RING E3 ligases constitute the largest family of E3 ligases, characterized by the presence of a RING finger domain. This domain facilitates the direct transfer of ubiquitin from the E2 enzyme to the substrate protein by bringing the E2-ubiquitin conjugate near the target. RING-type E3 ligases do not form a covalent intermediate with ubiquitin themselves; instead, they act as scaffolds that promote the transfer reaction. They are crucial regulators of the stability and activity of numerous interaction substrates [44].

This broad family encompasses single-subunit RING E3 ligases, including Mouse double minute 2 (MDM2), Casitas B-lineage lymphoma (CBL), and Tumor necrosis factor receptor-associated factor (TRAF) 6. However, the most significant subclassification of RING E3 ligases is the multisubunit CRLs. These complexes use a Cullin protein as a scaffold to assemble various proteins. Within the CRLs, several key types have been identified [39,45]. The Skp1-Cullin1-F-box (SCF) protein complex is a prime example; it comprises Skp1, Cullin1, Rbx1, and an F-box protein (e.g., β-TrCP and FBXW7) that provides substrate specificity. CRL2 ligases contain Cullin2, famously found in the VHL complex. CRL3 ligases utilize BTB domain proteins as both adaptor and substrate receptor. CRL4 ligases are exemplified by the DDB1-CUL4 complex, where the DDB1- and CUL4-associated factor (DCAF) family members, including CRBN, act as substrate-recognition subunits [45]. Another critical multisubunit RING E3 ligase is the anaphase-promoting complex/cyclosome, essential for cell cycle regulation. The broad substrate specificity of RING E3 ligases and their involvement in diverse cellular pathways make them significant targets for therapeutic intervention (Fig. 4).

HECT E3 ligases operate through a distinct two-step mechanism. Unlike RING ligases, HECT ligases directly accept ubiquitin from E2, forming a transient thioester bond with the ubiquitin molecule via a conserved cysteine residue in their HECT domain. Subsequently, ubiquitin is transferred from the HECT ligase to the lysine residue of the target substrate [46,47]. This unique catalytic mechanism enables more precise control over substrate ubiquitination. Approximately 28 HECT E3 ligases are found in humans, including E6AP (UBE3A), Neural precursor cell expressed developmentally down-regulated protein 4 (NEDD4), ITCH, and SMAD specific E3 ubiquitin protein ligase 1/2 (SMURF1/2). Prominent examples include members of the Nedd4 family, which are involved in regulating membrane protein trafficking and degradation, and Ube3a (E6-AP), which is implicated in Angelman syndrome and the human papillomavirus-mediated degradation of p53 [46].

RBR E3 ligases represent a hybrid class that integrates the functional features of both RING and HECT ligases. This family was relatively recently characterized, with approximately 14 RBR E3 ligases identified to date. Structurally, RBR ligases are defined by two RING domains separated by an in-between-RING (IBR) domain. Functionally, they recruit an E2 ubiquitin-conjugating enzyme via their first RING domain (RING1), a mechanism analogous to that of RING ligases. However, they subsequently form a thioester intermediate with ubiquitin on a conserved cysteine residue within their second RING domain (RING2)-akin to the catalytic process of HECT ligases-before finally transferring the ubiquitin to the target protein. This unique two-step mechanism provides an additional layer of regulatory complexity to the ubiquitination process [48,49]. Key examples include Parkin, a critical ligase involved in mitophagy and Parkinson’s disease, and HOIL-1-interacting protein (HOIP), a component of the linear ubiquitin chain assembly complex with a central role in immune signaling and inflammation [48].

A key advantage of TPD lies in its exploitation of the inherent flexibility of individual E3 ubiquitin ligases to recognize diverse substrates. This flexibility provides substantial opportunities for developing degraders that selectively target oncoproteins with precision across specific tissues, tumors, and subcellular locations [41]. Ongoing research in induced protein degradation is actively enhancing our understanding of these underlying mechanisms and identifying novel E3 ubiquitin ligase targets, thereby driving the development of more effective and selective therapeutic agents [41].

Despite the vast potential of E3 ubiquitin ligases in the drug discovery landscape, only a small fraction has been effectively harnessed by small-molecule degraders, including MGDs; currently, this remains predominantly limited to CRBN. This constraint underscores a critical need for the continued and expanded exploration of E3 ubiquitin ligase biology to fully establish induced protein degradation as a highly specific and efficient therapeutic strategy [50,51,52].

Significant efforts are underway to address this limitation, including the use of various natural and synthetic small molecules to modulate the activity and substrate recognition of diverse ligases [51,53]. While CRBN has historically been the primary E3 ligase exploited for MGD development, the field is rapidly expanding to include a broader array of E3 ligases [50,51,54]. This diversification is essential to unlocking the full therapeutic potential of MGDs, enabling the precise targeting of a wider range of “undruggable” target proteins and expanding their application across diverse disease contexts [54].

The advent of TPD represents a paradigm shift in pharmacology, offering a potent strategy to eliminate disease-causing proteins previously considered undruggable by conventional small-molecule inhibitors [75]. Among the primary drivers of this approach are MGDs, which, alongside PROTACs, hijack the cell’s natural protein disposal machinery, the ubiquitin-proteasome system. At the core of this process are E3 ubiquitin ligases, acting as crucial mediators that tag a target protein for degradation [23].

A significant challenge in the TPD field particularly regarding MGDs, is the underutilization of the vast E3 ligase family. Nearly all clinical and preclinical degraders overwhelmingly rely on a very small, well-characterized subset, most notably CRBN [76]. This heavy reliance on a limited number of E3 ligases imposes several critical constraints on the full potential of MGDs. First, it restricts the range of proteins that can be effectively degraded. Successful degradation requires the formation of a stable and productive ternary complex between the E3 ligase and the POI. However, many disease-relevant proteins are structurally or spatially incompatible with CRBN-mediated proximity, rendering them inaccessible to current MGD strategies [77].

