Event-driven, not occupancy-driven
A classic inhibitor works by occupancy: as long as the drug is sitting in the protein's pocket, the protein is blocked. Remove the drug and the protein wakes up. A targeted protein degrader works differently — by event. It doesn't just block the protein; it tags it for destruction, and once the protein is gone, the drug is free to go tag another. The mechanism of action is catalytic, so a small amount of drug can remove a lot of protein.
Why does this matter? Because you no longer need a deep, functional pocket to inhibit. You only need any place to grab the protein so you can flag it for disposal. That opens the door to targets long considered undruggable — scaffolding proteins, transcription factors, proteins whose harm comes from just being present rather than from an enzyme activity you could block.
Two flavours: PROTACs and molecular glues
A PROTAC is a dumbbell: one end binds the target protein, the other end binds the E3 ligase, joined by a linker. It is a deliberate, modular design — but the price is size. PROTACs are big for a small molecule, well beyond the comfortable rules of oral drug-likeness, so getting them into cells and making them orally available is a genuine struggle. A molecular glue (or molecular glue degrader) is smaller and sneakier: a compact molecule that subtly reshapes a protein surface so that the target and the E3 ligase suddenly stick to each other — gluing two proteins that wouldn't normally meet.
Glues keep the convenience of an ordinary small molecule — small, often orally available — but they are harder to design on purpose, since you must induce a brand-new protein–protein interaction rather than just stitch two known binders together. The field's holy grail is rational glue design; much of today's success still comes from a mix of careful chemistry and serendipity.
PROTAC (dumbbell): [ target binder ]---linker---[ E3 ligase binder ]
| |
TARGET ~~~ pulled close ~~~ E3 LIGASE -> ubiquitin tag -> proteasome
Molecular glue: [ small compact molecule ]
reshapes surface so TARGET + E3 LIGASE stick directly
-> ubiquitin tag -> proteasome
Key contrast:
Inhibitor = occupancy (block while bound; stoichiometric)
Degrader = event (destroy, then move on; catalytic)Choosing among modalities: a final map
Pull the whole track together. The honest way to pick a modality is to lay the question against each tool's strengths and costs — there is no universally 'best' one.
- Small molecule — best when the target has a good pocket and you want an oral, cheap, cell-penetrant drug; the default first choice.
- Antibody / biologic — best for surface or circulating targets needing extreme selectivity or immune recruitment; injected, costly, long-lived.
- Peptide — for protein–protein interfaces that need more reach than a small molecule, accepting delivery and stability work.
- Oligonucleotide / mRNA — when you want to turn a gene's protein down (oligo) or up (mRNA); programmable design, but delivery (often liver-limited) dominates.
- Degrader (PROTAC / glue) — when you want to remove an intracellular, pocket-poor, or 'undruggable' protein entirely, catalytically.