Design of Class I/IV Bromodomain-Targeting Degraders for Chromatin Remodeling Complexes

Abstract

Targeted protein degradation is an emerging technology that can be used for modulating the activity of epigenetic protein targets. Among bromodomain-containing proteins, a number of degraders for the BET family have been developed, while non-BET bromodomains remain underexplored. Several of these proteins are subunits in chromatin remodeling complexes often associated with oncogenic roles. Here, we describe the design of class I (BPTF and CECR2) and IV (BRD9) bromodomain-targeting degraders based on two scaffolds derived from pyridazinone and pyrimidine-based heterocycles. We evaluate various exit vectors and linkers to identify analogues that demonstrate selectivity within these families. We further use an in-cell NanoBRET assay to demonstrate that these heterobifunctional molecules are cell-permeable, form ternary complexes, and can degrade nanoluciferase–bromodomain fusions. As a first example of a CECR2 degrader, we observe that our pyrimidine-based analogues degrade endogenous CECR2 while showing a smaller effect on BPTF levels. The pyridazinone-based compounds did not degrade BPTF when observed through Western blotting, further supporting a more challenging target for degradation and a goal for future optimization.

Graphical Abstract

INTRODUCTION

Epigenetic regulation occurs through mechanisms that modify gene expression without changing the genomic sequence. One such process is chromatin remodeling, which involves alterations in the chromatin structure through changes in the nucleosome position or histone modification, eviction, or exchange.¹ Remodeling can occur via both ATP-dependent and ATP-independent mechanisms. The ATP-dependent processes are catalyzed by multidomain chromatin remodeling complexes classified into four families: SWI/SNF, ISWI, CHD, and INO80.² Dysregulation of these chromatin remodelers is often associated with oncogenic phenotypes.³ Several of these chromatin remodeling complexes also contain bromodomain-containing proteins. The mammalian SWI/SNF complexes GBAF (or ncBAF) and PBAF contain the class IV bromodomains BRD9⁴ and BRD7⁵ subunits, respectively. In addition, the GCN5 bromodomain stabilizes the SWI/SNF complex on chromatin.⁶ In the less-studied ISWI family,⁷ NURF recognizes chromatin through its largest subunit BPTF,⁸ and CERF contains the CECR2 bromodomain-containing protein.⁹ BPTF, CECR2, and GCN5 are members of the class I bromodomain family (Figure 1A), and their role in nucleosome remodeling makes them important targets for anti-cancer therapy.¹⁰

Figure 1.

(A) Part of the bromodomain phylogenetic tree, showing class I, II (BET), and IV bromodomains. Adapted in part with permission from ref 19. Copyright 2019 Springer Nature.¹⁹ (B) Previously reported non-BET bromodomain (BD)-targeting degraders (bromodomain binding moiety in blue; E3 ligase ligand in green) and their biological activity. (C) Pyridazinone- and pyrimidine-based scaffolds with their affinity values for BPTF.

Although not as well-characterized as BET inhibitors, several small-molecule inhibitors for class I and class IV bromodomains have been recently developed.¹¹ However, for many non-BET bromodomains, it remains unclear whether bromodomain inhibition alone will be effective to induce a significant phenotypic effect. In the case of BPTF, we¹² and others¹³ have shown that bromodomain inhibitors, when used as single agents, may be insufficient for anticancer therapeutic applications. We previously reported that BPTF inhibitors sensitize 4T1 breast cancer cells to the chemotherapeutic doxorubicin. Therefore, combination therapy may be a useful option in cases where bromodomain inhibition on its own proves to be ineffective.^12,14

An alternative pharmacological modality is targeted protein degradation, which has progressed rapidly for BET bromodomains.¹⁵ In contrast, few non-BET bromodomain targeting degraders have been reported (Figure 1B). These degraders have in some cases proven more effective than monovalent inhibitors. For example, the BRD9 inhibitor BI7273 demonstrated only modest effects in synovial sarcoma cells, but a degrader generated from the same scaffold (dBRD9) led to a greater therapeutic response.¹⁶ Recently, an orally bioavailable compound ACBI2 has also been developed against the BAF chromatin remodeling complex ATPase SMARCA2.¹⁷ Similar to dBRD9 development, inhibition of the highly homologous PCAF/GCN5 bromodomains by the small molecule GSK4027 was insufficient to recapitulate the effects of genetic knockouts in macrophages, motivating the development of the degrader GSK699.¹⁸ These studies highlight the significance of using protein degradation to target class I and class IV bromodomain-containing proteins in chromatin-remodeling complexes. Within class I, degraders for BPTF and CECR2 have yet to be reported prior to this study.

Heterobifunctional molecules also provide a new way to establish selectivity across different bromodomains. Gadd et al. showed that the BRD4 degrader MZ1 can induce protein–protein interactions with the E3 ligase, leading to more stable and cooperative ternary complexes for BRD4 over other BET bromodomains.²⁰ While designing selective small-molecule inhibitors for class I bromodomains remains challenging,²¹ forming distinct ternary complexes is a potential alternative for targeting specific members of the family.

Targeted protein degradation is a useful therapeutic tool given that it allows the full protein to be removed through sub-stoichiometric treatment of degrader compounds in an event-driven process. For BPTF, genetic knockdown studies have shown strong downstream phenotypic effects. Richart et al. demonstrated that BPTF knockdown resulted in decreased c-Myc recruitment to DNA.²² BPTF knockdown also decreased high-grade glioma growth in adult and pediatric models.²³ Another class I bromodomain, CECR2, was recently shown to drive breast cancer metastasis by regulating NF-κB activity, making CECR2 a possible target for treating metastatic breast cancer.²⁴ Given the therapeutic utility demonstrated by genetic knockdown studies, we anticipate that protein degradation approaches would be valuable for modulating the activity of these proteins and the larger nucleosome remodeling complexes they form.

Here, we describe the design and evaluation of first-generation degraders for several class I bromodomains and BRD9 in class IV using two different scaffolds (Figure 1C). We use a pyridazinone-based scaffold featured in BZ1, previously developed for binding to PCAF/GCN5²⁵ and BPTF,¹² and a pyrimidine-based scaffold derived from TP-238 as a dual BPTF/CECR2 chemical probe.^26,27 We establish exit vectors for linker attachment and explore ternary complex formation through both in vitro assays and in-cell NanoBRET. Focusing our efforts on BPTF, we use our in-cell NanoBRET assay to demonstrate the degradation of a designed nanoluciferase–bromodomain construct through both scaffolds and, subsequently, a full length nanoluciferase-BRD9 construct. Finally, we show the first examples of degradation of endogenous BPTF and CECR2 using the TP-238-based degraders. Surprisingly, the pyridazinone degraders used in this study did not degrade endogenous BPTF, indicating room for further optimization. Future work will look at further structure–activity relationships (SAR) with these new degraders for tailoring selectivity and efficacy within class I bromodomains to enable subsequent cellular studies.

RESULTS AND DISCUSSION

We previously reported compounds 1 and 2 as high-affinity pyridazinone-based ligands for the BPTF bromodomain.¹² Through X-ray cocrystal structures, we showed that with different positions of the amine in the tetrahydroisoquinoline ring, we can engage D2957, D2960, and E2954 in BPTF (PDB: 7RWQ and 7RWO). Therefore, for our first-generation pyridazinone degraders, we chose to attach linkers from the tetrahydroisoquinoline analogues, allowing us to explore two distinct exit vectors (Figure 2A). We chose alkyl and PEG-based linkers to assess any potential effects on ternary complex formation and cell permeability.²⁸ Degraders 3–6 were synthesized according to Scheme 2 using pomalidomide-based cereblon-targeting E3 ligase ligands.

Figure 2.

BPTF BD (gray) with (A) compound 1 (yellow) and 2 (orange) overlaid; (B) compound 1 (yellow) and TP-238 (green) overlaid. (C, D) AlphaScreen binding isotherms for small molecules in Table 1, indicating that BPTF bromodomain binding activity is retained.

Scheme 2.

Synthesis of Compounds 3–6

We had also reported a crystal structure of TP-238 (7) with BPTF (PDB ID: 7KDZ).²⁹ Overlaying the cocrystal structures of TP-238 and compound 1 (Figure 2B) indicated that the pendant N(CH₃)₂ group in TP-238 provides a similar exit vector to the amine in compound 1, although the longer chain in TP-238 may provide more conformational flexibility. Using TP-238, we synthesized pomalidomide conjugates 8–10 (Schemes 6 and 7). Compounds 8 and 9 are composed of more rigid and polar piperazine linkers, with alkyl and PEG versions, respectively. To study the effect of an alkyl linker in lieu of the piperazine ring, we designed compound 10 as the closest analogue of TP-238.

Scheme 6.

Synthesis of Compounds 8 and 9

Scheme 7.

Synthesis of Compound 10

Biophysical Characterization of Pyridazinone and TP-238-Based Degraders.

To validate our chosen exit vectors, we used an AlphaScreen assay to measure the binding affinity of our degraders with the BPTF bromodomain. This AlphaScreen competition assay has been previously described by Ycas et. al²⁹ and Olson et al.³⁰ The assay conditions were previously optimized using an H4Kac₄ histone peptide, where the IC₅₀ was found to be comparable to the reported K_d value in the literature.³⁰ In a subsequent study, this assay was used to determine the IC₅₀ of the lead compound BZ1 and compared to the K_d value obtained through BROMOscan.¹² Compounds 1 and 2 were also tested in this study and can be compared to the IC₅₀ of BZ1. For all AlphaScreen experiments for this manuscript, BZ1 was run as a positive control along with the test compounds to ensure that the relative IC₅₀ values are representative of the affinity under the assay conditions. A representative trace of BZ1 is shown in Figure 2C. The CECR2 assay was benchmarked using TP-238 (compound 7 in Figure S2B), and the BRD9 assay conditions were optimized using BZ1 (Figure S1A; K_d previously determined in ref 12).

