Hydrophobic Tag Degraders Overcome Endocrine‐Resistant Breast Cancer by Recruiting HSP27‐Mediated E3 Ligase Complex for ERα Proteasomal Degradation

ABSTRACT

Hydrophobic tag (HyT)‐mediated protein degradation has emerged as a pivotal tool for targeted protein degradation (TPD), yet its underlying degradation mechanism remains incompletely elucidated. Herein, we designed structurally optimized HyT‐based degraders by covalently conjugating hydrophobic amino acid tags to ERα‐targeting ligands via alkane linkers of varying lengths, identifying the lead compound VI‐10h. VI‐10h exhibited potent antiproliferative activity and efficient ERα degradation in endocrine‐resistant breast cancer cells (LCC2, MCF‐7^D538G, MCF‐7^Y537S, and MCF‐7^EGFR) and superior antitumor activity compared to the clinical drug fulvestrant (Ful) in MCF‐7 and tamoxifen‐resistant LCC2 xenograft models. To elucidate the HyT‐mediated degradation mechanism, we synthesized biotin‐conjugated HyTs (biotin‐Lys and biotin‐Trp) and performed pull‐down assays combined with mass spectrometry. Our results unveiled that VI‐10h selectively recruits heat shock protein 27 (HSP27) as a non‐canonical E3 ligase adaptor protein, forms an ERα–HSP27–RING1 ternary complex to promote ERα degradation, disrupts estrogen‐dependent oncogenic networks, and circumvents the drug resistance associated with conventional CRBN‐ or VHL‐dependent E3 ligase‐recruiting degraders. This study clarifies a HyT‐mediated ERα degradation mechanism and supports the feasibility of using HyT degraders to overcome resistance to conventional E3 ligase‐recruiting strategies and endocrine‐resistant breast cancer, thereby establishing a molecular design strategy for next‐generation targeted degraders.

A degrader with a hydrophobic tag was discovered that recruits HSP27 (atypical E3 adaptor) to form an ERα‐HSP27‐RING1 complex, induces ERα degradation in endocrine‐resistant breast cancer cells/xenografts, and circumvents conventional CRBN‐ or VHL‐dependent degrader resistance.

1. Introduction

Estrogen receptor‐positive (ERα⁺) breast cancer accounts for approximately 70% of clinical cases, with a key bottleneck in its treatment being the widespread development of endocrine resistance [1]. Previous studies have confirmed that acquired mutations in the ligand‐binding domain (LBD) of estrogen receptor 1 (ESR1), particularly the D538G and Y537S, are key drivers of this resistance [2, 3, 4]. Current strategies to address endocrine resistance include next‐generation oral selective estrogen receptor degraders (SERDs), such as fulvestrant (Ful) and elacestrant; combination therapies; and targeted protein degradation exemplified by the proteolysis‐targeting chimera (PROTAC) ARV471, with the latter demonstrating distinct advantages [1, 5]. Notably, long‐term use of selective estrogen receptor modulators (SERMs) may induce antagonist‐agonist conformational changes in the ER binding pocket, leading to the emergence of endocrine resistance [6]. This suggests that developing ERα‐targeted degraders with dual antagonist and degradative functionalities could offer a more promising therapeutic strategy for endocrine‐resistant breast cancers.

Recently, hydrophobic tag (HyT)‐mediated protein degradation has emerged as an innovative alternative to traditional targeted protein degradation (TPD) platforms [7]. This technology capitalizes on rationally designed bifunctional molecules (typically <500 Da) that covalently attach hydrophobic pharmacophores (e.g., adamantane, Boc‐protected amino acids) to target proteins via warhead‐directed conjugation [8, 9, 10, 11, 12, 13]; the solvent‐exposed hydrophobic surfaces induce structural destabilization of the target protein via “hydrophobic collapse,” subsequently activating cellular quality control mechanisms [14, 15]. Although it is generally accepted that the cellular protein quality control system plays a key role in the degradation of HyT‐modified proteins, the precise molecular mechanisms remain largely unclear. Recent evidence indicates that HyT degraders markedly degraded ALK and EZH2 by HSP70 rather than traditional E3 ligases (CRBN or VHL) [8]. These HyTs not only achieve sub‐micromolar degradation efficiency but also exhibit favorable pharmacokinetic properties in xenograft models. Additionally, the small heat shock protein HSP27 regulates protein quality control by recognizing hydrophobic patches on misfolded proteins, which is achieved via synergistic or competitive interactions with E3 ubiquitin ligases [16, 17, 18], and under stress conditions, HSP27 can also recruit E3 enzymes (RING1 [19] or RING1B [20]) via a “chaperone‐ubiquitination” cascade. Despite significant advances in HyT technology [21, 22], how HyT‐target protein interactions are converted into ubiquitination and proteasomal degradation remains unclear. Furthermore, optimization of the linker, identification of novel targetable proteins, and rational design of ideal HyT moieties all rely on a clear understanding of this mechanism [23]. These challenges highlight the urgent need for a mechanism‐driven optimization strategy based on the elucidation of HyT‐mediated target degradation.