Second, the ubiquitous expression of CRBN across diverse human tissues, while beneficial for some applications, significantly hinders the achievement of tissue- or cell-type specificity. For diseases requiring selective action within particular organs or cell populations, this broad expression can lead to on target but off-tissue toxicity, limiting the development of safer, more precise MGD therapeutics [78,79]. In contrast, exploiting E3 ligases with restricted or disease-specific expression patterns holds promise for enhancing selectivity and therapeutic index [79].

Third, as MGD-based therapies advance into clinical use, the emergence of acquired resistance presents a growing challenge. Cancer cells, for example, can escape degradation pressure by downregulating or mutating the recruited E3 ligase, such as CRBN, ultimately leading to therapeutic failure [77]. Expanding the repertoire of druggable E3 ligases would provide crucial alternative options to counteract these resistance mechanisms and sustain clinical efficacy. Studies using clustered regularly interspaced short palindromic repeats (CRISPR)-suppressor scanning and haploid genetics have revealed that resistance to MGDs targeting neosubstrates, such as G1 to S phase transition 1 (GSPT1) and RBM39, which interact with CRBN and DCAF15, respectively, is directly linked to mutations altering the ternary complex heterodimerization surface [77]. These hotspot mutations, validated in patients who relapse from degrader treatment, disrupt the intended degradation process [80]. Several studies have reported loss-of-function mutations or reduced expression of E3 ligase substrate receptors as a major pathway of resistance to degraders. This mechanism is exemplified by the connection between IKZF1/3 and CRL4^CRBN E3 ubiquitin ligase mutations and resistance to immunomodulatory drugs in multiple myeloma, which is consistent with the positive correlation of CRBN expression levels with response to lenalidomide and pomalidomide in multiple myeloma patients [81,82].

Research in the field is currently focused on multiple strategies to explore and recruit novel E3 ligases. These efforts include the systematic, multiomics characterization of the entire E3 ligase family to rank individual ligases based on their chemical ligandability, expression patterns, and known PPIs [16]. Advanced discovery platforms are guiding this rational discovery process, ranging from chemoproteomic covalent screening techniques that identify new ligandable “hotspots” on E3 surfaces [83] to computational deep-learning (DL) models capable of predicting degrons and E3-substrate binding interfaces [84]. This approach necessitates a deeper understanding of E3 ligase biology, encompassing factors such as substrate selection modes, catalytic specificity, regulatory mechanisms, expression dynamics, and physiological functions [85].

Despite being smaller and simpler molecules (typically <500 Da) than PROTACs-thereby offering superior cellular permeability and the potential for oral administration-designing effective MGDs remains highly complex. This complexity arises from the unpredictable nature of the PPIs they must modulate [10]. Consequently, many MGDs have been discovered serendipitously, and their development continues to rely on innovative screening methods to identify compounds that can effectively stabilize or induce these interactions [10]. While the current MGD landscape is constrained by its heavy dependence on a narrow set of E3 ligases, this limitation simultaneously represents the field’s greatest opportunity. By developing innovative strategies to engage the vast, untapped majority of the E3 ligase family, researchers can significantly expand the degradable proteome, overcome therapeutic resistance, and design the next generation of highly specific and potent MGDs [10,86].

An ideal E3 ligase for MGD development would possess characteristics that enable efficient and selective protein degradation. Key features include strong and specific binding to the target protein, the ability to induce ubiquitination and subsequent proteasomal degradation, and a favorable pharmacological profile for drug development [86,87,88].

When selecting an ideal E3 ligase for MGD development, several key criteria must be considered:

First, specificity is essential. The E3 ligase should exhibit high affinity and selectivity for the target protein, crucial for minimizing off-target effects and maximizing the desired degradation. It must also be able to recognize and bind to specific degradation signals (degrons) on the target protein [87,89].

RING, HECT, and RBR constitute mechanistically and structurally diverse enzyme families that orchestrate the selective ubiquitination and degradation of proteins. Understanding their catalytic modes is fundamental for therapeutic exploitation: RING ligases (e.g., CRLs, UBR family) act as scaffolds that facilitate direct ubiquitin transfer from E2 to the substrate; HECT ligases (e.g., NEDD4, UBE3A) form a transient thioester intermediate allowing tighter catalytic control; and RBR ligases (e.g., Parkin, HOIP) integrate features of the two former groups, allowing hybrid regulation. In TPD and MGD development, E3 ligases form molecular hubs that dictate substrate specificity and degradation efficiency. However, current therapeutics rely heavily on a narrow subset, mainly CRBN, which limits tissue selectivity and leads to potential resistance. Expanding the usable E3 repertoire, including DCAF15, DCAF16, or other context-specific ligases (MDM2, RNF146, DTX1, UBR7), is key to diversifying degrader strategies and accessing undruggable targets.

An ideal E3 ligase for MGD applications should combine substrate selectivity, efficient ubiquitin transfer, structural tractability, and favorable pharmacological properties. Integrating structural biology, chemoproteomics, and computational design allows the rational engagement of new E3 ligases, transforming the conceptual understanding of E3 enzymology into practical frameworks for next-generation therapeutic development.

Cell-based phenotypic screens are highly effective in identifying molecules that induce protein degradation, even without prior knowledge of the specific E3 ubiquitin ligase involved. These screens use functional readouts or reporter systems to detect changes in cellular behavior or protein levels resulting from protein degradation [141,142].

One well-defined process for identifying potential degraders involves cell line preparation followed by high-throughput degradation assays [141]. For instance, researchers engineered HeLa cell lines to stably express HaloTag-GFP, alongside a control line harboring a HaloTag D106A mutation to counter-screen for false positives. Test com-pounds were dispensed into 384-well plates, and after two days of incubation, GFP fluorescence was measured to quantify protein degradation. Compounds that induced over 50% degradation of HaloTag-GFP, while causing less than 50% degradation of the D106A mutant, were identified as hits [141]. Alternatively, these assays can utilize cellular responses or reporter systems affected by protein degradation rather than direct protein level measurements. This approach may involve monitoring a decrease in the activity of a specific protein or employing a fluorescent protein fused to the target protein to track its degradation in real-time.