Heterobifunctional molecules from both scaffolds were high-affinity binders of BPTF (Figure 2C,D and Table 1), establishing that our exit vectors do not significantly perturb binding to the BPTF bromodomain. Notably, the bifunctional pyridazinone compounds exhibited 3- to 10-fold stronger binding affinity than the monovalent ligands. We tested the warhead attached to the linker (without the CRBN-recruiting ligand) counterparts of compounds 3 and 4 and found them to be stronger in affinity than the warheads alone (33 and 64 nM, respectively; data not shown). This may be the effect of adding the linker, which results in a tertiary amine and also provides another amine group for potential interactions with acidic side chains in the protein. This would be consistent with the structural hypothesis of amine groups engaging the acidic triad in BPTF previously described.¹² While we would expect some of these interactions to be mitigated by conversion to an amide upon pomalidomide addition, there may be additional interactions due to the pomalidomide end of the compound finding a surface on the protein to engage with. Similar increases in affinity for full PROTAC structures versus parent inhibitors have been previously observed.³¹

Table 1.

AlphaScreen IC₅₀ Values for the Inhibition of BPTF, BRD9 and CECR2 Bromodomains

		BPTF BD IC50 (nM)a	BRD9BD IC50 (nM)b	CECR2BD IC50 (nM)b
1		25012	556	1670
2		37012	300	1300
3		24 ±9	21	53
4		158 ±55	98	485
5		70 ±23	63	225
6		121 ±38	18	1460
7		43028	ND	33
8		146 ±12	844	24
9		267 ±8	996	33
10		360 (N=l)	ND	ND

^aAverage of two technical replicates with N = 3 and SD.

^bAverage of two technical replicates with N = 2 unless otherwise indicated. ND = not determined. Binding isotherms are shown in Figure 2C and Figures S1 and S2.

We further characterized the affinity of these compounds against the BRD9 and CECR2 bromodomains, which are significant off-targets of the two scaffolds (Figure S1, Figure S2, and Table 1). As observed previously with pyridazinone-based compounds, they retained binding to BRD9 and CECR2, although compound 4 and 6 demonstrated some attenuation in affinity to CECR2. Consistent with the previously reported TP-238 binding profile, our TP-238-derived conjugates displayed the highest affinity to CECR2 and, in the case of 8 and 9, slightly weaker (6–8-fold) binding to BPTF.^26,29 In contrast to the pyridazinones, these compounds were weaker affinity binders for BRD9.²⁶ The pyridazinone- and pyrimidine-based scaffolds therefore provided us with two unique selectivity profiles to target representative members of the class I and IV bromodomain families.

In Vitro Ternary Complex Formation via AlphaScreen.

The stability of the ternary complex is critical for degradation by heterobifunctional molecules.³² We therefore sought to study ternary complex formation in vitro using an AlphaScreen-based assay (Figure 3A). In this experiment, glutathione-coated donor beads and nickel acceptor beads were paired with GST-tagged bromodomains (BPTF, BRD9, CECR2, and PCAF) and His-tagged CRBN-DDB1. The AlphaScreen signal was measured at varying degrader concentrations, and the amplitude of the maximum point on the curve was used as an indicator of ternary complex stability. Small-molecule potency was indicated by the concentration at the mid-point of the bell curve.

Figure 3.

(A) In vitro AlphaScreen ternary complex formation assay workflow. Ternary complex formation with (B) BPTF, (C) BRD9, (D) CECR2, (E) PCAF bromodomains.

In this assay format, all pyridazinone-based degraders showed high ternary complex formation with BPTF (Figure 3B), consistent with the high affinity of the pyridazinone for this target (Table 1). To our surprise, while compound 4 seemed to be promiscuous, compound 3 showed a drop in signal for BRD9, CECR2, and PCAF (Figure 3C–E), indicating that the exit vector 1 (Figure 2A) in the alkyl linker series may help bias selectivity for BPTF over the other bromodomains tested in this assay. A similar albeit minor effect was observed in the PEG linker series in the case of PCAF (Figure 3E) but not for BRD9 or CECR2. Comparing the alkyl and PEG linkers, compounds 5 and 6 with PEG linkers generally seemed to form less stable ternary complexes for CECR2 and PCAF compared to BPTF and BRD9. Since these compounds do not demonstrate any significant loss of affinity with the bromodomain targets (Table 1), we hypothesize that there may be additional positive or negative cooperativity effects³³ during ternary complex formation contributing to this selectivity.

In contrast, we did not observe a significant difference in ternary complex stability between the alkyl and PEG linkers for the TP-238 analogues. Both compounds 8 and 9 showed similar ternary complex behavior with BPTF, BRD9, and CECR2 (Figure 3B–D). The signals for compound 10 were generally lower across all the proteins tested, highlighting the importance of the more rigid piperazine linker framework as opposed to alkyl chains in this scaffold. All TP-238 based degraders showed limited ternary complex formation with PCAF (Figure 3E). Based on these results, we focused on compounds 8 and 9 for further studies.

Comparing our pyridazinone and pyrimidine scaffolds, the pyridazinone degraders typically showed more potent ternary complex formation for BPTF and BRD9, while the TP-238 degraders were more potent for CECR2. In particular, compound 8 showed the highest potency for CECR2. This was consistent with the selectivity profile of the bromodomain-targeting moiety of these degraders. Encouragingly, apart from compound 4, all of the other degraders demonstrated a significant loss of ternary complex stability and potency for PCAF (and most likely GCN5 due to their high sequence similarity), indicating selectivity over this important class I bromodomain target.

In-Cell Ternary Complex Formation and Degradation of BPTF-BD Via NanoBRET.

After establishing in vitro ternary complex formation, we optimized a NanoBRET ternary complex assay to rank-order degraders in cells (Figure 4A).³⁴ This assay also serves as an indirect indicator of cellular permeability, which is often a major challenge for heterobifunctional degraders.³⁵ We designed both N- and C-terminal fusions of nanoluciferase (Nluc) and BPTF-BD with 11-amino acid linkers (see the Supporting Information for the design of the plasmid constructs). The vectors were transfected into HEK293T cells, and compounds 4 and 5 were dosed at 1 μM for 2 h in cells pre-treated with the proteosome inhibitor MG-132 to first determine ternary complex formation. We found that the C-terminal fusion construct showed ternary complex formation, while the N-terminal fusion did not with either compound (Figure S3). We thus used the C-terminal Nluc-BPTF-BD fusion construct for subsequent experiments.

Figure 4.

(A) NanoBRET assay format used to monitor ternary complex formation and degradation in HEK293T cells. (B) In-cell ternary complex formation was rank-ordered by determining fold-change in BRET ratio compared to the DMSO control. (C) Degradation of the Nluc-BPTF-BD construct was studied by measuring the total donor luminescence in the absence of MG-132. Degradation was rescued on pre-treatment with 10 μM MG-132.

We tested our degraders for ternary complex formation at 1 μM (Figure 4B). In this assay, we observed a trend within the pyridazinone degrader series, where the alkyl linker compounds (3 and 4) demonstrate a higher BRET ratio (3.5- and 3.2-fold increase over DMSO, respectively) compared to the PEG linkers (2.3- and 1.6-fold for compounds 5 and 6, respectively). For the TP-238 series, a higher BRET signal was observed for compound 8 (3.1-fold over DMSO control) compared to compound 9 (1.8-fold), providing more evidence that the alkyl linkers appear to be favored. These data show that our compounds are cell-permeable in this assay, making them useful starting points for further SAR.

We then monitored the degradation of the Nluc-BPTF construct by measuring the total donor luminescence after treatment of transfected HEK293T cells with 1 μM of the degraders for 6 h. We chose compounds 4 and 5 from the pyridazinone series as representative examples of alkyl and PEG linkers. We observed a significant decrease in donor luminescence with both our pyridazinone and TP-238 degraders compared to the untreated DMSO control, indicating degradation of the construct (Figure 4C). To validate the proteosome dependence, we pre-treated cells with 10 μM proteosome inhibitor MG-132. This pre-treatment resulted in rescue of degradation, with some additional stabilization of the donor signal, supporting a proteosome-dependent mechanism.

As luciferase enzyme-based assays are prone to interference due to potential inhibitors,³⁶ we were concerned about any off-target effects of our molecules on the Nluc-BPTF construct. To demonstrate that these degraders bind to the BPTF bromodomain and not Nluc, we pre-treated transfected HEK293T cells with the monovalent BPTF inhibitor 19 that we had previously characterized.¹² We tested this in the ternary complex assay format and observed a dose-dependent dissociation of the ternary complex with increasing concentrations of 19 (Figure S4). This further validated that the ternary complex formation is BPTF bromodomain-dependent.

In-Cell Ternary Complex Formation and Degradation of Full-Length BRD9 Via NanoBRET.

We further characterized our compounds for ternary complex formation and degradation of full-length BRD9 using NanoBRET. BRD9 is a reported off-target for the pyridazinone scaffold^12,37 and TP-238.²⁶ Due to its smaller size (75 kDa) compared to full-length BPTF (~300 kDa), we expected full-length BRD9 (BRD9-FL) to be more amenable to transfection and easier to study in this assay compared to full-length BPTF. We could also use dBRD9³⁸ as a positive control to validate our assay in this system. We used compounds 4, 5, 8, and 9 as examples of alkyl- and PEG linker-containing degraders. For the pyridazinone degraders, we observed a similar trend to our BPTF ternary complex data, where compound 4 showed a higher BRET signal than 5, which was comparable in magnitude to dBRD9 (2-fold over DMSO control) (Figure 5A). For the TP-238 series, the BRET ratios were slightly lower than dBRD9 (~1.5-fold over DMSO), which is consistent with the lower affinity of the TP-238 ligand for BRD9, although given the data in Table 1, we expected it to show a much lower signal. It is unclear whether the full-length BRD9 shows some additional affinity for this scaffold in the context of this experiment.

Figure 5.

In-cell NanoBRET experiments with BRD9-FL, indicating (A) ternary complex formation and (B) degradation of the Nluc-BRD9-FL fusion.

