Merck and Princeton scientists create method to map cell-surface microenvironments

An interdisciplinary team of scientists from Merck & Co. and Princeton University have devised a new method for making better maps of protein interactions on cell surfaces. The technique, dubbed MicroMapping, allows researchers to stick molecular tags on proteins within a tight radius, about 4 nm, around a protein of interest. Those molecular tags allow researchers to easily identify a target’s protein neighbors.

Merck wanted to develop a technique to study microenvironments on the surfaces of immune cells and cancer cells. Scientists think understanding these microenvironments could yield new drug targets for cancer immunotherapy, or reveal clues for how to make existing immunotherapies, such as Merck’s multibillion dollar checkpoint blocker Keytruda, work better for larger numbers of people with cancer. Mapping protein interactomes—the networks of interacting proteins—is key to this goal.

“The toolkit for identifying protein interactomes in cells is woefully inadequate,” says Lyn Jones, chief scientist of the Center for Protein Degradation at the Dana-Farber Cancer Institute. Jones was taken aback when he saw Merck present parts of the work at a chemical biology conference in Boston in September 2019. Now, after seeing the new publication, he calls MicroMapping “a very elegant solution.”

“We’ve all been anxiously waiting for the paper to come out,” says Christopher am Ende, a chemist from Pfizer who was also at the Boston meeting. “We can definitely envision using something like this.”

The project arose from what Princeton chemist David W. C. MacMillan calls a “disruptive chemistry meeting” at Merck, a brainstorming session that gathered biologists and chemists into the same room to hash out challenges in drug discovery. During the exercise, MacMillan heard Olugbeminiyi O. Fadeyi and Rob C. Oslund, a chemist and chemical biologist at the Merck Exploratory Science Center in Cambridge, Massachusetts, describe the challenge of mapping protein interactomes on cells.

Existing protein mapping methods attach enzymes to a protein of interest to generate reactive compounds that tag neighboring proteins. But these methods are “too messy,” Oslund says. “They label too much and are hard to control.” Sometimes the molecular tags themselves diffuse far away from the protein. As a result, these methods provide only fuzzy spatial information, the cellular equivalent of knowing whether two individuals are in the same country. Oslund and Fadeyi wanted to devise a technique that would identify a protein’s next-door neighbors.

About 4 months later, one of MacMillan’s graduate students, Stefan McCarver, stumbled on an obscure paper from 1980 that gave him and MacMillan an idea: they could use a small molecule photocatalyst, activated by blue light, to make molecular tags instead of an enzyme.

The Princeton and Merck teams synthesized and tested different iridium-containing photocatalysts, which they then attached to antibodies. Those photocatalyst-studded antibodies target a second antibody that binds a cell surface protein of interest. The photocatalysts act like antennae, capturing energy from blue light and transferring it to nearby compounds containing a diazirine group through a physical process called Dexter energy transfer. The energy kicks out a pair of nitrogen atoms and transforms the diazirine into a highly reactive carbene group, which will immediately react with any protein nearby.

By attaching a molecular tag like biotin to the diazirine, researchers can easily isolate the tagged proteins and identify them with standard mass spectrometry approaches. The final result is a neighborhood-level list of proteins that potentially interact with each other (Science 2020, DOI: 10.1126/science.aay4106).

This new MicroMapping system gives researchers two extra levels of precision compared to older techniques. First, the length of the reaction is easily controlled by turning the blue light on and off. Second, carbene has a short half-life of about 1 ns. “It is one of organic chemistry’s most reactive molecules,” MacMillan says. Carbenes will diffuse only about 4 nm away from the photocatalyst before attaching to a neighboring protein or reacting with water, which deactivates them, he adds.

The Merck team says their MicroMapping technique will immediately be useful for identifying new targets for drug discovery. As a demonstration, the team used MicroMapping to find a list of proteins that neighbor PD-L1, a well-studied protein that cancer cells use to evade detection from the immune system. The group also showed that MicroMapping can be used to study protein interactions at the interface between two different immune cells: a B cell and a T cell. They expect the technique will give new clues for drug discovery in immunology, neuroscience, and other fields where cell-to-cell interactions are important.

John Tallarico, US head of chemical biology and therapeutics at the Novartis Institutes for BioMedical Research, says that existing enzyme-based methods are “cumbersome and blunt instruments” compared to Merck’s MicroMapping. He’s especially excited about the potential to map protein interactomes between different immune cells, such as macrophages and dendritic cells.

MicroMapping doesn’t prove that neighboring proteins actually have important interactions with each other. It just gives biologists and drug discoverers a shortlist of clues to investigate. Another limitation is that MicroMapping requires adding two antibodies to a protein of interest, and the resulting system may be too large to fit between some cells. The Merck team presented data at the Boston chemical biology conference last fall that suggests that this challenge can be overcome by using a small molecule ligand in place of the antibody to deliver the photocatalyst to a protein of interest. That work is not published yet.

Although both groups didn’t want to discuss that new work on the record, they made it clear that other groups have been excited about the technique before even the first MicroMapping paper was published last week. “I am getting approached by all of these companies who’ve heard about it through the grapevine,” MacMillan says. And academic scientists have reached out too, looking to strike collaborations in fields as diverse as oncology, neuroscience, and HIV biology.

MacMillan says that the new MicroMapping paper is just the first example of what’s become a big focus in his lab. He had only two people on the MicroMapping project when it started, but now 14 people in his 40-person lab are working on additional MicroMapping approaches. He expects that number to grow to 20 by the end of the year.