Development of a Novel 18F-labeled Probe for PET imaging of Estrogen Receptor β

Purpose: Estrogen receptors beta (ERβ) is an important ER subtype and plays crucial roles in many physiological and pathological disorders. Herein, we aim to develop a probe, 18 F-PVBO, for in vivo ERβ targeted PET imaging with promising results. Methods: 18 F-PVBO was synthesized using a two-step radiolabeling method. The relative binding a�nities of the reference compound PVBO towards ERα and ERβ were determined by a competitive radiometric binding assay using 3 H-estradiol. Cytotoxicity and cell uptake were evaluated in ERβ-positive DU145 cells. PET imaging, including blocking study, was performed in DU145 tumor-bearing nude mice (n = 3 per group), and the biodistribution study of 18 F-PVBO was also performed. Results: The non-radioactive PVBO showed 12.46-fold stronger binding a�nity to ERβ than to ERα in vitro. 18 F-PVBO was synthesized with a 15-28% radiochemical yield (n = 5) within 40 min, and the radiochemical purity was >98%. The uptake of 18 F-PVBO in DU145 cells was signi�cantly blocked by ERB-041 (p < 0.05). The uptake of 18 F-PVBO in DU145 xenografts increased during the 120 min dynamic scanning, with a maximum uptake of 2.80 ± 0.30% ID/g at 120 min. Based on time active curves (TACs), injection of 18 F-PVBO with unlabeled PVBO or ERB-041 resulted in a signi�cant signals reduction with the T/M ratio <1 at 30, 60, 75, and 120 min post-injection (p <0.05). Comparison of the %ID/g showed 18 F-PVBO had a higher T/M ratio compared to 18 F-FES in DU145 tumor-bearing mice at 60 min (1.65 vs. 1.28), 75 min (1.76 vs. 1.35), and 120 min (1.80 vs. 1.37) (p < 0.05). Conclusion: 18 F-PVBO shows 12.46-fold stronger binding to ERβ over ERα, with high radiochemical stability. It demonstrates the feasibility of noninvasively imaging ERβ positive tumors by small-animal PET and provides a new strategy for visualizing of ERβ in vivo.


Introduction
For decades, molecular imaging has brought signi cant impact to clinical diagnosis and treatment strategies.Functional receptors are the primary theranostic targets in nuclear medicine and have been actively investigated for targeted imaging and therapy in preclinical research and clinical practice.It is well established that estrogen receptor (ER), function as an essential nuclear receptor, is a valuable biomarker for noninvasive imaging of hormonal status [1].ERα is the most widely studied ER at present [2].Evidence proves the signi cance of ERα in determining the diagnosis and treatment of patients with estrogen-dependent cancers [3,4].Notably, ERα was regarded as the only ER form until ERβ was rstly cloned in 1996 [2].ERβ has a DNA-binding domain and ligand-binding domain that is 96% and 60% homologous with those of ERα, respectively [2], indicating that it may have similar but not identical functions.In addition to presenting in mammary epithelial and ductal cells, ERβ is detected in various tissues, such as subcutaneous adipose tissue [5], brain, and prostate [6].Furthermore, changes in ERβ expression levels and downstream pathways are involved in regulating many physiological and pathological processes [7][8][9], such as the occurrence and development of the Parkinson's disease [10], endometriosis [11], coronary heart disease [12], and diabetes [13].All these ndings highlight ERβ to be a promising biomarker for the early diagnosis of a variety of diseases.
With more evidence showing different functions of ERα and ERβ, independent imaging and quanti cation of them using molecular imaging technique are desired for de ning their roles in living subjects.The food drug administration (FDA) -approved positron emission tomography (PET) probe, 16α-18 F-17βuoroestradiol ( 18 F-FES), for ER targeting shows 6.3 times binding a nity for ERα than that of ERβ [14].It has been recognized as an ERα speci c PET probe [15] and widely used to screen and evaluate breast cancer patients receiving endocrine therapy [16].Given its much higher binding to ERα and relatively narrow indications, 18 F-FES has limitations in the molecular imaging of ERβ and ERβ related diseases.
At present, applications for ERβ PET imaging have been rarely studied.Some attempts have been spent to develop ERβ PET probes, while all the results are disappointing.For example, Yoo J. et al. developed 18 F-FEDNP, which has an 8.3-fold binding a nity preference for ERβ over ERα.Biodistribution studies were performed in the ER knockout mice, but the probe's speci c uptake levels were modest [17].Lee J. K. et al. reported 18 F-8BFEE 2 with high ERβ selectivity, but the attempts to improve in vivo targeting ability and biodistribution pro le failed [18].Antunes I. F. et al. reported 18 F-FHNP had 3.5 times higher binding a nity for ERβ over ERα.The PET imaging was performed in ERα positive SKOV3 xenograft model, and 18 F-FHNP showed 2-fold lower tumor uptake than 18 F-FES [19].The previous studies highlight the strong need to develop new generation PET probes for evaluating ERβ levels in vivo.