Another multistep approach for phenotypic screening incorporates the production of lentiviral particles, the establishment of stable cell lines, and flow cytometry-based screening [143]. Lentiviral virions encoding a GFP-tagged POI are produced and transduced into HeLa S3 cells. Cells exhibiting stable expression are selected and transfected with plasmids encoding various biodegraders-consisting of E3 ligases linked to a specific protein binder. The degradation of the POI is assessed by measuring GFP fluorescence intensity via flow cytometry. Hits are subsequently confirmed through retransfection and validated for proteasome-dependent degradation using specific inhibitors [143].

While phenotypic screens can identify molecules responsible for protein degradation, they do not initially identify the specific E3 ligase involved. Further experiments, often using target deconvolution strategies, become necessary [64]. For example, CR8, initially identified as a CDK inhibitor, was later found to be an MGD that degrades cyclin K. Researchers used drug resistance and reporter-based CRISPR screens to identify DDB1, an E3 ligase adaptor protein, as the key player in the degradation of CCNK [64]. Similarly, HQ461 was identified as an antitumor agent through a genome-wide CRISPR drug resistance screen, revealing the power of phenotypic screens in uncovering the mechanism of action of MGDs [64]. Site-specific ligand-incorporation-induced proximity technology enables the incorporation of unnatural amino acids into proteins, allowing the identification of PROTACable sites and facilitating the discovery of new TPD effector ligands [144].

CRISPR-based E3 ligase knockout/knockdown screens are systematically used to identify the E3 ligases required for MGD-induced degradation of specific substrates [38,145]. This method involves creating a comprehensive small guide RNA (sgRNA) library targeting over 700 human E3 ligases and DUBs, including newly designed sgRNAs targeting specific functional domains to enhance gene loss-of-function efficiency [38,145]. HAP1 cell lines stably expressing doxycycline-inducible Cas9 are infected with the sgRNA library, and gene knockout/knockdown is initiated [145]. Cells are cultured with or without selected MGD compounds, and each sgRNA’s relative abundance is measured by next-generation sequencing. Statistical analysis identifies E3 ligases whose ablation impacts MGD-induced degradation. CRISPR-based E3 ligase knockout/knockdown screens are validated with complementary experiments, including competitive growth assays, flow cytometry, cell cycle analysis, immunofluorescence, and western blot [146]. This robust approach is further validated through in vitro and in vivo experiments, which explore the functional significance of E3 ligases in complex biological contexts, such as tumor growth and immune modulation. Further insights into cellular and immunological roles are gained through additional assays, including western blot, flow cytometry, RNA-seq, ATAC-seq, ChIP-seq, and single-cell RNA-seq, all supported by advanced bioinformatics tools.

Quantitative proteomics is a powerful tool for understanding how MGDs influence protein degradation and the role of E3 ligases in this process. Methods such as tandem mass tag (TMT) mass spectrometry (MS), stable isotope labeling by amino acids in cell culture (SILAC), and label-free quantification (LFQ) with data-dependent and data-independent acquisition are employed to globally monitor protein degradation patterns [147]. These approaches accurately quantify protein expression across various samples, allowing researchers to measure protein degradation rates and observe how MGDs impact protein turnover. For example, the isotope-coded affinity tags method involves extracting proteome samples separately before stable isotope labeling, whereas the SILAC method labels proteomes within cells before extraction. Both methods rely on enzymatic digestion and MS for quantification and identification [147].

Subcellular proteomics facilitate the visualization of the spatial expression of E3 ligases and POIs within organelles, further pinpointing the involvement of specific E3 ligases. Additionally, ubiquitinome analysis is essential for identifying specific ubiquitination sites modified by degraders, thereby providing a direct link between MGDs and E3 ligase activity. Proximity labeling-based MS methods can also elucidate degrader-induced PPIs, offering critical insights into E3 ligase recruitment. While isobaric labeling techniques, such as tandem mass tags (TMT), offer significant advantages for multiplexing, achieving reliable quantification requires meticulous attention to every experimental step [148].

By combining quantitative proteomics with methods that enrich for ubiquitinated proteins, researchers can identify proteins that undergo ubiquitination and subsequent degradation in response to MGDs. This approach potentially implicates specific E3 ligases in the degradation process [149]. If a specific E3 ligase is suspected to be involved, its activity can be suppressed through siRNA-mediated knockdown or entirely eliminated via CRISPR-Cas9 knockout. The resulting alterations in protein degradation patterns are then monitored through quantitative proteomics [149,150]. Ultimately, specific E3 ligases are often associated with distinct protein degradation profiles; thus, observing changes in these patterns following the manipulation of E3 ligase activity provides critical insights into the ligase’s physiological function and its target proteins [4,149].

Ubiquitin-remnant profiling, thermal proteome profiling (TPP), and chemical proteomics are complementary techniques used to study protein ubiquitination [151,152,153]. Ubiquitin-remnant profiling (ubiquitome analysis) identifies ubiquitination sites on proteins [154,155]. It involves analyzing peptides generated after trypsin digestion, specifically looking for ubiquitin remnants, providing information on which proteins are ubiquitinated and at which specific sites. TPP identifies protein interactions by detecting changes in their thermal stability [156,157]. When proteins interact with MGDs, their thermal stability can be altered, providing insights into direct or indirect interactions between MGDs, ligases, and their substrates by observing shifts in protein melting temperatures. Chemical proteomics identifies the protein targets of small molecules, including ligases and their substrates [153,154]. This technique uses specialized small-molecule probes (covalent or affinity-based) to capture and identify specific proteins that have interacted with the probe, offering crucial insights into the direct interactors of E3 ligases and their potential substrates.

