PROTACS: Upgrading Small Molecule Inhibitors

Proteolysis-targeting chimeras (PROTACs) are target protein degraders that — rather than inhibit a single function of a target protein — cause its complete degradation. Unlike like small molecule inhibitors, they can bind to proteins without active sites, creating the potential to target proteins known to cause disease but previously thought to be undruggable. They also are effective at lower doses and have the potential to be more selective, with fewer off-target effects. Early clinical results are promising, and the outlook for PROTACs is exciting, leading to both growing interest and investment.

Designed for Special Delivery

One of the main mechanisms of action (MoAs) of small molecule drugs is binding to the active site of a protein and preventing it from performing a specific function involved in the disease pathway. This approach has been effective for the treatment of many indications but is limited to known proteins with identifiable binding sites. Many more proteins do not fit into this category than do, with the majority considered undruggable using this specific approach.

Target protein degradation (TPD) is an alternative strategy garnering growing interest. As the name implies, this approach involves degradation of an entire target protein through natural processes rather than binding to the active site to inhibit protein activity. The most advanced technology relies on proteolysis-targeting chimeras (PROTACs), which are bifunctional molecules with two distinct groups connected by a linker. The first group attaches to the target protein, and the second group binds to an E3 ubiquitin ligase, which tags the protein with ubiquitin molecules (ubiquitination), marking it for degradation by the 26S proteasome protease complex into short amino acid sequences.^1,2

The MoA of PROTACs is referred to as an event-driven MoA, involving “hijacking” the endogenous ubiquitin–proteasome system (UPS).² The role of the UPS is to eliminate misfolded, damaged, and overabundant proteins, but disease-causing proteins often evade it.³ PROTACs bring those disease-causing proteins (proteins of interest (POIs)) and E3 ubiquitin ligases in close proximity so that transfer of ubiquitin molecules to the POI can occur (at lysine residues), leading to degradation of malfunctioning proteins that otherwise would not be destroyed by the UPS.^3,4

The generalizable nature of the PROTAC MoA means this strategy can be applied for the treatment of many indications, including cancer, autoimmune disorders, and neurodegenerative diseases, among others. Indeed, PROTACs can theoretically be used to target any intracellular or transmembrane protein, including those lacking classic binding sites, including mutant and oncogenic proteins.⁴ In many cancers, in fact, E3 ligases are overexpressed and could therefore be ideal for use in TPD treatment strategies.

There is much still to be understood about what makes an ideal PROTAC-based therapy for a given indication, however. Early evidence does suggest that the formation of an appropriate ternary complex between the POI, the PROTAC, and the E3 ligase is essential to success and not always predicted by bifunctional complex formation (PROTAC to POI and PROTAC to E3-ligase).⁵

Many different chemical groups have been investigated as protein ligands within PROTACs, including existing small molecule inhibitors and newly designed compounds. As with the development of new small molecule inhibitors, ligands are designed by identifying a potential binding region and creating molecules with structural features and functional groups that enable some level of affinity to the region. High-throughput screening and computational methods are employed in this effort. Ligands binding E3 ligases are developed in a similar manner.

Alkenyl oxindoles have been recently reported as effective protein ligands for use in PROTACs.⁶

New Mechanism of Action with Many Advantages

Only a fraction of all proteins that could potentially be targeted for treatment of disease are considered to be druggable targets. In fact, of the approximately 3,000 proteins that have been identified as disease-related, most do not have active binding sites that would allow small molecule inhibitors to bind.⁷ Examples include nonenzymatic proteins, such as scaffolding proteins, transcriptional factors, and RAS oncoproteins associated with lung, colorectal, and pancreatic cancers, among others.⁴

PROTACs do not need to bind to proteins with high affinity because the MoA is based on bringing the E3 ligase close to the POI, not inhibition of the POI’s function. They thus expand the landscape of druggable proteins — and offer many advantages over traditional small molecule inhibitors.^2,3,7,8 Because PROTACs do not stay bound to the POI for an extended period and are released from a POI following its degradation, they are able to bind to other POIs, allowing for lower doses and thus reduced negative side effects. The lack of long-term protein binding also avoids potential and problematic buildup of the POI.

In addition, the formation of the ternary complex through dual binding and selective recognition of POIs at the whole protein level by E3 ligases affords increased specificity for PROTACs, enabling better tissue targeting with reduced off-target toxicity.^2,8 Furthermore, because PROTACs degrade entire proteins rather than just inhibit their enzymatic function, all potential undesired activity (e.g., structural activity) is eliminated, and the development of drug resistance through POI overexpression and mutation is avoided.^2,4

Wide Variety of Protein Targets

Since the PROTAC concept was first developed in the early 2000s, PROTAC-based molecules have been developed to target many different classes of proteins, including targeting nuclear receptors, kinases, G protein–coupled receptors (GPCRs), transmembrane proteins, small GTPases, epigenetic proteins, transcription factors, and protein aggregates.

