An exploration of technologies for targeted protein modification

  • Luke Martin Simpson

Student thesis: Doctoral ThesisDoctor of Philosophy


Targeted protein degradation (TPD) is a promising strategy for both therapeutics and research, providing new opportunities to overcome some of the current limitations associated with traditional small-molecule therapeutics, RNA interference (RNAi) and genome editing approaches. Proteolysis-targeting chimeras (PROTACs) are typically heterobifunctional molecules which bring a protein of interest (POI) into spatial proximity of an E3 ubiquitin ligase, triggering POI ubiquitylation and proteasomal degradation. The Sapkota lab previously developed an affinity-directed protein missile (AdPROM) system for targeted proteolysis by employing specific polypeptide binders of intracellular POIs conjugated to an E3 ligase or its recruiter. The key advantage of the AdPROM system is that it can rapidly inform the utility of degrading a POI through a specific E3 ligase, so that key E3 ligase and POI pairs can be prioritised for the more resource-intensive development of PROTACs.

To develop a ligand-inducible AdPROM (L-AdPROM) system, I combined AdPROM with Halo-tag/VHL-recruiting PROTAC (HaloPROTAC) technology. By expressing an L-AdPROM construct consisting of an anti-GFP nanobody conjugated to the Halo-tag, robust degradation of GFP-tagged POIs was achieved only upon treatment of cells with the HaloPROTAC. For GFP-tagged POIs, ULK1, FAM83D, and SGK3 were knocked in with a GFP-tag using CRISPR/Cas9 genome editing technology. Targeted GFP-tagged POI degradation using the L-AdPROM system led to loss of protein function. By substituting the anti-GFP nanobody for a monobody that binds H- and K-RAS, robust degradation of unmodified endogenous RAS proteins was observed only in the presence of the HaloPROTAC. Therefore, the L-AdPROM technology is versatile and adaptable, where, in principle, the small polypeptide binder can be substituted for any high-affinity POI-targeting binder.

Through the replacement of the anti-GFP nanobody component of the HaloPROTAC L-AdPROM system for an anti-mCh nanobody, it was possible to promote the inducible proteasomal degradation of individual mCh-tagged CK1 isoforms, specifically CK1α, CK1δ and CK1ε. Although the CK1 family of Ser/Thr protein kinases are considered constitutively active and promiscuous in nature, CK1 isoform-specific functions and regulation remain poorly understood. To expand our toolkit for studying CK1 isoform-specific biology in cells, CK1α, CK1δ and CK1ε homozygous N-termini mCh knockin cell lines were generated using CRISPR/Cas9 technology. By expressing an L-AdPROM construct consisting of an anti-mCh nanobody conjugated to the Halo-tag, robust degradation of a specific mCh-tagged CK1 isoform was achieved only upon treatment of cells with the HaloPROTAC. Following HaloPROTAC-mediated CK1 isoform degradation, global phospho-proteomic analyses will be employed as an unbiased approach to identify novel isoform-specific CK1 substrates.

PROTACs redirect and consequently rely on endogenous cellular machinery for target protein degradation. Therefore, differences between E3s, including expression and activity, and changes in the cellular environment, due to POI localisation for example, can potentially impact PROTAC efficacy for target protein degradation. To interrogate whether the nature of the recruited E3 ligase and the subcellular context of the POI impact PROTAC-mediated degradation, I designed and expressed either Halo or FKBP12F36V (dTAG) constructs consisting of varying localisation signals and tested the efficacy of their degradation by VHLor CRBN-recruiting PROTACs targeting either Halo or dTAG. Halo protein localised to the nucleus, cytosol, outer mitochondrial membrane (OMM), endoplasmic reticulum (ER), peroxisome and lysosome, but not the Golgi, was amenable to degradation with the Halo/VHL-recruiting PROTAC, HaloPROTACE. In contrast, only nuclear-, cytosol- and ER-localised FKBP12F36V was amenable to degradation with the FKBP12F36V/VHL-recruiting PROTAC, dTAGVHL, whilst OMM-, Golgi-, peroxisome- and lysosome-localised targets were less amenable. With the FKBP12F36V/CRBN-recruiting PROTAC, dTAG-13, FKBP12F36V localised to the nucleus, cytosol, ER, and OMM were amenable to degradation, while Golgi-, peroxisome- and lysosome-localised FKBP12F36V were less so. Given that differentially localised Halo and FKBP12F36V proteins display different levels of degradation by the same respective PROTACs, it is evident that the subcellular context of the POI can influence the efficacy of PROTAC-mediated POI degradation.

Finally, through the substitution of the AdPROM E3-recruiting moiety for a promiscuous phosphatase, I sought to test whether artificially recruiting a phosphatase to a POI could mediate targeted POI dephosphorylation. Reversible phosphorylation of proteins, catalysed by protein kinases and phosphatases, is a fundamental process that controls protein function and is responsible for finetuning intracellular signalling. Failure of phospho-control often accounts for many human diseases. Therefore, modulating the phosphorylation status of POIs by redirecting kinases and phosphatases has the potential to alter POI function in a desired manner. By employing the phosphatase catalytic subunits PPP1CA and PPP2CA conjugated to an anti-GFP nanobody, termed the affinity-directed phosphatase (AdPhosphatase) system, PPP1CA or PPP2CA was recruited to a GFP-tagged POI to mediate POI-GFP dephosphorylation. AdPhosphatase-mediated FAM83D-GFP dephosphorylation in mitosis inhibited the phosphorylation-mediated proteasomal degradation of FAM83D-GFP, whilst AdPhosphatase-mediated GFP-ULK1 dephosphorylation attenuated starvation-induced autophagy and DFP-induced mitophagy. Through simple substitution of the targeting enzyme of interest, AdPROM/AdPhosphatase technology could potentially be further modified to induce any desired POI post-translational modulation. Such approaches could rapidly inform the roles of specific posttranslational modifications on protein function and could be exploited for investigating different drug discovery approaches.
Date of Award2022
Original languageEnglish
SupervisorGopal Sapkota (Supervisor) & Ian Ganley (Supervisor)

Cite this