To explore the underlying degradation mechanisms of HyT technology and overcome resistance associated with traditional E3 ligase‐recruiting degraders, this study integrated Boc‐protected hydrophobic amino acid tags (L‐Boc₃Arg, L‐BocPhe, L‐BocLeu, L‐Boc₂Lys, and L‐BocTrp) into the 7‐oxabicyclo[2.2.1]heptene sulfonamide (OBHSA) scaffold, which possesses a three‐dimensional structure and specific ER antagonism, generating a series of novel HyT‐based ERα degraders (Scheme 1). The rationale of this design is to exploit the unique structure of OBHSA to induce H11–H12 conformational rearrangement in ERα [24], thereby exposing hydrophobic surface pockets and promoting ERα ubiquitination and degradation. Concurrently, local hydrophobic imbalance mediated by HyTs further destabilizes ERα folding, thus forming a dual‐action mechanism. Herein, we report the development of a novel series of HyT derivatives and validation of their antitumor activity against multiple resistant breast cell lines in vitro and in vivo. The mechanism of action of these HyT degraders to circumvent traditional E3 ligase resistance was also elucidated. Furthermore, the feasibility of HyT technology in reversing endocrine resistance was evaluated.

SCHEME 1.

HyT‐mediated ERα degradation: a structure‐guided design paradigm. The A‐ring of OBHSA mimics estradiol (E2), with its phenolic hydroxyl group forming strong hydrogen bonds to Arg394 (R394) and Glu353 (E353) in ERα Y537S; the E‐ring phenolic hydroxyl group interacts with Thr347 (T347).

2. Results and Discussion

Firstly, we synthesized a series of OBHSA‐derived degraders (VI‐10a–VI‐10n, Figure 1A) containing hydrophobic side chains via the amide coupling of OBHSA‐derived amino intermediates (VI‐8a–VI‐8d, Scheme S1) with Boc‐protected hydrophobic amino acid tags (VI‐9a–VI‐9e, Scheme S2). Fluorescence polarization competitive binding assays were performed to assess the binding affinity of these compounds for ERs (Table S1). The results demonstrated that most compounds exhibited significant selectivity for ERα, with those modified by L‐Boc₃Arg, L‐BocPhe, L‐BocLeu, L‐Boc₂Lys, and L‐BocTrp showing enhanced ERα binding affinity. Notably, VI‐10g (L‐BocPhe) exhibited the highest relative binding affinity (RBA = 10.2%, Table S1).

FIGURE 1.

Structural optimization and functional characterization of HyT‐based ERα degraders. (A) Synthetic route of HyT‐linked ERα degraders (VI‐10a–VI‐10n). (B) The antiproliferative effects of VI‐10h, VI‐10j, Ful, and OBHSA in MCF‐7 cells for 96 h. (C, D) Crystal violet staining (C) and quantitative analysis (D) of the colony‐forming capacity of MCF‐7 cells treated with VI‐10h, Ful, or OBHSA for 14 days. n = 3, mean ± SD, one‐way ANOVA, ***p < 0.001, ****p < 0.0001. (E, F) Western blot analysis of ERα degradation in MCF‐7 cells treated with VI‐10a–VI‐10n, Ful, and OBHSA (0.3 µM, 24 h). (G‐I) Dose‐dependent degradation of ERα by VI‐10h (G), Ful (H), and OBHSA (I) in MCF‐7 cells for 24 h. (J) Dose‐response curves and half‐maximal degradation concentration (DC₅₀) of VI‐10h, Ful, and OBHSA‐induced ERα degradation in MCF‐7 cells. (K) Volcano plot of the TMT‐based quantitative proteomic data showing specific ERα depletion in HyT‐treated cells versus controls.