We then monitored the degradation of the Nluc-BRD9-FL construct under similar conditions as our BPTF study. Encouragingly, both compound 4 and 5 caused degradation of full-length BRD9, comparable to the dBRD9 control (Figure 5B). Consistent with their ternary complex behavior, compounds 8 and 9 also degrade BRD9-FL, albeit not as potently as dBRD9. These studies confirmed that both series of degrader molecules can engage full-length bromodomain-containing proteins in cells and induce their degradation. They also established the robustness of our NanoBRET assay, indicating that it is applicable to multiple bromodomain-containing constructs.

Degradation of Endogenous BPTF, CECR2, and BRD9.

With well-characterized compounds in hand, we tested their ability to degrade endogenous BPTF in HEK293T cells. We used the 6 h treatment condition from our NanoBRET assay and monitored the total amount of BPTF in lysates through Western blotting. Surprisingly, our pyridazinone degraders did not show degradation of endogenous BPTF or two class I bromodomain-containing proteins, PCAF, and CECR2 or the class IV bromodomain BRD9 under these conditions of up to 10 μM of the compounds (Figure S7). In contrast, the TP-238 series demonstrated a moderate effect with partial degradation between 0.1 and 1 μM concentrations against BPTF, with compound 8 demonstrating the higher potency (Figure 6A); however, inconsistent results were observed where additional Western blot experiments showed inconsistent and sometimes lower levels of degradation of endogenous BPTF (see Figure S10). Additionally, the NanoLuc tag in the NanoBRET assay may be enhancing the degradability of tagged BPTF by providing additional or more accessible lysine residues for ubiquitination, compared to the endogenous untagged protein. This effect has been previously reported in other cases with tags such as dTag and GFP.³⁹

Figure 6.

Western blotting analysis of HEK293T cells with treatment of (A) compounds 8 and 9 against BPTF for 6 h, (B) compound 8 against CECR2 for 6 h, and (C) compound 8 against BRD9 for 6 h.

Taken together with our NanoBRET data, we hypothesize that the minimal effect of these compounds on full-length BPTF in endogenous systems may be a cell line-dependent effect and the levels of functional BPTF available for degradation. In cells, endogenous BPTF is a subunit of the NURF complex, which may hinder its accessibility by heterobifunctional degraders. The cellular stability of the pyridazinone series also warrants further evaluation. The moderate effect observed via the TP-238 scaffold hints toward longer, pre-organized linkers (such as the piperazine moiety in compounds 8 and 9), which may be able to improve the activity of BPTF degraders.

Due to the modest effects observed in degrading endogenous BPTF, we sought to examine the main off-target proteins of our scaffolds. TP-238 is a high affinity inhibitor of CECR2, so we selected compound 8 as a potential degrader of CECR2 over our pyridazinones. Encouragingly, we observed significant degradation of CECR2 between 5 and 50 nM (Figure 6B) in HEK293T cells after 6 h of treatment with compound 8. A hook-effect was observed at higher concentration. The appearance of the CECR2 band at 5–50 nM due to protein resynthesis was observed after 72 h, while CECR2 was degraded efficiently at 500–5000 nM (Figure S8). These effects may result from the breakdown of 8 via hydrolysis of the cereblon (CRBN) motif or cleavage of the alkyl linker under physiological conditions. Western blot analysis of compound 8 against off-target BRD9 showed no significant degradation across various concentrations (Figure 6C).

To verify the mechanistic pathways of degradation, competition experiments were performed with CECR2 and cereblon-selective inhibitors as well as proteasome inhibitors. TP-238 and POM (pomalidomide) rescued CECR2 degradation at 1000 nM; additionally, CECR2 degradation was also impeded with proteasome and neddylation inhibitors MG-132 and MLN4924 (Figure 7). Together, these results support that degradation was dependent on CECR2 and proceeds via the ubiquitination and proteasome-dependent pathways.

Figure 7.

Western blot analysis of competition experiment in HEK293T cells for CECR2 with treatment of (A) compounds 8 and various concentrations of TP238/7 and POM (pomalidomide) and (B) compound 8 with various concentrations of MLN4924 and MG-132. For the full blots of each, see Figure S9.

Additionally, 8 was tested in the breast cancer cell line LM2 and lung cancer cell line A549 (Figure S10). In CECR2-overexpressed LM2 cells, noticeable degradation of exogenous CECR2 was observed between 5 and 500 nM, while a similar degradation pattern of endogenous CECR2 between 1 and 100 nM was observed in A549 cells, supporting the activity of this degrader across multiple cell lines.

Together, these results support the first degraders of the class I bromodomain-containing protein CECR2, shown to degrade CECR2 at low nanomolar concentrations. Future studies will include optimization of the CECR2 ligand exit vector, enabling the study of the functional effects of regulating CECR2 and the CERF complex, where selectivity of the heterobifunctional will be analyzed by quantitative proteomics. Optimization of our degrader to improve BPTF degradation will also be conducted in future studies, including the use of additional E3-ligase targeting ligands. Degradation of both of these proteins will provide useful information about the physiological effects of these proteins in normal physiology and pathophysiology.

CONCLUSIONS

We describe the design of heterobifunctional degraders targeting class I and BRD9 bromodomains using two different scaffolds with distinct selectivity profiles. We explore exit vectors based on tetrahydroisoquinoline-substituted pyridazinones, evaluating both alkyl and PEG linkers, and a rigid piperazine linker tethered to the TP-238 scaffold. Using an in vitro AlphaScreen assay to assess ternary complex formation, we discover that several of the pyridazinone analogues demonstrate a selectivity bias for BPTF and BRD9 over PCAF and CECR2. For the TP-238 degraders, the piperazine-based linker design is found to be more potent in this ternary complex analysis. We further use an in-cell NanoBRET assay to show that our degraders are cell-permeable and form ternary complexes with nanoluciferase-fused BPTF bromodomain and full-length BRD9 constructs. In the absence of a proteosome inhibitor, these fusion constructs are also degraded when treated with our degrader scaffolds. However, we find that our pyridazinone-based degraders do not show degradation activity for endogenous BPTF and caution against determining degradation effects based solely on artificial bromodomain constructs using NanoBRET. In contrast, the TP-238 degraders demonstrate a moderate effect on BPTF levels and a strong dose-dependent attenuation of CECR2 levels via Western blotting. In the case of BPTF and CECR2, 8 and 9 represent the first degraders of either protein. These tool compounds could provide a starting point to evaluate the effects of targeted protein degradation in cancer models after further ligand optimization and protein selectivity evaluation are performed. These studies also highlight the benefit of using multiple scaffolds with different selectivity profiles for initial studies of previously undegraded proteins. Areas for future investigation include the recently demonstrated role of CECR2 in breast cancer metathesis²⁴ and pancreatic cancer driven by high levels of BPTF and MYC.⁴⁰

EXPERIMENTAL SECTION

Materials and Methods.

All commercially available reagents were used without further purification. Flash column chromatography was performed on a Teledyne-Isco Rf-plus CombiFlash instrument with RediSep columns. NMR spectra were collected on a Bruker Avance III AX-400 or a Bruker Avance III HD-500 equipped with a Prodigy TCI cryoprobe. Chemical shifts (δ) were reported in parts per million (ppm) and referenced to residual solvent signals for chloroform-d (¹H 7.26 ppm), dimethyl sulfoxide-d₆ (¹H 2.50 ppm, ¹³C 39.5 ppm) and methanol-d₄ (¹H 3.31 ppm, ¹³C 49.0 ppm). Coupling constants (J) are in Hz. Splitting patterns were reported as s (singlet), d (doublet), t (triplet), q (quartet), and m (multiplet). High resolution ESI-MS spectra were recorded on a Thermo Fischer Orbitrap Velos equipped with an autosampler. Where stated, compounds were purified by reverse-phase high-performance liquid chromatography (RP-HPLC) on a C-18 column using 0.1% TFA water and CH₃CN as solvents, and TFA salts were quantified using the procedure described by Carlson et al.⁴¹

Purity Analysis.

All compounds tested in cells were ≥95% pure by RP-HPLC. Compounds 3–6 and 8–10 were run on a RP-HPLC with a C-18 column. A gradient of 0–40% ACN in 0.1% TFA H₂O over 60 min was used for compounds 3, 5, and 6, and 0–60% ACN in 0.1% TFA H₂O over 60 min for compounds 4, 8, 9, and 10. Spectral traces are shown in Figure S9.

Synthetic Methods.

The synthesis and characterization of compounds 1 and 2 were described previously.¹² TP-238 (7) was purchased from Cayman Chemical.

General Procedure A for the Synthesis of Intermediates 13–16.

Compound 1 or 2 (1.0 eq) was stirred in 1,4-dioxane at room temperature, followed by addition of the N-Boc linker (1.1 eq) and N,N-diisopropylethylamine (1.5 eq). The reaction mixture was heated in a sealed tube at 110 °C for 18 h. Following completion of the reaction, the 1,4-dioxane was removed by rotary evaporation. The crude mixture was extracted into ethyl acetate and washed with a saturated sodium bicarbonate solution (3 × 20 mL) and brine (20 mL). The organic layer was dried over magnesium sulfate, filtered, concentrated in vacuo, and purified by flash column chromatography (CombiFlash Rf system: 4 g of silica, DCM/methanol, 0–20% methanol, 30 min) (Scheme 1).

Scheme 1.

Synthesis of Intermediates 13–16

General Procedure B for the Synthesis of Compounds 3–6.

Step 1.

Compounds 13–16 were stirred in DCM at room temperature, followed by addition of trifluoroacetic acid (5.0 eq), and stirred at room temperature for an additional 2 h. Following completion of the reaction, the mixture was blown dry under a stream of nitrogen, and the crude product was used without further purification.

Step 2.

The crude product from Step 1 (1.0 eq) was stirred in dry DMF and DIEA (4.0 eq) at room temperature. A mixture of 12 (1.2 eq) and HCTU (1.2 eq) in dry DMF was then added, and the reaction mixture was stirred at room temperature for 16 h. A portion of the crude material was then purified by reverse-phase HPLC (Scheme 2).

tert-Butyl 2-((2-(2,6-Dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)oxy)acetate (11).

