De novo Folding Mechanisms of Lasso Peptides

By integrating molecular dynamics simulations with deep learning, this study elucidates the universal uphill folding mechanism of lasso peptides, revealing that their low spontaneous folding probability is driven by loop stability and high entropy costs that can be mitigated by spatial confinement, thereby establishing fundamental principles for their rational engineering.

The Gist

The Big Picture: The "Impossible Knot"

Imagine you have a long, floppy piece of string. Now, imagine you need to tie a specific, complex knot where one end of the string loops through the middle of the string and gets locked in place. This is what Lasso Peptides are: tiny protein chains that twist themselves into a lasso shape (a loop with a tail threaded through it).

These "lassos" are incredible. They are so tough that bacteria can't digest them, and they can be used as powerful medicines to fight infections or cancer. But here's the problem: Nature makes them look easy, but math says it should be impossible.

The scientists in this paper asked: How does a floppy string know how to tie itself into this knot without a pair of hands (or an enzyme) to help it?

The Problem: The "Up-Hill" Struggle

Usually, when you drop a piece of string on the floor, it just lies there in a messy pile. It takes a lot of effort to pick it up and tie a knot. In the world of proteins, this is called entropy (disorder). Nature loves disorder; it hates order.

The researchers discovered that for these lasso peptides, tying the knot is like trying to roll a heavy boulder up a steep hill.

• The Bottom of the Hill: The messy, unfolded string (the most comfortable place for the peptide).
• The Top of the Hill: The perfect lasso knot.

Their simulations showed that if you just let these peptides sit in a test tube (water), they almost never tie themselves. The chance of a random peptide tying itself into a lasso is less than 0.8%. It's like flipping a coin and getting "heads" 10 times in a row. It's statistically unlikely to happen by accident.

The Solution: How They Figured It Out

Since the peptides are so stubborn, the scientists couldn't just watch them tie themselves in a normal computer simulation; it would take longer than the age of the universe to see it happen once.

To solve this, they used a clever trick called "Multi-Ensemble Markov Models" (a fancy way of saying they ran thousands of tiny, parallel simulations and used AI to stitch the results together).

• The Analogy: Imagine trying to see a rare bird that only flies once every 100 years. Instead of waiting in one spot, you hire 1,000 people to look in different spots at the same time, and then you use a computer to predict exactly when and where the bird will appear.

They also used a technique called Umbrella Sampling.

• The Analogy: Imagine you are trying to push a heavy box up a hill, but you can't do it alone. So, you use a "force" (like a giant umbrella pushing down on the box) to hold it in place at different points on the hill. This lets you map out the entire hill, even the parts you can't reach on your own.

The Two Keys to Success

After mapping the "hill" for 20 different types of lasso peptides, they found two main reasons why some knots are easier to tie than others:

1. The "Loop" Needs to Be Stiff (The Anchor)
For the tail to thread through the loop, the loop itself has to stay open and stable.

• The Analogy: Think of the loop as a hula hoop. If the hoop is made of wet, floppy clay, it will collapse, and the tail can't get through. But if the hoop is made of stiff wire (specifically a structure called a beta-hairpin), it stays open.
• The Discovery: The peptide that tied itself the best (Microcin J25) had a very stiff, wire-like loop. Peptides with floppy loops almost never succeeded.

2. The "Entropy Tax" (The Cost of Order)
Tying a knot requires the string to stop wiggling and stay still. This costs "energy" because nature prefers wiggling.

• The Analogy: Imagine a chaotic dance party. Getting everyone to stand still and hold a specific pose is hard work. The "tax" for stopping the dance is high.
• The Discovery: The biggest reason these peptides don't tie themselves is this "tax." They are too floppy.

The "Magic Pocket" (How Nature Solves It)

If the peptides are so bad at tying themselves, how does nature make them? The answer is the Cyclase Enzyme.

The researchers simulated what happens when the peptide is inside a tiny, tight pocket (the enzyme).

• The Analogy: Imagine trying to tie a knot in a giant, open field. It's hard because the string can go anywhere. Now, imagine trying to tie that same knot inside a tiny, cramped shoebox. The walls of the box force the string to stay close together. It's much harder for the string to get "lost" or messy.
• The Result: When they simulated the peptide inside this "shoebox" (the enzyme pocket), the knot formed much more easily. The enzyme acts like a chaperone or a molding tool. It doesn't just tie the knot; it forces the string into the right shape so the knot can happen.

The Experiment: Proving It Works

To prove their computer models were right, they went into the lab.

• They took the "stiffest" peptide (Microcin J25) and made tiny changes to its DNA.
• Some changes made the loop stiffer; others made it floppier.
• The Result: When they made the loop stiffer (more beta-hairpin), the peptide was produced much more efficiently. When they made it floppier, production dropped. This confirmed that loop stiffness is the secret sauce.

The Takeaway

This paper tells us that lasso peptides are naturally lazy. They don't want to tie themselves. They need a "push" (the enzyme) and a "mold" (the tight pocket) to get the job done.

Why does this matter?
Now that we know the rules (stiff loops + tight pockets = good knots), scientists can design new, custom-made lasso peptides in the lab. We can engineer them to be better medicines, stronger antibiotics, or even new materials, by tweaking the "stiffness" of their loops to make them easier to manufacture.

In short: Nature uses a tiny molecular "shoebox" to force a floppy string to tie a perfect knot. Once we understand the rules of that shoebox, we can build our own knots for the future.

