Targeted degradation of extracellular proteins: state of the art and diversity of degrader designs

Abstract

Selective elimination of proteins associated with the pathogenesis of diseases is an emerging therapeutic modality with distinct advantages over traditional inhibitor-based approaches. This strategy, called targeted protein degradation (TPD), is based on hijacking the cellular proteolytic machinery using chimeric degrader molecules that physically link the target protein of interest with the degradation effectors. The TPD era began with the development of PROteolysis TAtrgeting Chimeras (PROTACs) in 2001, with various methods and applications currently available. Classical PROTAC molecules are heterobifunctional chimeras linking target proteins with E3 ubiquitin ligases. This induced interaction leads to the ubiquitylation of the target protein, which is needed for its recognition and subsequent degradation by the cellular proteasomes. However, this technology is limited to intracellular proteins since the effectors involved (E3 ubiquitin ligases and proteasomes) are located in the cytosol. The related methods for selective destruction of proteins present in the extracellular space have only emerged recently and are collectively termed extracellular TPD (eTPD). The prototypic eTPD technology utilizes LYsosomal TArgeting Chimeras (LYTACs) that link extracellular target proteins (secreted or membrane-associated) to lysosome-targeting receptors (LTRs) on the cell surface. The resulting complex is then internalized by endocytosis and trafficked to lysosomes, where the target protein is degraded. The successful elimination of various extracellular proteins via LYTACs and related approaches has been reported, including several important targets in oncology that drive tumor growth and dissemination. This review summarizes current progress in the eTPD field and focuses primarily on the respective technological developments. It discusses the design principles and diversity of degrader molecules and the landscape of available targets and effectors that can be employed for eTPD. Finally, it emphasizes current open questions, challenges, and perspectives of this technological platform to promote the expansion of the eTPD toolkit and further development of its therapeutic applications.

Introduction

The term targeted protein degradation (TPD) encompasses a group of innovative approaches that enable the selective removal of unwanted proteins via designed degrader molecules. These technologies leverage natural cellular degradation pathways: proteasomal degradation (primarily utilized to eliminate intracellular proteins) and lysosomal degradation (also used to eliminate extracellular proteins; Fig. 1). TPD relies on proximity-induced pharmacology and greatly expands the potential for drug development against previously undruggable targets [1].

Fig. 1

The general principles of TPD approaches for intracellular and extracellular proteins. A Scheme of iTPD with a PROTAC degrader. The PROTAC molecule induces the proximity of the intracellular target protein (POI) and the E3 ubiquitin ligase. The latter catalyzes the attachment of a polyubiquitin chain to the POI (normally with the help of E2 ubiquitin-conjugating enzymes), forming the recognition signal for the proteasomes that ultimately degrade the POI. B Scheme of eTPD with a LYTAC degrader. The LYTAC molecule links the POI with the LTR on the cell surface, which induces the internalization of the formed complex and sorting to lysosomes, where the POI is finally degraded

Historically, the TPD concept was introduced in 2001 when a team of researchers headed by Craig Crews and Raymond Deshaies employed a synthetic chimera molecule that linked the methionyl aminopeptidase 2 (METAP2) target protein to E3 ubiquitin ligase for ubiquitylation and subsequent proteasomal degradation [2]. That degrader molecule was named a PROteolysis-TArgeting Chimera (PROTAC) and represented the first example of successful intracellular TPD (iTPD; Fig. 1A). Since then, the PROTAC field has greatly expanded, with degraders readily available for a wide range of targets, particularly proteins involved in the pathogenesis of severe diseases, most notably cancers.

A specific database of PROTAC degraders (PROTAC-DB, http://cadd.zju.edu.cn/protacdb/) listed 6111 molecules linking 442 target proteins to 20 different E3 ubiquitin ligases at the end of 2024 [3]. Notably, some of these molecules have already advanced to clinical trials, primarily for potential applications in oncology. Eighteen proteasome-targeting protein degraders had already entered early clinical trial stages by early 2023, with one entering stage III in 2022 [4]. To illustrate the rapid progress in the field, just one year later, at the beginning of 2024, the number of degraders in phase III clinical trials has risen to eight, with approximately 40 in earlier clinical trial stages, including classical PROTACs, molecular glues, and a related but mechanistically distinct class of selective estrogen receptor degraders (SERDs) [5]. The therapeutic applications are predominantly focused on malignancies, with blood cancers being the most targeted, followed by breast and prostate cancers, with the respective targets being estrogen and androgen receptors. Currently, the US Food and Drug Administration (FDA) has approved three degrader drugs to treat multiple myeloma and other hematological disorders, all belonging to the molecular glues class: thalidomide and its derivatives lenalidomide and pomalidomide. They were first approved for use in multiple myeloma in 2003–2005; however, their mechanism of action was revealed later in 2013, detailed as degradation of Ikaros and Aiolos transcription factors by inducing their interaction with the cereblon (CRBN) component of the E3 ubiquitin ligase complex [6].

An additional functionally distinct class of iTPD degraders employs the autophagy-lysosome system to degrade the protein of interest (POI) instead of the ubiquitin–proteasome system [7]. This group of approaches includes AuTophagosome-TEthering Compounds (ATTECs), AUtophagy-TArgeting Chimeras (AUTACs), and AUTOphagy-TArgeting Chimeras (AUTOTACs). ATTECs utilize degrader molecules linking POIs to the autophagosome protein LC3, resulting in the POI’s degradation after fusion with the lysosome [8]. However, evidence of a direct LC3 binding for several ligands used in the early studies still remains insufficient [9]. Both AUTACs and AUTOTACs employ the autophagy cargo receptor p62 as an effector instead of LC3, with AUTAC requiring the target protein ubiquitylation [10], while AUTOTAC is ubiquitylation-independent [11]. The hijacking of the autophagosome-lysosome pathway is especially promising for degrading pathogenic protein aggregates, which are unsuitable for proteasome-dependent iTPD due to their large size, such as aggregated tau and α-synuclein [12].

While iTPD and especially PROTACs/molecular glues have demonstrated remarkable progress over the past 25 years, their application remains limited mostly to intracellular proteins. However, nearly half of all known proteins are extracellular or cell membrane-associated [13]. This proteome comprises many promising therapeutic targets with distinct roles in severe human pathologies. While half of the FDA-approved small molecules and almost all biologics engage these proteins [14], most potential extracellular and secreted protein targets remain undruggable. Therefore, a dedicated technological platform is needed to eliminate these proteins, called extracellular TPD (eTPD).