Compound 11 was synthesized according to literature procedures.³⁸ 2-(2,6-Dioxopiperidin-3-yl)-4-hydroxyisoindoline-1,3-dione (22) (240 mg, 0.88 mmol, 1.0 eq) was dissolved in DMF (2 mL) at room temperature, followed by the addition of potassium carbonate (180 mg, 1.30 mmol, 1.5 eq) and tert-butyl 2-bromoacetate (186 mg, 0.95 mmol, 1.1 eq). The reaction mixture was stirred at room temperature. After 2 h, the mixture was extracted into ethyl acetate and washed with water (20 mL) and brine (20 mL). The organic layer was dried over magnesium sulfate, filtered, concentrated in vacuo, and purified by flash column chromatography (CombiFlash Rf system: 24 g of silica, hexanes/ethyl acetate, 0–100% ethyl acetate, 16 min) to obtain a white solid (198 mg, 58% yield). ¹H NMR (500 MHz, DMSO): δ 11.11 (s, 1H), 7.80 (dd, J = 8.5, 7.2 Hz, 1H), 7.48 (d, J = 6.9 Hz, 1H), 7.38 (d, J = 8.1 Hz, 1H), 5.10 (dd, J = 12.8, 5.5 Hz, 1H), 4.97 (s, 2H), 2.89 (ddd, J = 16.9, 13.8, 5.5 Hz, 1H), 2.63−2.56 (m, 1H), 2.55−2.51 (m, 1H), 2.09−2.00 (m, 1H), 1.43 (s, 9H), H₂O from solvent at 3.32 ppm, ethyl acetate impurity at 4.02, 1.99, and 1.17 ppm (4.8% by weight).

2-((2-(2,6-Dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)-oxy)acetic acid (12).

Compound 11 was stirred in DCM at room temperature, followed by addition of trifluoroacetic acid (5.0 eq), and stirred at room temperature for an additional 2 h. Following completion of the reaction, the mixture was blown dry under a stream of nitrogen, and the crude product 12 was used without further purification (Scheme 3).

Scheme 3.

Synthesis of 2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)oxy)acetic acid (12)

tert-Butyl (5-(6-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)pentyl)-carbamate (13).

Following general procedure A (1 (63 mg, 0.22 mmol, 1.0 eq), tert-butyl (6-bromohexyl)carbamate (67 mg, 0.24 mmol, 1.1 eq), N,N-diisopropylethylamine (57 μL, 0.33 mmol, 1.5 eq), and 1,4-dioxane (1 mL)), product 13 was obtained as a brown solid (42 mg, 40% yield). ¹H NMR (500 MHz, DMSO): δ 8.61 (s, 1H), 7.61 (s, 1H), 7.08 (d, J = 8.1 Hz, 1H), 6.99 (dd, J = 8.1, 2.3 Hz, 1H), 6.97 (d, J = 2.2 Hz, 1H), 6.76 (t, J = 5.8 Hz, 1H), 3.60 (s, 3H), 3.51 (s, 2H), 2.89 (q, J = 6.6 Hz, 2H), 2.79 (t, J = 5.9 Hz, 2H), 2.62 (t, J = 5.9 Hz, 2H), 2.42 (t, J = 7.3 Hz, 2H), 1.50 (q, J = 7.2 Hz, 2H), 1.37 (s, 9H), 1.33−1.21 (m, 4H), H₂O from solvent at 3.32 ppm, DCM impurity at 5.76 ppm (2.5% by weight). ¹³C NMR (126 MHz, DMSO): δ 157.0, 155.6, 142.6, 136.1, 135.5, 132.1, 127.5, 127.3, 123.6, 121.3, 107.8, 77.3, 57.7, 55.2, 50.4, 29.5, 28.3, 26.6, 26.5, 26.2 (one resonance obscured by solvent, two resonances overlapping). HRMS (ESI-TOF) calculated for C₂₅H₃₇ClN₅O₃⁺ [M + H]⁺: 490.2579, observed 490.2546.

tert-Butyl (6-(7-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)hexyl)-carbamate (14).

Following general procedure A (2 (60 mg, 0.21 mmol, 1.0 eq), tert-butyl (6-bromohexyl)carbamate (64 mg, 0.23 mmol, 1.1 eq), N,N-diisopropylethylamine (54 μL, 0.31 mmol, 1.5 eq), and 1,4-dioxane (1 mL)), product 14 was obtained as a yellow solid (28 mg, 26% yield). ¹H NMR (500 MHz, DMSO): δ 8.61 (s, 1H), 7.60 (s, 1H), 7.11 (d, J = 8.1 Hz, 1H), 7.00 (dd, J = 8.1, 2.3 Hz, 1H), 6.93 (d, J = 2.3 Hz, 1H), 6.75 (t, J = 5.8 Hz, 1H), 3.60 (s, 3H), 3.51 (s, 2H), 2.89 (q, J = 6.6 Hz, 2H), 2.78 (t, J = 5.9 Hz, 2H), 2.63 (t, J = 5.8 Hz, 2H), 2.41 (t, J = 7.3 Hz, 2H), 1.50 (t, J = 7.2 Hz, 2H), 1.36 (s, 9H), 1.27 (m, 4H), H₂O from solvent at 3.32 ppm. ¹³C NMR (126 MHz, DMSO): δ 157.0, 155.6, 142.6, 136.2, 135.7, 131.4, 129.3, 127.5, 121.9, 121.7, 107.7, 77.3, 57.6, 55.4, 50.5, 29.5, 28.3, 26.6, 26.5, 26.2 (one resonance obscured by solvent, two resonances overlapping). HRMS (ESI-TOF) calculated for C₂₅H₃₇ClN₅O₃⁺ [M + H]⁺: 490.2579, observed 490.2547.

tert-Butyl (2-(2-(2-(6-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)-ethoxy)ethoxy)ethyl)carbamate (15).

Following general procedure A, (1 (60 mg, 0.21 mmol, 1.0 eq), tert-butyl (2-(2-(2-bromoethoxy)ethoxy)ethyl)carbamate (71 mg, 0.23 mmol, 1.1 eq), N,N-diisopropylethylamine (54 μL, 0.31 mmol, 1.5 eq), 1,4-dioxane (1 mL)), product 15 was obtained as a brown solid (53 mg, 49% yield). ¹H NMR (500 MHz, DMSO): δ 8.62 (s, 1H), 7.60 (s, 1H), 7.06 (d, J = 8.1 Hz, 1H), 6.99 (dd, J = 8.1, 2.3 Hz, 1H), 6.97 (d, J = 2.1 Hz, 1H), 6.75 (t, J = 5.8 Hz, 1H), 3.61−3.58 (m, 7H, overlapping resonances), 2.73−2.68 (m, 4H), 3.38 (t, J = 6.2 Hz, 2H), 3.06 (q, J = 6.1 Hz, 2H), 2.79 (t, J = 5.9 Hz, 2H), 2.70 (t, J = 5.8 Hz, 2H), 2.64 (t, J = 5.9 Hz, 2H), 1.37 (s, 9H), H₂O from solvent at 3.32 ppm. ¹³C NMR (126 MHz, DMSO): δ 157.0, 155.6, 142.6, 136.1, 135.3, 132.1, 127.5, 127.3, 123.6, 121.4, 107.8, 77.6, 69.7, 69.5, 69.2, 68.5, 57.0, 55.4, 50.7, 28.7, 28.2 (one resonance obscured by solvent, two resonances overlapping). HRMS (ESI-TOF) calculated for C₂₅H₃₇ClN₅O₅⁺ [M + H]⁺: 522.2478, observed 522.2444.

tert-Butyl (2-(2-(2-(7-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)-ethoxy)ethoxy)ethyl)carbamate (16).

Following general procedure A, (2 (75 mg, 0.26 mmol, 1.0 eq), tert-butyl (2-(2-(2-bromoethoxy)ethoxy)ethyl)carbamate (89 mg, 0.28 mmol, 1.1 eq), N,N-diisopropylethylamine (67 μL, 1.5 mmol, 1.5 eq), 1,4-dioxane (1 mL)), product 16 was obtained as a brown solid (39 mg, 36% yield). ¹H NMR (500 MHz, DMSO): δ 8.61 (s, 1H), 7.60 (s, 1H), 7.11 (d, J = 8.2 Hz, 1H), 7.00 (dd, J = 8.1, 2.3 Hz, 1H), 6.92 (d, J = 2.3 Hz, 1H), 6.74 (t, J = 5.8 Hz, 1H), 3.63−3.56 (m, 7H, overlapping resonances), 3.53−3.49 (m, 4H), 3.38 (t, J = 6.1 Hz, 2H), 3.05 (q, J = 6.0 Hz, 2H), 2.78 (t, J = 5.7 Hz, 2H), 2.71 (t, J = 5.8 Hz, 2H), 2.64 (t, J = 5.9 Hz, 2H), 1.36 (s, 9H), H₂O from solvent at 3.32 ppm, DCM impurity at 5.75 ppm (2.4% by weight). ¹³C NMR (126 MHz, DMSO): δ 157.0, 155.6, 142.6, 136.1, 135.7, 131.2, 129.3, 127.5, 121.9, 121.7, 107.7, 77.6, 69.7, 69.5, 69.2, 68.5, 56.9, 55.6, 50.8, 28.2 (overlapping resonances), (one resonance obscured by solvent, two resonances overlapping). HRMS (ESI-TOF) calculated for C₂₅H₃₇ClN₅O₅⁺ [M + H]⁺: 522.2478, observed 522.2442.

N-(5-(6-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)pentyl)-2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)oxy)acetamide (3).

