How to Develop SMARCA2 PROTACs: Selective Targeting, Warhead Selection, and Linker Design

* Please be kindly noted that our services and products can only be used for research to organizations or companies and not intended for any clinical or individuals.

Consult with Our Experts

What Is SMARCA2, and Which Target Forms Are Suitable for PROTAC Development?

SMARCA2, also known as Brahma (BRM), is an ATP-dependent chromatin remodeling protein and a catalytic component of the SWI/SNF (switch/sucrose non-fermentable) protein complex. It uses energy from ATP hydrolysis to reposition nucleosomes, regulate chromatin accessibility, and coordinate transcription-associated processes. SMARCA2 belongs to the same protein family as SMARCA4 (BRG1), another catalytic ATPase subunit of SWI/SNF complexes. These two proteins are structurally related and generally occupy mutually exclusive positions within individual SWI/SNF complexes, allowing them to perform overlapping chromatin remodeling functions.

The relationship between SMARCA2 and SMARCA4 provides the main biological rationale for SMARCA2 PROTAC development. In particular, cells with loss-of-function SMARCA4 mutations or SMARCA4 deletion may become dependent on the remaining SMARCA2-containing chromatin remodeling machinery. This synthetic lethal relationship makes selective SMARCA2 degradation an attractive research strategy. Rather than inhibiting the catalytic activity of both paralogs, a selective PROTAC can potentially remove SMARCA2 protein while preserving SMARCA4 expression and activity. However, the presence of a SMARCA4 mutation alone does not establish target suitability. Researchers should first confirm whether SMARCA2 protein remains expressed, whether its relevant structural domains are intact, and whether the biological system demonstrates dependence on SMARCA2.

SMARCA2 contains several structural regions relevant to degrader development. Its N-terminal region contributes to protein interactions and complex assembly, while the helicase-related ATPase module contains the catalytic machinery responsible for ATP hydrolysis. The C-terminal bromodomain recognizes acetylated lysine residues on histones and other chromatin-associated proteins. This bromodomain has become the principal entry point for SMARCA2 PROTAC development because it provides a defined ligand-binding pocket with established small-molecule recognition chemistry. The ATPase domain offers an alternative binding strategy, although its degradation characteristics and paralog selectivity require separate consideration.

Table 1. SMARCA2 Structural Features and Their Implications for PROTAC Development.

What Have Previous SMARCA2 PROTAC Studies Revealed?

SMARCA2 PROTAC development has progressed from broad degradation of SWI/SNF ATPase subunits toward more selective degradation strategies based on ternary complex engineering. Early compounds established that bromodomain ligands could recruit SMARCA2 and SMARCA4 to VHL, but their similar binding profiles often resulted in degradation of both paralogs. Subsequent research demonstrated that the linker and the recruited E3 ligase can influence the relative orientation and stability of the target–PROTAC–E3 complex. More recent designs have used these interactions to achieve substantial SMARCA2 selectivity even when the original target-binding ligand recognizes SMARCA2 and SMARCA4 with similar affinity.

Table 2. Representative SMARCA2 PROTACs and Their Research Findings.

These studies reveal that SMARCA2 degradation efficiency and selectivity are separate optimization objectives. ACBI1 and AU-15330 established that strong bromodomain engagement and VHL recruitment can produce efficient degradation, but their activity against SMARCA4 and PBRM1 limits their suitability when selective SMARCA2 depletion is required. JQ-dS-4 further demonstrated that changing the target-binding region can alter which SWI/SNF components are affected, although ATPase-directed binding did not automatically provide paralog selectivity.

A947 and ACBI2 demonstrated a different design principle: the same general bromodomain recognition strategy can produce a more selective degrader when the linker and E3 recruitment geometry are optimized. YD23 extended this concept to CRBN recruitment, showing that a nonselective bromodomain ligand can be converted into a selective degrader through systematic linker structure–activity relationship studies. More recent compounds, including SMD-3236 and SMD-6346, further demonstrate how warhead structure, linker architecture, and E3 recruiter selection can be combined to improve degradation depth, paralog selectivity, and molecular properties.

• Binary affinity is not sufficient: similar binding affinity toward SMARCA2 and SMARCA4 does not prevent selective degradation if the complete ternary complexes have different stability and ubiquitination efficiency.
• Linker optimization is a selectivity strategy: linker length, branching, and conformational restriction can alter target–E3 orientation and determine whether a degrader favors SMARCA2 or SMARCA4.
• E3 ligase choice changes the design space: VHL and CRBN have both supported selective SMARCA2 degradation, but their recruiter structures and induced protein interfaces require different optimization strategies.
• Complex-associated effects require separate evaluation: PBRM1 degradation or changes in other SWI/SNF components may depend on the target-binding profile and the biological context rather than simply reflecting SMARCA2 degradation.