Since the protein degradation machinery is located inside the cell, the targeted elimination of extracellular proteins must employ an “outside-in” strategy that internalizes these proteins [15]. Therefore, the chimera molecules used for eTPD induce the proximity of a target protein (POI) with an effector protein that mediates internalization and endocytosis. Unlike PROTAC-targeted intracellular proteins that normally undergo proteasomal degradation, the internalized POIs are routed through the endocytic pathway and ultimately degraded by lysosomes (Fig. 1B). The first landmark paper on eTPD [16] successfully demonstrated degradation of four pathologically-relevant proteins, including three membrane proteins (epidermal growth factor receptor [EGFR], programmed cell death 1 ligand 1 [PD-L1] and cluster of differentiation 71 [CD71]) and one soluble protein (apolipoprotein E4). Analogous to PROTACs in iTPD, the respective degrader molecules were called LYsosome-TArgeting Chimeras (LYTACs).

Only emerging in 2020, the eTPD field is considerably younger than the iTPD field. However, given the importance and pathological significance of multiple soluble and cell surface-located proteins, it holds immense promise. An example of a promising target for the development of new eTPD degraders is the estrogen receptor (ER), regarding its well-established role in cancer pathogenesis and progression [17]. The ER degraders based on PROTAC technology developed by Arvinas are already in late-stage clinical trials [14]. However, targeting the extracellular part of the ER and hijacking the endosome-lysosome system for its degradation might be an alternative solution with certain clinical benefits.

Hematological oncology represents a particularly promising area for eTPD application. One of the key challenges in treating hematological malignancies is overcoming the limitations of conventional chemotherapy, including its severe side effects and acquired resistance. Generally, the available chemotherapy agents are toxic and often ineffective for many patients [18], while targeted gene therapy approaches remain expensive and unavailable for most patients. Therefore, there is a strong need for novel therapeutic modalities for blood cancers that address these issues. Since TPD has a catalytic mode of action that requires a relatively small dosage for efficacy, its application is associated with significantly lower drug-related toxicity [19]. Furthermore, TPD acts by degrading the entire POI, helping to overcome resistance due to protein mutations. Numerous iTPD degraders are currently in clinical trials for various hematological malignancies [20, 21]. The development of eTPD approaches will undoubtedly reveal more candidates for clinical trials, expanding the spectrum of pathologically relevant extracellular POIs that are undruggable by other methods.

This review focuses on recent technological advances in the eTPD field. It emphasizes the structural classification of related methods, their advantages and associated problems, and open questions in the field.

Advantages of TPD approaches

From a pharmacological perspective, TPD approaches offer several key advantages over “classical” occupancy-driven protein inhibition [22]. One notable advantage is illustrated by the case of receptor tyrosine kinases (RTKs), which are well-established targets of anti-cancer therapy [23]. Notably, classical inhibitors of certain RTKs inhibit only the respective kinase domains and often cause only a limited effect on immediate downstream signaling. This kinase inhibition does not block the scaffolding function of RTKs and the related compensatory signaling. In contrast, using specific degraders for RTKs overcomes this limitation by eliminating the entire protein, resulting in sustained inhibition of the related signaling pathways [24].

Another significant advantage is the possibility of using very low concentrations of the degrader molecules, as the induced degradation is both catalytic and sub-stoichiometric [25]. This advantage is achieved due to the recycling of degrader molecules that are not degraded along with the POI and thus can engage in multiple cycles of degradation. This property seems to be a common feature of iTPD degraders, where all involved molecules are initially present in the cytosol. The situation is more complicated for eTPD chimeras since these degrader molecules bind the POI outside the cell and then undergo internalization. While there is evidence that these degrader molecules can be recycled back to the extracellular compartment together with the receptor/effector proteins [26], further studies are needed to characterize the recycling mechanism and understand if it is universal.

Interestingly, several PROTACs that utilize known small molecule inhibitors as their warheads demonstrate fewer side effects and lower related toxicity than the respective inhibitors [27, 28]. While the molecular mechanism underlying this phenomenon remains unclear, it can likely be explained by improved target affinity due to a larger surface available for the interaction interface between the target protein and the PROTAC molecule, compared to small molecule inhibitors alone. In theory, the same effect can be expected for LYTACs and related molecules employed in eTPD approaches based on known small molecules’ interactions with protein targets. The increased stability of the PROTAC-induced ternary complex may also contribute to improved selectivity [29, 30].

The last but particularly important advantage to note is the broad applicability of the eTPD approach, considering its range of possible targets. Diverse lysosome-dependent TPD strategies have already been demonstrated to be applicable to membrane proteins, secreted proteins, and extracellular protein aggregates.

General principles of eTPD

The basic concept utilized to develop LYTACs and other eTPD degraders is induced proximity in the extracellular space [31]. The design of degrader molecules aims to introduce the recognition interface for the POI on one side and for an “effector” protein necessary to induce the degradation pathway on the other. The desired result is the formation of a ternary complex consisting of the POI, the targeting chimera, and the effector. In the case of the most advanced iTPD approach, the PROTAC technology, the effector is the E3 ubiquitin ligase, which ubiquitylates the POI and thus promotes its interaction with the proteasome and subsequent degradation. In the case of eTPD, the effectors are plasma membrane-associated proteins that induce POI internalization and lysosomal delivery. Therefore, the degrader molecule contains two distinct functional parts: one that interacts with the POI and another that interacts with the effector. These two parts are connected by a linker, creating a classical “warhead-linker-ligand” structure (Fig. 2). In this review, we use structural classification and naming of degrader parts by analogy to PROTACs. However, there is currently no “official” unified nomenclature, and establishing one based on a consensus of the TPD community would be helpful. Generally, PROTACs are heterobifunctional molecules comprising two recognition modules that act as ligands for a target protein and an E3 ubiquitin ligase, connected by a linker. In several early studies, the target protein was sometimes called the “prey,” and the part of the degrader molecule specifically binding to that protein was correspondingly called the “bait” [32]. The part interacting with E3 ubiquitin ligase was historically called the “ligand” [33]. However, recently, the term “handle” has increasingly been used for the E3 ubiquitin ligase-interacting module in bispecific molecules [34, 35].

Fig. 2

The modularity of eTPD degrader design. The chimeric degrader molecule comprises two recognition modules that bind to the target protein (warhead) and the effector protein (ligand), respectively, and are connected by a linker. The diversity of molecules that can constitute each of these modules is illustrated below

In many cases, the term “ligand” was also used for the part interacting with the POI (e.g., “a ligand that binds to the targeted protein” [36]), and the term “degron” was used for the part interacting with the E3 ubiquitin ligase [37], contributing to terminological ambiguity. Since 2016, the degrader part responsible for binding to the POI has increasingly been called the “warhead” [38, 39]. To our knowledge, the term “warhead” in the context of PROTAC molecules always referred to the component specific to the POI. Therefore, using this term to refer to the POI-interacting part in a universal nomenclature for degraders seems logical. Noteworthy, the exact meaning of “warhead” differs between the TPD context and classical covalent inhibitors, where warheads are the reactive groups of drugs that engage in covalent interactions with the enzyme/receptor residues [40]. The degraders’ warheads, which are based on entities different from small molecule inhibitors, in most cases bind to the POIs in a non-covalent manner. The term “ligand” refers to the part interacting with the E3 ubiquitin ligase in the iTPD context or the lysosome-targeting receptor (LTR) or similar internalization effectors in the eTPD context.

