Intermolecular β-sheet formation guides the interaction between ubiquitin-like modifier FAT10 and adapter protein NUB1L

Using magic-angle spinning NMR spectroscopy, this study reveals that the adapter protein NUB1L binds to the N-domain of the ubiquitin-like modifier FAT10 to induce intermolecular β-sheet formation and stabilize an unfolded state, thereby acting as a holdase to facilitate FAT10-mediated proteasomal degradation during inflammation.

The Gist

The Big Picture: The Cell's "Recycling Bin"

Imagine your cell is a bustling city. Like any city, it produces trash (damaged or unnecessary proteins) that needs to be thrown away. The city has a specialized recycling plant called the 26S Proteasome.

Usually, to get something into this recycling plant, you have to tag it with a "trash can" sticker called Ubiquitin. But there's a special, emergency version of this sticker called FAT10. FAT10 is used during times of stress or inflammation (like when you have an infection). It doesn't just tag the trash; it grabs the trash and drags it straight to the shredder, ensuring it's destroyed immediately.

The Mystery: Why is FAT10 so "Loose"?

Scientists have known for a while that FAT10 is a bit weird. Unlike the standard Ubiquitin sticker, which is a tight, sturdy ball of string, FAT10 is floppy and loose. It melts at a low temperature and tends to tangle up.

For years, researchers suspected this "looseness" was actually a feature, not a bug. They thought FAT10 needed to be floppy to work, but they couldn't see how because it was too messy to study with traditional microscopes or standard imaging tools.

The Detective Work: Using "Magic Angle Spinning"

To solve this mystery, the researchers in this paper used a special technique called MAS NMR (Magic-Angle Spinning Nuclear Magnetic Resonance).

• The Analogy: Imagine trying to listen to a conversation in a crowded, noisy room. If everyone is standing still, it's chaotic. But if you put everyone on a giant, fast-spinning carousel tilted at a specific angle, the noise smooths out, and you can hear the conversation clearly.
• The Result: This technique allowed the scientists to see the atomic-level details of FAT10, even when it was messy or bound to other proteins.

The Plot Twist: The "Shapeshifting" Sticker

The team studied how FAT10 interacts with a helper protein called NUB1L. NUB1L acts like a "delivery driver" that picks up the FAT10-tagged trash and brings it to the recycling plant.

Here is what they discovered, which was a huge surprise:

1. Before the Meeting: When FAT10 is alone, it looks like a folded origami crane (a specific shape called a $\beta$-grasp fold). It has a little bit of a "tail" that is floppy and disordered.
2. The Meeting: When NUB1L grabs FAT10, something dramatic happens. NUB1L doesn't just hold the folded crane; it unfolds it.
3. The Transformation: NUB1L grabs that floppy tail of FAT10 and stretches it out into a straight, rigid line. It then snaps this line onto NUB1L's own body, creating a new, shared bridge between the two proteins. This is called an intermolecular $\beta$-sheet.
4. The Aftermath: Once NUB1L grabs that tail, the rest of the FAT10 "crane" falls apart. The tight folds dissolve, and the whole protein becomes a messy, floppy string again—except for the specific spot where NUB1L is holding it.

The "Holdase" Analogy

The authors propose that NUB1L acts as a "Holdase."

• Think of it like this: Imagine you have a tightly wound spring (the folded FAT10). To get it into a narrow tube (the recycling plant), you need to pull it straight. NUB1L is the hand that grabs the spring, pulls it straight, and holds it in that stretched-out, vulnerable state.
• Why do this? Because the recycling plant (the proteasome) can only eat things that are unfolded. If FAT10 stayed folded, the plant couldn't chew it up. By unfolding FAT10, NUB1L is essentially saying, "Okay, I've got you. I'm keeping you stretched out so the shredder can eat you and your cargo immediately."

Why This Matters

This discovery explains how the cell handles emergency waste disposal:

• Speed: FAT10 is designed for speed. It doesn't need a helper to unfold it; the helper (NUB1L) forces it to unfold the moment they meet.
• Safety: This process is "VCP-independent." Usually, cells need a motor protein (VCP) to force things open before shredding. FAT10 skips this step, making the process faster and more direct, which is crucial during inflammation.
• Cancer Connection: FAT10 is often overactive in cancers. Understanding exactly how it unfolds and interacts with NUB1L gives scientists a new target for drugs. If we can stop NUB1L from grabbing FAT10, we might be able to stop cancer cells from hiding their "trash" or, conversely, stop them from destroying tumor-suppressing proteins too quickly.

Summary

In short, this paper reveals that FAT10 is a shapeshifter. It starts as a neat, folded package, but when it meets its partner NUB1L, it gets ripped apart and stretched into a straight line. This "fuzzy" interaction ensures that the cell can rapidly destroy dangerous proteins during an immune response, acting like a high-speed delivery system that unfolds its cargo right before throwing it into the shredder.

Technical Summary

1. Problem Statement

The 26S proteasome is central to cellular proteostasis, typically degrading proteins tagged with polyubiquitin chains. However, the ubiquitin-like modifier FAT10 (Human Leukocyte Antigen F Adjacent Transcript 10) represents a unique pathway where it targets substrates for degradation without requiring polyubiquitin chains or the segregase VCP/p97.