Chemistry-General Methods
All of the reagents and materials were purchased from Sigma Chemical Co.(St.Louis, MO, USA), Bide Pharmatech Co. (Shanghai, China), and Energy Chemical Co.(Shanghai, China).Reactions took place opening to the atmosphere unless otherwise speci ed.The reactions were monitored by thin-layer chromatography (TLC) using 200 μM silica gel (China National Pharmaceutical Group Co., China) or Dionex UltiMate 3000 HPLC. 1 H and 13 C NMR spectra were recorded on a Bruker Avance III 400 spectrometer.High-resolution mass spectra (HRMS) were performed on an Agilent G6520 Q-TOF mass spectrometer. 18F-Fluoride was produced using a Sumitomo 10MeV cyclotron. 18F-Fluoride was produced from 18 O-water and the 18 O(p,n) 18 F reaction.The synthesis and characterization of PVBO and the precursor are presented in supplemental materials.
In vitro relative binding a nity of the reference compound PVBO towards ERα and ERβ Relative binding a nities were determined by a competitive radiometric binding assay using 3 H-estradiol ([2,4,6,7-3 H]-17β-estradiol, 70-115 Ci/mmol, Perkin-Elmer, Waltham, MA).Data are presented as the relative binding a nity (RBA) with the RBA of estradiol towards ERα and ERβ set to 100%.The experiment and data analysis were performed according to a previously described binding assay protocol [21].

Computational Details
Molecular docking for PVBO was performed against ERα using the highest resolution X-ray crystal structure (PDB entry 6VIG, 1.45 Å) by Schrodinger suit [22].As for ERβ, molecular docking was performed with the crystal structure of ERβ complexed with ERB-041 (2-(3-uoro-4-hydroxyphenyl)-7-vinyl-1,3benzoxazol-5-ol) (PDB entry 1X7B), because PVBO was derived from ERB-041.All the water molecules as well as the ligand were removed by PyMOL.Before docking, the Schrödinger protein preparation wizard module was used to prepare the 3D structures of protein, optimize 3D structures with pH = 7.0, and restrain minimization converge heavy atoms to RMSD of 0.30 Å, using force eld OPLS_2005.The docking grid of 20 Å was generated over the co-crystallized ligand with the Receptor grid generation module.PVBO and ERB-041 were pre-processed by the Ligprep, with Epik to generate the proper protonation states at pH 7.0, with at most 32 stereoisomers generated.The compounds were then docked to the protein using the "extra precision" glide docking.

Radiosynthesis
Starting from unlabeled precursor 12, 18 F-PVBO was obtained using a two-step radiosynthetic method.
The 18 F-uoride solution was passed through a QMA SepPak Light anion exchange cartridge (Waters) and eluted from the cartridge into a vial with MeCN (aq) (1.1 ml) containing Krypoto x 222 (13 mg) and K 2 CO 3 (3 mg).The solvents were evaporated at 120 °C and then dried 3 more times by adding 1 ml of MeCN.Then, precursor 12 was dissolved in 1 ml of MeCN and added to the vial.The reaction system was sealed and heated at 120 °C for 20 min.After the solvents were evaporated, 1 ml of DCM and 0.25 ml of TFA were added, and the mixture was heated at 60 °C for 5 min to remove the methoxymethyl ether group.The product was puri ed by HPLC (Column: XBridge BEH C18 OBD Prep Column, 5 μm, 10 mm × 250 mm; eluent: acetonitrile in 0.3% phosphoric acid solution; ow: 3 ml/min. 18F-FES was produced as previously described [23]. Distribution coe cients (Log D 7.4 ) HPLC-puri ed 18 F-PVBO (100 μCi/tube) was added to a mixture of n-octanol/PBS pH 7.4 (0.5 mL/0.5 mL).After vortexing at room temperature for 10 min, the tubes were shaken in a water bath at 37°C for 30 min.After standing for strati cation, 0.2 mL of solution was drawn from both phases.The radioactivity was counted using an automated gamma counter (Wizard 2, model 2480, PerkinElmer).The experiments were performed in triplicate.

