🧬 biology

Evolution of the SARS-CoV-2 main protease under fitness constraints imposed by folding stability and activity

This study demonstrates that the SARS-CoV-2 main protease (Mpro) has remained evolutionarily constrained by structural and functional requirements since the Omicron era, preserving its substrate recognition and supporting its viability as a robust, long-term antiviral target.

Original authors: David Ferreiro, Miguel Arenas

Published 2026-07-22

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: David Ferreiro, Miguel Arenas

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 ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed.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

Imagine a microscopic world where viruses are like masterful forgers, constantly trying to copy their own blueprints to build new copies of themselves. To do this, they need a special machine called a "protease." Think of this protease as a pair of molecular scissors. The virus builds a long, tangled string of proteins, and the scissors must snip this string at very specific spots to release the functional tools the virus needs to survive and spread. If you can jam those scissors or break the handle, the virus can't build anything, and the infection stops.

Scientists have been hunting for a way to jam the scissors of the SARS-CoV-2 virus (the one that causes COVID-19). They found a perfect target: the virus's "main protease" (Mpro). It's a great target because it's essential for the virus, it looks very different from any scissors our own bodies have (so we don't accidentally hurt ourselves), and it's been used as a target for other viruses before. However, viruses are sneaky. They mutate, changing their shape slightly to dodge our defenses or the medicines we create. The big question is: Can these scissors change so much that our drugs stop working, or are they stuck in a shape they can't escape? This is where the story of a new study comes in, diving deep into the history and future of these viral scissors.

The Viral Scissors That Refuse to Change Shape

In this study, researchers David Ferreiro and Miguel Arenas from the Universidade de Vigo decided to play detective with the SARS-CoV-2 main protease. They wanted to see how this crucial "scissor" protein has changed over time as the virus evolved, and whether those changes would make it harder for our medicines to work. They looked at a massive collection of over 100,000 viral sequences, tracking the virus from the very beginning of the pandemic up to the second half of 2025.

The Timeline of the Scissors
The researchers found that for the first couple of years, the virus mostly kept its original "Wild Type" scissors. Then, around 2021, a few small changes appeared. But the real game-changer happened in 2022 with the rise of the Omicron variant. A specific mutation called P132H took over, becoming the dominant version of the scissors from 2022 all the way through 2025. While other mutations popped up occasionally, like L89F or K90R, they were like fleeting trends that didn't stick. The P132H mutation was the star of the show, found in more than 6% of the virus samples in 2022.

The "No-Go" Zones
Here is the most important part of the story: The virus tried to change its scissors, but it couldn't touch the most important parts. The researchers checked the "active site" (the sharp cutting edge of the scissors) and the "dimerization interface" (the handle where two halves of the scissors snap together to work). They found that mutations in these critical areas were incredibly rare—almost non-existent. It's as if the virus tried to repaint the handle or sharpen the blade, but every time it did, the scissors broke or stopped working. The virus is stuck; it can only change the parts of the scissors that don't affect how they cut.

Testing the Scissors in Action
To see if these small changes (like P132H) actually messed up the scissors' ability to cut, the team used powerful computer simulations. They built digital models of the virus's scissors and tested them against three different "strings" (natural substrates) that the virus normally cuts. They ran these simulations for 10 nanoseconds each, watching how the molecules moved and interacted.

The results were reassuring for drug developers. Even with the mutations, the scissors worked just fine.

Cutting Power: The ability of the mutated scissors to grab and cut the viral strings remained almost exactly the same as the original version.
Grip Strength: The researchers measured the "binding free energy" (how tightly the scissors hold the string). The changes were so tiny they barely mattered.
Movement: They used a technique called Principal Component Analysis (PCA) to watch the overall dance of the protein. The mutated scissors performed the same big, sweeping movements as the original ones. While there were tiny, subtle shifts in the smaller, secondary movements (measured by something called Kullback-Leibler divergence), the main "dance" of the protein stayed the same.

The Verdict
The study suggests that the SARS-CoV-2 main protease is under strict "fitness constraints." This is a fancy way of saying that the virus is so dependent on these scissors working perfectly that it simply cannot afford to change the parts that matter. The virus can shuffle its deck and change the color of the box the scissors come in, but the scissors themselves must stay the same.

Because the active site and the way the protein folds are so tightly locked down by the need to function, the virus hasn't found a way to evolve resistance against drugs that target this specific protein. The researchers found that while the virus is constantly diversifying, the main protease has remained remarkably stable since the Omicron-associated P132H lineage emerged.

Why This Matters
This discovery is a huge relief for anyone relying on antiviral drugs. It suggests that medicines designed to jam these specific scissors (like nirmatrelvir, ensitrelvir, and simnotrelvir) are likely to remain effective for a long time. The virus is trapped by its own biology; it can't change the scissors without killing itself. The study concludes that by watching how the virus evolves and understanding these physical limits, scientists can design better treatments that are robust against future changes. The viral scissors are still the same old scissors, just wearing a slightly different hat.

