I’ve been reading through research lab websites lately, trying to understand what these labs actually do. Not the abstract jargon. The real stuff. This is the first one: the Hui Lab at UCSD.
The problem in one sentence
Your immune system has T cells — think of them as soldiers that hunt down threats like infections and cancer. But soldiers need brakes. Without brakes, they’d attack your own body. That’s what immune checkpoints are: molecular brakes on T cells.
The problem? Cancer figured out how to exploit those brakes.
Tumors hijack the checkpoint system and basically slam the brakes on your T cells. The soldiers are right there, ready to fight, but the tumor is whispering “stand down.” And the T cells listen.
This is why checkpoint blockade immunotherapy works — drugs like Keytruda block those brakes so the T cells can attack again. But here’s the catch: it only works in a fraction of patients. The Hui Lab wants to understand why.
What they actually study
Three things. Let me break them down.
1. How do the brakes actually work inside the cell?
When a checkpoint receptor like PD-1 gets triggered, it sends a signal inside the T cell to shut things down. For years, the field thought this signal mostly ran through two enzymes called Shp1 and Shp2.
The Hui Lab showed that’s not the whole story. PD-1 still works even when you delete both Shp1 and Shp2. There are other mechanisms nobody’s fully understood yet.
Their most recent finding is wild: when PD-1 activates Shp2, the Shp2 proteins don’t just individually go do their job. They clump together into liquid droplets — like oil in water — through a process called liquid-liquid phase separation (LLPS). These droplets act as organizing hubs that pull in the T cell’s signaling machinery and shut it down.
This is a fundamentally different way to think about cell signaling. It’s not just enzyme meets substrate. It’s enzymes forming little blobs that reorganize the entire neighborhood.
2. Receptors talk to each other on the same cell
Here’s something the field largely missed: immune receptors and their ligands don’t just interact between cells (your T cell talks to the tumor cell). They also interact on the SAME cell surface — what biologists call “cis-interactions.”
The Hui Lab has found several of these:
- PD-L1 and CD80 interact in cis, changing how each one functions
- B7 and CD28 interact in cis at specific membrane structures
- CTLA-4 depletes B7 ligands it grabs from other cells (a process called trogocytosis)
These cis-interactions add a whole new regulatory layer that conventional immunology overlooked. It’s like discovering that the soldiers aren’t just listening to commands from headquarters — they’re also whispering to each other, and those whispers change everything.
3. The mouse model problem
Most of what we know about immune checkpoints comes from mouse models. The Hui Lab showed something uncomfortable: human PD-1 is significantly more inhibitory than mouse PD-1. It binds its ligands more strongly and engages Shp2 more effectively.
They found a conserved motif present in vertebrate PD-1 that’s actually ABSENT in rodents. When they swapped the mouse PD-1 intracellular domain for the human version, T cell behavior changed dramatically — and so did response to anti-PD-1 therapy.
Translation: the primary preclinical model the field relies on has a fundamentally weaker version of the drug target. Mouse models may systematically underestimate how PD-1 works in humans.
Why this matters
If you’re a patient with cancer and your doctor is considering immunotherapy, all three of these findings matter:
- Understanding the phase separation mechanism could lead to drugs that target the Shp2 condensates, not just the receptor.
- Cis-interactions mean the same drug might behave differently depending on what other receptors are sitting next to its target on the same cell.
- The mouse-human gap means we’ve been testing drugs in a system that doesn’t fully recapitulate the human biology. That could explain why some therapies that looked great in mice fail in people.
The Hui Lab is small but punches above its weight. Their work lands in Immunity, Science Immunology, Nature Immunology, JEM, and Cell. That’s the top tier.
The bottom line
Cancer immunotherapy works by releasing the brakes on your immune system. But we don’t fully understand how those brakes work, how they talk to each other, or whether our animal models even have the same brakes we do. The Hui Lab is systematically closing those gaps — one molecular mechanism at a time.