Genome-wide CRISPR-Cas9 screens serve as powerful tools for identifying genes that modulate cellular sensitivity or resistance to MGDs [145,158,159]. These screens employ the CRISPR-Cas9 system to knock out genes on a large scale, allowing researchers to observe subsequent changes in cellular behavior. This systematic approach can identify critical genes, such as E3 ligases and their essential cofactors, that influence therapeutic sensitivity to MGDs.

For example, in a study of glioblastoma (GBM), an E3 ligase sgRNA library was constructed to screen for genes affecting glioma cell growth. RNF185 was identified as a significantly depleted gene, correlating with decreased expression and favorable prognostic significance in patients. In vitro overexpression experiments further revealed that RNF185 functions as a tumor suppressor [159]. This observation highlights how genome-wide CRISPR-Cas9 screens can pinpoint specific E3 ubiquitin ligases whose inactivation alters cellular responses to various MGDs [145]. The experimental process involves constructing an extensive sgRNA library, packaging it into lenti-viral vectors, and transducing it into HAP1-Cas9 cells. Pooled CRISPR-Cas9 screens then track cell fitness in the presence or absence of specific small-molecule compounds. High-throughput sequencing of genomic DNA enables the precise quantification of changes in sgRNA abundance. These data undergo rigorous statistical analysis to identify genes whose loss significantly modifies the cellular response to the target protein degradation process [159,160,161].

RNA-seq serves as a powerful tool for uncovering transcriptional alterations associated with MGD treatment, potentially highlighting novel E3 ligases or affected signaling pathways. For instance, in Huntington’s disease, a proteogenomics approach coupled with alternative splicing analysis revealed widespread neuronal differentiation stage- and CAG length-dependent splicing changes. These findings improved the understanding of genes related to RNA processing, neuronal function, and epigenetic modifications associated with mutant HTT splicing [162]. Furthermore, RNA-seq analysis directly facilitates the development and mechanistic understanding of MGDs and other targeted protein de-graders. A systematic characterization of underexplored E3 ligases across seven dimensions-including chemical ligandability, expression patterns, and protein-protein interactions (PPIs)-through the analysis of 30 large-scale datasets led to the identification of 76 E3 ligases as promising candidates for degrader development [16]. In cancer research, developing new degraders like MDEG-541, a MYC PROTAC, has led to CRBN-dependent degradation of relevant cancer targets, implicitly relying on the understanding of target expression and E3 ligase activity often gleaned from RNA-seq data [163].

RNA-seq is also central to elucidating drug resistance mechanisms related to E3 ligases. In AML, transcriptome sequencing of wild-type and adriamycin-resistant HL60 cells identified differentially expressed genes associated with E3 ubiquitin ligases. This led to the development of a prognostic model comprising five genes, where high expression of UBE2L3 was identified as a reliable biomarker for drug resistance and poor prognosis in AML [164]. Similarly, in cervical cancer (CC), public databases providing mRNA expression and clinical patient data were utilized to develop a robust risk prediction model involving E3 ligase-associated genes, suggesting that targeting specific E3 ligases could be an efficient therapeutic strategy for CC [165] (Fig. 6).

In the context of GBM, analyzing RNA sequencing databases identified specific transcriptomic signatures. For instance, elevated expression of RNF7, TCEB1, SOCS1, and SOCS3, which encode components of the Cullin5-RING E3 ligase, was found to predict unfavorable GBM prognoses [166]. Recent advancements in MGDs, which leverage the CRBN E3 ubiquitin ligase to degrade GSPT1, emphasize the importance of understanding the cellular context for selective targeting, including the correlation between therapeutic response and cellular characteristics, such as CRBN expression, often assessed through transcriptomic analyses [167]. Beyond human diseases, the RNA-seq principles for E3 ligase characterization are observed in studies conducted in the halophyte Sesuvium verrucosum in response to salinity stress, providing a methodological blueprint for analyzing E3 ligase expression profiles under various conditions, including drug treatments [168].

High-throughput binding assays, including microscale thermophoresis (MST), surface plasmon resonance (SPR), and amplified luminescent proximity assay (AlphaLISA), are used to screen for direct interactions between MGDs, ligases, and substrates [169,170,171]. These methods rapidly assess binding affinities and kinetics, essential for understanding biological pathways and drug discovery.

MST measures the movement of molecules along a microscopic temperature gradient. Changes in thermophoretic movement upon binding allow for quantification of binding affinity [172,173]. MST is an immobilization-free technique that requires small sample volumes, making it suitable for precious or limited materials and versatile for various biological interactions [171,172,174,175].

SPR is a label-free technique that measures binding events by detecting changes in refractive index near a gold surface as molecules bind [176,177]. It provides real-time information on binding kinetics, including association and dissociation rates, and can be used in a high-throughput array format [176,178]. SPR is widely used in drug discovery to characterize drug-target binding and elucidate structure-activity relationships [178]. Recent advancements in understanding MGD highlight SPR’s utility, including in exploring the phytohormone auxin as an MGD for TIR1 and AFB protein [179].

AlphaLISA is a bead-based proximity assay used in high-throughput screening due to its homogeneous format, high sensitivity, and ability to detect various interactions [169,180,181,182,183]. It functions by bringing donor and acceptor beads into proximity upon binding, leading to a luminescence signal. In TPD, AlphaLISA is essential for measuring ternary complex formation mediated by MGDs. While AlphaLISA and TR-FRET are commonly applied, a comparative study revealed that AlphaLISA was more susceptible to chemotype-dependent interference [184].

