DUBTACs Explained: How Deubiquitinase-Targeting Chimeras Are Redefining Targeted Protein Stabilization
For nearly two decades, targeted protein degradation has dominated the conversation around proximity-inducing therapeutics. PROTACs and molecular glues taught the field how to hijack the cell’s ubiquitin-proteasome system to eliminate disease-driving proteins, and that success created an obvious question: could the same bifunctional logic run in reverse? Deubiquitinase-targeting chimeras, or DUBTACs, answer yes. Instead of tagging a protein for destruction, a DUBTAC recruits a deubiquitinating enzyme to strip away the ubiquitin marks that would otherwise send a protein to the proteasome — rescuing it from degradation instead of causing it [1]. That single mechanistic inversion opens an entire therapeutic territory that classical degraders cannot reach: diseases caused by too little of a protein, not too much.
This article explains what a DUBTAC is, how it works at the molecular level, how it compares with PROTACs and molecular glues, which targets have already been validated in the peer-reviewed literature, where the field’s real bottlenecks sit, and how DNA-encoded library (DEL) screening is positioned to accelerate discovery of the next generation of DUB-recruiting ligands.
What Is a DUBTAC?
A DUBTAC is a heterobifunctional small molecule: two ligands joined by a chemical linker, where one end binds a disease-relevant target protein and the other end binds a deubiquitinating enzyme (a “DUB”). Bringing the two together in a ternary complex lets the DUB act on the target, removing ubiquitin chains that would otherwise mark it for destruction. The name deliberately echoes PROTAC (PROteolysis TArgeting Chimera), the founding bifunctional degrader modality first described by Sakamoto and colleagues in 2001, which uses the identical bivalent design principle to recruit an E3 ubiquitin ligase instead [2].
To understand why that swap matters, it helps to recall how the ubiquitin-proteasome system (UPS) normally works. E3 ubiquitin ligases attach chains of ubiquitin — most consequentially chains linked through lysine 48 (K48) — onto substrate proteins. The 26S proteasome recognizes K48-linked polyubiquitin as a degradation signal and destroys the tagged protein. Deubiquitinating enzymes, of which the human genome encodes around 100, run this process in reverse: they hydrolyze the isopeptide bonds holding ubiquitin chains together, editing or erasing the degradation signal entirely. This bidirectional ubiquitin cycle is one of the cell’s primary mechanisms for controlling protein abundance, and it was characterized in foundational work by Hershko and Ciechanover, for which they later shared the Nobel Prize in Chemistry [3]. A DUBTAC is, in effect, a synthetic switch that lets a chemist redirect a chosen DUB’s natural editing activity onto a chosen target protein.
How DUBTACs Work: Mechanism of Action
The mechanism follows a clear sequence:
- Target engagement. One arm of the DUBTAC binds a surface pocket on the disease-relevant target protein — the same kind of ligand-discovery problem faced in any small-molecule drug discovery campaign.
- DUB recruitment. The other arm binds a deubiquitinase, typically at a ligandable cysteine or an allosteric pocket rather than the catalytic active site itself, since occupying the active site would block the enzyme’s function.
- Ternary complex formation. Because both ligands are tethered by a linker, the DUBTAC brings the target protein and the DUB into physical proximity, forming a three-way (ternary) complex analogous to the target–PROTAC–E3-ligase complex in degrader pharmacology.
- Deubiquitination. With the DUB’s catalytic domain now positioned next to the target’s ubiquitin chains, the enzyme hydrolyzes the K48-linked chains and removes the degradation signal.
- Stabilization. Stripped of its “destroy me” tag, the target protein escapes proteasomal recognition, its half-life increases, and it accumulates to higher steady-state levels than in the untreated cell.
