One Receptor, Two Opposite Effects: How PAR1 Switches Between Protecting and Harming Blood Vessels

University of California San Diego

A protein sitting on the surface of your blood vessel cells can do two completely opposite things. Cut it one way, and it protects the vessel wall, reducing inflammation and maintaining structural integrity. Cut it another way, and it triggers inflammation and vascular leakage. Same protein. Opposite outcomes. Until now, nobody understood how that was possible.

The dual life of PAR1

Protease-activated receptor-1 (PAR1) sits on the surface of endothelial cells, the thin layer lining every blood vessel in the body. Unlike most receptors that bind a separate signaling molecule, PAR1 is activated by being physically cut by enzymes called proteases. Two different proteases can do the cutting, and they produce opposite results.

One protease triggers a harmful response: inflammation, increased vascular permeability, and the kind of leakage that contributes to sepsis, heart attack damage, and stroke injury. The other triggers a protective response that counteracts inflammation and helps maintain the blood vessel barrier. Both responses come from the same receptor molecule, which raised a puzzling question: how can a single protein produce two contradictory signals?

Same enzyme, different address

The answer, according to a team led by JoAnn Trejo at UC San Diego School of Medicine, comes down to cellular geography. Both the protective and harmful PAR1 responses are orchestrated by the same intermediate enzyme, called GRK5. But the response depends entirely on where GRK5 is located within the cell when it acts.

For PAR1 to trigger its protective response, GRK5 must be anchored to the cell's plasma membrane, the outer barrier separating the cell from its surroundings. For PAR1 to trigger the harmful inflammatory response, GRK5 can act from the cytoplasm, the fluid interior of the cell.

The distinction is elegant in its simplicity. The same enzyme, performing the same biochemical reaction, produces opposite downstream effects based purely on its physical position within the cell. This spatial control mechanism provides a molecular explanation for how a single receptor manages to send dramatically different messages.

AlphaFold 3 reveals the structural basis

The team used AlphaFold 3, the Nobel Prize-winning AI protein structure prediction tool, to understand why the two different protease cuts lead to different GRK5 behaviors. The structural models revealed that different cuts to the extracellular portion of PAR1, the part exposed outside the cell, produce distinct conformational changes that propagate through the membrane-spanning region and alter how the receptor's intracellular domain interacts with GRK5.

In other words, the specific location of the cut on the outside of PAR1 determines its shape on the inside, which in turn determines whether it recruits GRK5 from the membrane or from the cytoplasm, which finally determines whether the cell mounts a protective or inflammatory response.

Therapeutic implications for vascular disease

"This opens the door to therapies that could harness the protective response of PAR1 without potentially triggering the opposite response," said co-corresponding author Irina Kufareva, professor at UC San Diego's Skaggs School of Pharmacy and Pharmaceutical Sciences.

The clinical relevance spans multiple conditions. In sepsis, uncontrolled vascular inflammation and leakage are primary drivers of organ failure and death. In heart attack and stroke, the inflammatory cascade following the initial event causes additional damage to tissue that might otherwise recover. A drug that selectively activates PAR1's protective pathway while leaving the inflammatory pathway dormant could theoretically address all of these.

Current anticoagulant therapies that target PAR1 block both pathways indiscriminately. The new findings suggest it might be possible to design more selective compounds that specifically promote the protective conformation, perhaps by mimicking the way the beneficial protease cuts the receptor, or by anchoring GRK5 to the membrane.

What this study does not establish

The research identifies the molecular mechanism in cell-based experiments but does not demonstrate a therapeutic application. Moving from understanding how a receptor switches between two states to designing a drug that reliably biases it toward one state is a substantial scientific and engineering challenge. The AlphaFold 3 models provide structural hypotheses, but computational predictions require experimental confirmation of the specific interactions involved.

The study was conducted primarily in endothelial cell cultures and structural modeling. How the spatial regulation of GRK5 operates in the complex environment of intact blood vessels, where cells experience blood flow, interact with immune cells, and respond to multiple simultaneous signals, adds layers of complexity not captured in the current work.

The balance between PAR1's protective and harmful pathways may also differ across vascular beds, disease states, and patient populations. A therapeutic strategy that works in one context might not translate directly to another.

Building on decades of PAR1 research

PAR1 was first identified as a thrombin receptor in the early 1990s. Its role in blood clotting made it an early drug target, leading to the development of the antiplatelet agent vorapaxar. But PAR1's dual signaling capacity has complicated therapeutic development, because blocking the receptor entirely eliminates both beneficial and harmful effects.

The current study provides the first detailed mechanistic explanation for how the two pathways diverge, offering a roadmap for more selective intervention. Whether that roadmap leads to viable therapies will depend on follow-up work in animal models and eventually clinical trials.