Diversity of degrader modules

Various classes of molecules can be used as recognition modules (either warheads or ligands) in degrader molecules (summarized in Fig. 2). Several can be used both as warheads and LTR ligands (antibodies, peptides, and aptamers). Warheads can also be small molecules, while ligands can be based on carbohydrate conjugates. Various modules differ significantly in their physicochemical characteristics (molecular weight and volume, polar surface area, and hydrophobicity) that ultimately influence the pharmacodynamics (PD) and pharmacokinetics (PK) properties of chimeric degrader molecules. The currently available spectrum of molecules does not exclude the possibility that other types will be employed as warheads or ligands in the future. The variable structural composition of different degrader molecules is illustrated in Tables 1, 2, 3, 4, 5.

Antibody-based modules

Antibodies were employed as warheads to specifically target several POIs in the first landmark study from Carolyn Bertozzi’s group, which introduced the LYTAC technology and the eTPD approach [16]. Antibodies currently represent the most widely used type of warhead in eTPD applications. Different types of antibodies and their derivatives have been successfully used to construct degraders, including fully assembled antibodies, fragment antigen-binding (Fab) fragments, and nanobodies. The recent development of antibody engineering methods has allowed the creation of more sophisticated approaches, including bi-specific antibodies that recognize both the POI and LTR [41, 42].

One of the main advantages of antibodies is their exceptional specificity that can be further fine-tuned using various advanced antibody engineering methods. Moreover, robust and well-established production techniques are readily available. Another significant factor is the availability of numerous monoclonal antibodies targeting clinically important proteins, including those already approved by the FDA and other regulatory authorities for therapeutic use. In principle, these antibodies can serve as recognition modules in the design of eTPD degraders.

The main disadvantages of using antibodies as potential therapeutics include their relatively high production cost (compared to peptides and small molecules) and their potential immunogenicity. One of the biggest challenges for all protein-based therapeutics, including antibodies, is their short persistence in serum due to elimination via several mechanisms (enzymatic degradation, renal clearance, and liver metabolism) [43]. Another disadvantage related explicitly to eTPD applications is their relatively large size and molecular weight, which might interfere with internalization efficiency. For example, efficient internalization via the asialoglycoprotein receptor (ASGPR)-dependent pathway requires that the overall size of the complex be smaller than 70 nm [44]. Therefore, smaller antibody fragments (e.g., Fab) generally work much better for internalizing the complex with the POI [45]. The poor pharmacokinetics of large molecules such as antibodies is also an important disadvantage, limiting their tissue penetration ability and delivery to the intended site of action (e.g., solid tumors) [46]. However, the intrinsic propensity of antibodies to be retained in the circulation can be particularly beneficial for their application in treating hematological malignancies.

Peptides

The first example of a peptide-based warhead used in eTPD was a polyspecific integrin-binding peptide used to degrade several integrins [47]. In this study, peptidic warhead moiety was conjugated with a triantennary N-acetylgalactosamine (tri-GalNAc) motif that served as a ligand for a novel type of liver-specific LTR, the asialoglycoprotein receptor (ASGPR). All-peptidic degraders (with both the warhead and LTR ligand parts being peptides), which have the important advantage of being entirely genetically encoded, have been recently reported [48].

Being relatively small, peptide-based degraders can theoretically exhibit greater efficiency in the internalization of the degrader-POI complexes than antibodies. However, potential immunogenicity remains an issue. Another challenge is the delivery method for peptide-based degraders in therapeutic applications. While many small molecules can be orally delivered, peptides typically require delivery via viral vectors or nanoparticle (NP)-based systems.

Aptamers

Several published studies have demonstrated the successful use of DNA- and RNA-based aptamers as recognition modules in eTPD degraders [49,50,51]. In principle, aptamers can be generated against virtually any protein target, and their high specificity—along with their relatively low cost and rapid production—makes them particularly attractive for use in eTPD. Similar to antibodies, many methods are readily available for aptamers design and synthesis. For instance, aptamers can be rapidly produced on a large scale by DNA solid phase synthesis and selected using the Cell-SELEX (systematic evolution of ligands by exponential enrichment) method [52]. However, aptamer selection is generally a multi-step process with several rounds of enrichment to achieve the desired specificity and affinity [53].

Aptamers are not immunogenic since they are composed of nucleotides. However, this chemical structure is associated with a serious issue: in vivo stability. Since aptamers are nucleic acids, they are susceptible to hydrolysis by nucleases (particularly relevant for RNA-based aptamers). Structure optimization and chemical modifications may help to address this problem, but these inevitably make the production of aptamers more complicated and costly. More importantly, the introduction of modifications raises the issues of biocompatibility.

Warhead-specific: small molecules

Small molecules are currently used in eTPD only for warheads. The first warhead employed in the history of the TPD field was a small molecule used in the seminal paper by Raymond Deshaies’ and Craig Crews’ groups in 2001 [2]. They used the small molecule ovalicin due to its ability to bind the active site of METAP2, which was the first example of a PROTAC-degraded protein. In the eTPD field, the small molecules employed as binding modules for target proteins include enzyme inhibitors [54] and vitamins such as biotin [55].

Small molecules have distinct advantages over larger biologics from a pharmacological perspective. Their small size often means favorable PD and PK, particularly regarding oral bioavailability, which is a considerable benefit for therapeutic applications. Moreover, small molecules can often cross the cell membrane, which is necessary for interacting with intracellular targets. However, this particular advantage is less relevant for eTPD degraders that act in the extracellular space. Finally, they are characterized by relatively low production costs.

The greatest disadvantage that significantly limits the applicability of small molecules in degrader design is that they typically need a defined pocket in the target structure to bind.

Ligand-specific: carbohydrates

The diversity of LTR ligands is generally comparable to that of warheads and includes antibody-based ligands, aptamers, and peptides. As described above, small molecules are another class of binders in the warhead context; however, this class differs in the LTR ligand context and could be better described as carbohydrate-based ligands.

The first article on LYTACs described using the insulin-like growth factor 2 receptor (IGF2R, also called the cation-independent mannose 6-phosphate receptor, or CI-M6PR) as the LTR and a ligand based on one of its physiological ligands, mannose-6-phosphate (M6P). M6P is an N-linked oligosaccharide synthesized in the trans-Golgi network and utilized by the cell for tagging proteins destined for the lysosomal compartment [56]. These M6P-marked proteins are specifically recognized by two M6P receptor isoforms, the cation-dependent receptor (CD-M6PR) in the Golgi membrane and the cation-independent receptor (CI-M6PR) present in both the Golgi and plasma membrane [57]. Inspired by this sorting mechanism, the first published eTPD ligand was a chemically synthesized glycopeptide with multiple serine-O-mannose-6-phosphonate (M6Pn) residues, specifically recognizing CI-M6PR on the cell surface [16].