The founding demonstration of this mechanism came from Nomura and colleagues at UC Berkeley and Novartis, who screened a covalent-fragment library against the K48-specific deubiquitinase OTUB1 and identified EN523, a ligand that binds a non-catalytic allosteric cysteine (C23) on OTUB1 without inhibiting its enzymatic activity. Linking EN523 to lumacaftor — an existing, clinically used corrector of the misfolded cystic fibrosis transmembrane conductance regulator (CFTR) — produced the first DUBTAC, which stabilized ΔF508-CFTR protein levels and restored chloride channel function in patient-derived cells more effectively than lumacaftor alone. The same study showed the approach generalizes: an OTUB1-recruiting DUBTAC also stabilized the tumor-suppressor kinase WEE1 in hepatoma cells, establishing that a single DUB recruiter can be paired with different target ligands to stabilize structurally unrelated proteins [1].
DUBTACs vs. PROTACs vs. Molecular Glues
DUBTACs belong to a broader family of induced-proximity, or “chemically induced proximity,” modalities that also includes PROTACs and molecular glues. All three exploit the same basic insight — that forcing two proteins into unnatural proximity can redirect one protein’s enzymatic activity onto the other — but they differ in design and in the pharmacological outcome they produce [4].
| Modality | Effector recruited | Outcome for the target | Representative example |
|---|---|---|---|
| PROTAC | E3 ubiquitin ligase (e.g., CRBN, VHL) | Ubiquitination → proteasomal degradation | Sakamoto et al., 2001 [2] |
| Molecular glue | Varies, often stabilizes a pre-existing or induces a novel protein–protein interface via a single small molecule | Degradation, stabilization, or altered activity, depending on the interface glued | Schreiber, 2021 [5] |
| DUBTAC | Deubiquitinase (e.g., OTUB1, USP7, USP28, USP1) | Deubiquitination → protein stabilization | Henning et al., 2022 [1] |
PROTACs and DUBTACs share a bivalent architecture — a target ligand and an effector ligand joined by a linker — so many of the same medicinal chemistry principles (linker length and geometry, ternary complex cooperativity, cell permeability, degrader/stabilizer “hook effect”) carry over between the two modalities [4][6]. Molecular glues are mechanistically distinct: a single, typically smaller molecule stabilizes or creates a protein–protein interface directly, without the explicit two-ligand-plus-linker design of a bifunctional chimera [5]. Where DUBTACs are unambiguous in outcome — they stabilize — some molecular glues can also drive degradation (as with the glues that recruit substrates to cereblon), so modality and outcome should not be assumed to map one-to-one.
Why Protein Stabilization Matters
The first two decades of targeted protein degradation were, almost by necessity, focused on diseases driven by a gain-of-function or overexpressed protein — an oncogenic kinase, a hormone receptor, a transcription factor active in the wrong context. PROTACs and molecular glues are extraordinarily well suited to that problem: remove the offending protein and the disease-driving signal disappears.
But a large share of human disease runs in the opposite direction. Loss-of-function mutations, accelerated turnover of an otherwise-functional protein, and premature degradation of tumor-suppressive proteins all cause disease through a deficit, not an excess, of protein activity. Classical small-molecule pharmacology struggles here, because there is often no enzymatic pocket to inhibit — the therapeutic goal is to have more of the protein around, correctly folded and functioning, not less. DUBTACs give medicinal chemists a direct chemical route to that goal: instead of searching for an activator of a protein’s existing function, they can intercept the degradation machinery upstream and let the stabilized protein do the rest [6]. As reviewed by Ma and colleagues, this reframes an entire category of previously intractable “undruggable” loss-of-function biology as addressable with the same bifunctional-molecule toolkit that made targeted degradation successful [6].
Validated DUBTAC Targets and Case Studies
Although the modality is only a few years old, a growing set of peer-reviewed studies has validated DUBTACs across genuinely different target classes and disease areas.
Cystic fibrosis — CFTR. The founding 2022 study remains the field’s clearest proof of concept: an OTUB1-recruiting DUBTAC built on the clinical CFTR corrector lumacaftor stabilized ΔF508-CFTR and restored chloride channel conductance in patient-derived cells, outperforming lumacaftor alone [1]. Because CFTR-directed correctors and potentiators are already an approved drug class, this example illustrates a direct, near-term path from an existing therapeutic ligand to a next-generation DUBTAC.