Following general procedure B, compound 13 was Bocdeprotected in Step 1, and the crude material was used for Step 2: (29 mg, 0.07 mmol, 1.0 eq), 11 (30 mg, 0.09 mmol, 1.2 eq), HCTU (37 mg, 0.09 mmol, 1.2 eq), N,N-diisopropylethylamine (52 μL, 0.30 mmol, 4.0 eq), and DMF (0.5 mL). A portion of the crude product was purified by reverse-phase semi-prep HPLC (5–30% ACN in 0.1% TFA water over 25 min, C18 column) to obtain product 3 as a white solid (2× TFA salt). ¹H NMR (500 MHz, DMSO): δ 11.12 (s, 1H), 9.76 (s, 1H), 8.79 (s, 1H), 8.02−7.92 (m, 1H), 7.82 (dd, J = 8.5, 7.3 Hz, 1H), 7.66 (s, 1H), 7.51 (d, J = 7.2 Hz, 1H), 7.41 (d, J = 8.5 Hz, 1H), 7.23 (d, J = 8.3 Hz, 1H), 7.16 (dd, J = 8.3, 2.3 Hz, 1H), 7.13 (d, J = 2.2 Hz, 1H), 5.12 (dd, J = 12.9, 5.4 Hz, 1H), 4.78 (s, 2H), 4.57−4.51 (m, 1H), 4.26 (dd, J = 15.5, 8.1 Hz, 1H), 3.75−3.67 (m, 1H), 3.62 (s, 3H), 3.36−3.24 (m, 1H), 3.24−3.13 (m, 4H), 3.13−2.97 (m, 2H), 2.95−2.81 (m, 1H), 2.65−2.56 (m, 1H), 2.54 (d, J = 4.4 Hz, 1H), 2.09−1.99 (m, 1H), 1.78−1.67 (m, 2H), 1.58−1.43 (m, 2H), 1.39−1.27 (m, 4H). ¹³C NMR (126 MHz, DMSO): δ 172.8, 169.9, 166.7, 165.5, 157.0, 155.0, 142.0, 138.1, 136.9, 133.0, 132.5, 127.7, 124.7, 122.6, 121.8, 120.4, 116.8, 116.1, 109.2, 67.7, 55.1, 51.8, 48.8, 38.1, 30.9, 28.8, 25.7, 25.6, 25.0, 23.4, 22.0 (one resonance obscured by solvent, three resonances overlapping). HRMS (ESI-TOF) calculated for C₃₅H₃₉ClN₇O₇⁺ [M + H]⁺: 704.2594, observed 704.2552.

N-(6-(7-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)hexyl)-2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)oxy)acetamide (4).

Following the general procedure B, compound 14 was Bocdeprotected in Step 1, and the crude material was used for Step 2: (13 mg, 0.03 mmol, 1.0 eq), 11 (13 mg, 0.04 mmol, 1.2 eq), HCTU (17 mg, 0.04 mmol, 1.2 eq), N,N-diisopropylethylamine (23 μL, 0.13 mmol, 4.0 eq), and DMF (0.5 mL). A portion of the crude product was purified by reverse-phase semi-prep HPLC (5–40% ACN in 0.1% TFA water over 25 min, C18 column) to obtain product 4 as a white solid (2× TFA salt). ¹H NMR (500 MHz, DMSO): δ 11.12 (s, 1H), 9.84 (s, 1H), 8.79 (s, 1H), 7.97 (t, J = 5.5 Hz, 1H), 7.82 (dd, J = 8.5, 7.3 Hz, 1H), 7.64 (s, 1H), 7.51 (d, J = 7.2 Hz, 1H), 7.40 (d, J = 8.5 Hz, 1H), 7.28 (d, J = 8.3 Hz, 1H), 7.18 (dd, J = 8.2, 2.3 Hz, 1H), 7.08 (d, J = 2.3 Hz, 1H), 5.12 (dd, J = 12.9, 5.4 Hz, 1H), 4.78 (s, 2H), 4.63−4.49 (m, 1H), 4.27 (dd, J = 15.7, 8.0 Hz, 1H), 3.72 (d, J = 12.4 Hz, 1H), 3.62 (s, 3H), 3.35−3.25 (m, 1H), 3.24−3.12 (m, 4H), 3.12−2.99 (m, 1H), 2.96−2.83 (m, 1H), 2.68−2.55 (m, 1H), 2.54 (d, J = 4.5 Hz, 0H), 2.10−1.95 (m, 1H), 1.81−1.66 (m, 2H), 1.55−1.40 (m, 2H), 1.36−1.22 (m, 4H). ¹³C NMR (126 MHz, DMSO): δ 172.8, 169.9, 166.7, 166.7, 165.5, 158.1, 157.0, 155.0, 142.1, 137.1, 136.9, 133.0, 129.6, 129.5, 127.7, 123.1, 120.9, 120.4, 116.8, 116.1, 109.0, 67.7, 55.1, 51.8, 49.0, 48.8, 38.1, 30.9, 28.8, 25.8, 25.6, 24.5, 23.4, 22.0 (one resonance obscured by solvent). HRMS (ESI-TOF) calculated for C₃₅H₃₈ClN₇O₉⁺ [M + H]⁺: 735.2420, observed 704.2554.

N-(2-(2-(2-(6-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)ethoxy)-ethoxy)ethyl)-2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)oxy)acetamide (5).

Following general procedure B, compound 15 was Boc-deprotected in Step 1, and the crude material was used for Step 2: (27 mg, 0.06 mmol, 1.0 eq), 11 (25 mg, 0.08 mmol, 1.2 eq), HCTU (32 mg, 0.08 mmol, 1.2 eq), N,N-diisopropylethylamine (45 μL, 0.26 mmol, 4.0 eq), and DMF (0.5 mL). A portion of the crude product was purified by reverse-phase semi-prep HPLC (5–30% ACN in 0.1% TFA water over 25 min, C18 column) to obtain product 5 as a white solid (1× TFA salt). ¹H NMR (500 MHz, DMSO): δ 11.13 (s, 1H), 10.10 (s, 1H), 8.77 (s, 1H), 8.00 (t, J = 5.7 Hz, 1H), 7.81 (dd, J = 8.5, 7.3 Hz, 1H), 7.65 (s, 1H), 7.50 (d, J = 7.2 Hz, 1H), 7.41 (d, J = 8.6 Hz, 1H), 7.21 (d, J = 8.3 Hz, 1H), 7.14 (dd, J = 8.3, 2.2 Hz, 1H), 7.11 (d, J = 2.2 Hz, 1H), 5.11 (dd, J = 12.9, 5.4 Hz, 1H), 4.79 (s, 2H), 4.53 (d, J = 15.4 Hz, 1H), 4.34 (dd, J = 15.6, 7.3 Hz, 1H), 3.84 (t, J = 5.0 Hz, 2H), 3.77−3.70 (m, 1H), 3.64−3.62 (m, 2H), 3.61 (s, 3H), 3.61−3.58 (m, 2H), 3.50 (t, J = 5.8 Hz, 2H), 3.46−3.37 (m, 3H), 3.34 (q, J = 5.8 Hz, 2H), 3.20−3.09 (m, 1H), 3.09−2.99 (m, 1H), 2.95−2.84 (m, 1H), 2.60 (s, 1H), 2.57−2.52 (m, 1H), 2.08−1.99 (m, 1H). ¹³C NMR (126 MHz, DMSO): δ 172.8, 169.9, 166.9, 166.7, 165.5, 157.0, 154.9, 142.0, 138.1, 137.0, 133.0, 132.4, 127.7, 127.7, 124.5, 122.6, 121.7, 120.4, 116.8, 116.1, 109.2, 69.6, 69.4, 68.8, 67.5, 64.4, 54.3, 52.1, 49.0, 48.8, 38.4, 30.9, 24.7, 22.0 (one resonance obscured by solvent). HRMS (ESI-TOF) calculated for C₃₅H₃₉ClN₇O₉⁺ [M + H]⁺: 736.2492, observed 736.2445.

N-(2-(2-(2-(7-((5-Chloro-1-methyl-6-oxo-1,6-dihydropyridazin-4-yl)amino)-3,4-dihydroisoquinolin-2(1H)-yl)ethoxy)-ethoxy)ethyl)-2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)oxy)acetamide (6).

Following general procedure B, compound 16 was Boc-deprotected in Step 1, and the crude material was used for Step 2: (27 mg, 0.06 mmol, 1.0 eq), 11 (26 mg, 0.08 mmol, 1.2 eq), HCTU (32 mg, 0.08 mmol, 1.2 eq), N,N-diisopropylethylamine (45 μL, 0.26 mmol, 4.0 eq), and DMF (0.5 mL). A portion of the crude product was purified by reverse-phase semi-prep HPLC (5–30% ACN in 0.1% TFA water over 25 min, C18 column) to obtain product 6 as a white solid (2× TFA salt). ¹H NMR (500 MHz, DMSO): δ 11.13 (s, 1H), 9.99 (s, 1H), 7.97 (q, J = 5.7 Hz, 1H), 7.81 (dd, J = 8.5, 7.3 Hz, 1H), 7.62 (s, 1H), 7.50 (d, J = 7.2 Hz, 1H), 7.40 (d, J = 8.5 Hz, 1H), 7.26 (d, J = 8.3 Hz, 1H), 7.17 (dd, J = 8.2, 2.3 Hz, 1H), 7.08 (d, J = 2.3 Hz, 1H), 5.11 (dd, J = 12.9, 5.4 Hz, 1H), 4.77 (s, 2H), 4.53 (d, J = 15.7 Hz, 1H), 4.40−4.31 (m, 1H), 3.82 (t, J = 5.0 Hz, 2H), 3.79−3.71 (m, 1H), 3.67−3.55 (m, 7H), 3.48 (t, J = 5.8 Hz, 2H), 3.45−3.36 (m, 3H), 3.33 (q, J = 5.5 Hz, 2H), 3.18−3.08 (m, 1H), 3.08−2.99 (m, 1H), 2.95−2.84 (m, 1H), 2.64−2.56 (m, 1H), 2.57−2.52 (m, 1H), 2.08−1.99 (m, 1H). ¹³C NMR (126 MHz, DMSO): δ 172.8, 169.9, 166.9, 166.7, 165.5, 157.0, 154.9, 142.1, 137.1, 137.0, 133.0, 129.6, 129.3, 127.6, 123.1, 121.0, 120.3, 116.7, 116.1, 108.9, 69.6, 69.3, 68.8, 67.5, 64.4, 54.4, 52.2, 49.3, 48.8, 38.3, 30.9, 24.3, 22.0 (one resonance obscured by solvent, two resonances overlapping). HRMS (ESI-TOF) calculated for C₃₅H₃₉ClN₇O₉⁺ [M + H]⁺: 736.2492, observed 736.2447.