The ligands for the second generation of eTPD-engaged LTRs were based on N-acetylgalactosamine (GalNAc), an amino sugar derivative of galactose. This carbohydrate moiety is a natural high-affinity ligand for the liver-expressed asialoglycoprotein receptor 1 (ASGPR) and has binding constants in the low nanomolar range. The association between GalNAc and ASGPR is pH-sensitive, resulting in dissociation at lower pH in the endo-lysosomal compartment, with the receptor normally recycled back to the cell surface [58]. These properties make GalNAc-based conjugates a widely used system for drug delivery into hepatocytes, particularly for oligonucleotide-based drugs [59].

Linkers

The linkers that connect two recognition modules in a degrader molecule are often overlooked in the degrader design due to their underestimated importance. Research in the field tends to start with simpler linkers, with their design mainly guided by chemical synthesis requirements. Notably, the linker influences both the distance between the warhead and ligand and their orientation; therefore, the optimal linker length is crucial for the successful degradation of the POI. In experiments with all-protein chimeras, longer linkers decreased the internalization efficacy, and the shortest linker composed of just two amino acids abolished the uptake [48]. In some cases, optimizing only the linker part of the targeting chimera can significantly increase its affinity and general degradation efficiency [60]. More complex and functional linkers have been synthesized in recent years, particularly in the PROTAC field, with the added advantage of stabilizing the ternary complexes. Moreover, in iTPD approaches, linkers have been used for additional functionalities beyond degradation, such as incorporating specific activators like photo-switches or conformational locks [61, 62]. Thus, similar developments can also be expected in the eTPD field. However, whether linker composition and structure are less important in eTPD degraders (e.g., in LYTACs compared to PROTACs) because of mechanistic differences remains an open question. Probably, the stability of the ternary complex is less important for LYTACs, because the target ubiquitylation is not as necessary for LYTAC-dependent degradation as for PROTACs.

Generally, computational methods and structural analysis provide a solid basis for the rational design of linkers [63, 64]. Readers interested in the diversity of linkers can find more information in a recent comprehensive review [65].

Diversity of LTRs

LTRs and the related effector proteins are key players in eTPD, as it requires inducing the internalization of a target protein that ultimately leads to its elimination via intracellular degradation pathways. These effectors can be classified according to their respective internalization mechanisms (Fig. 3). The conventional LTRs typically induce internalization upon either conformational changes (Fig. 3A) or receptor clusterization (Fig. 3B) induced by ligand binding. There are also special cases beyond these mechanisms that involve membrane-bound E3 ubiquitin ligases (Fig. 3C) and direct lysosomal targeting without classical receptors (Fig. 3D). This review aims to provide a comprehensive overview of available LTRs and other effectors that can be employed for degrader design. The details of the available degraders are provided in Tables 1, 2, 3, 4, 5. Note that if the authors do not use a specific term for the technology utilized in the respective study, it is referred to as “LYTAC-based” in the tables.

Fig. 3

The diversity of eTPD effectors and the respective degradation mechanisms. A IGF2R/CI-M6PR-based degraders: The binding of the degrader induces conformational changes in the LTR and endocytosis. B ASGPR-based degraders: The binding of the degrader induces LTR clustering and endocytosis. C Targeting membrane-associated E3 ubiquitin ligases as eTPD effectors instead of LTRs. D CPP-LSS-based degraders: Direct internalization and lysosomal targeting without LTRs or E3 ubiquitin ligase

First generation (IGF2R/CI-M6PR)

IGF2R/CI-M6PR was the first LTR employed for eTPD, as reported in the pioneering study by Carolyn Bertozzi’s group [16]. The large extracellular part of IGF2R contains several binding sites specific to different ligands [66]. The binding of these ligands induces a conformational change that leads to the internalization of the receptor via clathrin-dependent endocytosis (Fig. 3A). The adapter protein 2 (AP2) complex interacts with the cytoplasmic domain of the IGF2R, facilitating the formation of clathrin-coated vesicles that are then routed to early endosomes, late endosomes, and finally lysosomes. The acidic pH in these compartments allows IGF2R to dissociate from its cargo and be recycled back to the cell membrane or trans-Golgi network. This mechanism is employed by various degrader molecules developed to date, containing both physiological ligand-based binding modules for IGF2R (such as M6P or insulin-like growth factor 2 [IGF2]) or other recognition molecules (aptamers), as illustrated in Table 1.

Table 1 eTPD degraders utilizing IGF2R (CI-M6PR) as the LTR

Second generation (ASGPR)

Several research groups independently reported the use of the liver-specific ASGPR as a suitable LTR for eTPD in 2021 [45, 47, 54]. The ASGPR was chosen due to its importance as a physiological receptor already involved in the degradation of several circulating proteins and its being primarily expressed on hepatocytes, a cell type highly capable of catabolizing substantial amounts of protein. ASGPR internalization and endocytosis are induced by its oligomerization/clusterization. The trimeric nature of the ASGPR prompted the creation of branched ligands for this receptor (Fig. 3B). While the ASGPR exhibits a low affinity for monomeric GalNAc (dissociation constant [K_D] ∼ 40 μM), it exhibits a much greater affinity for its branched trimeric form tri-GalNAc (K_D ∼ 3 nM) [74].

To date, only ligands based on GalNAc and its derivatives have been reported for this type of LTR (Table 2). However, further development of ASGPR-based degraders with alternative architectures can also be expected.

Table 2 eTPD degraders utilizing ASGPR as the LTR

Other conventional LTRs

The question of LYTAC receptors alternative to IGF2R (CI-M6PR) and ASGPR emerged immediately [78], with substantial progress made over subsequent years. Alternative LTRs currently available for eTPD are listed in Table 3. However, this list is not exhaustive because novel LTRs will likely be identified as eTPD studies progress.

Table 3 eTPD degraders utilizing LTRs other than IGF2R and ASGPR

LTR expression patterns

One of the key principles of the eTPD approach is the interaction of the degrader molecule with the effector protein that mediates the internalization of the entire cargo complex (POI-degrader-effector), which is required for the subsequent lysosomal degradation of the POI. Thus, degradation efficacy depends on the expression level of the effector proteins (LTRs or particular E3 ubiquitin ligases) on the cell surface. In this regard, two issues are common for several known LTRs: low expression and lack of tissue specificity. Since the outset of the PROTAC era, one of the main issues has been identifying E3 ubiquitin ligases with particular expression profiles suitable for tissue- and cell-type-specific target degradation [91]. The same issue impacts the eTPD field: finding specifically expressed LTRs.