Fragment-based drug discovery (FBDD) identifies small, low-molecular-weight fragments that bind weakly but specifically to target proteins, including E3 ubiquitin ligases. These fragments can then be optimized into functional MGDs [185,186,187,188,189]. FBDD is particularly promising for MGD development as it excels at identifying weak interactions within shallow or cryptic pockets, which are crucial for mediating induced PPI [23,51]. Its synergy with structural biology techniques, such as X-ray crystallography and NMR spectroscopy, and computational chemistry tools provides atomic-level insights and optimizes lead compounds, central to MGD activity [185,190] (Fig. 7).

Covalent ligand discovery focuses on identifying electrophilic compounds that covalently modify reactive residues (e.g., cysteines) on E3 ligases, creating new opportunities for proximity-induced protein degradation [38,83,87,191,192]. These ligands form strong, often irreversible bonds, providing increased potency, a prolonged duration of action, and enhanced selectivity [38,193,194,195]. This strategy expands the range of targetable E3 ligases [28,195,196]. Fragment-based screening with reactive groups is a common method [197,198]. An example is the structure-based design targeting the SH2 domain of SOCS2, leading to the covalent inhibitor MN551 [199]. Covalent ligands can also be incorporated into PROTACs [191,196]. Activity-based protein profiling identifies cysteine-reactive small molecules that interact with E3 ubiquitin ligases like RNF4, displaying potential for broadening E3 ligase recruiters for TPD applications [200].

Live-cell E3 ligase activity assays using ubiquitin-binding domain (UBD) reporters offer a powerful tool for dynamic, real-time monitoring of polyubiquitin chain formation in an E3-dependent way, making them useful for screening E3 ligase inhibitors or activators [201,202,203]. These UBD reporters bind specifically to polyubiquitin chains, allowing the direct visualization and tracking of ubiquitin chain formation kinetics and dynamics within living cells [201,202]. The assays are E3 ligase-dependent and responsive to high-throughput screening for the rapid identification of compounds that modulate E3 ligase activity [204].

Scalable assays have been developed to identify compounds that alter the interactome of a specific E3 ligase by tracing its abundance after pharmacologically induced autodegradation and address the challenge of co-opting only a small fraction of E3 ligases for degrader development; this was validated for CRBN [77]. Cell-based target engagement assays evaluate the binding affinity of E3 ligase ligands in living cells, overcoming issues like poor cell permeability encountered in vitro [205]. Newer technologies, including NanoLuciferase- and HaloTag-based screenings, enable the determination of kinetics and stability of small-molecule-induced ternary complex formation with live cells [206].

Covalent functionalization and electroporation (COFFEE) are a novel method that involves the covalent labeling of recombinant E3 ligases via maleimide-thiol chemistry, followed by their introduction into cells via electroporation [28,36]. This technique allows the functional analysis of engineered or previously uncharacterized E3 ligases directly within a cellular context [36,37], bypassing the conventional requirement for identifying specific binders. Understanding the roles of E3 ligases in cellular regulation and developing TPD strategies is highly valuable, as demonstrated by the application of this method to various E3 ligases [207]. Furthermore, COFFEE aligns with broader efforts in TPD where covalent recruitment strategies are increasingly utilized. For instance, the discovery of a cysteine-reactive covalent recruiter (EN884) against the SKP1 adapter protein of the SCF complex illustrates that core components of Cullin-RING E3 ubiquitin ligase complexes can be exploited for PROTAC and MGDs applications [208]. These advancements collectively emphasize the potential of utilizing both functionalized E3 ligases and covalent recruiters to target diverse target proteins, thereby expanding the druggable proteosome.

Computer-aided drug design and artificial intelligence (AI)-driven drug discovery are central to modern MGD research, as they allow the structural exploration of ternary complexes that historically lacked crystallographic resolution. Accurate protein structures predicted by AlphaFold [209,210,211] and RoseTTAFold [212], along with experimental approaches like X-ray crystallography and cryo-electron microscopy (cryo-EM) [17], offer the structural framework required to model degrader-target-E3 ligase assemblies. Using these structures, computational methods, such as molecular docking, molecular dynamics simulations, and free-energy calculations (MM/PBSA, FEP), are used to position degrader molecules within PPI, resolve steric clashes, assess conformational flexibility, and estimate binding affinities [213]. These approaches help identify key interface residues and predict geometries that favor ubiquitin transfer, thus guiding rational degrader design [214].

AI and machine learning (ML) extend these capabilities by analyzing large degradomics and ligandability datasets to predict novel E3-substrate interactions and prioritize compound candidates. Tools like MetaDegron [215] and UbiBrowser [216] allow proteome-wide predictions of degron motifs and ligase-substrate compatibilities, while DL models trained on structural and chemical features predict degrader binding affinity, specificity, and off-target interactions [217,218,219]. Notably, ML models trained on cellular degradation data bridge the gap between initial ternary binding events and functional outcomes by predicting whether binding will translate into productive ubiquitination and target degradation [220,221]. Additionally, AI platforms assess physicochemical and pharmacokinetic properties to filter unsuitable compounds and accelerate lead optimization [222,223].

Beyond simple prediction, generative AI models represent a paradigm shift by enabling the de novo design of MGD. Graph neural networks and variational autoencoders can generate entirely new chemical scaffolds that are not merely derivatives of existing drugs. These models tailor candidate molecules to precisely match the structural features of E3 ligase binding pockets and target surfaces [224,225]. Furthermore, these models conduct multi-parameter optimization to balance potency, selectivity, solubility, and ADME/Tox characteristics [222,226,227], while AI-powered retrosynthesis planning ensures synthetic feasibility by proposing practical chemical routes [228]. Such generative strategies are particularly powerful for designing MGDs against novel E3 ligases, expanding beyond canonical systems like CRBN and DCAF15 [62,229]. This includes exploring unconventional mechanisms, such as CR8-mediated CDK12/cyclin K-DDB1 recruitment and BI-3802-induced BCL6 polymerization [110,114] (Fig. 8).

