Skip to main content
Premium Trial:

Request an Annual Quote

ShanghaiTech Team Develops New Proximity Labeling Approach


NEW YORK (GenomeWeb) – Researchers from ShanghaiTech University have developed a new proximity-labeling technique for studying protein-protein interactions.

In a study published this week in Nature Methods, they used the approach to identify interaction partners of the protein CD28, a membrane receptor expressed on the surface of T cells that is involved in T cell activation.

Proximity labeling typically uses a target protein to tag other nearby proteins with a molecule that allows them to be extracted from a cell and analyzed. The method is increasingly popular with researchers using it to, for instance, identify the protein components of specific subcellular compartments or identify interaction partners of proteins that are difficult to isolate via conventional immunoprecipitation mass spec experiments.

Among the most common proximity-labeling approaches are APEX and BirA*. The former uses engineered ascorbic acid peroxidase (APEX) tags genetically inserted in into proteins of interest. Upon stimulation with hydrogen peroxide, this tag releases biotin-phenoxyl radicals that tag nearby proteins in the cell, and these tagged proteins can then be pulled out of the sample using streptavidin-based enrichment and analyzed using mass spec.

BirA* labeling works much the same way, using a biotin ligase inserted into proteins of interest to then tag nearby molecules, allowing them to be pulled down.

In their work, the ShanghaiTech researchers presented a new proximity-labeling system, which they called PUP-IT, for pupylation-based interaction tagging. In this approach, proteins of interest are genetically fused with the pafA gene, which codes for the Pup ligase PafA. When exposed to ATP, PafA ligates a second protein, Pup(E), to lysines on both the target protein and nearby proteins that may be interactors.

Min Zhuang, a ShanghaiTech researcher and senior author on the study, said the idea for the labeling approach came out of work she was doing as a grad student and post-doc studying ubiquitin systems in bacteria. She noticed that ubiquitin modifications were highly specific to lysine residue but could be non-specific in terms of the specific lysine residues and substrates they modified.

"In some cases it can be quite promiscuous in terms of what substrate can be labeled as long as the lysine is there at the right moment and in the right position," she said. "So I though maybe this could be used in a way similar to the APEX method."

One potential advantage of the PUP-IT method is that all components of the labeling system can be inserted genetically, so researchers don't have to add reagents to their cell cultures, Zhuang said, noting that the addition of hydrogen peroxide required by APEX labeling leads to oxidative stress, which could alter the behavior of the cell being studied.

Additionally, she said that it could in some cases be challenging to introduce these molecules into the parts of a tissue or cell being studied at the desired concentration. The fact that PUP-IT components can all be added genetically could also make the approach suitable for use in live organisms, the authors wrote.

On the other hand, APEX labeling occurs much more quickly than PUP-IT, which allows researchers to look at protein interactions with greater temporal resolution. Additionally, the PUP-IT approach requires researchers to insert genes for two proteins, PafA and Pup(E) into the system of interest.

The authors also noted that Pup(E) is larger than the molecules used for labeling in the APEX and BirA* approaches and that this makes it poorly suited to studies of protein-protein interactions within organelles because it can't enter those compartments. Additionally, the large size of the PafA compared to BirA* or APEX could change the function or localization of the target protein, a question that the authors wrote "needs to be examined in each case."

To explore whether the PUP-IT system could work for proximity labeling, Zhuang and her colleagues first searched previous mass spec studies identifying Pup-modified proteins to see if there was a specific amino acid sequence required for the modification to occur, as this could limit the substrates the system could label. They found no such sequence and determined that a lysine residue alone was sufficient for a protein to be labeled.

Additionally, they tested whether the PUP-IT labeling was proximity-based by mixing PafA with free GST and finding that while PafA was self-modified, the free GST had low levels of modification. The researchers also created two different forms of Pup(E), one fused with carboxylase to allow biotinylation in cells and another fused with biotin, which they then used for the rest of the study.

To test whether the system could identify weak protein-protein interactions, they applied PUP-IT to the MATH domain of the protein SPOP, which has known weak interactions with the peptides pep1, pep2, and pep3. They found the approach detected these interactions in both in vitro and in vivo experiments.

They then used PUP-IT to identify interaction partners of CD28, transfecting Jurkat cells with PafA-fused CD28 and biotin-tagged Pup and then using streptavidin pulldown to extract Pup-labeled proteins. Analyzing these proteins by mass spec, they identified roughly 200 proteins as potential CD28 interactors.

In a follow-up experiment, they identified 41 proteins that were highly enriched. That group included a number of known CD28 interactors and 33 of the 41 of were "well connected to the CD28 signaling pathway" based on a gene ontology analysis, the authors wrote.

Zhuang and her colleagues also determined that they could use the approach to identify cells containing CD28 ligands like CD86 and CD80 that are expressed on antigen-presenting cells that interacted with Jurkat cells with PafA-fused CD28.

Zhaung said her ShanghaiTech colleague Haopeng Wang, an immunologist and co-author on the paper, is doing research to further validate the CD28 interactors identified in the study. She said her lab is now interested in using the technique to study the interactomes of different transmembrane proteins.

The researchers have filed a patent application covering the approach.