Different LTRs used to date in experimental studies have distinct tissue expression profiles [48]. While IGF2R is the most ubiquitous and present in various tissues, other LTRs have more specific expression patterns. For example, ASGPR is a liver-specific receptor, sortilin is primarily expressed in the central nervous system, and transferrin receptor (TfR) is present in the brain, liver, and muscles. The specific expression patterns of currently known LTRs are shown in Fig. 4. These data are particularly crucial for designing tissue-specific degraders. Another important consideration is the relative expression of certain effectors in normal versus pathological tissues, such as cancer, as targeting cancer-specific or enriched receptors would be a beneficial therapeutic strategy, minimizing off-target effects.

Fig. 4

The expression patterns of different LTRs. The expression data of the indicated LTRs across selected tissues were analyzed using the GTEx Portal (https://gtexportal.org/home/) and reported as a heatmap with tissue clustering. The gene names for conventional LTRs reported in Tables 1, 2, 3 were used as input

Membrane-associated E3 ubiquitin ligases as effector proteins in eTPD

An interesting class of effectors that can be co-opted to induce internalization and degradation of the POI cargo is the plasma membrane-located E3 ubiquitin-protein ligases. These ligases are not classical LTRs but can nonetheless promote lysosomal clearance of target proteins in a similar manner via a mechanism that is currently poorly understood [92]. The first example comes from Cotton et al. [41], who created fully recombinant bispecific antibodies recruiting the membrane-bound E3 ligase ring finger protein 43 (RNF43) to degrade PD-L1. Under normal conditions, RNF43 is known to ubiquitylate specific proteins, such as frizzled receptors, for endocytosis and proteasomal degradation [93]. This particular E3 ligase was chosen as a potential eTPD effector because of its structural characteristics that facilitate antibody generation (single-pass membrane protein with structured ectodomain), and its ubiquitous expression. Although E3 ligases typically catalyze the ubiquitylation of their target proteins, and several forms of appended ubiquitin signal serve as a marker for proteasomal degradation [94], the induced interaction with RNF43 resulted in lysosomal degradation of the cargo in this case, as demonstrated by the use of specific pathways inhibitors. Notably, lysosomal acidification inhibitor bafilomycin blocked the degradation of PD-L1, while proteasomal inhibitor MG-132 had no such effect [41].

Interestingly, attempts are now being made to downregulate RNF43 using an iTPD approach, as illustrated by a recent article from the biotechnology company Surrozen [95]. The RNF43-mediated degradation of frizzled receptors blocks Wnt signaling that is important for tissue homeostasis and biogenesis. Thus, degrading RNF43 by linking it to hepatocyte-expressed ASGPR is considered a way to reactivate the Wnt signaling pathway in the liver to promote tissue-specific regeneration.

In addition to RNF43, several recent studies have used another closely related E3 ligase, zinc and ring finger 3 (ZNRF3), as a membrane receptor for cargo internalization and degradation [96, 97]. The first utilized bispecific antibodies to induce the proximity of the membrane-localized target protein (cargo) and the E3 ligase that induces cargo internalization and degradation. There is evidence that ZNRF3 engages both the lysosomal and proteasomal degradation pathways, with this process also involving the ubiquitylation of the cargo (Fig. 3C). This approach was further extended to several other E3 ligases with similar structures (RNF128, 130, 133, 149, 150), suggesting the universality of the observed mechanism for this class of membrane proteins. Notably, several of the additionally identified cell-surface E3 ubiquitin ligases demonstrated discrete tissue expression patterns, potentially allowing for the tissue-specific degradation of the target proteins. Moreover, both RNF43 and ZNRF3 exhibit higher expression in human colon adenomas and colorectal cancer than in normal tissue [96], a characteristic that can be exploited for therapeutic purposes.

An open question regarding the involvement of ubiquitylation in the case of ZNRF3 and related ligases is the architecture of the attached ubiquitin chain. Further studies are needed to understand whether it is Lys48-linked polyubiquitin (a classical degradation tag – an “eat me” signal for the 26S proteasome), or other forms of ubiquitin signal such as mono-ubiquitin that can engage some lysosome-routing adaptors. Several forms of ubiquitin signal can be present simultaneously, explaining the engagement of two distinct degradation pathways.

The second study utilized bispecific R-spondins (RSPOs), another example of chimeric molecules that engage E3 ligases for cargo internalization and subsequent degradation [97]. RSPOs are stem cell-derived cytokines and natural ligands for RNF43 and ZNRF3, playing important roles in regulating the Wnt signaling pathway. They induce the downregulation of these E3 ubiquitin ligases, thereby upregulating their physiological targets and thus promoting the Wnt signaling [98]. RSPOs also have distinct domains that bind other signaling proteins and target them to the same E3 ligases, followed by the degradation of the resulting ternary complex. The modularity of this interaction allows the engineering of RSPOs to target other non-physiological substrates using the domain-swapping approach. Thus, RSPOs can be engineered to lose their natural signaling capacity but gain the ability to bind a new substrate while maintaining their ability to bind RNF43/ZNRF3 and induce the degradation of the new target in an RNF43/ZNRF3-dependent manner. As a proof-of-principle, this study created an R-spondin 2 (RSPO2) chimera (R2PD1) that could bind to PD-L1 and induce its lysosomal degradation in three melanoma cell lines at a picomolar concentration by forming a ternary complex with RNF43/ZNRF3 [97]. This approach represents a novel modality in the induced proximity pharmacology and has several advantages over bispecific antibodies due to the smaller size of chimeric molecules, such as facilitating delivery via viral vectors [99]. However, since RSPO-based chimeras are still new and not extensively studied compared to antibodies, open questions remain about the stability and immunogenicity of these targeting molecules, which require further investigation.

Using membrane-embedded E3 ubiquitin ligases as effector proteins instead of classical LTRs for target internalization represents a novel eTPD modality (Table 4), different from the conventional LYTAC approach and its modifications. One of the important issues remains the lack of information on the endocytosis mechanism involving RNF43/ZNRF3 and similar E3 ligases.

Table 4 Degraders targeting E3 ubiquitin ligases as effectors for eTPD

Direct endosome/lysosome targeting without LTRs

Several recent studies have reported the development of alternative eTPD approaches that do not rely on specific LTRs or membrane-bound E3 ubiquitin ligases. Instead, these approaches utilize the general endocytosis machinery for intracellular delivery of the POI cargo (Fig. 3D), using peptide sequences that directly interact with the plasma membrane (called a cell-penetrating peptide, or CPP) and later with the endocytic sorting complexes (lysosome-sorting sequence, or LSS). These approaches overcome one issue with traditional LYTAC: the relatively low expression levels of specific LTRs (or membrane-bound E3 ubiquitin ligases) used to promote cargo internalization. Another advantage of these approaches is the smaller size of the complex to be internalized, as endocytosis of large molecular assemblies can be problematic. Direct endocytosis-dependent methods (summarized in Table 5) also have broader applicability because they are not cell type-restricted and, in principle, should work with any cell capable of endocytosis. Since these approaches are particularly promising for further development as potential therapeutics, technologically very different from classical LYTAC-related eTPD, and not sufficiently covered in the available literature, we will discuss them in more detail here.