Because MGDs ultimately function by modulating PPIs, PPI computational modeling is also indispensable. Protein-protein docking predicts possible binding interfaces between targets and ligases, while hotspot analyses highlight residues that contribute most to binding free energy, and molecular dynamics simulations test the stability and ubiquitin accessibility of ternary complexes [213,227,230,231,232]. These models suggest which E3 ligases are structurally compatible with a given target [14] and support virtual screening pipelines that evaluate thousands of candidate molecules in silico, prioritizing those most likely to induce degradation [233,234].

Integrating SBDD, AI/ML-driven prediction, generative design, and PPI modeling offers a comprehensive computational toolkit for MGD discovery. These in silico methods collectively allow the rational construction of ternary complexes, the prediction of degradation outcomes, and the identification of new ligase-substrate pairs, while also accelerating lead optimization through early filtering of unfavorable compounds. By reducing experimental burden and development timelines, computational and AI-based pipelines are now indispensable for expanding the therapeutic potential of MGDs.

Degradation-focused quantitative proteomics tracks protein abundance changes upon MGD treatment to identify substrates that are selectively degraded. Stable isotope labeling through SILAC is a robust and efficient metabolic labeling technique, widely used for its accuracy in quantifying protein abundance and determining protein half-lives. It precisely identifies and quantifies isotopically labeled proteins and peptides, making it ideal for studying proteostasis and protein turnover [235,236,237]. Its precision is particularly valuable in studies focusing on cellular signaling dynamics [238]. TMT provides high multiplexing capabilities, allowing for simultaneous, time-resolved analyses of protein degradation and synthesis kinetics, even when combined with SILAC in hyperplexing methodologies [239]. While TMT enables the analysis of more samples concurrently, some studies note that it can have lower coverage than LFQ [238], which provides accurate and robust proteome-wide quantification, especially with methods like MaxLFQ. LFQ is particularly useful for detecting even subtle changes in protein levels across thousands of proteins, making it valuable for understanding cellular proteome remodeling [238,240].

Beyond confirming degradation, quantitative proteomics is crucial in the discovery and mechanistic understanding of MGDs. By profiling protein changes, researchers can pinpoint the specific targets of novel MGDs and unravel their exact mechanisms of action. For instance, chemoproteomic approaches combined with quantitative profiling have been central in identifying new covalent MGDs and their targets, including the oncogenic transcription factor NFKB1 [111]. Similarly, quantitative proteomics has been essential in confirming the degradation activity of MGDs toward specific targets, like cyclin K [110,225], and in elucidating the mechanisms of novel CRBN modulators, including their selective degradation of proteins like GSPT2 [18]. These quantitative methods are also fundamental for evaluating the potency and depth of protein degradation during the lead optimization phase of MGD development [149,241].

Ubiquitome profiling, utilizing ubiquitin-affinity purification coupled with MS, identifies proteins modified by ubiquitin. This method enables researchers to pinpoint potential substrates of ubiquitin ligases, particularly in response to stimuli such as MGDs. By enriching for ubiquitin-modified peptides and analyzing them via MS, researchers can identify specific proteins that undergo ubiquitination. This approach provides a comprehensive map of the ubiquitinated proteome, allowing for a deeper understanding of the biological processes and signaling pathways in which these target proteins are involved [242,243].

The BioE3 method exemplifies this strategy by using E3 ligase fusions, such as CRBN, to detect both existing and novel substrates. Specifically, this approach identifies neosubstrates like CSDE1 upon treatment with MGDs such as pomalidomide. This technique effectively confirms a major rearrangement of the endogenous ubiquitination landscape, providing direct evidence of how MGDs reprogram cellular machinery. By monitoring these shifts, researchers can map the broader impact of induced proximity on the ubiquitin-proteasome system [244]. Polyubiquitin enrichment (via TUBEs) was combined with LC-MS/MS to monitor how small molecules like MGDs alter cellular ubiquitination. This approach is crucial because it identifies specific ubiquitinated proteins and can pinpoint the exact lysine residues to which ubiquitin is attached, as indicated by the detection of non-degradative ubiquitination on UBE3A. It enables the precise characterization of compound-induced ubiquitin-mediated bioactivity [245,246].

Crosslinking mass spectrometry (XL-MS) is a robust technique used to identify direct or near-proximity interactions between proteins, especially transient complexes like those formed by E3 ligases and their substrates [247,248,249,250]. By using crosslinking agents to stabilize these interactions, XL-MS allows researchers to capture and analyze these normally short-lived complexes, providing valuable insights into PPIs and protein structures. XL-MS can pinpoint these newly formed interactions; for example, if an MGD brings a protein (protein A) not typically recognized by an E3 ligase (ligase X) into proximity, XL-MS can identify the cross-links between them, demonstrating that protein A has become a neosubstrate for ligase X.

Reverse proteomics begins with a known or candidate protein and aims to identify the specific E3 ubiquitin ligase(s) responsible for attaching ubiquitin to it. Analytical methods such as substrate trapping [251,252], proximity-dependent biotinylation [253,254,255], and CRISPR-Cas9 screening [89] allow for the identification of weak and transient interactions between E3 ligases and their target proteins. These techniques are particularly instrumental in uncovering the specific enzymatic machinery that regulates the stability and degradation of a given protein within the ubiquitin-proteasome system (Fig. 9).

Targeted functional screens are powerful tools for identifying proteins susceptible to MGD-induced degradation and understanding the regulatory mechanisms involved. CRISPR activation (CRISPRa) and CRISPR inhibition (CRISPRi) libraries, at the forefront of these genetic screens, are used to assess how susceptible proteins are to MGD-induced degradation by precisely controlling their expression levels [256]. These methods use modified Cas9 enzymes (dCas9). CRISPRa uses activating domains to enhance gene expression, while CRISPRi uses repressing domains to suppress it, all without altering the underlying DNA sequence [256]. The process involves introducing a library of CRISPR guide RNAs (gRNAs) into cells. Each gRNA is designed to target a specific gene, allowing for the systematic modulation of thousands of genes in parallel [257,258]. In the context of MGD research, cells are treated with MGD, and the impact of these CRISPR-mediated changes in gene expression on protein levels is measured [256].