General Procedure C for the Synthesis of Intermediates 20 and 21.

To a mixture of the sulfone 17²⁹ (1.0 eq) in DMF (0.1 M) was added the bromide 18 or 19 (2.0 eq) and K₂CO₃ (2.0 eq). The resulting mixture was heated to reflux and stirred. Upon completion, ethyl acetate and water were added. The layers were separated, and the organic layer was washed sequentially with water (2×), 10% aq LiCl, and brine. The organic layer was dried over MgSO₄, and the volatiles were removed via rotary evaporation. The resulting residue was purified by column chromatography (CombiFlash Rf system: 24 g of silica, hexanes/ethyl acetate, 0–100% ethyl acetate, 16 min) (Scheme 4).

Scheme 4.

Synthesis of Intermediates 20 and 21

tert-Butyl 4-(3-(4-(6-(3-(1H-Pyrazol-1-yl)propyl)-2-(methylsulfonyl)pyrimidin-4-yl)phenoxy)propyl)piperazine-1-carboxylate (20).

Following general procedure C, sulfone 17 was alkylated with bromide 18, sulfone 17²⁹ (309 mg, 0.828 mmol, 1.0 eq), bromide 18⁴² (510 mg, 1.66 mmol, 2.0 eq), K₂CO₃ (229 mg, 1.66 mmol, 2.0 eq), and DMF (8.0 mL). Product 20 was obtained as a white solid (445 mg, 90% yield). ¹H NMR (400 MHz, CDCl₃): δ 7.96 (d, J = 8.5 Hz, 2H), 7.56 (d, J = 2.3 Hz, 1H), 7.42 (d, J = 2.3 Hz, 1H), 6.96 (d, J = 8.5 Hz, 2H), 6.69 (s, 1H), 6.28 (s, 1H), 5.94 (br. s, 1H), 4.27 (t, J = 6.3 Hz, 2H), 4.09 (t, J = 6.3 Hz, 2H), 3.46−3.41 (m, 6H), 3.34 (s, 3H), 2.54 (t, J = 7.3 Hz, 2H), 2.42 (t, J = 5.1 Hz, 4H), 2.22−2.14 (m, 2 H), 1.97 (t, J = 6.9 Hz, 2H), 1.44 (s, 9H).¹³C (100 MHz, CDCl₃): δ 165.9, 163.9, 161.7, 154.9, 139.7, 129.7, 128.8, 128.3, 114.8, 106.1, 79.8, 68.0, 66.4, 60.5, 55.2, 53.2, 38.9, 29.7, 28.6, 26.8, 14.3; HRMS (ESI-TOF) calculated for C₃₀H₄₁N₅O₅S⁺ [M + H]⁺: 585.2854, observed 584.2836.

tert-Butyl (3-(4-(6-(3-(1H-Pyrazol-1-yl)propyl)-2-(methylsulfonyl)pyrimidin-4-yl)phenoxy)propyl)(methyl)-carbamate (21).

Following the general procedure C, sulfone 17 was alkylated with bromide 19, sulfone 17²⁹ (280 mg, 0.822 mmol, 1.0 eq), bromide 19⁴³ (258 mg, 1.03 mmol, 1.5 eq), K₂CO₃ (227 mg, 1.64 mmol, 1.5 eq), and DMF (8.0 mL). Product 21 was obtained as a white solid (275 mg, 65% yield). ¹H NMR (400 MHz, CDCl₃): δ 7.90 (d, J = 8.5 Hz, 2H), 7.51 (d, J = 2.3 Hz, 1H), 7.40 (d, J = 2.3 Hz, 1H), 6.90 (d, J = 8.5 Hz, 2H), 6.69 (s, 1H), 6.27 (br. s, 1H), 6.24 (s, 1H), 4.23 (t, J = 6.3 Hz, 2H), 4.00 (t, J = 6.3 Hz, 2H), 3.45−3.40 (m, 4H), 3.32 (s, 3H), 2.86 (s, 3H), 2.14 (t, J = 6.5 Hz, 2H), 2.00 (t, J = 6.5 Hz, 2H), 1.42 (s, 9H); ¹³C NMR (100 MHz, CDCl₃): δ 165.7, 163.9, 161.4, 155.9, 139.5, 129.7, 128.7, 128.2, 114.6, 105.9, 79.5, 65.8, 65.3, 49.3, 45.9, 38.9, 34.6, 29.6, 28.5, 27.8; HRMS (ESI-TOF) calculated for C₂₆H₃₅N₅O₅S⁺ [M + H]⁺: 530.2432, observed 530.2435.

General Procedure D for the Synthesis of Intermediates 23 and 24.

Step 1.

To a solution of 6-chlorohexan-1-ol or 2-(2-(2-chloroethoxy)ethoxy)ethan-1-ol (1.0 eq) in pyridine (2.0 M) was added TsCl (1.1 eq) and DMAP (0.1 eq) at 0 °C. The resulting mixture was allowed to warm to 23 °C and stirred for 2 h. Upon completion, EtOAc and H₂O were added, and the layers were separated. The organic layer was washed sequentially with 1 M aq HCl, H₂O, and sat. aq NaHCO₃ and then dried over MgSO₄. The volatiles were removed via rotary evaporation (Scheme 5).

Scheme 5.

Synthesis of Intermediates 23–26

Step 2.

The resulting crude residues were dissolved in DMF (0.15 M) followed by the addition of 2-(2,6-dioxopiperidin-3-yl)-4-hydroxyisoindoline-1,3-dione (22) (1.0 eq), KHCO₃ (1.6 eq), and KI (0.1 eq). The resulting mixture was heated to 80 °C and stirred for 22 h. Upon completion, ethyl acetate and water were added, and the layers were separated. The organic layer was washed sequentially with 10% aq LiCl, water, and brine, and then dried over MgSO₄. Volatiles were removed via rotary evaporation, and the resulting residues were purified via column chromatography (CombiFlash Rf system: 24 g of silica, hexanes/ethyl acetate, 0–100% ethyl acetate, 16 mins).

4-((6-Chlorohexyl)oxy)-2-(2,6-dioxopiperidin-3-yl)-isoindoline-1,3-dione (23).

Following general procedure D, tosylation was performed with 6-chlorohexan-1-ol (1.00 g, 7.35 mmol, 1.0 eq), TsCl (1.54 g, 8.08 mmol, 1.1 eq), and DMAP (90.0 mg, 0.735 mmol, 0.1 eq) in pyridine (3.5 mL). The product was obtained as a white solid and taken on crude to the next step. Continuing with general procedure D, crude tosylate (assumed 3.21 mmol, 1.1 eq), 2-(2,6-dioxopiperidin-3-yl)-4-hydroxyisoindoline-1,3-dione (22) (800 mg, 2.92 mmol, 1.0 eq), KHCO₃ (468 mg, 4.67 mmol, 1.6 eq), and KI (48.2 mg, 0.292 mmol, 0.1 eq) in DMF (20 mL). Product 23 was obtained as a yellow solid (337 mg, 29% yield. Note that minor impurities were observed in Cl/I exchange. ¹H NMR (400 MHz, DMSO-d₆): δ 11.1 (br. s, 1H), 7.80 (t, J = 7.6 Hz, 1H), 7.51 (d, J = 7.3 Hz, 1H), 7.44 (d, J = 8.2 Hz, 1H), 5.08 (dd, J = 12.3, 5.4 Hz, 1H), 4.21 (t, J = 6.4 Hz, 2H), 3.64 (t, J = 6.6 Hz, 2H), 2.95−2.82 (m, 1H), 2.63−2.50 (m, 2H), 2.17−1.95 (m, 1H), 1.81−1.69 (m, 4H), 1.53−1.37 (m, 4H). ¹³C NMR (100 MHz, DMSO-d₆): δ 172.8, 169.9, 166.8, 165.3, 162.1, 156.0, 137.0, 133.2, 119.8, 115.1, 68.7, 48.7, 45.3, 31.9, 30.9, 28.2, 25.9, 24.5, 22.0; HRMS (ESI-TOF) calculated for C₁₉H₂₁N₂O₅Cl⁺ [M + H]⁺: 393.1212, observed 393.1222.

4-(2-(2-(2-Chloroethoxy)ethoxy)ethoxy)-2-(2,6-dioxopiperidin-3-yl)isoindoline-1,3-dione (24).

Following general procedure D, tosylation was performed with 2-(2-(2-chloroethoxy)ethoxy)ethan-1-ol (1.00 g, 5.95 mmol, 1.0 eq), TsCl (1.36 g, 7.14 mmol, 1.2 eq), and DMAP (219 mg, 1.79 mmol, 0.3 eq) in pyridine (10 mL). Product was obtained as a colorless oil and was taken on crude to the next step. Continuing with general procedure D, crude tosylate (assumed 1.61 mmol, 1.1 eq), 2-(2,6-dioxopiperidin-3-yl)-4-hydroxyisoindoline-1,3-dione (22) (400 mg, 1.46 mmol, 1.0 eq), KHCO₃ (240 mg, 2.40 mmol, 1.6 eq), and KI (24.2 mg, 0.146 mmol, 0.1 eq) in DMF (10 mL). Product (24) was obtained as a white solid (399 mg, 58% yield). Spectral data matched the reported literature.⁴⁴

General Procedure D for the Synthesis of Intermediates 25 and 26.