The first example is the GlueTAC technology [100], which is based on a combined CPP-LSS construct previously characterized as an efficient internalization signal promoting the lysosomal delivery of labeled proteins [101]. The LSS part is a well-studied NPXY motif found in many transmembrane proteins that undergo clathrin-mediated endocytosis. NPXY-type motifs bind to the phosphotyrosine-binding (PTB) domain of cell surface clathrin-associated proteins, such as Dab2 and ARH. Interestingly, this study used a specifically engineered version of a nanobody as a warhead, capable of covalent interaction with the target proteins. This approach efficiently overcomes a problem of low binding affinity that may result in complex instability during endocytosis and off-target effects. However, producing such a covalent nanobody involves the site-specific incorporation of proximity reactive uncanonical amino acids (PrUAAs) and, thus, is somewhat complex and cannot be easily genetically encoded. The covalent nature of the interaction between the targeting chimera and the POI also suggests that the recycling of the chimera may be impossible, and it is likely degraded in the lysosome along with the POI (although this possibility was not addressed in this study). Therefore, for efficient therapeutic use, the concentration of the targeting chimera should be matched to that of the POI to be eliminated. Other issues that must be addressed are the safety and stability of PrUAA-containing degrader molecules.

Another study used the same CPP-LSS as in the GlueTAC approach [102] but with a set of small molecules (known binders of target proteins) as warheads. It demonstrated the lysosomal degradation of PD-L1 as an extracellular target. It also used the same approach to degrade several intracellular proteins (phosphodiesterase δ, nicotinamide phosphoribosyl transferase NAMPT and Bruton’s tyrosine kinase BTK) by targeting them to the lysosome. It demonstrated the greater efficacy of this TPD approach compared to traditional PROTACs available for NAMPT and BTK. Thus, using CPP-LSS may be promising not only for eTPD but also for iTPD modalities.

An alternative direct eTPD approach called SignalTAC [103] utilizes different LSS sequences based on the transplantable internalization signal from IGF2R, dileucine-based peptide SFHDDSDEDLLHI, which effectively promotes clathrin-mediated endocytosis and lysosomal degradation of the POI cargo. In addition to chimera molecules utilizing antibodies/nanobodies as a warhead, this study also constructed and tested peptide-based SignalTACs to degrade EGFR. In this case, the warhead was a six-residue EGFR-targeting peptide (LARLLT) previously designed using virtual screening methods [104]. Interestingly, these constructs presumably can be internalized more efficiently than antibody-based SignalTACs. The lower molecular weight and smaller size of the peptide chimeras likely allow improved interaction between the signaling motif and the adaptor proteins. Importantly, SignalTACs can be entirely genetically encoded, making this technology applicable to many pathologically relevant POIs with relatively less effort than traditional LYTAC approaches based on bioconjugation or complex synthesis.

This study further developed the SignalTAC approach using an improved internalization signal derived from the IGF2R sequence, a 10-amino acid endocytic signaling peptide (P3). Degraders based on P3 fusion were successfully used for the internalization and lysosomal degradation of PD-L1 and human epidermal growth factor receptor 2 (HER2) [105]. The tyrosine-based P3 internalization motif differs from the previously used glycine-based motif and is characterized by a strong net positive charge, eliminating the need for an additional CPP. The two motifs are believed to play different roles in the internalization of two distinct subpopulations of IGF2R present in the trans-Golgi network and plasma membrane. Indeed, using different pathway inhibitors suggested that the P3-activated internalization process depended on caveolae but not clathrin. This mechanism differs from previously observed clathrin-mediated endocytosis with P1, a glycine-based internalization motif used in the first generation of SignalTACs.

Clathrin- and caveolae-mediated endocytosis is not the only pathway that can be used to achieve the desired lysosomal degradation of extracellular proteins. An alternative is chaperone-mediated autophagy (CMA), a lysosomal proteolytic pathway that does not involve the formation of vesicles, unlike endocytosis-related pathways. It relies on chaperones and co-chaperones that recognize the degradation tags (short sequence motifs) in the cargo proteins and then deliver them to the lysosomes. The target proteins are internalized across the lysosomal membrane via the receptor lysosome-associated membrane protein type 2A (LAMP2A) [106]. A recent study [107] fused a peptide containing three different CMA-targeting motifs with commercial monoclonal antibodies to several cell surface proteins. The resulting Ab-CMA chimeras significantly decreased EGFR, HER2, and PD-L1 levels in vitro and in vivo. This study represents a proof-of-principle that CMA may be employed to degrade cell surface proteins. However, open questions remain regarding the underlying mechanism, especially the efficiency of the outside-in trafficking of the chimera-cargo complex. While the use of different pathway inhibitors suggests that Ab-CMA chimeras are internalized via macropinocytosis, the details of this process require further investigation.

Finally, another interesting example of direct lysosomal targeting in eTPD is modified nanoparticle with targeting binders (MONOTAB) [108]. This approach relies on efficient endocytosis and lysosomal targeting of specific NPs functionalized with antibodies to the POI, overcoming known problems with classical LYTAC approaches, such as specific receptor dependency and the hook effect (further discussed in "Challenges" section). There is also evidence that the MONOTAB approach promotes lysosomal biogenesis without impairing lysosomal health. Interestingly, this study also demonstrated the specific degradation of non-protein targets, such as extracellular vesicles (EVs), using NP functionalization with annexin V that binds phosphatidylserine (PS), a lipid molecule characteristic of EV membranes [108].

Table 5 Degraders for direct endosome/lysosome targeting in eTPD

eTPD state of the art: challenges, perspectives, and clinical applications

Challenges

The challenges currently faced by the eTPD field are largely shared with TPD in general and well-known from over 20 years of iTPD development [109]. However, there are several additional challenges specific to eTPD. Being a relatively young field, eTPD encounters certain problems mainly due to insufficient availability of experimental data. For example, a current lack of mechanistic studies makes it impossible to explain the observed differences in degradation efficacy between different targets, ligands, and warheads.

Many questions important for further developing eTPD as a therapeutic modality also remain unanswered. For instance, since contradictory evidence has been reported in the literature, whether the internalization effector (LTR or particular E3 ubiquitin ligase) is recycled via a universal mechanism or degraded along with the POI remains unknown. While further mechanistic studies are needed to clarify this issue, we can hypothesize that the choice between recycling or degradation for effectors is largely protein-specific and dependent on the nature of a particular effector and its involvement in cellular metabolism. Indeed, LTRs differ significantly in their intrinsic ability to be recycled back to the plasma membrane, with shuttling receptors such as TfR and sortilin demonstrating much greater recycling efficiency than IGF2R and ASGPR [48, 87].