In vitro stability
The stability of 18 F-PVBO was measured in PBS or fetal bovine serum (FBS).Brie y, 18 F-PVBO (100 μL, 37 MBq/mL) was added to 500 μL of PBS or FBS and incubated at 37 °C for 1 h or 2 h.Stability was evaluated by HPLC.

Cell culture and animal models
The human prostate cancer cell line DU-145 and the mouse embryonic broblast cell line NIH/3T3 were purchased from the American Type Culture Collection (Manassas, USA).Cells were cultured in the Minimum Essential Medium (MEM, Gibco, Carlsbad, USA) or Dulbecco's modi ed Eagle's medium (DMEM, Gibco, Carlsbad, USA) supplemented with 10% FBS (Gibco, Carlsbad, USA) in a 37°C incubator with 5% CO 2 .
Athymic female nude mice (6-8 weeks, n = 10, Slaccas, Shanghai, China) and female BALB/c mice (10-12 weeks, n = 15, Slaccas, Shanghai, China) were used in this study.All animal experiments were approved by the Institutional Animal Care and Use Committee of the Shanghai Institute of Materia Medica, Chinese Academy of Sciences.After one week of acclimatization, DU-145 cells (1 ×10 6 cells/mouse) were injected into the right thigh subcutaneously.

Cytotoxicity assay
The DU-145 cell line and NIH/3T3 cell line were cultured in MEM or DMEM supplemented with 10% FBS in a 37°C incubator with 5% CO 2 .MTT was used to assess the cytotoxicity of different concentrations (0, 10, 25, 50, 75, and 100 μM) of the unlabeled PVBO on these two cell lines.Brie y, 100 μL of cells were seeded in a 96-well plate at a cell density of 5000 cells/well and incubated for 24 h.The cells were treated with PVBO at different concentrations (0, 10, 25, 50, 75, and 100 μM) and incubated for 24 h.Ten microliters of MTT solution were added to each well and incubated for another 4 h.Then, the MTT solution was discarded, 150 μL of DMSO was added to each well, and the absorbance at 490 nm was measured with a microplate reader (Synergy H1 Hybrid, Biotek, Winooski, USA).

Cell uptake
Cell uptake was performed using DU145 cells in the presence and absence of the ERβ selective drug ERB-041 [20].DU145 cells were seeded in 12-well plates at a cell density of 1×10 5 cells/well and incubated for 24 h.Fresh serum-free medium (1 ml) containing 18 F-PVBO (37 MBq) was added to each well.To inhibit the uptake of 18 F-PVBO, DU145 cells were pretreated with the ERβ-selective ERB-041 for 1 h and then incubated with 18 F-PVBO (37 MBq) at 37°C for 15, 30, 60, and 120 min.Each well was washed with PBS three times before cell lysis with NaOH.The results were measured using an automatic gamma counter (Wizard 2, model 2480, PerkinElmer).Cell uptake is presented as the percentage of the total added radioactivity dose.

Immunohistochemical (IHC) staining
DU145 tumors were xed with formalin.IHC staining was performed on 4 μm sections taken from para n-embedded tumor tissues.After rehydration, the sections were incubated with a 1:200 dilution of the primary antibody, rabbit anti-estrogen receptor alpha antibody (ab3575, Abcam, Cambridge, UK) and rabbit anti-estrogen receptor β antibody (ab3576, Abcam, Cambridge, UK) at 4°C overnight.Then, the sections were incubated with a biotinylated secondary antibody (ab205718, Abcam, Cambridge, UK) and treated with Avidin-biotin-peroxidase.The 3-amino-9-ethylcarbazole substrate chromogen was used, followed by tissue counterstaining with hematoxylin.Sections were examined and photographed using a confocal microscope (Olympus, Tokyo, Japan)