Bromodomain-Directed SMARCA2 PROTAC Development

Target Site: The Acetyl-Lysine Pocket and the BC Loop Surface Presented to the E3 Ligase

The SMARCA2 bromodomain is a compact structural module of approximately 110 amino acid residues. It adopts a left-handed four-helix bundle consisting of αZ, αA, αB, and αC helices. Two connecting loops, the ZA loop and BC loop, form the acetyl-lysine recognition pocket. This pocket normally recognizes acetylated lysine residues on histones and other chromatin-associated proteins, allowing SMARCA2 to associate with acetylated chromatin. Its defined shape, hydrophobic environment, and conserved hydrogen-bonding interactions provide a suitable starting point for small-molecule ligand development.

Several residues around the pocket are important for ligand recognition. Structural studies of SMARCA2 bromodomain complexes have highlighted Asn1464, Phe1409, and Tyr1421 as important interaction residues for representative aminopyridazinyl phenol and tricyclic ligand scaffolds. For example, the aminopyridazine group of ACBI1 forms key interactions around Asn1464, while its terminal phenol group contributes aromatic interactions with Phe1409 and Tyr1421. These contacts help anchor the target-binding component within the acetyl-lysine pocket.

For PROTAC development, however, the bromodomain should not be considered only as a ligand-binding cavity. The surrounding BC loop and adjacent protein surfaces can contribute to the protein–protein interface formed after E3 ligase recruitment. Depending on the exit vector and linker trajectory, these surfaces may become positioned toward VHL or another recruited ligase. This creates an opportunity to use the target surface itself as part of the selectivity mechanism. A suitable warhead should therefore preserve its essential pocket interactions while leaving enough space for the attached linker to project toward the E3 ligase without blocking the surrounding protein interface.

Warhead Selection: High-Affinity SMARCA2/4 Bromodomain Ligands and Their Paralog Binding Profile

The main challenge in bromodomain warhead selection is the high structural similarity between SMARCA2 and SMARCA4. Their bromodomains share approximately 96% sequence homology, and many ligand-binding residues are conserved. As a result, conventional structure-based optimization may improve affinity toward both proteins rather than produce a genuinely SMARCA2-selective ligand. For this reason, current degrader development has frequently used high-affinity ligands that recognize both paralogs and relied on the complete PROTAC architecture to generate selective degradation.

The aminopyridazinyl phenol scaffold represented by Gen-1 is an important example. Its 2-(6-aminopyridazin-3-yl)phenol framework provides a hydrogen-bonding and aromatic interaction pattern compatible with the acetyl-lysine pocket. The scaffold was used in YD23, where it was connected to a CRBN-recruiting component through linkers of different lengths and compositions. The resulting degrader showed strong SMARCA2 preference even though the parent ligand was not intrinsically selective between SMARCA2 and SMARCA4.

Tricyclic bromodomain-binding scaffolds provide another design direction. The structural development of PRT3789 used a tricyclic ligand that binds deeply within the bromodomain pocket and establishes interactions with Phe1409, Tyr1421, and Asn1464. The scaffold provided a suitable starting point for connecting the target ligand to a VHL recruiter. Rather than relying exclusively on improved binary affinity, subsequent optimization focused on the target-facing and E3-facing interactions of the complete degrader.

High-affinity spirocyclic SMARCA ligands provide an additional starting point for modern degrader development. Structural studies indicated that a piperidinyl substituent on the ligand could be exposed toward the solvent, making it a practical attachment region for linker installation. Related designs have been incorporated into potent VHL- and CRBN-recruiting SMARCA2 degraders. When selecting among these chemotypes, researchers should compare pocket occupancy, ligand orientation, solvent exposure, and the effect of derivatization on target engagement rather than ranking warheads only by their affinity values.

• Aminopyridazinyl phenol: consider when a well-established acetyl-lysine recognition pattern is desired. Its nonselective binding profile can still support selective degradation after linker optimization.
• Tricyclic bromodomain scaffold: consider when the ligand provides extensive pocket interactions and a suitable attachment position for VHL recruitment.
• Spirocyclic ligand scaffold: consider when the solvent-exposed piperidinyl region offers a practical linker exit vector and an opportunity to optimize molecular properties.
• Warhead selection criterion: confirm that linker-bearing intermediates retain bromodomain binding and that the complete PROTAC produces a different degradation profile from the parent ligand.

Linker Design: Exit Vector, Length, and Conformational Restriction in Bromodomain Degraders

Linker design is one of the most important determinants of SMARCA2-selective degradation. The linker must connect the bromodomain-binding ligand to the E3 recruiter without disrupting either binding interaction. More importantly, it must position the two proteins so that a productive ternary complex can form and expose suitable SMARCA2 lysine residues to the ubiquitination machinery. Because the SMARCA2 and SMARCA4 bromodomains are highly similar, small changes in linker geometry can substantially alter the relative stability of their respective ternary complexes.