If a protein is typically degraded by MGD, lowering its gene expression with CRISPRi would likely reduce its levels. In contrast, if a protein is not affected by the MGD, its levels should remain relatively stable regardless of whether CRISPRa or CRISPRi is employed [256]. These screening strategies offer crucial insights by facilitating the identification of several key components. First, they help pinpoint MGD substrates, which are the specific target proteins directly recognized and degraded by the MGD [256]. Second, these screens reveal regulatory factors, encompassing proteins that influence the kinetics or efficiency of the degradation process itself [256,259]. Finally, they highlight potential therapeutic targets, as identifying the cellular components critical for MGD function can reveal entirely new avenues for therapeutic intervention [256].

Protein array-based screening is a high-throughput method that allows researchers to test thousands of purified proteins for MGD-induced PPIs with a known E3 ligase, directly facilitating substrate discovery [260,261]. This method uses protein arrays, which are solid surfaces, typically glass slides, displaying a high density of immobilized proteins [260,261]. By simultaneously testing a large number of proteins against a specific E3 ligase, researchers can rapidly identify potential binding partners [260,262]. The exact definition of MGDs may vary depending on the research context; however, it generally signifies a specific cellular process or pathway in which E3 ligases are fundamentally involved. The ability to directly observe interactions between E3 ligases and their potential substrates on the array allows a faster identification of these critical protein partners [261,262,263]. Identifying E3 ligase substrates is crucial for understanding various cellular processes and disease mechanisms, as well as for potentially developing new therapeutic targets [202,263] (Fig. 10A).

Structural biology strategies provide atomic-level insights into the intricate interactions within MGD-induced complexes. Cryo-EM and X-ray crystallography are powerful methods that provide high-resolution visualization of ternary complexes formed by E3 ligases, MGDs, and neosubstrates [264,265,266,267,268]. These structural insights clarify the specific interface responsible for substrate recruitment and guide the rational design of new MGDs, as well as the selection of appropriate neosubstrates for TPD [269,270]. These two techniques are complementary: cryo-EM is excellent for visualizing large, flexible complexes and conformational changes induced by MGDs, while X-ray crystallography provides atomic details of smaller, well-ordered structures and high-resolution structures of glue-protein complexes [264,266,271,272].

Structural analysis of these ternary complexes reveals the precise interface where the MGD, E3 ligase, and neosubstrate interact. Visualizing this interface allows researchers to identify key residues or structural features vital for substrate binding and recognition [19,273,274]. The knowledge gained from these analyses can then be used for rational MGD design and substrate selection. Specifically, if a particular region of the neosubstrate is determined to be critical for binding, MGDs can be strategically designed to target that precise area. Furthermore, by understanding the binding preferences of the E3 ligase, researchers can identify additional neosubstrates that are more likely to be targeted for degradation [19,273] This structural insight transitions MGD development from empirical screening to a more predictable, SBDD paradigm (Fig. 10B).

Recent advancements in these fields have revealed how MGDs engage the E3 ligase and target protein for degradation at an atomic level [272]. For example, the CRBN E3 ligase’s recognition of the G-motif in target proteins is essential for how MGDs facilitate target degradation, effectively reprogramming the ligase’s substrate specificity to create a neosubstrate or degron [229]. This knowledge is widely applicable in drug discovery and development, where MGDs are emerging as a promising new class of therapeutics [264]. The two techniques are indispensable for understanding how MGDs bind to their target proteins and E3 ligases, and how this binding leads to conformational changes that trigger protein degradation or other cellular effects [19,273,274].

The therapeutic efficacy and safety of MGDs critically depend on their ability to mediate precise E3 ligase-substrate interactions. However, achieving this specificity remains significantly challenging. Many E3 ligases exhibit broad or incompletely characterized substrate repertoires, making it difficult to predict or prevent off-target degradation [285]. Structurally, this challenge arises from the shallow, solvent-exposed, and often flexible binding interfaces characteristic of E3 ligases, such as CRBN and DCAF15. These features permit promiscuous molecular recognition and small energy barriers between productive and nonproductive ternary complexes [286].

From a thermodynamic perspective, the cooperativity driving ternary complex formation is frequently marginal, with ΔΔG values small enough that subtle changes in MGD conformation or cellular conditions can redirect binding toward unintended substrates. Moreover, the dynamic nature of protein surfaces and the entropic contributions of weak hydrophobic contacts complicate the prediction of selectivity. These molecular properties explain why off-target effects persist even in rationally designed MGDs [286].

Recent efforts to address these limitations include high-throughput mapping of E3-substrate networks, structure-based design to enhance binding cooperativity, and computational modeling to predict alternative interaction geometries [35,51]. However, these approaches remain hindered by incomplete structural data and the lack of a comprehensive E3-substrate interaction atlas [35]. Thus, despite notable progress, achieving true selectivity remains a central translational obstacle for MGD therapeutics.

E3 ligase expression and activity are highly variable across different tissues, developmental stages, and disease states, profoundly influencing both the efficacy and safety of MGDs. For instance, a ligase that is abundant in hematologic malignancies may be inactive or absent in solid tumors [38,287]. This heterogeneity extends beyond expression levels to include subcellular localization, post-translational modifications, and competition with endogenous substrates [263,288]. Consequently, MGDs optimized in one cellular context may not function effectively in another.

