Positive Selection Screen Identifies Natural Product β-Catenin Inactivators

Through a large-scale positive selection screen of natural product mixtures using mammalian cells, researchers identified a compound that inactivates oncogenic β-catenin by activating novel protein kinase Cs to relocate the protein to juxtamembrane vacuolar structures, demonstrating the potential of phenotypic screening to target previously undruggable cancer drivers.

The Gist

The Big Problem: The "Undruggable" Villain

Imagine your body is a bustling city. Inside the cells of this city, there are workers called proteins that keep everything running. Sometimes, a worker gets a "glitch" in their software and goes rogue. This rogue worker is called β-catenin (or β-cat).

When β-cat goes rogue, it acts like a hyperactive construction foreman. It tells the cell to build too many new buildings (cells), leading to tumors and cancer. For decades, scientists have tried to build a "stop sign" (a drug) to tell this foreman to stop. But the problem is that β-cat is shaped like a smooth, round marble. It has no cracks, no handles, and no pockets for a drug to grab onto. Scientists call these "undruggable" targets because it's like trying to stop a greased pig with a pair of tweezers.

The Detective's Trick: The "Up" Assay

Instead of trying to find a drug that fits into a hole (which doesn't exist), the scientists at Dana-Farber Cancer Institute decided to play a different game. They set up a survival trap.

1. The Bait: They took cells and gave them a special "suicide switch" (a protein called DCK*). This switch is harmless on its own.
2. The Poison: They added a harmless-looking chemical (BVdU) to the cells. Normally, this chemical is just a snack. But if the suicide switch is present, the switch turns the snack into a deadly poison.
3. The Connection: They glued the rogue foreman (β-cat) to the suicide switch.
   • If β-cat is present: The switch is active. The poison is made. The cell dies.
   • If β-cat is gone or stopped: The switch is inactive. The poison is never made. The cell survives.

The Goal: They wanted to find a natural product (a plant or fungus extract) that would kill the poison-maker (the cell with the active switch) but leave the safe cells alone. If a cell survived the poison, it meant the drug had successfully stopped the rogue foreman.

The Search: Digging Through Nature's Library

Instead of testing synthetic chemicals made in a lab, the scientists decided to dig through nature's library. They tested over 326,000 samples of natural products—extracts from plants, fungi, and marine life from around the world.

Think of this like searching for a needle in a haystack, but the haystack is made of 300,000 different types of hay, and the needle is a specific chemical that stops a specific protein.

To make the search easier, they didn't just test the whole plant extract (the "hay"). They broke the extracts down into smaller, sorted piles (fractions). This helped them avoid the "toxic trash" (compounds that kill cells just because they are poisonous, not because they stop the specific target) and concentrate on the "good stuff."

The Discovery: The "Vacuum Cleaner" Effect

Out of the hundreds of thousands of samples, they found a few winners. One specific lineage of a plant (from the Euphorbiaceae family, which includes spurge) caught their eye.

Inside this plant, they found a tiny molecule they called Compound 2.

Here is the magic trick Compound 2 pulled off:

• It didn't destroy the foreman: Usually, scientists hope a drug will break the protein into pieces (degrade it). Compound 2 didn't do that. In fact, it made more of the protein appear!
• It moved the foreman: Instead of destroying the protein, Compound 2 acted like a vacuum cleaner or a magnet. It grabbed the rogue foreman (β-cat) and dragged him out of the main office (the nucleus) and shoved him into a storage closet near the door (a vacuole at the edge of the cell).
• The Result: Even though the foreman was still there, he was locked in the closet. He couldn't shout orders to the construction crew anymore. The cell stopped building tumors.

The Mechanism: The "Key" and the "Lock"

How did Compound 2 do this? The scientists discovered it was a Protein Kinase C (PKC) activator.