ACBI2 provides a useful example of linker branching. Its development incorporated an additional methyl group into an apolar C5 linker. This modification restricted the degrader's conformational behavior and reduced cellular efflux, contributing to improved molecular exposure. ACBI2 achieved approximately 1 nM DC₅₀ against SMARCA2 and approximately 32 nM against SMARCA4 in the reported evaluation. The resulting approximately 30-fold degradation window demonstrates that a relatively small linker modification can affect both degradation selectivity and pharmacokinetic properties.

PRT3789 provides another example of conformational control. The introduction of a methyl substituent into the linker was designed to favor a productive SMARCA2–PROTAC–VHL arrangement over the corresponding SMARCA4 complex. This modification restricted the accessible degrader conformations and helped establish a more favorable interaction network with the SMARCA2-specific protein surface. The result illustrates why linker optimization should examine the three-dimensional arrangement of both proteins rather than treating the linker as an interchangeable spacer.

Spirocyclic and heterocyclic linkers provide additional opportunities to control molecular conformation. SMD-6346 illustrates how an oxygen-containing spirocyclic architecture can modify the basicity of a piperidine-containing region while retaining potent SMARCA2 degradation. In practical development, the most useful strategy is to establish a solvent-exposed attachment position first, prepare a focused linker series, and then compare flexible, branched, and conformationally restricted structures. Linker length should be evaluated together with polarity, ionization, solubility, and the direction in which the E3 recruiter approaches the target.

• Exit-vector selection: prioritize solvent-exposed substituents on the bromodomain ligand. Structural models should confirm that linker installation does not obstruct essential acetyl-lysine pocket interactions.
• Length optimization: compare short, intermediate, and extended linkers while keeping the warhead and E3 recruiter constant. Avoid assuming that longer linkers necessarily improve degradation.
• Branching and rigidity: introduce methyl branches, cyclic groups, or heterocyclic elements when the initial flexible linker produces weak selectivity or unfavorable molecular properties.
• Polarity and basicity: evaluate ether incorporation, nitrogen substitution, and heterocyclic composition to balance solubility, permeability, and linker conformation.
• Matched-pair evaluation: compare linker analogues using both SMARCA2 and SMARCA4 degradation curves to distinguish improvements in general potency from genuine paralog selectivity.

E3 Ligase Selection: VHL and CRBN Recruitment for Bromodomain-Directed Degradation

VHL has extensive precedent in bromodomain-directed SMARCA2 PROTAC development. ACBI1, AU-15330, A947, ACBI2, and PRT3789 demonstrate that VHL recruitment can support efficient degradation of chromatin-associated SMARCA proteins. VHL-recruiting structures generally use a hydroxyproline-containing ligand to engage the VHL protein complex. The attachment position and linker trajectory determine how the recruited ligase is oriented relative to the bromodomain surface. This is particularly important when the objective is to exploit SMARCA2-specific interactions that are not equally available in SMARCA4.

CRBN provides an alternative E3 recruitment strategy. YD23 uses a pomalidomide-derived CRBN-binding component connected to the Gen-1 bromodomain ligand through a chemically optimized linker. The study demonstrated that CRBN recruitment could produce more than 100-fold degradation selectivity for SMARCA2 over SMARCA4. SMD-6346 further demonstrated that a CRBN-based architecture can achieve potent and selective degradation with improved oral exposure. These examples establish CRBN as a practical alternative rather than merely a secondary comparator to VHL.

The two E3 systems should not be compared solely by their individual ligand affinities. VHL and CRBN present different protein surfaces, interaction geometries, and linker attachment opportunities. A VHL recruiter may support a favorable interaction with the SMARCA2 bromodomain surface in one architecture, while a CRBN recruiter may perform better with a different ligand orientation and linker composition. Researchers should therefore assess ternary complex stability, target ubiquitination, and cellular degradation for each complete molecule. CRBN-dependent changes in other proteins should also be monitored to distinguish intended SMARCA2 depletion from recruiter-associated effects.

Table 3. Design Strategies for Bromodomain-Directed SMARCA2 PROTACs.

ATPase-Directed SMARCA2 PROTAC Development

Target Site: Catalytic and Allosteric Surfaces of the SMARCA2 ATPase Module

The SMARCA2 ATPase module is the catalytic motor responsible for converting ATP hydrolysis into nucleosome movement and chromatin remodeling. Its structure contains two related helicase-like lobes that form an ATP-binding cleft at their interface. ATP binding and hydrolysis induce conformational changes within the catalytic module, allowing the protein to interact with DNA and reposition nucleosomes. These structural features create potential recognition sites for ATPase-directed ligands, including compounds that occupy the nucleotide-binding region or interact with nearby regulatory surfaces.