While this variability poses a significant translational challenge, it also offers unique therapeutic opportunities for achieving tissue-specific degradation and reducing systemic toxicity [289]. Future studies integrating single-cell transcriptomics, proteomics, and spatial proteome mapping will be essential to precisely match E3 ligase availability with appropriate disease contexts [16,288,289,290,291]. By leveraging these insights, we can design MGDs that utilize ligases exclusively expressed in diseased tissues, thereby enhancing the therapeutic index and minimizing off-target effects in healthy cells.

A persistent limitation in MGD development is the incomplete mechanistic understanding of ternary complex formation. Biochemical assays can reveal degradation but not the structural determinants that govern selectivity or cooperativity [292]. High-resolution methods, including X-ray crystallography and cryo-EM, remain indispensable to define how MGDs reorganize E3 ligase-substrate interfaces at the atomic level [265,293,294,295,296].

Recent structural studies have revealed that minimal chemical modifications-such as altering hydrogen bond donors or aromatic substituents-can significantly impact the stability and productivity of ternary complexes [18]. Without such precise structural information, many promising MGDs may fail to achieve efficient ubiquitination of the target protein. This underscores the critical necessity for parallel structural and biochemical validation during the preclinical development phase.

Despite the vast potential of E3 ligases, their intrinsic structural properties-namely their large, featureless, and flexible protein surfaces-render most of them poorly druggable by conventional small molecules, traditionally limiting the usable repertoire to a handful of well-characterized ligases such as CRBN and MDM2 [28,29,289]. To circumvent these constraints, innovative approaches including covalent ligand discovery, FBDD and the use of engineered or viral E3 ligases with restricted tissue expression have been explored; however, each of these carries distinct translational challenges in terms of selectivity, reversibility, and clinical safety [23,51,185,297]. Consequently, improving the “ligase druggability landscape” remains a critical priority for the development of next-generation MGDs, implying the utilization of a wider array of E3 ligases for clinical application. While the limitations of CRBN-based MGDs are well established, the E3 ligase platform is rapidly expanding with promising alternatives, notably other DCAF ligases, among which the VHL ligase has rapidly emerged as a critical and actionable platform. The discovery of a small molecule that binds the HIF1-α-binding pocket on VHL and acts as an MGD by recruiting CDO1 as a neosubstrate into the VHL E3 ligase complex, clearly demonstrates that VHL has moved beyond a merely potential target [65]. Furthermore, an independent report identifying dGEM, a novel MGD that recruits the VHL E3 ligase to the neosubstrate GEMIN [298], further reinforces the reproducibility and clinical relevance of the VHL MGD approach. These two distinct, well-characterized VHL MGD examples suggest that claims regarding VHL’s addressability via MGDs are accurate at this juncture, which carries the significant implication of expanding the range of clinically targetable ligases beyond CRBN to novel E3 platforms.

Computational modeling has begun to inform the rational design of MGDs by predicting degron motifs, E3-substrate compatibilities, and binding cooperativity [14,213,227,299]. However, these models remain largely descriptive due to several critical limitations, such as incomplete training datasets, a lack of diverse structural templates, and the highly dynamic nature of cellular protein networks [35,39,84].

To improve predictive accuracy, future computational frameworks must integrate context-specific parameters. First, these models should incorporate cell type-specific E3 ligase expression profiles. Second, they must account for the impact of post-translational modifications on protein interactions. Finally, the models need to reflect the conformational flexibility inherent in ternary complex formation [300]. Until such integrative models are fully realized, computational predictions must be empirically validated through quantitative proteomics and structural analysis to ensure translational reliability. This synergy between in silico prediction and experimental validation is essential for the robust development of next-MGDs.

Resistance to MGDs has already been observed and may arise through several mechanisms, including mutations in the E3 ligase (e.g., CRBN) or the neosubstrate, the loss of essential cofactors such as DDB1 or CUL4, or the compensatory activation of redundant degradation pathways [21,58,80,167,301]. Comparative analysis of strategies to overcome such resistance reveals distinct advantages and trade-offs. First, ligase switching-the recruitment of alternative ligases-can bypass mutation-induced resistance; however, this approach requires compatible degron motifs and rigorous structural validation [29,302]. Second, ligand re-engineering aimed at improving binding cooperativity offers a modular approach to restore efficacy, though it may inadvertently broaden the substrate profiles. Third, combination therapies involving chemotherapy, targeted agents, or immunotherapy can mitigate adaptive resistance, yet they carry the risk of cumulative toxicity [121]. Finally, systems-level approaches, such as CRISPR-based functional genomics, can systematically identify resistance nodes and suggest potential compensatory pathways [89].

Critically, these strategies differ mechanistically and in terms of translational feasibility and safety. This diversity highlights the necessity for context-dependent optimization tailored to specific disease environments, rather than relying on one-size-fits-all solutions.

Despite over 600 putative human E3 ligases, most MGD research remains confined to a few canonical examples [36,67,303]. Expanding the ligase repertoire is thus essential to broaden therapeutic reach, overcome resistance, and achieve tissue- or pathway-specific control. Recent advances, including BioE3, COFFEE, and degron mapping, have enhanced the detection of transient ligase-substrate interactions [28,36,255,304]. Integrating these tools with multiomics datasets, including proteomics and single-cell transcriptomics, promises to accelerate the discovery of new degradable targets.

Moreover, the continuous identification of neosubstrates, such as CK1 α and other CRBN-dependent oncogenic proteins, extends the potential of MGDs beyond hematologic malignancies to solid tumors [16,17,121,289,305]. However, many newly reported targets lack validation of ternary complex formation or degradation kinetics, underscoring the ongoing need for stringent structural and biochemical confirmation.

Overall, while MGDs represent a transformative paradigm in cancer therapy, their successful clinical translation requires a deeper mechanistic understanding of E3-substrate specificity, structural cooperativity, and the evolution of resistance. A critical synthesis of successes and limitations, encompassing thermodynamics, ligand design, and cellular context, will be crucial for guiding the development of the next generation of safe and effective MGD-based anticancer agents.