Western blot
The membranes were incubated with 1:1000 dilutions of primary rabbit anti-estrogen receptor alpha (ab3575, Abcam, Cambridge, UK), rabbit anti-estrogen receptor β (ab3576, Abcam, Cambridge, UK) and anti-GAPDH (1:50000, 60004-1-lg, Proteintech, Rosemont, USA) antibodies at 4 °C overnight.The next day, after washing with PBS three times for 10 min each, the membranes were incubated with the secondary antibodies for 1 h at room temperature.Finally, the target proteins were visualized using a LAS4000 enhanced chemiluminescence system (GE Healthcare, Wisconsin, USA).

Micro-PET/CT
Micro-PET/CT (Inveon, Siemens Company, Germany) studies were carried out in DU145-bearing athymic female nude mice using 18 F-PVBO or 18 F-FES.The in vivo targeting speci city of 18 F-PVBO was evaluated in a blocking study with the administration of unlabeled PVBO (5 μmol/kg) or ERB-041 (5 μmol/kg) 30 min before 18 F-PVBO injection.The dose of 18 F-PVBO was 7.02 ± 0.93 MBq per mouse.The dose of 18 F-FES was 6.51 ± 0.64 MBq per mouse.A 120 min dynamic PET scan was immediately performed for 2 h after injecting the radiotracer through the tail vein.A 10 min CT scan was acquired for anatomic localization.The relative position of the organ was de ned based on the anatomy of CT.The uptake values were expressed as the percentage of the injected dose per gram of tissue (%ID/g).

Biodistribution
BALB/c mice were anesthetized via inhalation of 2% iso urane and injected with 3-5 MBq radiotracer through the tail vein for the biodistribution study.The mice were sacri ced at 30 min, 60 min and 120 min after injection.Blood was withdrawn by cardiac puncture, and the organs/tissues were collected, weighed and evaluated using an automatic gamma counter (Wizard 2, model 2480, PerkinElmer).Uptake values were expressed as percentage of injected dose per gram (%ID/g).

Histological analysis of major organs
The mice were sacri ced at 7 days after 18 F-PVBO PET imaging.Tissue samples from the heart, liver, spleen, lung, kidney, and brain were collected for pathological examination.H&E staining of the major organs of nude mice was performed based on a standard procedure [24].

Probe synthesis
The tosylate-precursor compound 12 for radio uorination and the non-radioactive reference compound PVBO were obtained by multiple organic synthesis steps including MOMBr and TBSCl hydroxyl group protection, nitration reaction, Wittig reaction, reduction reaction, aldoamine condensation, and uorination reaction (Fig. 1).The products were characterized and con rmed by mass spectroscopy (MS) and nuclear magnetic resonance (NMR), including 1 H-NMR and 13 C-NMR as listed in the experimental section.The puri cation of the compounds was performed by Semi-Preparative HPLC.The purity of the product was above 98%.

In vitro binding a nity and cytotoxicity
The relative binding a nity (RBA) of the reference compound PVBO for ERα and ERβ was 0.52% and 6.48% relative to estradiol, respectively.PVBO showed a 12.46-fold higher selectivity for ERβ than for ERα.Cell viability was found to be over 90% after treatment with PVBO at concentrations as high as 100 μΜ, suggesting PVBO displayed low cytotoxicity towards DU145 cells and NIH/3T3 cells (Supporting g.S1).

Molecular docking study
The docking score of PVBO against ERα (Fig. 3a) was -8.11 Kcal/mol, weaker than that of ERβ (Fig. 3b), -8.25 Kcal/mol, which was in consistent with the relative binding a nity data.Docking of ERB-041 against its own crystal structure receptor (self-docking) yielded the docking score of -8.30Kcal/mol with the binding mode comparatively consistent with the crystal conformation (RMSD = 0.33) (Fig. 3c).

ERβ expression in DU145 cells and cell uptake of 18 F-PVBO
The results of IHC showed that in DU145 xenografts, the expression of ERα was barely detectable, while the expression of ERβ was much higher (Supporting g.S2).Western blotting results showed that compared with NIH/3T3 cells, DU145 cells had a signi cantly higher ERβ expression (Fig. 4a).The uptake of 18 F-PVBO in DU145 cells increased with extending incubation times and reached 4.17±0.29%after 120 min.Importantly, in the presence of ERB-041, there was signi cant inhibition uptake of 18 F-PVBO at 15, 30, 60, and 120 min (Fig. 4b, p < 0.05).