Innate immunity — cGAS. In 2025, Deng and colleagues reported the first-in-class DUBTAC directed at cyclic GMP-AMP synthase (cGAS), the cytosolic DNA sensor that initiates type-I interferon signaling. Their molecule stabilized and activated cGAS, demonstrating that DUBTACs can be used not only to rescue disease-associated loss-of-function variants but also to pharmacologically amplify a signaling protein’s normal activity by protecting it from turnover [7].
Metabolic signaling — AMPK. Liu and colleagues built a USP7-recruiting DUBTAC to stabilize AMP-activated protein kinase (AMPK), a central regulator of cellular energy metabolism, showing that the modality extends beyond OTUB1 to other well-characterized, ligandable deubiquitinases [8].
Oncology — tumor suppressors via USP28. Wang and colleagues designed USP28-based DUBTACs to stabilize tumor-suppressive substrates as an anticancer strategy, extending the modality’s DUB toolbox further and reinforcing that multiple, chemically distinct DUBs can each anchor a viable DUBTAC platform [9].
Oncology — tumor-suppressive E3 ligases. In a mechanistically elegant twist, Chen and colleagues used DUBTACs to stabilize E3 ligase proteins that are themselves tumor suppressors, showing that the modality can rescue components of the ubiquitin system itself when their loss — rather than a downstream substrate’s loss — is what drives disease [10].
Broadening the DUB toolbox — USP1. Most recently, Qian and colleagues demonstrated that USP1 can be harnessed as a DUBTAC effector for targeted protein stabilization, adding a fourth validated deubiquitinase (alongside OTUB1, USP7, and USP28) to the field’s expanding recruiter toolkit [11].
Taken together, these studies span cystic fibrosis, innate immunity, metabolic disease, and oncology, and they validate at least four structurally distinct DUBs as viable recruitment handles — evidence that the mechanism is general rather than a one-off curiosity tied to a single enzyme or target.
Design Challenges and Open Questions
DUBTACs inherit the standard medicinal chemistry liabilities of any bifunctional molecule — molecular weight, cell permeability, and oral bioavailability all need active management, just as they do for PROTACs [4][6]. Two challenges are more specific to this modality:
DUB ligand scarcity. E3 ligase pharmacology benefited enormously from a small number of highly ligandable, well-characterized ligases (cereblon, VHL) that anchored the first wave of PROTACs. The human deubiquitinase family is considerably larger — on the order of 100 enzymes across several structural subfamilies — but far fewer of them have a validated, selective small-molecule binder. Each new DUB recruiter (OTUB1, USP7, USP28, USP1, and others) currently represents a genuine discovery campaign in its own right rather than a reusable, off-the-shelf handle [6][11].
Selectivity and catalytic-site avoidance. Because occupying a DUB’s catalytic site would simply inhibit the enzyme rather than redirect it, useful DUB recruiters generally need to bind an allosteric or non-catalytic site — as EN523 does at OTUB1’s cysteine 23 — while still enabling productive positioning of the ternary complex. Finding such sites, and confirming they don’t cause off-target deubiquitination of unrelated substrates, adds a layer of complexity beyond standard ligand discovery [1][6].
Where the Field Stands
As of 2026, DUBTACs remain a preclinical, tool-compound-stage modality: no DUBTAC has yet advanced into human clinical trials. What has grown rapidly is the depth of peer-reviewed mechanistic validation — from a single 2022 proof of concept to independent, multi-target confirmation across at least four DUBs and four therapeutic areas within about three years [1][7][8][9][10][11]. That trajectory mirrors the early years of PROTAC development, where a comparable run of academic validation preceded the modality’s move into industry pipelines and, eventually, the clinic [4]. The current bottleneck is not proof of mechanism — it is the pace at which new, selective DUB-recruiting chemical matter can be discovered [6].