The catalytic ATP-binding cleft is the most straightforward starting point for ATPase-directed degrader design because it has a defined functional role and can be targeted by inhibitor-derived ligands. However, the ATPase domains of SMARCA2 and SMARCA4 are highly related. A ligand that binds the catalytic pocket may therefore recognize both paralogs, limiting the ability to achieve selective SMARCA2 degradation through warhead affinity alone. In addition, ATPase inhibition and protein degradation are different outcomes. An inhibitor can block ATP hydrolysis without necessarily positioning the target for efficient ubiquitination.

Allosteric surfaces around the two catalytic lobes and regions involved in nucleosome or DNA interaction provide additional exploratory opportunities. Such sites may allow ligand binding without directly competing with ATP, potentially changing the conformational state or accessibility of the ATPase module. However, these surfaces have not established the same level of PROTAC development precedent as the bromodomain pocket. Their suitability should be supported by direct ligand-binding evidence, structural accessibility, and experimental confirmation that ligand engagement can support productive E3 recruitment.

Warhead Selection: ATPase-Directed Ligands and the Consequences of Chromatin Retention

ATPase inhibitor-derived ligands provide the main starting point for this development route. JQ-dS-4 is a reported example of an ATPase-directed degrader that connects an inhibitor-derived target-binding fragment to a CRBN-recruiting component. This strategy demonstrated that ATPase-domain recognition can be converted into degradation of SMARCA proteins while avoiding the direct PBRM1 bromodomain recognition associated with some bromodomain-directed compounds. However, the reported approach did not establish strong selectivity between SMARCA2 and SMARCA4.

One important observation is that ATPase-directed degradation can proceed more slowly than bromodomain-directed degradation. A proposed explanation is that ATPase-bound SMARCA2 and SMARCA4 may remain associated with chromatin, whereas bromodomain engagement can alter their chromatin association and make the proteins more accessible to the degradation machinery. This difference suggests that target accessibility and complex context may influence the efficiency of ATPase-directed PROTACs. The precise contribution of chromatin retention should be examined experimentally rather than assumed for every ATPase-binding scaffold.

When selecting an ATPase warhead, researchers should prioritize a ligand with a confirmed binding mode and a chemical position that can tolerate linker installation. ATPase inhibitors with strong activity against both SMARCA2 and SMARCA4 may still be useful for developing dual degraders or establishing an initial structure–activity relationship. However, a selective SMARCA2 program requires additional evidence that the complete degrader can distinguish the two paralogs. If a candidate inhibits ATPase activity but produces little protein depletion, increasing the parent ligand affinity may not solve the problem. The more relevant questions are whether the conjugated ligand retains target engagement, whether the linker presents the E3 recruiter correctly, and whether the target remains accessible for ubiquitination.

• ATP-binding cleft ligands: prioritize inhibitor-derived scaffolds with established catalytic-pocket recognition and accessible positions for chemical derivatization.
• Allosteric ligand candidates: consider only when structural or biochemical evidence supports a defined binding surface and a reproducible binding mode.
• Paralog binding profile: measure SMARCA2 and SMARCA4 engagement separately. Similar ATPase-domain recognition should be treated as a selectivity challenge rather than evidence of selective degradation.
• Chromatin accessibility: distinguish catalytic inhibition from target removal and investigate whether chromatin-associated target pools show slower degradation kinetics.

Linker Design: Exit Vectors on ATPase Ligands and the Effect of Complex Context

ATPase-directed linker design must account for the position of the ligand within the catalytic cleft and the surrounding chromatin-associated protein environment. Unlike an isolated soluble domain, the ATPase module may be incorporated into a large SWI/SNF complex and positioned near nucleosomal DNA. A linker that appears geometrically suitable in an isolated protein structure may therefore encounter steric restrictions when the complete complex is considered. The attachment position should ideally project toward an accessible solvent-facing region rather than into the catalytic interface or a densely packed protein–DNA contact surface.

The JQ-dS-4 architecture provides an early example of connecting an ATPase-binding inhibitor fragment to a CRBN-binding moiety through an extended chemical connector. It demonstrates that ATPase recognition can be retained within a heterobifunctional degrader, but it does not establish a universally optimal linker length or geometry for SMARCA2. Because degradation is relatively slow in this design, future optimization should compare the effect of linker length and composition on ternary complex formation, intracellular exposure, and degradation kinetics rather than assuming that the linker is already optimized.

