PROTACs: A Practical Guide

A proteolysis targeting chimera (PROTAC) is a bifunctional compound designed to trigger degradation of a protein of interest. PROTACs are comprised of two active domains—one targeting an E3 ubiquitin ligase and the other targeting the desired protein—joined by a linker molecule. PROTACs are often used to target proteins that were previously undruggable when using traditional inhibitor approaches. Because of this, PROTACs hold much promise in the therapeutic space and are allowing researchers to explore basic biological questions they weren’t able to before.

Below is a practical guide to designing and evaluating the efficacy of a PROTAC, as well as some troubleshooting tips for when things don’t go as planned.

Considerations before designing PROTACs

Before considering PROTAC design, Benedict Cross, Ph.D., Chief Technology Officer at PhoreMost, advises researchers to take a step back and ask, “Is there a need or a rationale for a degrader to be developed?” Cross says that, over the past few years, he has seen some people designing PROTACs just for the sake of it. “Many of the first wave clinical programs are looking to degrade targets for which traditional enzymatic inhibitors already exist,” he says. “This can be important to overcome some weaknesses in those molecules but is arguably not capitalizing on the benefits of the degrader mechanism of action.”

Kristin Riching, Ph.D., Senior Research Scientist at Promega, starts by thoroughly investigating the native protein target biology in terms of expression, localization, degradation, and turnover rate, as well as the expected phenotype after the protein is degraded. She adds that understanding the expression, localization, and tissue distribution of the E3 ligase is also essential. Breanna Zerfas, Ph.D., Research Scientist at Dana-Farber Cancer Institute, echoes a similar sentiment. “It’s crucial to pick ligases that actually exist in the disease-relevant system or the types of cells being targeted,” she says.

PROTAC design

Leveraging structural data

When beginning to design a PROTAC, Riching leverages structural information gained from crystallographic or biophysical approaches combined with computational modeling. Zerfas says she also typically starts with computational modeling that will provide information on the best way to perform the ternary complex between the protein of interest and the E3 ligase. “But that is of course dependent on there being either a crystal structure or a cryo-EM structure of your protein of interest with the ligands that you're interested in turning into a PROTAC,” she adds. If there isn’t a structure to work with, Zerfas begins by very broadly designing a linker. From there, she says, you can use a basic set of compounds that will cover a lot of chemical space to begin homing in on what a potential PROTAC would look like.

Picking a target

Inga Shchelik, Ph.D., a Postdoctoral Fellow at Yale University, says the ideal targets for PROTACs are disease-driving proteins that are underrepresented or not expressed in healthy cells, and that the easiest-to-target proteins have known ligands that bind to them. Zerfas says that, typically, the best PROTAC targets are ones that inhibitors haven't been successful with in the past. “That’s because those proteins are so well studied that you know they're disease relevant,” she says. Cross notes that, when choosing the right target for degradation, it’s crucial to first understand the relationship between the degradation mechanism of action and the resulting phenotype. “Large-scale loss-of-function data provides a strong starting point to coordinate protein-of-interest with a degrader modality, but it does not tell the whole story,” he says, noting that kinetics, resistance, and selectivity all need to be interrogated. He suggests coupling-based tools like dTag, HaloTag, and DHFR and using high-content stability assay readouts, for example kinetic-enabled microscopy.

The importance of linker design

According to Zerfas, one of the most important aspects of linker design is the distance it creates between the E3 ligase and the protein of interest. “For some proteins you want that linker to be really short, because you need some protein-protein interaction across the two proteins in order to get good ubiquitination,” she says. But in the case of larger complexes, a longer linker may be essential to prevent clashing of the proteins that would interfere with a ternary complex forming. Shchelik agrees that the linker should be optimized for length, but also flexibility. “You can estimate the right linker by computer analysis when the ligase-protein of interest complex is available, but sometimes it just takes screening,” she explains.

Riching says that, in addition to stability, linkers can influence cellular permeability and solubility, which can impact PROTAC efficacy. “Though ‘linkerology’ is an area of continual study, early PROTACs are often initially synthesized with generic polyethylene glycol (PEG)-based linkers of varying length to identify hits,” says Riching. From there, they’re further refined. At PhoreMost, Cross says that they simulate linker requirements using their SITESEEKER^® screening platform, “so we can enter the drug design phase knowing the crucial properties of the linker—like length and flexibility—and how that might change on an E3-by-E3 basis,” he says.

Is your PROTAC working?

To evaluate efficacy, Zerfas starts with a biochemical assay—typically using purified proteins and doing time-resolved fluorescence energy transfer (TR-FRET) to check binding efficiency. These assays are high-throughput, allowing for a huge number of compounds to be tested at once. When Zerfas finds a good protein-protein interaction with TR-FRET, she moves on to western blotting. “Always follow it up with a western blot to make sure there is actually protein degradation,” she says, “because you can get good association of your two proteins, but that doesn't mean it'll be functional and lead to degradation.” Riching adds that, although western blotting is useful, she opts for a more quantitative, sensitive method that involves tagging endogenous proteins via CRISPR with HiBiT and then monitors the loss of luminescent signal over time. Shchelik says proteomics is another valuable method, where proteins are isotope-labeled and then degradation can be evaluated by mass spectrometry.

Riching explains that if weak degradation or no degradation is observed she goes back upstream to evaluate what else could have gone wrong, looking at potential issues with cellular permeability or ubiquitination, for example. Cross and Shchelik both say a good place to look is the E3 ligase. “If everything else is looking good but there is no degradation then this is where the exploration of additional E3 ligases becomes so important,” says Cross. “This can be a shortcut to masses of medicinal chemistry and drug design if we can just identify a more productive ternary complex.”