Discovering DUBTAC Components with DNA-Encoded Library Screening
That bottleneck is precisely the kind of problem DNA-encoded library (DEL) screening was built to solve. A DEL couples each small molecule in a combinatorial library to a unique DNA barcode, allowing libraries of hundreds of millions to billions of compounds to be pooled, affinity-selected against a target in a single experiment, and deconvoluted by sequencing rather than one-compound-at-a-time testing [12]. That scale is a direct answer to the DUB ligand scarcity problem described above: rather than starting from a small, hand-curated fragment set and hoping to find an allosteric binder, a DEL campaign can sample chemical space broadly enough to find rare, non-obvious pockets on a chosen deubiquitinase — exactly the kind of site EN523 occupies on OTUB1.
DEL screening is useful on both arms of a DUBTAC. Against the DUB itself, a selection can be designed to enrich for ligands at a specific allosteric or non-catalytic surface rather than the active site, and to counter-screen against related DUB family members for selectivity up front. Against the target protein, the same technology that has already been applied to PROTAC and molecular glue discovery for degrader development applies unchanged, since the target-binding ligand-discovery problem is identical in either modality.
Vipergen’s DEL platform is built specifically for this kind of target-first screening. YoctoReactor® synthesizes DEL libraries of hundreds of millions of drug-like small molecules with a purification step after each synthesis round, ensuring a reliable match between a library member’s chemical structure and its DNA barcode — a property that matters enormously when the hit rate against a difficult allosteric pocket is expected to be low. Binder Trap Enrichment® (BTE) performs solution-based selection against a purified protein target with a low false-positive rate, well suited to screening directly against a recombinant DUB domain. Cellular Binder Trap Enrichment® (cBTE) goes a step further, enabling selection directly in living cells without requiring a purified target protein at all — relevant for a deubiquitinase whose relevant conformation or post-translational state may be difficult to reconstitute outside a cellular context [13].
For a discovery program pursuing a next-generation DUBTAC — whether against one of the four validated DUBs or a novel one — this combination of massive library scale, code-compound fidelity, and both in-solution and in-cell selection formats addresses the central rate-limiting step the DUBTAC field currently faces: finding selective, allosteric, non-inhibitory ligands for deubiquitinating enzymes.
For the broader landscape of next-generation induced-proximity modalities beyond PROTACs — including RIPTACs, PHOTACs, and other emerging chimeras alongside DUBTACs — see Next-Generation Targeted Protein Degradation: Beyond Classical PROTACs. For the classical degrader side of the same design logic, see PROTACs and E3 Ligases: Expanding the Toolbox for Targeted Protein Degradation and How Molecular Glue is Transforming Drug Discovery.
How Vipergen Can Help
Vipergen applies its YoctoReactor, BTE, and cBTE technologies to proximity-inducing molecule discovery across the full spectrum of induced-proximity modalities — PROTACs, molecular glues, and emerging chimeras such as DUBTACs — as well as to traditionally difficult target classes including protein–protein interactions, transcription factors, and membrane proteins. For a DUBTAC or other stabilization-focused program, that means the ability to screen directly against a deubiquitinase target, in solution or in a cellular context, at a library scale designed to find the rare allosteric binders this modality depends on.
If you are exploring a targeted protein stabilization or DUBTAC discovery program, get in touch with Vipergen’s team to discuss how DEL screening can be applied to your target.
Frequently Asked Questions
A DUBTAC (deubiquitinase-targeting chimera) is a heterobifunctional small molecule that links a target-protein ligand to a deubiquitinase-recruiting ligand, bringing the two proteins together so the deubiquitinase removes degradation-signaling ubiquitin chains from the target and stabilizes it [1].
Both are bivalent molecules that form a ternary complex with an effector enzyme, but a PROTAC recruits an E3 ubiquitin ligase to add ubiquitin and degrade the target, while a DUBTAC recruits a deubiquitinase to remove ubiquitin and stabilize the target — opposite pharmacological outcomes from a similar design principle [1][2].