Subsequently, we evaluated the antiproliferative activity of HyT‐based degraders against various breast cell lines with respect to different linker lengths and tag types. In ERα⁺ breast cancer cell lines (MCF‐7 and T47D), the antiproliferative efficacy of these HyTs was dependent on linker length. Compounds with n = 6 exhibited superior potency over those with n = 3. Notably, the L‐Boc₂Lys‐ and L‐BocTrp‐tagged derivatives (VI‐10h: 3.16 nM; VI‐10j: 5.77 nM) showed potency comparable to Ful in MCF‐7 (Figure 1B) and T47D (Figure S1A) cells and outperformed OBHSA (Figure 1B, Table S2). Notably, VI‐10h significantly suppressed the colony‐forming capacity of MCF‐7 cells at 0.005 to 5 µM (Figure 1C,D), consistent with its antiproliferative activity measured by the CCK‐8 assay (Figure 1B). Further linker optimization revealed that VI‐10h (n = 6) and VI‐10j (n = 6) exhibited potent growth inhibition in MCF‐7 and T47D cells, outperforming HyT degraders with n = 4 and n = 8 (Table S2). Safety evaluations demonstrated that these compounds exhibited negligible toxicity against normal mammary epithelial cells (MCF‐10A) and showed no inhibitory activity against triple‐negative breast cancer cells (MDA‐MB‐231, ERα⁻), thus confirming their ERα‐dependent tumor‐targeted specificity and pharmacological safety (Table S2).

Moreover, we systematically evaluated the ERα degradation capacity of HyTs and investigated the effect of linker length and amino acid tag type on their ERα proteolytic degradation activity via Western blot analysis. The results revealed that most compounds exhibited potent ERα degradation capacity. The degradation activity of VI‐10a–VI‐10j varied with the amino acid tag type and linker length. Notably, L‐BocLeu, L‐Boc₂Lys, and L‐BocTrp tags markedly enhanced ERα degradation efficiency in MCF‐7 (Figure 1E) and T47D (Figure S1B) cells. Linker optimization confirmed that compounds with n = 6 outperformed those with n = 3. Further evaluation of n = 4, 6, and 8 modified with L‐Boc₂Lys or L‐BocTrp revealed that VI‐10h (n = 6) and VI‐10j (n = 6) exhibited optimal degradation activity, which diminished with excessively long or short linkers (Figure 1F). VI‐10h (DC₅₀ = 1.08 nM, Figure 1G,J) and Ful (DC₅₀ = 1.45 nM, Figure 1H,J) induced dose‐dependent ERα degradation in MCF‐7 cells with significantly stronger potency than OBHSA (DC₅₀ = 0.73 µM, Figure 1I,J), while VI‐10j (DC₅₀ = 20.0 nM, Figure S1C,E) and VI‐10m (DC₅₀ = 31.6 nM, Figure S1D,F) also mediated dose‐dependent ERα degradation. Concentration‐dependent effects were similarly observed in T47D cells (Figure S1G,H).

Compared with the parent OBHSA scaffold [25, 26], VI‐10h showed markedly stronger low‐dose ERα degradation potency, confirming that incorporation of the HyT module improved the degradation activity of the parent skeleton (Figure 1G–J). Tandem mass tag (TMT) labeling and mass spectrometry confirmed the targeted ERα degradation activity of VI‐10h and VI‐10j (Figure 1K). Given the superior antiproliferative and degradation activities of VI‐10h, its ER transcriptional regulatory activity was assessed in HEK‐293T cells via dual‐luciferase reporter assays with Ful and 4‐hydroxytamoxifen (4‐OHT) as controls. The results showed that VI‐10h effectively antagonized ERα transcriptional activity (EC₅₀ = 0.82 nM, Figure S1I) while significantly activating ERβ transcriptional activity (EC₅₀ = 0.089 nM, Figure S1J) without exerting an antagonistic effect on ERβ. Additionally, VI‐10h inhibited cell cycle progression (Figure S2A) and induced apoptosis (Figure S2B,C) in MCF‐7 cells. Taken together, these findings suggest that VI‐10h may block the transcriptional function of ERα by either degrading or antagonizing it, thereby inhibiting ERα‐dependent cell proliferation or abnormal signal transduction.