• The Knowledge Gap: While FAT10 is known to be degraded along with its substrates and requires the adapter protein NUB1L (NEDD8-ultimate buster 1 Long isoform) for proteasomal activation, the atomic-level mechanism of this interaction was unknown.
• Structural Challenges: FAT10 is intrinsically prone to aggregation and has a low melting temperature (41°C). Previous structural studies (X-ray and liquid-state NMR) struggled with the N-domain of FAT10 (N-FAT10), often requiring N-terminal deletions or stabilizing mutations that altered its degradation kinetics. It was suspected that the "loose folding" of FAT10 is functionally critical, but the specific conformational changes upon binding to NUB1L had not been resolved.

2. Methodology

The authors employed Magic-Angle Spinning (MAS) NMR spectroscopy, a technique uniquely suited for studying large complexes, insoluble aggregates, and intrinsically disordered proteins (IDPs) in the solid state.

• Sample Preparation:
  • Proteins: They utilized a Cysteine-free mutant of the N-domain of FAT10 (N-FAT10-C0, residues 5–86) uniformly labeled with $^{13}$C and $^{15}$N. Natural abundance NUB1L was used as the binding partner.
  • States: Samples were prepared in three forms: (1) Microcrystalline N-FAT10-C0, (2) Lyophilized/rehydrated N-FAT10-C0 (near-physiological pH), and (3) The N-FAT10-C0/NUB1L complex formed via co-sedimentation.
• Spectroscopic Techniques:
  • Resonance Assignment: Extensive 2D and 3D $^{13}$C-detected MAS NMR experiments (DARR, ZF-TEDOR DARR, NCOCX) were used to sequentially assign residues in the isolated and complexed states.
  • Secondary Structure Prediction: Backbone torsion angles ($\Phi, \Psi$) were predicted using TALOS-N based on chemical shifts.
  • Interface Mapping: Double-REDOR (Rotational-Echo Double-Resonance) experiments were performed to distinguish residues at the protein-protein interface from those in disordered regions. This filter relies on the transfer of magnetization from natural abundance NUB1L protons to labeled N-FAT10 nuclei.
  • Computational Modeling: AlphaFold-Multimer was used to predict the complex structure, which was then validated against experimental NMR data.

3. Key Results

A. Structural Characterization of Isolated N-FAT10-C0

• The isolated N-domain adopts a canonical ubiquitin-like $\beta$-grasp fold.
• MAS NMR successfully assigned 69 of 83 residues in microcrystalline samples and 56 residues in lyophilized samples.
• Regions that remained unassigned (disordered) in the isolated state included the flexible N- and C-terminal tails and a specific stretch (I68–I74/L76) preceding the $\beta$5-strand.

B. Conformational Changes upon Binding to NUB1L

Upon complex formation with NUB1L, the NMR spectra of N-FAT10-C0 changed drastically:

• Signal Loss: The majority of resonances disappeared, indicating that most of the N-domain became disordered (dynamic or heterogeneous) and invisible to the solid-state NMR pulse sequences.
• Signal Retention: Only specific regions remained visible and assigned:
  • Residues I68–V81 (specifically I68–K72 forming a loop and T73–V81 forming a $\beta$-strand).
  • The N-terminus (G4) and a few scattered anchor residues (e.g., W17).
• Intermolecular $\beta$-Sheet: The chemical shifts of residues T73–V81 confirmed they form a $\beta$-strand. Combined with AlphaFold-Multimer predictions, this provided experimental evidence that N-FAT10 and NUB1L interact via an intermolecular, anti-parallel $\beta$-sheet.
  • The $\beta$-sheet consists of residues T73–V81 of N-FAT10 and Y212–N217 of NUB1L.
  • This is longer than the AlphaFold prediction (which suggested only L76–V81), indicating the experimental $\beta$-sheet extends three residues further.

C. The "Fuzzy Complex" Model

• Double-REDOR Validation: The retained signals in the complex were confirmed to be at the interface with NUB1L, as they showed incomplete dephasing (magnetization replenishment from NUB1L).
• Holdase Mechanism: The data supports a model where NUB1L acts as a holdase. It captures the intrinsically disordered region of N-FAT10 (specifically the $\beta$5-strand precursor), stabilizing it in an unfolded state via the intermolecular $\beta$-sheet.
• Disorder: The rest of the N-domain, which was folded in the isolated state, becomes disordered upon binding, except for the anchor points and the N-terminus.

4. Significance and Conclusion

• Mechanism of Degradation: The study reveals that FAT10 utilizes a "shapeshifting" mechanism. To be conjugated to substrates, FAT10 must be folded. However, to be degraded by the proteasome, the N-domain must be unfolded. NUB1L facilitates this transition by capturing the $\beta$-strand region, forcing the rest of the domain into a disordered state. This "unfolded" presentation is likely essential for the rapid, VCP-independent entry of FAT10 and its substrates into the proteasome.
• Methodological Advance: This work demonstrates the power of MAS NMR in characterizing fuzzy complexes (complexes with both structured and disordered regions) and transient interactions that are difficult to capture with crystallography or liquid-state NMR.
• Biological Implications: Understanding this interaction is critical for developing therapeutics targeting FAT10 dysregulation in cancer and inflammatory diseases. The study highlights that the "loose folding" of FAT10 is not a defect but a functional feature enabling its role in immune response and proteostasis.
• AlphaFold Limitations: The study notes that while AlphaFold-Multimer correctly predicted the intermolecular $\beta$-sheet, it failed to predict the extensive disorder and the specific anchor residues, underscoring the necessity of experimental data for intrinsically disordered systems.

In summary, the paper establishes that NUB1L acts as a molecular chaperone/holdase that stabilizes the N-domain of FAT10 in a largely unfolded, fuzzy complex via an intermolecular $\beta$-sheet, thereby priming the system for rapid proteasomal degradation.