Key Takeaways
DUBTACs invert the logic that made PROTACs successful: rather than recruiting an E3 ligase to destroy a target, they recruit a deubiquitinase to protect one. That single change opens a therapeutic space — loss-of-function and premature-degradation disease biology — that classical degraders cannot address. Peer-reviewed validation has moved quickly, from a single 2022 proof of concept in cystic fibrosis to independently confirmed stabilization of targets across oncology, innate immunity, and metabolic disease, using at least four chemically distinct deubiquitinases as recruitment handles. The modality’s central bottleneck today is chemical matter: selective, allosteric, non-inhibitory ligands for deubiquitinating enzymes remain scarce relative to the well-trodden E3 ligase toolkit. DNA-encoded library screening — searching hundreds of millions of compounds against a DUB target in solution or directly in cells — is built for exactly that discovery problem, and is well positioned to help move DUBTACs from academic proof of concept toward drug candidates.
References
[1] Henning NJ, Boike L, Spradlin JN, et al. Deubiquitinase-targeting chimeras for targeted protein stabilization. Nat Chem Biol (2022), 18(4), 412-421. https://doi.org/10.1038/s41589-022-00971-2
[2] Sakamoto KM, Kim KB, Kumagai A, et al. Protacs: chimeric molecules that target proteins to the Skp1-Cullin-F box complex for ubiquitination and degradation. Proc Natl Acad Sci U S A (2001), 98(15), 8554-8559. https://doi.org/10.1073/pnas.141230798
[3] Hershko A, Ciechanover A. The ubiquitin system. Annu Rev Biochem (1998), 67(1), 425-479. https://doi.org/10.1146/annurev.biochem.67.1.425
[4] Békés M, Langley DR, Crews CM. PROTAC targeted protein degraders: the past is prologue. Nat Rev Drug Discov (2022), 21(3), 181-200. https://doi.org/10.1038/s41573-021-00371-6
[5] Schreiber SL. The rise of molecular glues. Cell (2021), 184(1), 3-9. https://doi.org/10.1016/j.cell.2020.12.020
[6] Ma Z, Zhou M, Chen H, Shen Q, Zhou J. Deubiquitinase-targeting chimeras (DUBTACs) as a potential paradigm-shifting drug discovery approach. J Med Chem (2025), 68(7), 6897-6915. https://doi.org/10.1021/acs.jmedchem.4c02975
[7] Deng Z, Chen L, Qian C, et al. The first-in-class deubiquitinase-targeting chimera stabilizes and activates cGAS. Angew Chem Int Ed (2025), 64(3), e202415168. https://doi.org/10.1002/anie.202415168
[8] Liu J, Hu X, Luo K, et al. USP7-based deubiquitinase-targeting chimeras stabilize AMPK. J Am Chem Soc (2024), 146(16), 11507-11514. https://doi.org/10.1021/jacs.4c02373
[9] Wang Z, Qian C, Xiong Y, et al. USP28-based deubiquitinase-targeting chimeras for cancer treatment. J Am Chem Soc (2025), 147(16), 13754-13763. https://doi.org/10.1021/jacs.5c01889
[10] Chen L, Deng Z, Xiong Y, et al. Deubiquitinase-targeting chimeras mediated stabilization of tumor suppressive E3 ligase proteins as a strategy for cancer therapy. J Am Chem Soc (2025), 147(33), 29875-29883. https://doi.org/10.1021/jacs.5c06306
[11] Qian C, Wang Z, Xiong Y, et al. Harnessing the deubiquitinase USP1 for targeted protein stabilization. J Am Chem Soc (2025), 147(17), 14564-14573. https://doi.org/10.1021/jacs.5c01662
[12] Goodnow RA Jr, Dumelin CE, Keefe AD. DNA-encoded chemistry: enabling the deeper sampling of chemical space. Nat Rev Drug Discov (2017), 16(2), 131-147. https://doi.org/10.1038/nrd.2016.213
[13] Vipergen. We present 3 technologies for DNA encoded libraries (DEL). Vipergen A/S. Accessed September 2026. https://www.vipergen.com/technology/