To analyze the subcellular distribution and protein level of ERα in MCF‐7 cells following treatment with VI‐10h, we utilized confocal laser scanning microscopy (CLSM). The results demonstrated that VI‐10h induced dose‐dependent downregulation of ERα protein over the concentration range of 0.5–5 µM (Figure 2A,B). Previous research has shown that HyT‐induced protein degradation mainly involves the ubiquitin‐proteasome system (UPS), lysosomal pathway, unfolded protein response, and endoplasmic reticulum‐associated degradation [27, 28]. As the detailed mechanism of HyT‐mediated degradation remains incompletely elucidated, we explored the pathway underlying VI‐10h‐induced ERα degradation in MCF‐7 cells via inhibitor blockade assays. Results revealed that 5 nM VI‐10h significantly induced ERα degradation. Notably, co‐treatment with the N‐degron inhibitor alanine‐arginine (Ala‐Arg) [29] and VI‐10h failed to block ERα degradation (Figure 2C), whereas co‐treatment with the proteasome inhibitor bortezomib and VI‐10h completely abolished ERα degradation. These findings suggest that VI‐10h‐induced ERα degradation may rely on the proteasome pathway. Although MG132 inhibits 20S proteasome activity [30] and blocks the proteolytic function of the 26S proteasome complex, it did not fully restore ERα protein levels, likely due to its potential autophagy‐activating properties, which may promote compensatory activation of autophagy‐related degradation pathways (Figure 2D). Furthermore, treatment with the autophagy inhibitors [31] 3‐methyladenine (3‐MA) [32] and ammonium chloride (NH₄Cl) [33] also failed to inhibit VI‐10h‐induced ERα degradation (Figure 2E), further indicating that autophagy is not the primary mechanism underlying VI‐10h‐induced ERα degradation. Taken together, VI‐10h induced ERα degradation in MCF‐7 cells via a proteasome‐dependent pathway, while the partial inhibitory effect of MG132 may stem from its nonspecific induction of autophagic upregulation. For the control OBHSA, we verified its degradation mechanism to exclude parent scaffold interference (Figure S2D–F), confirming UPS‐dependent ERα degradation consistent with our previous work [34].

FIGURE 2.

VI‐10h induced ERα degradation in MCF‐7 cells via the UPS pathway. (A) CLSM analysis of VI‐10h‐induced dose‐dependent (0.5–5 µM) changes in the subcellular distribution and protein levels of ERα in MCF‐7 cells for 24 h. (B) Quantitative analysis of ERα protein levels in MCF‐7 cells following VI‐10h treatment. n = 6, mean ± SD, one‐way ANOVA, ***p < 0.001. (C) Effect of N‐degron inhibitor Ala‐Arg on VI‐10h‐induced ERα degradation. Western blot analysis of ERα protein levels in MCF‐7 cells treated for 24 h with Ala‐Arg (5 µM) alone or in combination with VI‐10h (5 nM). (D) Effects of proteasome inhibitors (MG132 and bortezomib) on VI‐10h‐induced ERα degradation. Western blot analysis of ERα protein levels in MCF‐7 cells following treatment with MG132 (10 µM), bortezomib (10 µM), or their combination with VI‐10h (5 nM) for 24 h. (E) Effects of autophagy inhibitors (3‐MA and NH₄Cl) on VI‐10h‐induced ERα degradation. Western blot analysis of ERα protein levels in MCF‐7 cells treated with 3‐MA or NH₄Cl (5 µM) alone or combined with VI‐10h (5 nM) for 24 h.

To define the mechanism of HyT‐driven degradation, we first generated biotinylated HyTs (biotin‐Lys and biotin‐Trp, Scheme S3) and profiled their interacting proteins by pull‐down followed by LC‐MS/MS. In MCF‐7 lysates, both probes reproducibly enriched the small heat shock protein HSP27 as the dominant interactor, together with several additional proteins including tubulins and the E3 ligase RING1 (Figure 3A). Preferential capture of HSP27 by biotin‐Lys and biotin‐Trp was confirmed by immunoblotting (Figure 3B). To further confirm that these interactions occur in intact cells, we performed live‐cell labeling in MCF‐7 cells. Proteomic analysis again highlighted HSP27 among the most enriched proteins, with RING1 also detected in the probe‐specific fraction (Figure S3A–C).

