Direct-to-Biology Enables Rapid Identification of Potent FBXO22 Degraders

Although a direct-to-biology screening approach using 175 compounds failed to identify FBXO22-dependent degraders for four target proteins, it successfully revealed potent FBXO22 self-degrading homo-PROTACs and CRBN- or VHL-mediated FBXO22 degraders.

The Gist

The Big Idea: The "Trash Can" Strategy

Imagine your body is a bustling city, and your cells are the buildings. Inside these buildings, there are millions of proteins doing jobs. Sometimes, a protein gets "stuck" in a broken state, or it starts acting like a villain (causing cancer). Usually, the cell has a built-in trash disposal system called the proteasome. It's like a giant industrial shredder that recycles old or broken proteins.

However, the shredder is picky. It only takes out trash that has a specific "recycling tag" (a ubiquitin tag) stuck on it. If a bad protein doesn't have that tag, the shredder ignores it, and the bad protein stays around to cause trouble.

PROTACs (the stars of this story) are like molecular double-sided tape. One side sticks to the bad protein, and the other side sticks to the "tagging machine" (an E3 ligase). By holding them together, the PROTAC forces the tagging machine to slap a recycling tag on the bad protein, tricking the shredder into destroying it.

The Problem: The "Linker" Lottery

To make a PROTAC, you need to connect the "bad protein sticker" and the "tagging machine sticker" with a string called a linker.

• If the string is too short, the two stickers can't reach each other.
• If it's too long, they flop around and can't hold tight.
• If it's the wrong shape, they won't snap together.

Traditionally, scientists had to make one PROTAC, test it, make another, test it, and so on. It was like trying to find the perfect key for a lock by making one key a day. It took forever and cost a fortune.

The Solution: "Direct-to-Biology" (The Speed Run)

This paper describes a new, super-fast way to find the perfect keys. The team used a method called Direct-to-Biology (D2B).

The Analogy: Imagine you are a chef trying to find the perfect spice blend for a soup.

• Old Way: You cook a small pot, taste it, wash the pot, cook the next one, taste it... this takes weeks.
• The D2B Way: You set up a giant tray with 175 tiny cups. In each cup, you mix a different spice combo. You don't even bother filtering out the lumps or measuring the exact grams perfectly. You just dump the whole messy mixture into the soup pot and taste it immediately. If a cup tastes amazing, then you go back and make a perfect, clean version of that specific spice blend.

In this study, the scientists synthesized 175 different PROTAC candidates in parallel, without purifying them first. They tested the "messy mixtures" directly in living cells to see if they could destroy their targets.

The Experiment: The "Swap Meet"

The researchers wanted to see if they could use a specific tagging machine called FBXO22 to destroy four different bad proteins (BRD4, BTK, CRBN, and VHL). They also wondered if the other tagging machines (CRBN and VHL) could destroy FBXO22.

They built two huge libraries of these "double-sided tapes" using different lengths of string (linkers) and different stickers.

The Results:

1. The Surprise: They failed to destroy the four bad proteins (BRD4, BTK, etc.) using the FBXO22 machine. It turns out FBXO22 is picky; it didn't want to tag those specific targets, no matter how they tried to connect them.
2. The Win: However, they found something even cooler. The "messy mixtures" contained compounds that acted as homo-PROTACs.
   • Analogy: Imagine a person who is supposed to be the trash collector, but they get so tangled up with their own rope that they accidentally tag themselves for recycling.
   • The team discovered compounds that made FBXO22 destroy itself (self-degradation).
   • They also found compounds where the CRBN and VHL machines grabbed FBXO22 and shredded it.

Why This Matters

1. Speed: They found potent drugs in a fraction of the time it usually takes. They identified the winners from a "dirty" library and only purified the winners. This is a huge time-saver.
2. New Tools: They created powerful tools to destroy FBXO22. Why do we want to destroy FBXO22? Because this protein is a "double agent." In some cancers, it helps the tumor grow; in others, it stops the tumor from spreading. Scientists need a way to turn FBXO22 "off" to study exactly what it does. These new compounds are the "off switches."
3. Validation: They proved that even "dirty" chemical mixtures work in cells. This means scientists can skip the expensive purification step for the initial search, saving money and time.

The Bottom Line

This paper is a victory for efficiency. The scientists tried a "shotgun approach" (making hundreds of messy mixtures at once) to find a needle in a haystack. They didn't find the needle they were originally looking for (destroying BRD4/BTK), but they found a better needle: a way to destroy the tagging machine itself (FBXO22) and prove that this fast, messy method works perfectly for drug discovery.

It's like trying to find a specific car in a parking lot, but instead of checking every car one by one, you spray paint all the cars that look promising. You then go back and wash the paint off the winners to get the perfect car. It's fast, it's messy, but it gets the job done.