• The Analogy: Imagine PKC is a security guard who usually sleeps on the job. Compound 2 is a very specific alarm clock that wakes up only the "Novel" security guards (a specific type of PKC), not the "Classical" ones.
• The Chain Reaction: When these specific guards wake up, they start phosphorylating (tagging) the rogue foreman. This tag acts like a "Do Not Disturb" sign that forces the foreman to leave the office and go to the storage closet.
• The Shape Matters: The scientists found that Compound 2 has a tiny "handle" (a primary alcohol group) that fits perfectly into the guard's hand. If you cover that handle with a cap (acetylation), the guard can't grab it, and the drug stops working. This explains why a very similar chemical (Compound 3) didn't work.

Why This Matters

1. New Way to Fight Cancer: This proves you don't need to break a protein to stop it. You can just move it to a place where it can't do any harm.
2. Nature is a Goldmine: Even though big pharmaceutical companies stopped looking at natural products years ago, this study shows that nature still holds secrets that synthetic labs can't easily replicate.
3. The "Undruggable" is Druggable: They found a way to stop a target that everyone thought was impossible to hit, simply by changing the rules of the game.

In a nutshell: The scientists built a trap to find a natural substance that could sneak up on a cancer-causing protein, grab it, and lock it in a closet, effectively neutralizing it without destroying it. They found this "key" in a plant extract, proving that nature still has the best blueprints for new medicines.

Technical Summary

1. Problem Statement

• Undruggable Targets: Many genetically validated cancer targets, including the transcription factor $\beta$-catenin ($\beta$-cat), are considered "undruggable" because they lack deep hydrophobic pockets for small-molecule binding.
• Limitations of Current Approaches: While progress has been made with K-Ras inhibitors and "molecular glues" (e.g., IMiDs) that induce degradation, robust methods to degrade or inactivate $\beta$-cat remain elusive. Existing inhibitors often lack specificity or fail to degrade the protein.
• Screening Challenges: Traditional "down assays" (measuring loss of signal) are prone to false positives from general cytotoxicity. Furthermore, screening natural product mixtures is difficult due to the presence of toxic compounds and the complexity of isolating active principles from crude extracts.

2. Methodology

The authors developed a robust, mechanism-agnostic cell-based positive selection ("up") assay to identify compounds that inactivate $\beta$-cat.

• Reporter System:
  • Engineered 293FT cells (which are $\beta$-cat independent) to express a fusion protein: Oncogenic $\beta$-cat (S37C) fused to a modified Deoxycytidine Kinase (DCK)*.
  • Mechanism: DCK* converts the non-natural nucleoside BVdU into a toxic metabolite. Cells expressing the fusion protein die in the presence of BVdU.
  • Selection Logic: Compounds that degrade $\beta$-cat or inactivate the fusion protein reduce DCK* levels, allowing cells to survive BVdU treatment.
  • Controls: The system includes a GFP reporter (via IRES) to sort viable cells and a TK1 (Thymidine Kinase 1) knockout background. TK1 competes with BVdU; knocking it out sensitizes cells to BVdU, enhancing the assay's dynamic range.
• Screening Library:
  • Screened 326,304 samples from the NCI Program for Natural Product Discovery (NPNPD).
  • Included 40,744 crude extracts (F0) and 285,560 prefractionated subfractions (F1–F7).
  • Used prefractionation to sequester toxic compounds and concentrate minor active agents.
• Hit Validation & Deconvolution:
  • Primary Screen: Identified hits based on cell survival (GFP+ objects) in $\beta$-cat-DCK* cells vs. control DCK* cells.
  • Secondary Screening: Counterscreened against DCK*-IKZF1 to filter out general transcription/translation inhibitors.
  • Western Blotting: Validated hits by measuring $\beta$-cat-DCK* abundance and Axin2 (a $\beta$-cat target gene) levels.
  • Chemical Isolation: Isolated pure compounds from active lineages using HPLC and characterized them via NMR and Mass Spectrometry.
  • Mechanistic Studies: Used immunofluorescence, RNA-seq, kinase assays, microscale thermophoresis (MST), and molecular docking to determine the mode of action.