FIGURE 3.

HyT‐mediated degradation of ERα involves HSP27 and E3 ligase RING1. (A) Schematic workflow of biotinylated HyTs (biotin‐Lys or biotin‐Trp) used for pull‐down assays in MCF‐7 lysates, followed by LC‐MS/MS. HSP27 is the top enriched interactor. (B) Validation of HSP27 enrichment by biotin‐Lys and biotin‐Trp using immunoblotting. (C) Flow cytometry analysis of HaloTag‐ERα degradation in 293T cells after treatment with hydrophobic degraders (VI‐10h or VI‐10j, 0.1 µM) in control versus HSPB1 knockdown cells. HSPB1 knockdown reduces degrader‐induced ERα degradation. (D) Quantification of HaloTag signal intensities from (C). n = 3, mean ± SD, one‐way ANOVA, ***p < 0.001. (E) Western blot analysis showing that HSPB1 knockdown impairs ERα degradation induced by VI‐10h and VI‐10j (0.1 µM) in 293T‐HaloTag‐ERα cells. (F) Ranking of E3 ubiquitin ligases enriched in biotin pull‐down samples from proteomic analysis. RING1 is among the top candidates. (G) Co‐immunoprecipitation assay showing the interaction between HSP27 and RING1. (H) Flow cytometry analysis of HaloTag‐ERα degradation in 293T cells after treatment with hydrophobic degraders (VI‐10h or VI‐10j, 0.1 µM) in control versus RING1 knockdown cells. RING1 knockdown reduces degrader‐induced ERα degradation. (I) Quantification of HaloTag signal intensities from (H). n = 3, mean ± SD, two‐sided Student's t‐test, ***p < 0.001. (J) Western blot analysis confirming that RING1 knockdown reduces ERα degradation by VI‐10h and VI‐10j (0.1 µM). (K) Knockdown of CRBN has no effect on ERα degradation induced by hydrophobic degraders (VI‐10h or VI‐10j, 0.1 µM). (L) Knockdown of VHL has no effect on ERα degradation induced by hydrophobic degraders (VI‐10h or VI‐10j, 0.1 µM). (M) Proposed model: The HyT‐containing degrader recruits both HSP27 and the E3 ligase RING1 to promote ERα ubiquitination and subsequent proteasomal degradation.

We next asked whether the HyTs directly engage HSP27. Using biolayer interferometry, both biotin‐Lys and biotin‐Trp showed nanomolar binding affinity to HSP27, with KD values of 11.6 ± 0.55 nM and 15.1 ± 1.41 nM, respectively (Figure S3D,E). Together, these data indicate that HyT motifs can bind HSP27. To investigate the functional relevance of HSP27, we knocked down HSPB1 (encoding HSP27) in 293T‐HaloTag‐ERα cells and assessed the activity of HyT‐based degraders using flow cytometry. Silencing HSPB1 markedly impaired ERα degradation induced by two representative hydrophobic degraders, VI‐10h and VI‐10j (Figure 3C,D), consistent with results from immunoblot analysis (Figure 3E).

Although several additional proteins were enriched in proteomic datasets, excess hydrophobic degraders (VI‐10h or VI‐10j) markedly reduced capture of HSP27 and concomitantly diminished the co‐enrichment of other abundant proteins in competition pull‐down experiments (including HNRNPF, PCBP1, and DDX46) (Figure S3F). Knockdown of selected candidates (e.g., HNRNPF, PCBP1, and DDX46) did not phenocopy HSPB1 depletion (Figure S3G–I). Notably, although HSP70 has been previously implicated in degradation pathways involving other HyTs [8, 35], knockdown of HSP70 had little effect on ERα degradation in our system, suggesting a specific role for HSP27 (Figure S3J).