Technical Summary

1. Problem Statement

• PROTAC Development Bottleneck: Proteolysis Targeting Chimeras (PROTACs) are heterobifunctional molecules that recruit E3 ubiquitin ligases to degrade target proteins. However, their development is hindered by the "linker problem." Rational optimization of the chemical linker connecting the target ligand and the E3 ligase is difficult and typically requires the synthesis, purification, and testing of hundreds of candidates, which is time-consuming and resource-intensive.
• FBXO22 as an E3 Ligase: While the E3 ligases CRBN and VHL are well-established for targeted protein degradation (TPD), the utility of the E3 ligase FBXO22 remains under-explored. FBXO22 can be recruited via a simple primary amine warhead (which bio-converts to an aldehyde to bind Cys326), but it is unknown if this mechanism can effectively degrade a broad range of targets (like BRD4 or BTK) or if FBXO22 itself can be targeted by other E3 ligases.
• Need for Rapid Screening: There is a critical need for methodologies that can rapidly explore chemical space (linker and warhead combinations) without the bottleneck of traditional purification before biological testing.

2. Methodology: Direct-to-Biology (D2B)

The authors employed a Direct-to-Biology (D2B) approach, which involves synthesizing compound libraries in parallel without purification and testing the crude reaction mixtures directly in cellular assays.

• Library Design:
  • Libraries 1 & 2: Generated 140 compounds by coupling commercially available ligands for BRD4, BTK, CRBN, and VHL with various primary alkyl amines (13–14 variants each) or amino acids.
  • Library 3: A focused library of 35 compounds designed to optimize the most active hits from the initial screens, including variations in CRBN ligand cores (replacing isoindoline-1,3-dione with 1-oxoisoindoline) and linker lengths.
  • Synthesis: Reactions were performed on a 10-micromole scale in 96-well plates using EDC/OxymaPure coupling agents. Crude products were treated with HCl to deprotect Boc groups, yielding primary amines.
• Screening Platform:
  • Cell Lines: Stable HEK293 cell lines were engineered to inducibly express HiBiT-tagged versions of the proteins of interest (POIs): BRD4, BTK, CRBN, VHL, and FBXO22.
  • Assay: The HiBiT-LgBiT system was used. When the POI is degraded, the HiBiT tag is lost, preventing the reconstitution of NanoLuc luciferase activity. This allows for real-time, quantitative measurement of protein abundance via luminescence.
  • Workflow: Crude libraries were screened at a theoretical 1 µM concentration. Hits were identified, resynthesized to >95% purity, and re-tested to confirm potency and rule out false positives from impurities.

3. Key Contributions

• Validation of D2B for PROTACs: Demonstrated that crude, unpurified PROTAC libraries can effectively identify potent degraders, with a 94% true-positive rate (16/17 resynthesized compounds validated).
• FBXO22 Recruitment Failure for Targets: Despite extensive screening, the study found no evidence that FBXO22 could degrade the target proteins BRD4, BTK, CRBN, or VHL using the tested linker geometries. This suggests FBXO22 recruitment via simple alkyl amines may not be universally applicable to all targets.
• Discovery of FBXO22 Degraders: The study successfully identified potent CRBN-mediated and VHL-mediated degraders of FBXO22 itself.
• Discovery of FBXO22 Homo-PROTACs: The research identified potent FBXO22 self-degraders (homo-PROTACs) where a single molecule recruits two FBXO22 molecules, leading to their mutual degradation.

4. Key Results

• Screening Outcomes:
  • Library 1 & 2: No degradation of BRD4, BTK, CRBN, or VHL was observed via FBXO22 recruitment.
  • Unexpected Hits: Several compounds designed to recruit FBXO22 to targets (e.g., L1A8, a BRD4-FBXO22 linker) actually induced the degradation of FBXO22 itself.
  • Potent Degraders:
    • CRBN-mediated FBXO22 degraders: Compounds L1C8 and L1D8 showed DC50 values of ~1.8 µM and ~0.25 µM (HiBiT assay), respectively.
    • VHL-mediated FBXO22 degraders: Compounds L2G8 and L2G12 showed DC50 values of ~0.04 µM and ~0.09 µM (HiBiT assay), respectively.
• Validation:
  • Purified compounds confirmed the degradation of endogenous FBXO22 via the proteasome pathway (blocked by MG132).
  • Cell viability assays confirmed that degradation was not due to non-specific cytotoxicity.
• Optimization (Library 3):
  • Refinement of the linker chemistry led to the identification of FBXO22 homo-PROTACs.
  • Compounds L3A6 and L3E6 (simple diamine linkers without a specific POI ligand) induced FBXO22 self-degradation with exceptional potency: DC50 of 8.77 nM and 1.75 nM, respectively.
  • These compounds function by oxidizing to dialdehydes that cross-link two FBXO22 molecules.

5. Significance

• Accelerated Discovery: The study proves that D2B workflows can compress the timeline for PROTAC discovery from months to weeks, significantly reducing resource consumption (using <100 mg of building blocks).
• New Chemical Tools: The identified CRBN- and VHL-mediated FBXO22 degraders provide essential chemical tools to study the dual role of FBXO22 in cancer (where it can act as both an oncogene and a tumor suppressor).
• Mechanistic Insight: The discovery of potent FBXO22 homo-PROTACs reveals that simple alkyl diamines can induce self-degradation of the E3 ligase, a phenomenon with implications for understanding E3 ligase regulation and potential off-target effects in TPD campaigns.
• Strategic Limitation: The failure to degrade BRD4/BTK via FBXO22 highlights that while FBXO22 is a viable E3 ligase, its recruitment geometry is highly specific and may not be as broadly applicable as CRBN or VHL without more extensive linker optimization.