A practical development strategy is to establish the warhead attachment position first and then prepare a focused series of linkers with different lengths and conformational properties. Flexible PEG segments may help accommodate the spatial distance between the ATPase domain and the E3 ligase, while alkyl or heterocyclic segments can modify flexibility, polarity, and molecular properties. Rigid linkers may improve the orientation of the recruited ligase in some structures, but they can also prevent productive complex formation if the attachment position is not suitable.

• Exit-vector mapping: identify substituents directed toward solvent in the ATPase ligand complex and confirm that derivatization preserves catalytic-pocket engagement.
• Complex-aware modeling: evaluate the linker in the context of the larger SWI/SNF assembly and chromatin-associated structure, not only an isolated ATPase domain.
• Flexible linker screening: compare a limited series of PEG and alkyl spacers to identify an appropriate distance and allow the E3 recruiter to approach the target.
• Conformational restriction: introduce heterocyclic or rigid segments only after identifying a productive attachment position and linker-length range.
• Kinetic evaluation: monitor early and late degradation time points because a slowly degrading ATPase-directed compound may be underestimated by a single endpoint measurement.

E3 Ligase Selection: Matching Recruitment Geometry to a Chromatin-Associated Target

CRBN is the documented E3 ligase recruitment system for the ATPase-directed degrader JQ-dS-4. Its glutarimide-containing ligand provides a recognizable E3-binding element, while the connected linker positions CRBN near the ATPase-bound SMARCA protein. This example establishes CRBN as a practical starting point for ATPase-directed degrader development. However, the slower degradation profile indicates that successful binary recruitment does not necessarily produce rapid target ubiquitination, particularly when the target remains associated with a larger chromatin remodeling complex.

VHL can be considered as a complementary E3 recruitment option, but its suitability should be established for the specific ATPase ligand and linker architecture. VHL-based success in bromodomain-directed SMARCA2 degraders cannot automatically be transferred to ATPase-directed designs because the target-binding region and exposed protein surfaces are different. A VHL recruiter may provide a favorable ternary interface in one target domain but an unfavorable orientation in another.

The first E3 comparison should therefore focus on the complete molecular architecture. Researchers can compare CRBN and VHL recruiters using matched warhead and linker structures, then evaluate ternary complex formation, target ubiquitination, and degradation kinetics. E3 expression and activity should also be confirmed in the selected biological system. When degradation is weak, it is important to determine whether the limitation comes from insufficient recruiter engagement, an unfavorable ternary orientation, limited target accessibility, or inadequate presentation of ubiquitination sites.

Table 4. Design Strategies for ATPase-Directed SMARCA2 PROTACs.

Paralog-Selective and Bystander-Controlled SMARCA2 Degrader Design

Target Site: Non-Conserved Contact Residues and the SMARCA2-Specific Ubiquitination Zone

The major structural challenge in selective SMARCA2 degradation is distinguishing SMARCA2 from its highly homologous paralog SMARCA4. Their bromodomains share a similar acetyl-lysine pocket, so a conventional warhead may recognize both proteins. Nevertheless, the surrounding protein surfaces are not completely identical. These differences can become important after the PROTAC recruits an E3 ligase and creates a new protein–protein interaction interface.

One important structural feature is the SMARCA2-specific Gln1469 residue, which has been implicated in interactions with VHL. Structural analysis showed that this residue can form hydrogen-bond interactions with VHL residues Phe91 and Asp92. The corresponding region in SMARCA4 contains Leu1545 rather than the SMARCA2 glutamine, preventing the same interaction pattern. This finding provides a direct structural explanation for how a degrader can achieve preferential SMARCA2 degradation even when its target-binding ligand recognizes both bromodomains.

Another important feature is an extended loop adjacent to the SMARCA2 bromodomain pocket, approximately corresponding to residues 1400–1416 in the reported structural analysis. This region is not equivalently represented in SMARCA4 and can contribute additional contacts with the recruited VHL complex. It also contains Lys1405, an important ubiquitination site identified in the development of PRT3789. Lys1445 was identified as another relevant ubiquitination site. These lysine residues provide a structural basis for understanding why the complete SMARCA2 ternary complex can support more effective ubiquitination than the corresponding SMARCA4 complex.

These observations suggest that target selection should consider two connected but distinct structural requirements. The first is a ligand-binding region that anchors the PROTAC to the bromodomain. The second is a protein surface that can interact with the recruited E3 ligase and expose suitable lysine residues for ubiquitination. A useful degrader does not simply bind SMARCA2 more tightly; it must position the ligase in a way that favors productive ubiquitination of SMARCA2 over SMARCA4. Species-specific sequence differences in the extended loop should also be considered when selecting protein constructs and experimental models.

