to 2.5 kcal·mol⁻¹ for wild-type).

3. Structural Basis for Stability (N131)

Computational analysis attributed the exceptional stability of N131 to a 6% increase in surface Glutamate (Glu) residues compared to the wild type.

Hydration Shell: MD simulations revealed these surface Glu residues form a protective, ordered hydration shell that shields the protein surface.
Salt Bridges: At elevated temperatures (360 K), N131 maintained more intramolecular hydrogen bonds and formed heat-induced salt bridges, whereas the wild-type lost these interactions.
Solubility: The increased surface polarity reduced the hydrophobic SASA ratio, preventing thermal aggregation.

4. Preparative Scale Performance

In a gram-scale reaction, N131 converted substrate 2a to near-completion in 31 minutes (vs. slower conversion by wild-type).

Productivity: N131 achieved a specific productivity of 96.2 g product per g enzyme per hour, a 3.4-fold improvement over BbAMDase.
Total Turnover Number (TTN): N131 achieved a TTN of $3.0 \times 10^8$ , representing a 633-fold improvement over BbAMDase.

Significance and Claims

The paper claims that Ancestral Sequence Reconstruction is a powerful strategy for uncovering naturally evolved improvements in biocatalysts that are difficult to achieve through traditional directed evolution.

Decoupling Stability and Activity: The study demonstrates that ancestral enzymes can simultaneously possess higher thermal stability and higher catalytic activity, challenging the notion that stability often comes at the cost of activity.
Mechanistic Insight: The identification of surface Glu residues as a key driver for thermal stability via hydration shielding offers a new design principle for protein engineering.
Practical Utility: The N131 ancestor is presented as a superior biocatalyst for the industrial synthesis of optically pure $\alpha$ -chiral carboxylic acids, offering significant reductions in production costs and environmental footprint due to higher TTNs and faster reaction times.
Guidance for Engineering: The phylogenetic analysis revealed that key mutations previously identified through rational design (e.g., in the hydrophobic pocket) naturally occur in the evolutionary tree, suggesting that ASR can guide future engineering campaigns by highlighting "natural" solutions to stability and selectivity challenges.

The authors conclude that while not all reconstructed ancestors showed improvements, the successful identification of N131 and N31 validates ASR as a robust method for expanding the catalytic diversity and robustness of enzyme families.