Optimal “PROTACable” proteins are changed from their natural state due to overexpression, mutation, aggregation, isoform expression or localization that results in disease development, have a region that can be bound by a ligand and allow access by an E3 ligase, and have an unstructured region accessible by the proteasome.⁹ In addition, because the proteasome resides within the cell, cell-surface and extracellular proteins are not considered optimal POIs for PROTACs.

A few of the proteins for which PROTACs have been reported include TK, BRD4, AR, ER, STAT3, IRAK4, and tau. A more comprehensive list (including proteins reported through 2022) can be found in Table 1.⁷

Potential Applications in Many Indications

PROTACs in development target a wide variety of diseases. The majority degrade known cancer therapeutic targets or serve as immunotherapies, but there is also significant interest in the use of the TPD approach to degrade tau α-synuclein, poly-glutamine-expanded mHTT, and other proteins associated with neurodegenerative disorders, such as Alzheimer’s, Parkinson’s, and Huntington’s diseases; IRAK4 and other proteins that play important roles in autoimmune diseases, such as rheumatoid arthritis; proteins involved in challenging viral infections, including hepatitis B and C and COVID-19; and proteins implicated in metabolic diseases.^2,9 A list of different PROTAC targets and the diseases they are associated with can be found in Table 2.⁷

Table 2. Diseases and associated PROTAC targets⁷

Many Approaches to Increasing Target Specificity

While the design of PROTACs makes then inherently more specific than general small molecule inhibitors, they can in some cases bind to and promote degradation of proteins similar to the target POI. Researchers have therefore investigated several approaches for improving the target specificity of PROTACs, including photochemically controllable PROTACs (PHOTACs), hypoxia-activated PROTACs, folate-caged PROTACs, antibody–PROTAC conjugates (Ab-PROTACs), and aptamer–PROTAC conjugates (APCs).²

PHOTACs contain linkers that only adopt the necessary conformation for brining both end groups in proximity when exposed to some form of irradiation (often ultraviolet light) or are created as prodrugs bound in a “cage” that is only degraded when exposed to irradiation (such as X-rays). Radiation is applied only to the target area, leading to spatiotemporal control with increased targeting and reduced side effects.^2,7

Hypoxia-activated and folate-caged PROTACs work to increase the targeting ability of anti-cancer PROTACs due to the hypoxic conditions in solid tumors and the high expression of the folate receptor 1 (FOLR1) in many human malignancies, respectively.²

Ab-PROTACs are analogous to antibody–drug conjugates, leveraging the ability of antibodies to bind to specific antigens. In this case, however, they are delivering not a cytotoxic payload but a ligand for an E3 ligase. In addition, using an antibody as the protein ligand enables targeting of not only intracellular proteins but extracellular and membrane proteins as well.² For instance, antibodies that recognize specific cancer cell membrane receptors have been used to facilitate entry of PROTACs into these target cells.⁴

One downside of Ab-PROTACs is their larger size. Another option is to use small peptides that contain or mimic the ligand recognition sites of larger antibodies. Stapled peptides are of particular interest owing to their greater stability compared with unmodified peptides.⁴ PROTACs containing peptides are referred to by some as bioPROTACs and can also include bifunctional peptides, fusion proteins, and bispecific antibody mimics.⁹

Aptamers are small, single-stranded nucleic acids that bind selectively to proteins and have the advantages of being easy to synthesize, stable, and non-immunogenic. When used in APCs, they increase the targeting ability of PROTACs and have been demonstrated to improve the anticancer ability of traditional PROTACs.

Yet another approach to increasing the targeting of PROTACs involves the design of PROTACs that target multiple proteins simultaneously.⁴ PROTACs have also been formulated as nanoparticles to enable both active and passive delivery into target cells. RNA–PROTACs target RNA-binding proteins, while oligonucleotide-based PROTACs (O’PROTACs) and transcription factor–targeting chimeras (TRAFTACs) target transcription factors involved in different cancers, obesity, and cardiovascular and neurological diseases.⁹

Precursor molecules that self-assemble into PROTACs within cells have been explored as a means of increasing target specificity and overcoming the poor solubility/permeability/bioavailability of some larger PROTACs. CLIPTACs (in-cell click-formed proteolysis-targeting chimeras) form in the cell via click reactions and have been shown to induce protein degradation in different cell lines in vitro.⁷