3. Key Contributions

• Novel Screening Platform: Demonstrated the efficacy of a positive selection "up assay" using DCK* fusions for identifying protein inactivators/degraders, minimizing false positives from general toxicity.
• Natural Product Strategy: Successfully applied prefractionated natural product libraries to overcome the historical challenges of screening complex mixtures, yielding a high rate of validated hits.
• Discovery of a New Mechanism: Identified a class of natural products that inactivate $\beta$-cat not by degradation, but by relocalization via novel Protein Kinase C (PKC) activation.
• Specific Chemical Probe: Isolated and characterized Compound 2 (Lathyranol), a lathyrane natural product with a unique C19 primary alcohol, as a selective activator of novel PKCs.

4. Key Results

• Screening Outcomes:
  • From 326,304 samples, 353 hits (0.11% hit rate) were confirmed.
  • Hits were categorized into two groups:
    1. "Downregulators" (24 lineages): Reduced $\beta$-cat abundance (e.g., via degradation). One hit was a neolignan mixture, but isolation was difficult.
    2. "Inactivators" (12 lineages): Increased $\beta$-cat abundance but rendered it inactive (cells survived BVdU). These were the focus of the study.
• Characterization of Inactivators (Lineage L90865):
  • Isolated two compounds: Compound 2 (Lathyranol) and Compound 3 (Lathyranol-19-acetate).
  • Compound 2 (with a free primary alcohol at C19) was active; Compound 3 (acetylated) was inactive.
  • Phenotype: Compound 2 caused $\beta$-cat to relocalize from the nucleus/cytoplasm to juxtamembrane vacuolar structures, rendering it transcriptionally inactive (Axin2 downregulation).
• Mechanism of Action:
  • PKC Activation: Compound 2 specifically activates novel PKCs (PKC$\delta$, PKC$\theta$) but spares classical PKCs (PKC$\alpha$).
  • Evidence:
    • Induced phosphorylation of PKC substrates and PKC$\delta$ activation markers.
    • Required the Serine 715 residue on $\beta$-cat (S715A mutation abolished the effect).
    • Binding: MST showed Compound 2 binds PKC$\delta$ with ~20x higher affinity than PKC$\alpha$.
    • Molecular Modeling: The C19 hydroxyl group forms a critical hydrogen bond network in the C1b domain of PKC$\delta$, stabilizing the active conformation. The acetylated Compound 3 cannot form this bond.
  • Selectivity: Unlike PMA or Ingenol-3-angelate (which activate both classical and novel PKCs and are highly toxic), Compound 2 selectively targets novel PKCs, avoiding the toxicity associated with classical PKC activation.
• Genetic Validation: Constitutively active PKC$\delta$ and PKC$\theta$ phenocopied Compound 2's effects, while classical PKC$\alpha$ inactivation did not block the effect, confirming the role of novel PKCs.

5. Significance

• Therapeutic Potential: Provides a novel path to inactivate oncogenic $\beta$-cat, a target previously deemed difficult to drug. The mechanism (relocalization via novel PKC activation) offers a new therapeutic strategy distinct from proteasomal degradation.
• Drug Development: Compound 2 serves as a superior starting point for developing PKC agonists for cancer therapy compared to existing tools (PMA/ING) because of its selectivity for novel PKCs and lower toxicity profile.
• Methodological Impact: Validates the power of phenotypic positive selection screens combined with prefractionated natural products to discover complex biological mechanisms and high-quality chemical probes. It highlights that natural products remain a vital source of structurally diverse, bioactive molecules that synthetic libraries often miss.
• Biological Insight: Reinforces the emerging view that PKC activation can act as a tumor suppressor in certain contexts (e.g., colorectal cancer) by inhibiting Wnt/$\beta$-cat signaling, challenging the historical view of PKCs solely as oncogenic drivers.