Furthermore, other factors undoubtedly influence the fate of both the effectors and the POIs during endocytosis and subsequent intracellular trafficking. Post-translational modifications (e.g., variable forms of attached ubiquitin signals that are further recognized by the respective ubiquitin receptors) and specific interactions with components of the endocytic machinery and various sorting complexes likely play an important role in this mechanism and, thus, should be explicitly elucidated in the eTPD context. One good example is the study of structural determinants for eTPD with LYTACs employing IGF2R (CI-M6PR) as the LTR, demonstrating that retromer complex recycles the IGF2R-degrader-POI assembly from the endosomes back to the cell surface, interfering with the degradation pathway [68].

Another open question in the eTPD field is how hijacking particular LTR effectors interferes with their physiological function. One important example is targeting the glucagon-like peptide 1 receptor (GLP-1R) that may interfere with glucose homeostasis [110]. Using native ligands as parts of degrader chimeras reported in several studies can potentially promote off-target signaling [42]. Moreover, some classical LTRs, such as ASGPR and IGF2R, are prone to be occupied by endogenous ligands, limiting their efficacy as effectors in POI degradation [68]. Therefore, designing alternative orthogonal ligands for sites not occupied by native ligands would constitute a more effective strategy for eTPD, potentially avoiding interference with natural signaling pathways. Indeed, using degraders that bind to TfR sites not known to interact with its natural ligand transferrin has not shown any effect on physiological transferrin uptake [48]. Moreover, regarding IGF2R as the LTR, fine-tuning the binding affinity of degraders with binding sites orthogonal to its natural ligand IGF2 has demonstrated a lack of interference with the native internalization of IGF2 by IGF2R and its trafficking to the lysosome for degradation [48].

Regarding the general challenges shared with iTPD, one of the most significant is potential off-target or on-target off-tissue effects [111]. Evidently, further efforts to improve the degraders’ specificity and to develop better delivery systems are necessary, such as prioritizing delivery to cancer over normal tissues. Another known issue that arises from the complex nature of degrader molecules is their large size and high hydrophobicity, resulting in non-optimal PD/PK properties [112] and, thus, the poor oral bioavailability of degrader-based therapeutics. Therefore, the degraders’ structure must be optimized. Indeed, given their modular and often complicated structure, most chimeric degrader molecules hardly comply with Lipinski’s classical rule-of-five [113], a set of criteria applied to the physicochemical properties of drugs and widely used to predict their bioavailability. However, the rule-of-five is often irrelevant for PROTAC molecules [114], which is expected to also be true for eTPD degraders (LYTACs and similar). For example, practically all PROTACs that have already demonstrated oral bioavailability in clinical studies have physicochemical properties that exceed the rule-of-five limits [115]. Notably, unlike PROTAC molecules, eTPD degraders are not required to cross the cell membrane (as standalone molecules before they become a part of the “POI-degrader-effector” cargo complex that undergoes internalization); therefore, the stringency of Lipinski’s classical rule-of-five applies even less to them.

The particular properties of chimeric degraders also define the so-called hook effect, a phenomenon observed for both PROTACs and LYTACs [116, 117] when the degradation efficacy for some POIs decreases with increasing degrader concentration. This phenomenon is explained by the nature of bispecific degraders that preferentially form binary complexes (degrader-POI and degrader-effector, where the effector is an LTR or E3 ubiquitin ligase) over the functional ternary complex (POI-degrader-effector). The key to overcoming this hurdle is the extensive characterization of degraders’ PD/PK, which would allow optimal non-saturating concentrations to be selected.

Perspectives

Here, we highlight three trends in eTPD technology that largely also apply to TPD in general. These trends are evidently connected to the abovementioned challenges in the field. Therefore, the first is expanding mechanistic studies to better characterize the modes of action of various degraders and other components involved in POI internalization and degradation. The second is further optimizing degraders and improving their properties to develop potential therapeutics. Finally, the third is expanding the eTPD toolkit, meaning searching for new target proteins and new effectors and accordingly creating novel classes of degraders. Here, we will focus on the latter two trends, highlighting some examples that illustrate the recent achievements in the field.

New targets

Expanding eTPD to new targets (and new clinical applications) is one of the hottest topics in the field. Generally, the efficacy and clinical applicability of eTPD approaches (and TPD in general) depend strongly on the selection of the target protein. Many published studies have reported only a limited set of targets (e.g., EGFR and PD-L1), as their primary goal was to demonstrate the proof-of-principle of various eTPD approaches. Therefore, further progress in this field is connected to targeting less obvious proteins that can represent even better clinical candidates in particular cases. The challenge is finding protein targets that are more specifically related to pathological processes and less involved in normal physiology. Their use would also help to avoid potential side effects associated with “switching off” certain central signaling hubs, such as EGFR. In oncology, potential new targets include the many other growth factor receptors and their ligands that are critical for signaling in cancer cells. Considering the recent trend of the eTPD expansion beyond oncology, new applications emerge in neurological disorders (particularly targeting protein aggregation), immunology (addressing cytokines and their receptors), diabetes, cardiovascular diseases, and infections (several examples are discussed in "Clinical applications" section). Moreover, one possible direction for further research in this field is exploring completely new extracellular POIs that have never been targeted. Developing eTPD approaches for proteins that are not ideal candidates from a mechanistic perspective but remain clinically important and undruggable by other approaches, due to a lack of surface pockets for small molecules or evolved mutations causing resistance, may be particularly beneficial.

New degrader architectures

Diverse research efforts are currently underway to improve existing degraders and design novel ones. One such direction is developing better ligands for known LTRs. For example, evolved IGF2 was recently used as a ligand for IGF2R, likely the most studied and most utilized LTR for eTPD [72]. One particular advantage of using IGF2 as the ligand is that it is a human protein, potentially overcoming immunogenicity issues associated with proteins from other origins, including those designed in silico.

Another promising approach to constructing new degraders is reusing previously characterized molecules with demonstrated affinity to clinically relevant POIs as warheads. In this regard, some monoclonal antibodies developed as potential therapeutics for various cancers that demonstrated good affinity and specificity to pathologically relevant proteins but failed in clinical trials [118, 119] represent particularly interesting opportunities. Alternatively, many peptides have been previously demonstrated to bind promising POIs, such as peptides developed to prevent amyloid aggregation in Alzheimer’s disease [120].