Technical Summary: Evolution of the SARS-CoV-2 Main Protease Under Fitness Constraints

Problem Statement
Despite the success of vaccines, SARS-CoV-2 continues to evolve, generating immune-evasive variants that necessitate complementary antiviral strategies. The main protease (Mpro, also known as 3CLpro or nsp5) is a critical therapeutic target due to its essential role in viral replication, high conservation across the Coronaviridae family, and lack of close human homologs. However, the evolutionary consequences of Mpro variation on substrate recognition, folding stability, and potential drug resistance remain insufficiently explored. While Mpro is highly conserved, it accumulates amino acid substitutions over time, particularly within specific viral lineages (e.g., Omicron-associated P132H). The central question addressed is whether these prevalent mutations compromise the enzyme's structural stability, catalytic function, or substrate binding affinity, and how evolutionary constraints shape the protein's fitness landscape.

Methodology
The study employed a multi-scale computational approach integrating evolutionary genomics with structural biophysics:

Data Collection and Variant Characterization: The authors analyzed 101,000 SARS-CoV-2 ORF1ab amino acid sequences downloaded from NCBI. Sequences were filtered for length and identity (>75% to the Wuhan reference) and aligned using MAFFT. Mpro sequences were extracted and grouped by mutation profiles and temporal frequency (2019–2025).
Structural Modeling: The wild-type (WT) Mpro and the ten most frequent variants were modeled in complex with three natural substrates (nsp4-5, nsp5-6, and nsp10-11) using AlphaFold3 (v3.0.1).
Molecular Dynamics (MD) Simulations: All-atom MD simulations were performed using AMBER22 with the ff14SB force field and TIP3P solvent. Systems underwent energy minimization, equilibration (NVT and NPT), and 10 ns production runs. Trajectories were analyzed for backbone RMSD and hydrogen bond (HB) occupancy.
Binding Energy Analysis: Binding free energies ( $\Delta G$ ) were estimated using the Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) method. Per-residue energy decomposition was conducted to identify key residues contributing to substrate binding.
Conformational Dynamics Analysis: Principal Component Analysis (PCA) was applied to the C $\alpha$ atomic coordinates to characterize dominant collective motions. The conformational landscapes were visualized using two-dimensional Kernel Density Estimation (KDE) on the first two principal components (PC1 and PC2).
Quantitative Divergence: The Kullback-Leibler (KL) divergence was calculated to quantitatively compare the probability distributions of conformational sampling between mutant variants and the WT, assessing shifts in conformational entropy.

Key Results

Evolutionary Patterns: The WT Mpro predominated until late 2021. The Omicron-associated P132H variant became dominant from early 2022 through 2025. Other mutations (e.g., L89F, K90R) showed transient or low-frequency increases. Crucially, mutations in the active site and dimerization interface were extremely rare or absent (frequencies generally <0.001), indicating strong purifying selection in these functional regions.
Substrate Binding Affinity: MM-GBSA calculations revealed that prevalent substitutions produced no substantial changes in overall substrate binding affinity across the three natural substrates. While statistically significant differences were found in specific variant-substrate pairs (e.g., V73I with nsp4-5), these effects were localized and insufficient to suggest meaningful fitness differences.
Interaction Networks: Detailed analysis of hydrogen bonds and per-residue contributions showed that the overall binding patterns and HB networks were largely conserved. While specific local perturbations occurred (e.g., reduced interactions between nsp4-5 V4 and the V73I variant), the core interaction network involving key active site residues (G23, T24, T25, T26, H41, C145, etc.) remained stable.
Conformational Dynamics: PCA revealed that both WT and mutant variants sample multiple metastable conformational substates, reflecting intrinsic protein flexibility.
- PC1 (Global Motions): The KL divergence along PC1 was zero for all mutants, indicating that the dominant, large-amplitude global collective motions are strictly conserved across all variants.
- PC2 (Secondary Motions): Modest KL divergences were observed along PC2 (ranging from ~0.08 to ~0.76), suggesting that specific substitutions subtly tune secondary motions and local dynamics without disrupting the overall dynamic architecture.

Key Contributions

Comprehensive Fitness Assessment: The study provides a systematic evaluation of how the most prevalent SARS-CoV-2 Mpro variants affect folding stability and substrate binding, confirming that current circulating variants maintain high functional fitness.
Dynamic Conservation: By combining PCA with KL divergence, the authors demonstrate that while Mpro evolution is constrained, it preserves the global dynamical behavior (PC1) essential for function, with only minor adjustments in secondary motions (PC2).
Epistatic Consideration: The analysis included variants with multiple mutations (e.g., K90R_P132H), allowing for the detection of potential epistatic effects that single-mutation studies might miss.
Methodological Framework: The paper establishes a workflow integrating evolutionary surveillance with biophysical modeling (AlphaFold3, MD, MM-GBSA, PCA) to monitor viral protein adaptation.

Significance and Claims
The paper claims that SARS-CoV-2 Mpro evolution is strongly constrained by structural and functional requirements, specifically the need to maintain substrate recognition and catalytic competence. The findings support the suitability of Mpro as a robust antiviral target, as the virus has not yet evolved variants that significantly compromise its interaction with natural substrates or its global conformational dynamics.

The authors assert that while current variants do not threaten the efficacy of Mpro inhibitors, the identification of specific, localized contacts that are vulnerable to future mutations is critical for anticipating resistance pathways. They propose that integrating evolutionary dynamics with biophysical fitness evaluations (stability, binding energetics, and conformational ensembles) offers a practical framework for monitoring viral evolution and designing therapeutic strategies with improved long-term robustness. The study does not claim that resistance is imminent but emphasizes the necessity of continued surveillance to detect variants that might shift the conformational ensemble or binding affinity in ways that favor drug escape.

The Viral Scissors That Refuse to Change Shape

More like this