The Importance of the Linker

As with ADCs, the linkers in PROTACs play an important role in determining the effectiveness of these novel therapies. The linker does more than connect the ligands for the POI and the E3 ligase enzyme; it also helps determine the physicochemical properties of the PROTAC7 and its solution conformation, which is key to bringing the E3 ligase and the POI together in the appropriate spatial arrangement.⁹

Several aspects of the linker must be considered when designing PROTACs, including its length, chemical composition, level of flexibility, and attachment sites.⁷ Linkers that are too short prevent ligands from binding to their targets, while linkers that are too long do not lead to effective ternary complex formation. Commonly used linkers for PROTACs include polyethylene glycol derivatives and alkyl chains (saturated and unsaturated) of different lengths. Plain alkyl chains can reduce cell permeability, however, and thus heteroatoms, such as oxygen and nitrogen, are often incorporated to improve their hydrophilicity.

In addition, because linear linkers often have too much flexibility, aromatic groups are introduced to help promote formation of more stable ternary complexes. This strategy can also increase the solubility and cell permeability of PROTACs.

Looking for New E3 Ligases

While there are approximately 600 known E3 ligases, very few of these enzymes have been recruited for PROTAC-mediated ubiquitination of target proteins. Examples include CRBN (cereblon), VHL (Von-Hippel-Lindau), MDM2 (mouse double minute 2), IAPs (inhibitors of apoptosis), and DCAF (DDB1 and CUL4 associated factor) 15 and 16.⁹

There are several different subclasses of E3 ligases defined on the basis of the mechanism by which they cause ubiquitination.⁹ RING ligases are the most prevalent and act by simultaneously binding to the POI and an E2 ligase, which transfers the ubiquitin to a POI lysine residue. HECT ligases bind the POI, then transfer ubiquitin onto themselves before transferring it to the POI. Not all E3 ligases perform the same with different POIs, and thus the choice of E3 ligase can directly impact PROTAC efficacy. They can also impact tissue and cell-type specificity.⁹

There is, therefore, significant interest in identifying/development other E3 ligases deployment in PROTAC therapeutics.⁹ For instance, Kelch-like family member 40 (KLHL40) and KLHL41 have been shown to target skeletal muscle, while RNF182 and tripartite motif-containing protein 9 (TRIM9) have been shown to prefer neurons and could be useful for the development of PROTACs targeting disorders of the central nervous system. Others have been found to exist preferentially in tumor cells, often at high expression levels, such as cell division cycle 20 (CDC20), cytosolic iron–sulfur assembly component 1 (CIAO1), WD repeat–containing protein 82 (WDR82), and the MAGE (melanoma antigen genes) family of ubiquitin E3 ligases. Many others are under investigation.

Early Days, but Many PRROTACS Now in the Clinic

The first PROTACs (ARV-110 and ARV-471 from Arvinas Therapeutics) entered human clinical trials in 2019 as oncology treatments. They have been sufficiently successful to reach phase II.³ Many other candidates from startups and big pharma companies alike are now at the preclinical and early-phase clinical stages. In the first half of 2022 alone, 14 new companies were developing TPDs, and 88 new PROTAC candidates were annnounced.¹ Candidates that have advanced to clinical trials include therapies for cancer (more than 20), autoimmune disorders, and neurodegenerative diseases.^1,4 Many have received orphan drug designation. Companies at the forefront include Bristol Myers Squibb (BMS), Nurix Therapeutics, Kymera Therapeutics, Dialectic Therapeutics, and Foghorn Therapeutics.⁹ Table 3 lists selected PROTAC candidates and their stages of clinical devleopment.⁷

Table 3. Selected PROTAC candidates in or close to entering the clinic⁷

Not Without Challenges

While PROTACs provide an entirely new MoA and therefore hold great promise for treating many different diseases, there are challenges to their development. On a fundamental level, there is a need to learn more about which protein targets will be most effectively degraded by PROTACs, and their rational design is not a simple matter.⁷

One of the biggest challenges to the advance of PROTACs is to design these TPDs so that they have sufficient solubility, cell permeability, and bioavailability combined with high target specificity to yield highly efficacious and safe new therapeutics.² While off-target effects and drug resistance development are expected to be reduced compared with traditional small molecule inhibitors, these issues can arise with PROTACs.