Because chaperones can facilitate substrate ubiquitination through E3 ligase coupling, we focused on RING1, which showed strong enrichment in the pull‐down proteome (Figure 3F and Figure S3F). Heat shock proteins have been reported to interact with E3 ubiquitin ligases and facilitate substrate ubiquitination and subsequent degradation [36, 37]. We further confirmed the interaction between HSP27 and RING1 by co‐immunoprecipitation (Figure 3G). Reconstitution and dependency tests supported that HSP27 is important for efficient RING1 association with HyT probes, suggesting that HSP27 may act as a co‐adaptor linking HyT engagement to the ubiquitin‐proteasome pathway (Figure S3K,L). Consistently, knockdown of RING1 diminished ERα degradation in response to VI‐10h and VI‐10j, as demonstrated by both flow cytometry and Western blotting analysis (Figure 3H–J). In parallel, VI‐10h and VI‐10j increased ERα ubiquitination under proteasome blockade (MG132) (Figure S3M), supporting a ubiquitin‐dependent proteasomal route.

We next examined whether VI‐10h and VI‐10j could retain degradation activity in cellular contexts where conventional VHL‐ or CRBN‐based PROTACs fail due to intrinsic or acquired resistance. For instance, downregulation or point mutations of CRBN frequently occur after prolonged IMiD or PROTAC exposure, resulting in loss of E3 ligase engagement and reduced efficacy [38]. Remarkably, both VI‐10h and VI‐10j effectively degraded ERα protein even in cells with VHL or CRBN knockdown (Figure 3K,L). These results demonstrate that the HyT‐based degraders maintain robust activity independent of canonical CRBN or VHL engagement, thereby supporting HSP27–RING1‐associated degradation axis that may remain effective in biological settings where traditional ligase‐recruiting PROTACs encounter resistance. Importantly, RNA‐seq analysis indicated that HyT‐induced degradation did not broadly activate stress response programs. GSEA revealed no significant enrichment of gene sets associated with cellular response to heat stress, unfolded protein response, or proteasome function (Figure S4A–C). Consistent with these findings, GSVA detected no significant differences in the activity of these pathways among the three groups, and representative genes showed only limited, gene‐specific changes rather than coordinated induction (Figure S4D–G). Collectively, these findings support a model in which HyT‐based degraders promote ERα degradation by recruiting HSP27 and the E3 ligase RING1, leading to ubiquitination and subsequent proteasomal degradation (Figure 3M).

To clarify the efficacy and advantages of HyT‐mediated degraders in overcoming endocrine resistance in ERα⁺ breast cancer, we evaluated the activity of lead compounds using cell proliferation assays (Table S3). Results showed that VI‐10h exhibited significant antiproliferative activity against both the tamoxifen‐resistant breast cancer cell line LCC2 (Figure 4A) and MCF‐7 resistant sublines, including MCF‐7^D538G (Figure 4B), MCF‐7^Y537S (Figure 4C), and MCF‐7^EGFR (Figure 4D). Building on this, we further evaluated the capacity of compounds VI‐10h and VI‐10j to degrade ERα in endocrine‐resistant breast cancer cells. We found that both compounds induced dose‐dependent ERα degradation (0.1–10 µM) in MCF‐7^D538G (Figure 4E), MCF‐7^Y537S (Figure 4F), and MCF‐7^EGFR (Figure 4G) cells, while markedly inhibiting colony‐forming capacity (Figure S5) and blocking cell cycle progression in these mutant cell lines (Figure S6A–C). Notably, even at high concentrations (10 µM), neither VI‐10h nor VI‐10j exhibited the “Hook effect” characteristic of PROTACs. This suggests that the HyT strategy not only enhances targeting specificity to reduce off‐target toxicity, but also stably maintains degradation activity over a wide concentration range, laying a solid foundation for its clinical translation. Mechanistic studies revealed that VI‐10h (Figure S7A–C) and VI‐10j (Figure S7D–F) significantly induced ERα degradation, whereas the proteasome inhibitor MG132 alone had no significant effect on ERα levels in MCF‐7 mutant cell lines. Co‐treatment with MG132 attenuated VI‐10h‐ or VI‐10j‐induced ERα degradation, supporting a proteasome‐dependent mechanism. Additionally, CLSM revealed that VI‐10h downregulated overall cellular ERα signal in a dose‐dependent manner (5–0.5 µM) in MCF‐7^D538G (Figure 4H,I), MCF‐7^Y537S (Figure 4J,K), and MCF‐7^EGFR (Figure 4L,M) cells, further supporting the UPS‐dependent degradation mechanism.