Technical Summary

1. Problem Statement

Lasso peptides are a class of ribosomally synthesized and post-translationally modified peptides (RiPPs) characterized by a unique [1]rotaxane topology, where the N-terminus is threaded through a macrocyclic ring and locked by bulky "plug" residues. This structure confers exceptional stability against proteases and harsh conditions. However, the de novo folding mechanism (spontaneous folding in solution without enzymatic assistance) remains poorly understood.

Key challenges include:

• Thermodynamic Unfavorability: The formation of the threaded conformation is kinetically trapped and thermodynamically unfavorable in solution.
• Sampling Limitations: Standard Molecular Dynamics (MD) simulations struggle to capture these rare folding events due to the vast disparity between simulation timescales (nanoseconds to microseconds) and biological folding timescales (microseconds to seconds).
• Lack of Generalizability: Previous studies focused almost exclusively on a single model peptide (microcin J25), failing to capture the universal principles governing the diverse library of lasso peptides.

2. Methodology

The authors employed an integrative approach combining extensive computational simulations, deep learning, and experimental validation to study 20 distinct lasso peptides lacking secondary post-translational modifications.

• Molecular Dynamics (MD) Simulations:
  • Unbiased MD: ~200 µs of simulation per peptide to explore conformational space.
  • Biased MD (Umbrella Sampling): Used to enhance sampling of high-energy pre-folded states that are rarely visited in unbiased simulations.
• Statistical Modeling:
  • Multi-Ensemble Markov Models (MEMMs): Constructed using the Transition-based Reweighting Analysis Method (TRAM). This framework integrates data from both biased and unbiased ensembles to enforce local equilibrium and accurately estimate thermodynamics and kinetics, overcoming the sampling deficiencies of standard Markov State Models (MSMs).
• Deep Learning & Pathway Analysis:
  • TS-DAR: A neural network used to identify transition states between unfolded and pre-folded basins.
  • Variational Autoencoder (VAE) & Latent-space Path Clustering (LPC): Used to cluster thousands of transition pathways into distinct "folding channels" and identify representative kinetic routes.
• Experimental Validation:
  • Cell-Free Biosynthesis (CFB): Used to produce wild-type and mutant microcin J25 variants.
  • Competition Assays: Wild-type and variant precursors were co-expressed to compare lasso peptide production yields via MALDI-TOF-MS.
  • Structural Analysis: Circular Dichroism (CD) and NMR were attempted but yielded inconclusive results due to low signal intensity; CFB production served as the primary functional readout.

3. Key Contributions & Results

A. Universal Uphill Folding Landscape

• The study revealed a universal uphill free-energy profile for all 20 lasso peptides. The unfolded state is the global minimum, while the pre-folded (threaded) state is a high-energy, sparsely populated basin.
• Spontaneous folding probability is consistently low (< 0.8%) across all peptides. Microcin J25 showed the highest propensity (0.75%), followed by rubrivinodin (0.63%), while the other 18 peptides had probabilities below 0.2%.
• There is a strong negative correlation ($r = -0.81$) between the folding free energy ($\Delta G_f$) and the probability of the pre-folded state.

B. Loop Stability as a Critical Determinant

• $\beta$-Hairpin Formation: The stability of the loop region is the primary driver of folding efficiency. Peptides with stable $\beta$-hairpins in the loop (e.g., microcin J25) exhibit longer loop relaxation times and higher pre-folded populations.
• Experimental Validation: Mutants of microcin J25 predicted to enhance $\beta$-hairpin content showed increased production in CFB competition assays, while those predicted to disrupt it showed reduced production. This confirms that loop pre-organization facilitates the threading process.

C. Entropic Cost and Spatial Confinement

• Entropic Penalty: The transition from a flexible linear chain to a rigid threaded structure incurs a massive entropic cost ($-T\Delta S$), which is the dominant barrier to folding.
• Cyclase Role: Simulations mimicking the spatial confinement of the cyclase enzyme's active site (using spherical potentials) demonstrated that restricting conformational freedom significantly reduces the entropic penalty.
• Mechanism: The cyclase acts not just as a catalyst for ring closure but as a folding chaperone, stabilizing the pre-folded intermediate by limiting the accessible conformational space.

D. Folding Pathways and Kinetics

• Pathway Channels: Using LPC, the authors identified distinct folding channels. For microcin J25, the dominant channel (74% flux) involves early $\beta$-hairpin formation in the loop, followed by N-terminal turning and threading.
• Kinetic Bottlenecks: In less efficient peptides like klebsidin, the $\beta$-hairpin is less stable, leading to local unfolding and a significantly prolonged threading step (138 µs vs. ~30 µs for microcin J25).
• Sequence Specificity: Specific motifs (e.g., Pro-Gly) and rigidifying residues (e.g., Threonine methyl groups) were found to accelerate folding by stabilizing tight turns and reducing flexibility.

4. Significance

• Fundamental Insight: The study establishes that lasso peptide folding is intrinsically unfavorable in solution and universally requires enzymatic assistance (cyclase) to overcome entropic barriers.
• Rational Engineering: By identifying loop stability ($\beta$-hairpins) and entropic cost as the two governing principles, the authors provide a blueprint for the rational design of novel lasso peptides. Engineers can now predict folding efficiency by optimizing loop sequences for secondary structure formation.
• Methodological Advance: The successful application of TRAM-based MEMMs and VAE-based pathway clustering to a diverse set of peptides demonstrates a robust framework for studying rare-event biomolecular folding processes that are inaccessible to standard simulation techniques.
• Biological Implication: The findings clarify the dual role of lasso cyclases: catalyzing the chemical reaction (adenylation/macrolactamization) and acting as a chaperone to overcome the thermodynamic hurdles of the [1]rotaxane topology.