Fully genetically encoded targeting molecules is another very promising approach for developing degraders. Notably, the successful engineering of human primary T cells to produce genetically encoded lysosome-targeting chimeras (GELYTACs) has been recently demonstrated [72], opening new perspectives in applying this approach in a site-specific manner, such as delivering potential LYTAC-based therapeutics directly to the desired areas, especially in tumors in combination with chimeric antigen receptor T-cell (CAR-T) therapy.

An important achievement is the recent introduction of modular solutions for constructing degrader chimeras. For example, using a universal antibody adaptor (originally derived from staphylococcal protein A) as a part of the degrader molecule opens the possibility of combining it with any existing antibody targeting an extracellular protein [69]. Modularity also allows the simultaneous degradation of multiple target proteins. A recent study achieved the simultaneous degradation of two membrane proteins with a degrader based on modular aptamer assembly [70].

From a technological perspective, the search for new degraders often relies on recent developments in in silico methods for binder design. This is especially relevant for protein-based degraders; indeed, in recent years we have seen tremendous progress in related technology, also highlighted by the awarding of the 2024 Nobel Prize in Chemistry to David Baker, Demis Hassabis, and John Jumper for computational protein design and structure prediction [121]. Readers interested in computer-aided degrader design can find further information in recent comprehensive reviews [122, 123].

New approaches for degrader delivery

Advanced delivery methods are often needed to successfully apply degraders at pathological foci, helping to avoid off-target effects. Several important examples of NP-based degrader delivery methods have recently been published [108]. Very sophisticated delivery methods are beginning to emerge, such as nanodevices based on DNA origami technology with an incorporated pH-sensitive activation switch capable of simultaneously degrading EGFR and PD-L1 [124]. As recently reported, the self-organization of peptide-based degraders in NPs offers interesting possibilities [125]. Notably, this study demonstrated a combined eTPD/iTPD approach that employs both lysosomal and proteasomal degradation of the POI.

NP-based degraders also open new possibilities for combination therapy. For example, using NPs modified with CD24-targeting antibodies and loaded with glucose oxidase (GOx), an enzyme that accelerates glucose utilization, achieved the targeted release of GOx to effectively deplete endogenous glucose in cancer cells. Combining this starvation therapy with CD24 degradation demonstrated a synergistic therapeutic effect against hepatocellular carcinoma both in vitro and in vivo [76].

Another interesting technique for controlled degrader delivery is engineered degrader platelets (DePLTs) bearing heat shock protein 90 (HSP90) linked to a POI-specific warhead [126]. Activated DePLTs accumulated at hemorrhagic areas and released the HSP90-based degrader to promote POI degradation by either eTPD (via HSP90’s interaction with LRP1 as the LTR) or iTPD (via an interaction with the intracellular ubiquitylation machinery), depending on the POI and specific warhead utilized. Due to the specific tropism of the platelets used, this approach appears particularly promising for conditions involving blood coagulation and related vascular pathologies.

Clinical applications

The range of known and future clinical applications for eTPD degraders is determined by the variety of target proteins that are important for the pathogenesis of specific diseases. Generally, there are well-determined criteria for optimal targets for eTPD-based therapy: disease-causing gain-of-function that can be attributed to specific alteration of the normal protein state due to mutation, aggregation, changes in expression levels or localization [91]. The multitude of well-characterized protein targets involved in tumorigenesis makes cancer the most promising area for degrader-based drug development. The most studied pro-oncogenic targets include central signaling hubs like EGFR, HER2, and several other RTKs, as well as immune checkpoint protein PD-L1 (details on the available degraders for these and other proteins are reported in Tables 1, 2, 3, 4, 5). Thus, in most cases, the eTPD degraders aim to disrupt the signaling cascades that drive tumor growth and dissemination. However, studies have appeared recently that also target tumor immune response, such as degrading CD24, which is important for protecting cancer cells from recognition by macrophages [76].

Clinical applications beyond oncology have also started to emerge, including severe human pathologies such as atherosclerosis, diabetes, and neurodegenerative diseases. We will briefly discuss some examples of these applications below. A recent paper from Novartis [74] demonstrated an interesting approach that aims to decrease low-density lipoprotein (LDL) levels in the plasma since elevated LDL is associated with the risk of atherosclerotic cardiovascular disease. Their chosen degradation target is proprotein convertase subtilisin/kexin type 9 (PCSK9), a soluble plasma protein that acts as a negative regulator of the LDL receptor (LDLR) [127]. Since the LDLR is responsible for removing LDL from the circulation, downregulating the LDLR increases LDL levels in the plasma. Based on this concept, Bagdanoff et al. successfully used heterobifunctional degraders of PCSK9 in multiple formats, combining different types of warheads that bind to PCSK9 with several variants of ASGPR binders.

A recent study has demonstrated the degradation of misfolded extracellular human islet amyloid polypeptide (hIAPP) associated with type-2 diabetes in vitro [82]. This study used metallohelices binding α/β-discordant segments of hIAPP as warheads. Metallohelices are metal-coordinated complexes of helical organic molecules, resembling α-helical peptides in structure and able to bind amyloid proteins and prevent their aggregation [128]. While the specificity of these molecules is still an issue and their respective mechanisms require further investigation, they can potentially be used in eTPD approaches to eliminate several amyloid-like proteins involved in diabetes and neurodegenerative diseases, including Alzheimer’s and Parkinson’s. One study has already reported an amyloid-targeting degrader that shows promise as the basis for Alzheimer’s therapy [83]. The ability of these chimeras to cross the blood–brain barrier is critical, given the aim of treating amyloid-related neurodegenerative disorders [129].

Another promising direction for eTPD development is in the medical microbiology field. One could envision applying targeted degraders to remove microbial pathogen-related proteins, such as secreted virulence factors and bacterial membrane-bound proteins [130]. Designing specific targeting approaches opens new opportunities to decrease pathogenicity and potentially cure related infections. Developing potential eTPD approaches for antiviral applications, such as targeting receptors crucial for viral entry or viral proteins, also looks very promising. Multiple examples of antiviral PROTACs targeting intracellular targets already exist [131].

Last but not least, there are interesting novel approaches to treating chronic inflammatory and autoimmune pathologies by degrading pro-inflammatory cytokines, such as tumor necrosis factor (TNF) [73].

Concluding remarks

Despite its relatively young age (with the first seminal article appearing only in 2020), the eTPD field has already demonstrated impressive expansion and achievements. Given its immense therapeutic potential, we anticipate continued exponential growth, accompanied by the development of innovative methods and applications, both in oncology and beyond. New target proteins are constantly explored, with novel smart degrader architectures appearing and the underlying molecular mechanisms becoming progressively clearer. The future of eTPD appears exceptionally promising, with the potential to revolutionize therapeutic modalities for treating cancers and other serious pathologies. Since the respective toolbox needs to be expanded, this review aims to contribute to this process by summarizing recent progress in the eTPD field and the construction principles of degrader molecules.