To a solution of chloro-linker (25 or 26) (1.0 eq) was added DMF (0.05 M) and NaI (10 eq), and the resulting mixture was heated to 60 °C for 20 h. Upon completion, ethyl acetate and water were added, and the layers were separated. The organic layer was washed sequentially with 10% aq LiCl, water, and brine and then dried over MgSO₄. Volatiles were removed via rotary evaporation, and the resulting residue was used crude in the next step of the synthesis (Scheme 5).

2-(2,6-Dioxopiperidin-3-yl)-4-((6-iodohexyl)oxy)-isoindoline-1,3-dione (25).

Following general procedure D, halogen exchanged was performed with 23 (123 mg, 0.313 mmol, 1.0 eq) and NaI (470 mg, 3.13 mmol, 10 eq) in acetone (7.5 mL). The product was obtained as a white solid (crude yield 140 mg, 92% yield). The crude product was used in the next step of the synthesis.

2-(2,6-Dioxopiperidin-3-yl)-4-(2-(2-(2-iodoethoxy)ethoxy)-ethoxy)isoindoline-1,3-dione (26).

Following general procedure D, halogen exchange was performed with 24 (348 mg, 0.744 mmol, 1.0 eq), NaI (1.11 mg, 7.44 mmol, 10 eq) in acetone (15 mL). Product was obtained as a white solid (crude yield 331 mg, 86% yield). The crude product was used in the next step of the synthesis. Spectral data matched the reported literature.⁴⁴

General Procedure E for the Synthesis of Compounds 8 and 9.

Step 1.

Sulfone 20 (1.0 eq) was dissolved in a 1:1 mixture of TFA/DCM (0.08 M), and the resulting mixture was stirred at 23 °C for 2 h. Upon competition, volatiles were removed via rotary evaporation.

Step 2.

The resulting residue was taken up in DMF (0.05 M). DIEA (5 eq) and linker 25 or 26 was added, and the mixture was stirred at 80 °C for 18 h. The reaction was allowed to cool to 23 °C, ethyl acetate and water were added, and the layers were separated. The organic layer was washed sequentially with 10% aq LiCl, water, and brine and then dried over MgSO₄. Volatiles were removed via rotary evaporation, and either the resulting residues were purified via column chromatography (CombiFlash Rf system: 24 g of silica, hexanes/ethyl acetate, 0–100% ethyl acetate, 16 min) or a portion of the crude material was then purified by reverse-phase HPLC (Scheme 6).

4-((6-(4-(3-(4-(6-((3-(1H-Pyrazol-1-yl)propyl)amino)-2-(methylsulfonyl)pyrimidin-4-yl)phenoxy)propyl)piperazin-1-yl)hexyl)oxy)-2-(2,6-dioxopiperidin-3-yl)isoindoline-1,3-dione (8).

Following general procedure E, sulfone 20 (49.1 mg, 0.0820 mmol, 1.0 eq) in TFA (0.50 mL) and DCM (0.50 mL). Resulting residue (assumed 0.0820 mmol) in DMF (1.80 mL) was added DIEA (71.0 μL, 0.409 mmol, 5 eq) and 25 (51.2 mg, 0.106 mmol, 1.3 eq). Product 8 was purified via column chromatography (CombiFlash Rf system: 24 g of silica, DCM/methanol, 0–20% ethyl acetate, 16 min) followed by reverse-phase HPLC 0–60% ACN in 0.1% TFA H₂O over 60 min, affording a yellow solid (18 mg, 25% yield). Note that HPLC purification was performed for characterization purposes of purity; the compound 8 used in the Western blot studies and other binding studies was only purified by column chromatography.

¹H NMR (500 MHz, DMSO-d₆): δ 11.12 (br. s, 1H), 8.16 (s, 1H), 7.99 (d, J = 8.3 Hz, 1H), 7.83 (dd, J = 8.3, 7.2 Hz, 1H), 7.76 (s, 1H), 7.52 (d, J = 8.5 Hz, 1H), 7.46 (d, J = 8.5 Hz, 2H), 7.09 (d, J = 8.5 Hz, 2H), 7.01 (s, 1H), 6.24 (s, 1H), 5.08 (dd, J = 12.9, 5.5 Hz, 1H), 4.26−4.18 (m, 2H), 4.15−4.08 (m, 2H), 3.32 (s, 3H), 2.94−2.83 (m, 2H), 2.66−2.56 (m, 2H), 2.36 (br. s, 1H), 2.12−1.95 (m, 6H), 1.79 (quintet, J = 5.6 Hz, 2H), 1.64 (br. m, 2H), 1.50 (quintet, J = 5.6 Hz, 2H), 1.39 (quintet, J = 5.6 Hz, 2H), 1.24 (br. m, 2H); ¹³C NMR (100 MHz, DMSO-d₆): δ 172.8, 170.0, 166.9, 165.4, 165.1, 138.7, 137.1, 133.3, 128.2, 119.8, 116.2, 115.3, 115.2, 114.8, 105.0, 68.8, 60.7, 59.8, 53.6, 48.8, 41.8, 32.5, 31.0, 28.5, 28.1, 25.2, 24.9, 22.0, 20.8, 18.1, 16.7, 14.1, 12.5. HRMS (ESI-TOF) calculated for HRMS (ESI-TOF) calculated for C₄₃H₅₃N₉O₈S⁺ [M + H]⁺: 856.3811, observed 856.3801.

4-(2-(2-(2-(4-(3-(4-(6-((3-(1H-Pyrazol-1-yl)propyl)amino)-2-(methylsulfonyl)pyrimidin-4-yl)phenoxy)propyl)piperazin-1-yl)ethoxy)ethoxy)ethoxy)-2-(2,6-dioxopiperidin-3-yl)-isoindoline-1,3-dione (9).

Following general procedure E, sulfone 20 (36.8 mg, 0.0601 mmol, 1.0 eq) in TFA (0.30 mL) and DCM (0.30 mL). Resulting residue (assumed 0.0601 mmol) in DMF (1.5 mL) was added DIEA (53.0 μL, 0.301 mmol, 5 eq) and 26 (34.0 mg, 0.0661 mmol, 1.1 eq). Product (9) was purified via column chromatography (CombiFlash Rf system: 24 g of silica, DCM/methanol, 0–20% ethyl acetate, 16 min).

¹H NMR (400 MHz, DMSO-d₆): δ 11.11 (s, 1H), 7.96 (d, J = 7.3 Hz, 1H), 7.81 (dd, J = 8.4, 7.3 Hz, 1H), 7.77 (s, 1H), 7.53 (d, J = 8.4 Hz, 1H), 7.46 (d, J = 8.5 Hz, 2H), 7.07 (d, J = 8.5 Hz, 2H), 6.23 (s, 1H), 5.08 (dd, J = 12.8, 5.5 Hz, 1H), 4.34 (t, J = 5.2 Hz, 2H), 4.21 (t, J = 6.8 Hz, 1H), 4.08 (t, J = 5.2 Hz, 2H), 3.80 (t, J = 5.6 Hz, 2H), 3.66 (t, J = 6.0 Hz, 2H), 3.54 (t, J = 5.8 Hz, 2H), 3.31 (s, 3H), 3.04 (q, J = 7.3 Hz, 10H), 2.95−2.81 (m, 2H), 2.63−2.51 (m, 2H), 1.19 (t, J = 7.3 Hz, 14H); ¹³C NMR (100 MHz, DMSO-d₆): δ 172.8, 169.9, 166.8, 165.3, 138.6, 137.0, 133.2, 130.0, 128.2, 120.0, 116.3, 115.4, 114.8, 104.9, 70.1, 69.7, 68.9, 68.7, 48.8, 45.4, 31.0, 29.6, 22.0, 8.5; HRMS (ESI-TOF) calculated for HRMS (ESI-TOF) calculated for C₄₃H₅₃N₉O₁₀S⁺ [M + H]⁺: 888.3709, observed 888.3724.

.4-(2-(2-(2-((3-(4-(6-((3-(1H-Pyrazol-1-yl)propyl)amino)-2-(methylsulfonyl)pyrimidin-4-yl)phenoxy)propyl)(methyl)-amino)ethoxy)ethoxy)ethoxy)-2-(2,6-dioxopiperidin-3-yl)-isoindoline-1,3-dione (10).

Following general procedure E, sulfone 21 (21.8 mg, 0.0400 mmol, 1.0 eq) in TFA (0.50 mL) and DCM (0.50 mL). To the resulting residue (assumed 0.0400 mmol) in DMF (1.50 mL) was added DIEA (36.2 μL, 0.200 mmol, 5 eq) and 26 (54.9 mg, 0.0800 mmol, 1.3 eq). Product 10 was purified via reverse-phase HPLC 0–60% ACN in 0.1% TFA H₂O over 60 min, affording a yellow solid (18 mg, 25% yield) (Scheme 7).

¹H NMR (500 MHz, DMSO-d₆): δ 11.10 (s, 1H), 8.17−8.08 (m, 1H), 7.95 (d, J = 8.3 Hz, 1H), 7.78 (dd, J = 8.6, 7.3 Hz, 1H), 7.76 (d, J = 2.0 Hz, 1H), 7.49 (d, J = 8.6 Hz, 1H), 7.46−7.41 (comp. m, 2H), 7.05 (d, J = 8.6 Hz, 2H) 7.02−6.97 (br. m, 1H), 6.23 (br. s, 1H), 5.07 (dd, J = 12.9, 5.5 Hz, 1H), 4.33−4.29 (comp. m, 2H), 4.21 (t, J = 6.7 Hz, 2H), 4.06 (t, J = 6.7 Hz, 2H), 3.80−3.76 (comp. m, 2H), 3.62 (t, J = 5.2 Hz, 2H), 3.52−3.45 (comp. m, 4H), 2.87 (td, J = 12.9, 5.5 Hz, 1H), 2.65−2.55 (comp. m, 2H), 2.24−2.15 (comp. m, 2H), 2.11−1.97 (comp. m, 2H), 1.89−1.81 (comp. m, 2H), 1.29−1.20 (comp. m, 2H); ¹³C NMR (125 MHz, DMSO-d₆): δ 172.8, 169.9, 166.8, 165.2, 155.8, 138.6, 136.9, 133.2, 129.9, 128.1, 127.7, 120.0, 116.3, 115.4, 114.7, 105.0, 101.3, 70.1, 69.7, 68.8, 68.7, 66.0, 56.4, 53.7, 48.7, 42.4, 30.9, 29.6, 26.5, 22.0; HRMS (ESI-TOF) calculated for HRMS (ESITOF) calculated for C₄₀H₄₉N₈O₁₀S⁺ [M + H]⁺: 833.3287, observed 833.3276.