FIGURE 4.

VI‐10h significantly inhibits cell proliferation and induces ERα degradation in ERα⁺ endocrine‐resistant breast cancer. (A–D) The antiproliferative effects of VI‐10h in LCC2 (A), MCF‐7^D538G (B), MCF‐7^Y537S (C), and MCF‐7^EGFR (D) for 96 h. (E–G) VI‐10h and VI‐10j significantly induce ERα degradation in MCF‐7^D538G (E), MCF‐7^Y537S (F), and MCF‐7^EGFR at different concentrations (G) for 24 h. (H–M) CLSM analysis of VI‐10h‐induced dose‐dependent (0.5–5 µM) changes in the subcellular distribution and protein levels of ERα in MCF‐7^D538G (H), MCF‐7^Y537S (J), and MCF‐7^EGFR (L) cells. Quantitative analysis of ERα protein levels in MCF‐7^D538G (I), MCF‐7^Y537S (K), and MCF‐7^EGFR (M) cells following VI‐10h treatment. n = 6, mean ± SD, one‐way ANOVA, ***p < 0.001, ****p < 0.0001.

To evaluate the drug‐like property of compound VI‐10h, this study systematically characterized its pharmacokinetic (PK) properties. Following intravenous (iv) and intraperitoneal (ip) administration, VI‐10h exhibited half‐lives (T _1/2) of 0.88 h and 14.98 h, respectively (Table S4), and a high clearance rate (CL = 4.54 mL/min/kg) in rats, suggesting the need for further structural optimization to improve PK properties. Subsequent pharmacodynamic studies employed ip dosing to assess the antitumor activity of VI‐10h in a Balb/c nude mouse MCF‐7 breast cancer model. Tumor models were further established by implanting ERα⁺ MCF‐7 cells and tamoxifen‐resistant LCC2 cells. In the MCF‐7 model, VI‐10h (0.5 and 2 mg/kg) significantly suppressed tumor growth compared to positive control Ful (2 mg/kg) (ip every other day). The VI‐10h‐treated group markedly reduced tumor volume (Figure 5A,C) and weight (Figure 5B) without significant alterations in mouse body weight (Figure S8A), demonstrating superior antitumor efficacy to Ful. H&E staining indicated normal histological features in the heart, liver, spleen, and kidney (Figure S8C), suggesting no systemic toxicity of VI‐10h. Western blot analysis revealed that both VI‐10h and Ful decreased ERα protein levels in tumor tissues, with VI‐10h showing higher degradation efficiency at equivalent doses (Figure 5D). Immunohistochemistry (IHC) demonstrated a reduction in Ki67‐positive cells in VI‐10h‐treated groups, confirming its in vivo antiproliferative activity (Figure S8D). In the tamoxifen‐resistant LCC2 model, VI‐10h (4 mg/kg) also significantly inhibited tumor growth compared to tamoxifen (4 mg/kg) and Ful (4 mg/kg) controls (ip every other day). VI‐10h markedly reduced tumor volume and weight more effectively than Ful (Figure 5E–G) and exhibited enhanced ERα degradation activity (Figure 5H). IHC revealed a lower proportion of Ki67‐positive cells in VI‐10h‐treated groups (Figure 5I), with no pathological abnormalities in major organs (Figure S8B,E), further supporting its antiproliferative activity and safety profile. Collectively, VI‐10h exerted potent tumor‐suppressive effects and a favorable safety profile in both MCF‐7 and LCC2 tumor models, supporting its potential as an anticancer drug candidate.

FIGURE 5.