Warhead Selection: Nonselective Binders That Produce Selective Degradation Outcomes

A key lesson from SMARCA2 PROTAC research is that a nonselective warhead can produce a selective degrader. This strategy is especially useful when the target and its paralog share highly conserved binding pockets and selective ligand discovery is difficult. Instead of attempting to create a warhead that completely excludes SMARCA4, researchers can select a ligand that binds both bromodomains and optimize the remaining molecular structure to favor SMARCA2-specific ternary complex formation.

A947 provides an early quantitative example. Its target-binding component recognizes the bromodomains of SMARCA2 and SMARCA4 with similar affinity, with reported K_d values of approximately 93 nM and 65 nM, respectively. Despite this similar binary recognition, A947 achieved a SMARCA2 DC₅₀ of approximately 39 pM and a SMARCA4 DC₅₀ of approximately 1.1 nM. This approximately 28-fold degradation window demonstrates that the E3-facing interface can create selectivity that is not present in the original warhead.

YD23 provides a complementary example using a CRBN-recruiting architecture. Its Gen-1-derived bromodomain ligand recognizes both SMARCA2 and SMARCA4, but systematic optimization of linker length and composition generated a compound with more than 100-fold degradation selectivity for SMARCA2 in the reported cellular evaluations. This result demonstrates that warhead selection should be based on the combination of target engagement, linker compatibility, and the resulting degradation profile.

PRT3789 provides a more direct example of structure-guided paralog selectivity. Its tricyclic bromodomain ligand anchors the compound within the target pocket, while the linker and VHL recruiter establish additional interactions with the SMARCA2-specific extended loop. The resulting ternary complex has a slower dissociation rate for SMARCA2 than for SMARCA4, supporting more effective target ubiquitination. This design illustrates why ligand affinity, ternary complex stability, and ubiquitination efficiency should be evaluated as separate but interconnected parameters.

• Nonselective aminopyridazinyl phenol ligands: use as a starting point when the bromodomain binding mode is well established, then optimize the linker and E3-facing interface to create selectivity.
• Tricyclic bromodomain ligands: consider when the scaffold provides strong pocket interactions and can position the linker toward the SMARCA2-specific surface.
• High-affinity spirocyclic ligands: consider when solvent-exposed attachment positions and suitable molecular properties support further linker optimization.
• Selection principle: prioritize the degradation selectivity of the complete molecule rather than requiring the parent ligand to be intrinsically selective.

Linker Design: Tuning Ternary Complex Geometry to Bias Degradation Toward SMARCA2

For paralog-selective SMARCA2 degradation, linker optimization should be directed toward a productive SMARCA2–PROTAC–E3 complex rather than simply maximizing the distance between the two binding components. The linker determines how the bromodomain-bound warhead presents the E3 recruiter to the surrounding protein surface. A small change in linker length, branching, or conformational preference can change the relative orientation of the two proteins and the accessibility of target lysine residues. This is particularly important when a nonselective ligand is used to distinguish two highly homologous paralogs.

The methyl-substituted linker used in PRT3789 illustrates conformational control. The methyl group was introduced to restrict the accessible degrader conformation and favor a productive interaction between the SMARCA2 bromodomain and VHL. The resulting complex showed a more favorable protein–protein interaction network and stronger SMARCA2 ubiquitination. This example suggests that conformational restriction can be more useful than indiscriminately increasing linker flexibility.

ACBI2 demonstrates the complementary value of linker branching. Its additional methyl group improved the degrader's conformational behavior and reduced efflux, while retaining potent SMARCA2 degradation and a measurable selectivity window over SMARCA4. SMD-6346 further illustrates how spirocyclic and oxygen-containing linker architecture can improve the balance between degradation potency, selectivity, and molecular properties. These examples indicate that linker optimization should consider both the protein interface and the physicochemical behavior of the complete degrader.

For a new SMARCA2 program, a practical strategy is to build a matched linker matrix using the same warhead and E3 recruiter. The first series can compare short, intermediate, and longer linkers. A second series can introduce methyl branching or heterocyclic elements around the most promising length. A third series can evaluate rigid or spirocyclic structures if the initial compounds show favorable target engagement but inadequate selectivity. This sequence helps identify whether the limiting factor is linker distance, conformational preference, or overall molecular properties.

• Control linker conformation: use methyl substitution or cyclic elements to restrict unproductive conformations when structural modeling supports a defined ternary orientation.
• Optimize the target-facing interface: ensure that the linker projects toward the SMARCA2-specific surface rather than interfering with the acetyl-lysine pocket.
• Preserve E3 accessibility: position the recruited ligase so that the extended loop and accessible lysine residues can participate in productive ubiquitination.
• Balance selectivity and exposure: compare branching, spirocyclic structures, ether incorporation, and nitrogen basicity to avoid improving ternary complex formation at the expense of cellular availability.
• Use matched paralog assays: evaluate SMARCA2 and SMARCA4 with the same concentration and time-course design to identify linker-dependent changes in the degradation window.