Choosing the right protein ligand, designing the optimum linker, selecting the best E3 ligase to recruit, and designing the best ligand of that ligase to afford PROTACs with desirable physicochemical and functional properties requires a wide range of knowledge and expertise and can be a lengthy process.² Indeed, greater understanding is needed regarding how PROTACs enter cells and are dispersed and metabolized, the 3D structures of E3 ligases and POIs, and the tissue-specific expression, localization, half-lives, and ubiquitination capacities of different E3 ligases.^2,4

Nonspecific delivery of PROTACs resulting from an inability to penetrate target cells/tissues can result in undesired side effects.⁴ For cancer therapies targeting overexpressed oncogenic proteins, there is also the potential to induce secondary malignancies. Mutation of ubiquitination genes, meanwhile, can lead to resistance to PROTACs owing to reduced expression of ubiquitination proteins and/or overexpression of deubiquitination enzymes.

PROTACS also present analytical challenges.³ Their pharmacokinetic and pharmacodynamic properties can be difficult to assess. Practical issues relate to the propensity of PROTACs to bind to glass and plastic surfaces, which can impact analytical results.

Exciting Outlook

Despite these challenges and the fact that long-term safety, tolerability, and efficacy data have yet to be generated, early clinical results are promising and the outlook for PROTACs is exciting, leading to both growing interest and investment.¹ In just two decades, the field has evolved from peptide-based PROTACs leveraging a very limited number of E3 ligases to those based on fully synthetic small molecules that recruit a much wider array of ubiquitin-recruiting enzymes.⁴ The advances achieved to date have allowed for demonstration of proof of concept for PROTACS while also revealing the limited information gathered to date. The latter fact represents tremendous opportunities to identify means for improving PROTACs, widening the portfolio of protein targets, and potentially further expanding PROTAC strategies.

Several More TPD Technologies

PROTACs are not the only target protein therapeutics under development today. Others of note include molecular glues, lysosome-targeting chimeras (LYTACs), and autophagy-targeting chimera (AUTACs).

Unlike PROTACs, with their two functional groups connected by a linker, molecular glues comprise a single bifunctional molecule that can bind to both the POI and the ligase recruiter to form the ternary complex.⁵ Molecular glues were first discovered when it was learned that thalidomide acts as a TPD.⁹ Since then, purposeful development of molecular glues has been pursued, with examples including small molecules and multi-specific antibodies, among other motifs. It is also worth noting that molecular glues can work by mechanisms other than protein degradation.

By employing the endosome–lysosome route, LYTACs enable the degradation of extracellular and membrane proteins, which play roles in many diseases.² As with PROTACs, LYTACs form a ternary complex with the POI and a lysosome-targeting receptor (TLR) on the cell surface, which causes degradation of the POI via clathrin-mediated endocytosis. They consist of a small molecule or antibody ligand for the POI and a synthetic ligand for the TLR. LYTACs have been developed that target EGFR and PD-L1 proteins.

AUTACs, meanwhile, use 8-nitro-cyclic guanosine monophosphate (cGMP) as a degradation tag linked to a small molecule ligand for the POI.² While PROTACs induce K48-linked polyubiquitination and proteasome-mediated degradation, AUTACs induce K63-linked polyubiquitination and lysosome-mediated degradation.

Beyond the TPDs highlighted here, there are numerous others under investigation, including many that involve modifications of the initial PROTAC strategy.

References

1. Renza, Gregory and Yinglu Zhang. “A pivotal year for PROTACs?” RBC Capital Markets Insights. 8 Feb. 2023.
2. Sincere, Nuwayo Ishimwe, Krishnan Anand, Sumel Ashique, and Chongge You. “PROTACs: Emerging Targeted Protein Degradation Approaches for Advanced Druggable Strategies.” Molecules. 28:4014 (2023).
3. Collins, Romesh. “PROTACs, A Promising Class of Novel Therapeutic Modalities.” Charles River Eureka. 25 May 2023.
4. Rutherford, Kailee A. and Kirk J. McManus. “PROTACs: Current and Future Potential as a Precision Medicine Strategy to Combat Cancer.” Mol. Cancer Ther. 23: 454–463 (2024).
5. Lowe, Derek. “Demystifying PROTACs.” Science. 29 Jun. 2023.
6. Wang Y et al. “Alkenyl oxindole is a novel PROTAC moiety that recruits the CRL4DCAF11 E3 ubiquitin ligase complex for targeted protein degradation.” PLoS Biol. 22: e3002550 (2024). do
7. Liu, Zi et al. “An overview of PROTACs: a promising drug discovery paradigm.” Mol. Biomed. 3: 46 (2022).
8. Casan, Joshua M. L. and John F. Seymour. “Degraders upgraded: the rise of PROTACs in hematological malignancies." Blood. 143: 1218–1230 (2024).
9. Békés, M., D.R. Langley, and C.M. Crews. “PROTAC targeted protein degraders: the past is prologue.” Nat. Rev. Drug Discov. 21: 181–200 (2022).