In vivo antitumor efficacy of ERα degraders in MCF‐7 and LCC2 breast cancer models. (A, C) Tumor growth in MCF‐7 xenograft mouse models treated with control, Ful (2 mg/kg), VI‐10h (0.5 mg/kg), and VI‐10h (2 mg/kg). n = 5, mean ± SD, one‐way ANOVA. *p < 0.05, **p < 0.01. (B) Quantification of final tumor weights in MCF‐7 xenograft tumor models. n = 5, mean ± SD, one‐way ANOVA. *p < 0.05, **p < 0.01. (D) Western blot analysis of ERα protein levels in MCF‐7 tumors. (E, G) Comparative efficacy in LCC2 tamoxifen‐resistant models treated with vehicle, tamoxifen (Tam, 4 mg/kg), Ful (4 mg/kg), VI‐10h (4 mg/kg), or VI‐10j (4 mg/kg). n = 5, mean ± SD, one‐way ANOVA. **p < 0.01, ***p < 0.001. (F) Quantification of final tumor weights in LCC2 xenograft tumor models. n = 5, mean ± SD, one‐way ANOVA. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (H) ERα degradation validation in LCC2 tumors. (I) IHC staining of Ki67 (a cell proliferation marker) in LCC2 tumors (I); scale bars: 50 µm.

3. Conclusion

Despite advances in TPD, HyT‐based degraders remain limited by ambiguous degradation mechanisms and E3 ligase‐related resistance, limiting their wider application. Herein, we designed and synthesized a series of HyT‐based ERα degraders by incorporating Boc‐protected amino acid‐derived HyT moieties into ERα‐targeting ligands. These degraders showed potent antiproliferative and degradation activity in ERα⁺ cell lines with retained robust efficacy in multiple endocrine‐resistant breast cancer models. Compound VI‐10h (n = 6, L‐Boc₂Lys) was identified as the lead candidate with optimal ERα degradation potency, and inhibitor blockade assays verified its UPS‐dependent mechanism. Biotinylated probes (biotin‐Lys and biotin‐Trp) for pull‐down assays and mass spectrometry analysis further revealed that VI‐10h selectively recruits HSP27 as an atypical E3 ligase adaptor, supporting a model in which VI‐10h promotes an ERα–HSP27–RING1 degradation axis. This process disrupts ERα signaling and the estrogen‐dependent oncogenic network, thereby overcoming resistance linked to conventional E3‐dependent degraders. Notably, VI‐10h displayed markedly improved antitumor activity versus Ful in MCF‐7 and LCC2 xenografts, supporting its potential as a novel therapeutic candidate for breast cancer.

This work highlights three core innovations: the identification of HSP27 as a non‐canonical chaperone adaptor that facilitates RING1‐associated ERα degradation; the circumvention of conventional E3 ligase‐associated resistance, offering a new therapeutic strategy for endocrine‐resistant breast cancer; and the establishment of a HyT‐based design platform for engaging non‐canonical degradation axes, a paradigm distinct from the canonical E3 ligand‐dependent strategy.

Notably, the HyT strategy holds some clear merits versus conventional PROTACs recruiting CRBN or VHL E3 ligases. First, HSP27 is widely overexpressed in multiple cancers but minimally expressed in normal tissues, conferring favorable tumor specificity [39, 40]. Second, unlike CRBN and VHL, which can be downregulated during acquired drug resistance, HSP27 is upregulated under cellular stress, potentially enabling HyT degraders to retain activity in resistant contexts [39, 40]. Third, the small molecular weight of HyT ligands may enhance druggability compared with CRBN‐ or VHL‐recruiting motifs [7]. Despite these strengths, the HyT platform harbors several potential limitations. First, the structure‐activity relationship underlying HyT‐mediated HSP27 recruitment remains incompletely elucidated, restricting further rational structural optimization. Second, the in vivo pharmacokinetic profiles and tissue distribution of HyT degraders remain to be fully characterized to advance clinical translation. Third, as HyT co‐opts HSP27, it could in principle influence the stability of certain HSP27 client proteins. Nevertheless, our RNA‐seq analysis did not show broad activation of cellular stress‐response pathways, suggesting that HyT does not induce a generalized proteotoxic or stress‐associated program under the conditions tested.

Collectively, our findings validate the feasibility of HyT technology to overcome E3 ligase‐associated resistance in ERα⁺ breast cancer and lay a molecular foundation for the development of next‐generation degraders to reverse endocrine resistance.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Experimental details, materials and methods, supporting figures (Figures S1−S8), supporting tables (Tables S1−S5), WB source data, ¹H and ¹³C NMR spectra, HRMS and HPLC, and associated references within the Supporting Information.

Supporting File: anie72865‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The raw and processed mRNA‐seq datasets have been deposited in the Gene Expression Omnibus (GEO) repository under accession number GSE322892. All data that support the findings of this study are available from the corresponding author upon reasonable request.