E3 Ligase Selection: How VHL and CRBN Recruitment Shift the Selectivity Window

VHL has particularly strong structural precedent for SMARCA2-selective degradation. A947, ACBI2, PRT3789, and SMD-3236 demonstrate different ways of combining bromodomain recognition with VHL recruitment. The most important advantage is not simply that VHL can ubiquitinate SMARCA2, but that its recruited surface can form additional interactions with SMARCA2-specific regions. The Gln1469 interaction and the extended-loop contacts provide examples of how the target itself contributes to the selectivity of the complete ternary complex.

PRT3789 further demonstrated that the stability of the ternary complex can affect the extent of target ubiquitination. Its SMARCA2 complex showed a slower dissociation rate than the corresponding SMARCA4 complex, allowing more favorable presentation of SMARCA2 lysine residues to the ubiquitination machinery. Lys1405, located within the SMARCA2-specific extended loop, and Lys1445 were identified as important ubiquitination sites. This observation provides a useful design principle: E3 recruitment should be optimized not only for protein proximity but also for the positioning of accessible ubiquitination sites.

CRBN offers a complementary route to paralog selectivity. YD23 demonstrated that a Gen-1-derived ligand connected to a CRBN recruiter through an optimized linker can selectively degrade SMARCA2. SMD-6346 further showed that a CRBN-based degrader can achieve a strong degradation window while improving molecular exposure through scaffold and linker optimization. These examples indicate that CRBN can provide an alternative interaction surface when a particular warhead and linker combination does not form a favorable VHL ternary complex.

The selection between VHL and CRBN should therefore be made using matched experimental comparisons. A candidate with a favorable VHL-based interface should not automatically be redesigned around CRBN, and a CRBN-based compound should not be rejected simply because an earlier VHL design performed better. Both systems can produce selective degradation, but the outcome depends on the target-binding scaffold, recruiter attachment position, linker geometry, and the availability of the E3 ligase in the selected biological context. Off-target protein changes should also be evaluated to distinguish direct degradation from secondary changes in SWI/SNF complex composition.

Table 5. Design Strategies for Paralog-Selective SMARCA2 PROTACs.

Common Failure Causes and Optimization Strategies for SMARCA2 PROTAC Development

Insufficient Selectivity Over SMARCA4 and PBRM1

Insufficient paralog selectivity is one of the most important challenges in SMARCA2 PROTAC development. Because the bromodomains of SMARCA2 and SMARCA4 share a highly conserved binding pocket, many warheads recognize both proteins. A compound may therefore show strong SMARCA2 binding but also induce substantial SMARCA4 degradation. PBRM1 presents an additional challenge because its fifth bromodomain can recognize related ligand scaffolds. A degrader designed without considering this binding profile may affect multiple SWI/SNF components rather than produce the intended target-specific response.

The optimization strategy should distinguish ligand selectivity from degradation selectivity. First, measure the binding profiles of the parent warhead against SMARCA2, SMARCA4, and relevant PBRM1 bromodomains. Next, compare matched PROTACs with different linker architectures and E3 recruiters. If the parent ligand binds both paralogs similarly but the complete molecule preferentially degrades SMARCA2, the selectivity is likely being generated by the ternary complex rather than the initial ligand interaction. Proteomic analysis and protein-level measurements can further identify direct and context-dependent bystander effects.

Unproductive Ternary Complex Formation and Inefficient Ubiquitination

Strong target binding and E3 recruitment do not guarantee efficient degradation. A PROTAC may form two stable binary complexes but fail to position SMARCA2 and the E3 ligase in a productive orientation. Even when a ternary complex forms, the relevant lysine residues may remain inaccessible to the ubiquitination machinery. This distinction explains why small changes in linker geometry can produce substantial differences in degradation activity even when the target-binding ligand and E3 recruiter remain unchanged.

Optimization should combine structural modeling with direct ternary complex and ubiquitination evaluation. Compare alternative linker lengths and attachment positions, assess whether the recruited E3 can approach the SMARCA2-specific surface, and investigate the accessibility of lysine residues such as K1405 and K1445 where the relevant structural mechanism applies. If a candidate shows strong binary binding but weak protein depletion, changing the ternary geometry may be more informative than simply increasing warhead affinity.

Poor Cellular Permeability, Solubility, and Intracellular Exposure

SMARCA2 PROTACs combine a relatively large target-binding ligand, an E3 recruiter, and a chemical linker. The resulting molecular size, polarity, hydrogen-bonding capacity, and ionization state can limit solubility and cellular permeability. Excessive linker flexibility or a highly basic nitrogen-containing region may also contribute to unfavorable distribution or efflux. These limitations can produce a compound with good biochemical activity but insufficient intracellular exposure to degrade SMARCA2 effectively.