General Procedure for AlphaScreen Assay.²⁹

His₉-tagged BPTF bromodomain was expressed and purified as described previously.²⁹ His-tagged BRD9 and CECR2 were purchased from Reaction Biology (cat. nos. RD-11–214 and RD-11–210 for BRD9 and CECR2, respectively). The AlphaScreen assay procedures were adapted from the manufacturer’s protocol (PerkinElmer, USA). Nickel chelate (Ni-NTA) acceptor beads and streptavidin donor beads were purchased from PerkinElmer (cat. no. 6760619 M). The biotinylated histone H4 KAc5,8,12,16 peptide was purchased from EpiCypher (cat. no. 12–0034), with the sequence Ac-SGRGK(Ac)-GGK(Ac)GLGK(Ac)GGAK(Ac)RHRKVLR-Peg(Biot).

All reagents were diluted in the assay buffer (50 mM HEPES-Na⁺ (ChemImpex), 100 mM NaCl (SigmaAldrich), 0.05% CHAPS (RPI), and 0.1% BSA (SigmaAldrich), pH 7.4). The final assay concentrations of the protein and biotinylated peptide used are shown in Table 2.

Table 2.

Concentrations of Protein and Peptide Used in AlphaScreen Assays

protein	[protein] nM	[peptide] nM
BPTF	30	100
BRD9	60	100
CECR2	60	100

Three-fold serial dilutions were prepared with varying concentrations of the compounds and a fixed protein concentration, keeping the final DMSO concentration at 0.25%. Five microliters of these solutions were added to a 384-well plate (ProxiPlate-384, PerkinElmer). Five microliters of the biotinylated peptide solution was then added to the wells. Ten microliters of pre-mixed nickel chelate acceptor beads and streptavidin donor beads were added to each well under low light conditions (<100 lux) to a final concentration of 20 μg/mL. For each test compound, BZ1 was also run as a positive control on the same plate to ensure the accuracy of the assay. The plate was sealed and incubated for 30 min in the dark. It was then read in AlphaScreen mode (excitation time = 100 ms, integration time = 300 ms, output: counts/s) using a Tecan Spark plate reader. Each compound was run in two technical replicates, and the data was normalized against a 0 μM inhibitor signal to obtain the % normalized AlphaScreen signal. IC₅₀ values were calculated in GraphPad Prism 9 using a sigmoidal 4-parameter logistic (4PL) curve fit.

General Procedure for In Vitro Ternary Complex Assay.

His-tagged CRBN-DDB1 protein was prepared following the procedure reported by Matyskiela et al.⁴⁵ GST-BPTF was purchased from BPS Biosciences (cat. no. 31134), while GST-BRD9, GST-CECR2, and GST-PCAF were purchased from Reactive Biology Corp (cat. nos. RD-11–187, RD-11–194, and RD-11–259, respectively). All reagents were diluted in assay buffer comprising 25 mM HEPES (pH 7.4), 100 mM NaCl, 0.1% BSA, and 0.05% Tween-20. An ECHO 650 (Labcyte Inc.) acoustic dispenser was used to generate a 10-point dilution curve from DMSO stocks of the degraders directly into a 384-well OptiPlate (PerkinElmer, cat. no. 6007290), giving a final DMSO concentration of 0.3%. Final concentrations of His-tagged CRBN-DDB1 and GST-tagged bromodomains used in the assay are shown in Table 3. AlphaScreen glutathione-coated donor and nickel chelate acceptor beads were purchased from PerkinElmer (cat. nos. 6765300 and 6760141, respectively).

Table 3.

Concentration of Proteins Used in AlphaScreen Ternary Complex Assays

protein	[GST-tagged bromodomains] nM	[His-CRBN-DDB1] nM
BPTF	60	120
BRD9	240	80
CECR2	240	80
PCAF	240	80

Briefly, to a 384-well OptiPlate containing 5× degrader in triplicate was added 5 μL of a 5× solution of His-CRBN-DDB1 and GST-tagged bromodomain and then incubated at rt for 1 h. After incubation, 10 μL of nickel chelate acceptor beads (20 μL/mL final concentration) and 10 μL of glutathione donor beads (20 μL/mL final concentration) were added. The plate was sealed and mixed on a MixMate (Eppendorf) for 1 h at rt, and then luminescence detection was collected on an Envision plate reader (PerkinElmer).

In-Cell NanoBRET Assay.

NanoBRET experiments were carried out using the NanoBRET CRBN Ternary Complex Starter Kit (cat. no. ND2720) according to the manufacturer’s protocol (Promega NanoBRET CRBN Ternary Complex Assay TM615).⁴⁶ All assays were run in 384-well format. HEK293T cells were transfected to a 1:100 donor:acceptor (Nluc:Halotag vectors) ratio for both BPTFBD and BRD9-FL vectors (see the Supporting Information for the design of plasmids). The cells were re-plated, treated with the HaloTag 618 ligand, and treated with compounds as described in the manufacturer’s procedure. Data was collected on a Tecan Spark plate reader using the settings in Table 4.

Table 4.

Tecan Spark Settings for NanoBRET Experiments

(1) shaking (linear)	duration [s]: 30; amplitude [mm]: 1; frequency [rpm]: 1440
(2) mode: luminescence multicolor donor emission	wavelength start = 445 nm; wavelength end = 470 nm; integration time = 300 ms
acceptor emission	wavelength start = 610 nm; wavelength end = 700 nm; integration time = 300 ms

Competition with Monovalent Inhibitor for Ternary Complex Formation.

For competition experiments (Figure S4), 10× solutions of compound 19 and MG-132 were prepared in Opti-MEM I Reduced Serum Medium and pre-mixed in tubes. Ten microliters of this solution was then added to each well. The plates were incubated for 30 min, followed by the addition of degrader compounds. The rest of the assay was carried out according to the technical manual referenced above.

Cell Culture.

HEK293T and LM2 cells (MDA-MB-231 derivatives)⁴⁷ with pLIX-CECR2-V5-SFB-puro were grown in a humidified 5% CO₂ environment at 37 °C. The pLIX-CECR2-V5-SFB-puro vector allows for doxycycline (Dox)-inducible expression of human CECR2 with a C-terminal V5-SFB tag. Cells were cultured in DMEM media (high-glucose, Gibco, cat. no. 11965–092) supplemented with 10% fetal bovine serum (Cellgro, cat. no. QB-110–001-101), penicillin–streptomycin (50 U/mL penicillin, 50 μg/mL streptomycin, Cellgro, cat. no. 15,140–122). The cells were passaged to a 1:20 dilution by decanting suspended cells and dissociating adherent cells from cell culture flasks in 0.25% trypsin/EDTA (Gibco, cat. no. 25200056) after 1 min of incubation.

Western Blotting.

Cells were seeded in six-well plates in 2.5 mL of cell culture medium. Before the addition of degrader, 250 μL of the medium was removed from each well, and the cells were treated with compounds at the desired concentrations for indicated times to a final DMSO concentration of 0.1% v/v. For collecting the lysates, the medium was removed, and the wells were rinsed with 1 mL of ice-cold PBS, followed by the addition of 100 μL of cold RIPA buffer (ThermoFisher, cat. no. 89900) supplemented with cOmplete Mini Protease Inhibitor cocktail (Roche, cat. no. 11836153001). The plates were incubated on ice for 15 min. After high-speed centrifugation (15,000g for 15 min), the supernatant was collected, and protein concentrations were determined by the BCA assay (ThermoFisher, cat. no. 23227). The samples were normalized by total protein content, mixed with 4× NuPAGE LDS loading buffer (Invitrogen) and heated at 100 °C for 10 min.

Proteins were resolved by SDS–PAGE on NuPage 4–12% Bis-Tris (for CECR2, BRD9, and PCAF) or 3–8% Tris-acetate gels (for BPTF) (Invitrogen, cat. nos. NP0323BOX and EA03785BOX) and transferred to PVDF membrane (Bio-Rad, cat. no. 1620174) using wet transfer for 60 min. Membranes were dried and blocked in TBS-T (Tris-buffered saline-T) containing 5% w/v nonfat dry milk for 1 h at room temperature. They were subsequently incubated with primary antibodies at the dilutions and times in Table S1. After the membranes were washed five times with TBS-T, they were incubated with secondary antibodies at the dilutions and times listed in Table S1. Finally, the membranes were washed five times in TBS-T and treated with SuperSignal West Dura substrates (ThermoFisher, cat. no. 34095) and imaged using a Bio-Rad Chemi-Doc imaging system.

ABBREVIATIONS

BAF

BRG1/BRM-associated factor

BD

bromodomain

BET

bromodomain and extraterminal

BPTF

bromodomain PHD-finger transcription factor

BRD9

bromodomain-containing protein 9

BRET

bioluminescence resonance energy transfer

CECR2

cat-eye syndrome chromosome region candidate 2

CRBN

cereblon

FL

full-length

GBAF

GLTSCR1/like-containing BAF complex

GCN5

general control nondepressible 5

ISWI

imitation switch

ncBAF

non-canonical BAF

Nluc

nanoluciferase

NURF

nucleosome remodeling factor

PBAF

polybromo-associated BAF

PCAF

p300/CBP-associated factor

PEG

polyethylene glycol

SAR

structure–activity relationship

SMARCA2

SWI/SNF-related matrix-associated actin-dependent regulator of chromatin a2