The development of ACBI2 and SMD-6346 illustrates how linker and scaffold changes can address these issues. Methyl branching in ACBI2 contributed to reduced efflux and improved oral exposure, while oxygen incorporation into an optimized spirocyclic architecture in SMD-6346 helped reduce basicity and improve the overall property profile. For new candidates, optimization should consider molecular weight, topological polar surface area, hydrogen-bond donors and acceptors, ionization, solubility, permeability, and metabolic stability together. A structural change should be retained only when its effect on both degradation and molecular properties is supported by experimental data.

Incomplete Target Degradation and Context-Dependent Biological Responses

Incomplete SMARCA2 depletion can result from limited target accessibility, inadequate E3 recruitment, insufficient ubiquitination, rapid protein resynthesis, or differences in the composition of SWI/SNF complexes. Chromatin-associated SMARCA2 may also show different degradation kinetics depending on the target-binding region. In addition, the expression and activity of the recruited E3 ligase can vary between biological systems, meaning that a degrader optimized in one model may perform differently in another.

Evaluation should therefore extend beyond a single DC₅₀ measurement. Determine D_max, compare early and late degradation time points, and assess target recovery after compound removal where relevant. SMARCA2 and SMARCA4 should be quantified independently, while PBRM1 and other relevant SWI/SNF subunits can help identify broader complex-associated effects. Finally, distinguish the degree of protein depletion from the downstream biological response. A strong degradation profile does not necessarily produce the same functional outcome in every SMARCA4-deficient model because residual complex composition and compensatory transcriptional programs may differ.

BOC Sciences Support for SMARCA2 PROTAC Development

BOC Sciences provides integrated support for SMARCA2 PROTAC projects, covering target feasibility assessment, ligand identification, linker and E3 ligase system design, custom synthesis, and mechanistic evaluation. Whether your project focuses on bromodomain-directed degradation, exploratory ATPase targeting, or selective degradation over SMARCA4, we can help develop a suitable molecular design and evaluation strategy.

SMARCA2 Target Assessment and Paralog Degradability Evaluation

• Assess SMARCA2 domain integrity, target expression, and the structural suitability of the selected protein form
• Compare SMARCA2 and SMARCA4 bromodomain or ATPase-domain structures to identify potential binding and selectivity determinants
• Evaluate the accessibility of SMARCA2-specific interaction surfaces and potential ubiquitination regions
• Assess target turnover, E3 ligase availability, and complex-associated factors to support target feasibility decisions

SMARCA2 Warhead Identification and Ligand Design

• Evaluate bromodomain-binding and ATPase-directed ligand scaffolds according to the selected target region
• Identify solvent-exposed exit vectors and assess how linker installation affects ligand recognition
• Compare nonselective SMARCA2/SMARCA4 ligands for their potential to generate selective degradation through ternary complex engineering
• Support molecular docking, protein-ligand structural analysis, and ligand optimization for target-specific degrader design

Linker and E3 Ligase System Design With Custom PROTAC Synthesis

• Develop matched linker libraries varying length, branching, polarity, flexibility, and conformational restriction
• Compare VHL and CRBN recruiter structures, attachment positions, and compatibility with the selected SMARCA2 warhead
• Design and synthesize custom PROTAC candidates using PEG, alkyl, aromatic, heterocyclic, and spirocyclic linker structures
• Provide purification and analytical characterization of synthesized degrader candidates for subsequent evaluation

Ternary Complex, Selectivity, and Developability Evaluation

• Evaluate SMARCA2–PROTAC–E3 ternary complex formation and compare alternative linker and recruiter architectures
• Assess target ubiquitination and quantify SMARCA2 degradation using DC₅₀, D_max, and time-dependent measurements
• Compare SMARCA2 and SMARCA4 degradation and monitor PBRM1 and other relevant proteins for potential bystander effects
• Evaluate solubility, cellular permeability, molecular stability, and other properties affecting intracellular degrader exposure

Table 6. BOC Sciences Products and Building Blocks for SMARCA2 PROTAC Development.

Explore featured products that can expand your research options and accelerate your next discovery.

Access end-to-end service solutions that help bring efficiency, flexibility, and expertise to your research pipeline.

News

• Protein Degradation with New Chemical Modalities: Successful Strategies in Drug Discovery and Chemical Biology
• Degrader Building Blocks for Targeted Protein Degradation
• Recent Progress of PROTACs Targeting Non-enzymatic Functional Proteins
• Discovery of Molecular Glue Degraders

Technical Information

• Proteolysis-Targeting Chimeras (PROTAC)
• Targeted Protein Degradation: The Next Frontier in Drug Discovery
• Overview of 12 Targeted Protein Degradation Technologies
• Peptide and Small Molecule Based PROTACs Technology