Micro-PET/CT in vivo imaging
Representative PET/CT images of 18 F-PVBO and 18 F-FES in DU145 tumor-bearing mice are shown in Fig. 5a and 5b.Based on the time activity curves (TACs) of 18 F-PVBO, the tumor uptake showed a gradual upward trend over 120 min dynamic scan (Fig. 5d, n = 3).The maximum uptake was 2.80 ± 0.30% ID/g at 120 min.All images also showed high liver uptake along with intestinal excretion.As a comparison, 18 F-FES showed maximum accumulation in the tumor at 10 min post-injection (PI), with a gradual decrease during 120 min of scanning (1.80 ± 0.13% ID/g at 120 min) (Fig. 5e, n = 3).As shown in Fig. 5c, in the preinjected PVBO and ERB-041 groups, only weak tumor signals were recorded in PET images at 120 min.The TACs of the blocking experiment are shown in Fig. 5f and 5g (n = 3 per group).Based the T/M ratio, co-injection of 18 F-PVBO with unlabeled PVBO or ERB-041 resulted in a signi cant reduction of the T/M ratio at each time point (p < 0.05).Importantly, T/M ratios of 18 F-PVBO were higher than 18 F-FES in DU145 tumor-bearing mice at 60 min (1.65 vs. 1.28), 75 min (1.76 vs. 1.35), and 120 min (1.80 vs. 1.37)PI (Fig. 5h, p < 0.05).

Biodistribution of 18 F-PVBO
To further evaluate the in vivo characteristics of 18 F-PVBO, a biodistribution experiment was performed in BALB/c mice. 18F-PVBO showed high accumulation in the liver and heart, with a slow clearance in 120 min.The liver uptake was 21.21 ± 9.79% ID/g at 30 min PI and decreased to 14.02 ± 8.65% ID/g at 120 min PI.The uptake of the small intestine was 2.00 ± 0.89% ID/g at 30 min and increased to 12.02 ± 11.86% ID/g at 60 min PI, indicating clearance of 18 F-PVBO through the liver-hepatobiliary system.The radioactive signals of the kidney and bladder were at a low level.Accumulation of 18 F-PVBO in the brain was observed.Uptake in bone was found and reached 6.01 ± 4.58% ID/g at 120 min PI, indicating moderate de uorination of 18 F-PVBO in mice (Fig. 6, Supporting table S1).

H&E staining
The were sacri ced 7 days after 18 F-PVBO PET imaging.Tissue samples from the heart, liver, spleen, lung, kidney, and brain were collected for pathological examination.Based on H&E staining, no acute pathological changes were observed in these tissues (Supporting g.S3).

Discussion
The discovery of PET was a milestone in the development of modern imaging technology.At present, it is a highly powerful imaging method to detect cellular and molecular events in vivo, which reveals metabolic function through uptake of the radionuclide-labeled probe [25].Although research on hormone ligands in nuclear medicine is in the ascendant, ERβ-targeted radioactive probes are not intensively investigated.Meanwhile, because of limited pathological information of ERβ, the choice of animal models for PET imaging of ERβ is di cult.Researchers even performed in vivo ERβ PET imaging using ERα positive SKOV3 xenograft model, but the result was not satisfactory because of the high expression of ERα in SKOV3 cells [19].Similarly, the previously reported PET probe 18 F-FEDNP tested in ER knockout mice, was regarded as not suitable for ERβ PET imaging [17].Another study introduced 18 F-8BFEE 2 to achieve good ERβ selectivity, it not conduct further in vivo imaging [18].To the best of our knowledge, no in vivo imaging study using ERβ positive model has been carried out, although some ligands have been reported to possess good ERβ selectivity [18].DU145 cells were reported to express only ERβ [26,27], and in our study, ERβ expression was veri ed by western blot and immunohistochemistry, implying DU145 xenografts can be used a model for ERβ PET imaging.
The goal of our work is to develop an 18 F-labeled probe to enable noninvasive imaging of ERβ expression in living tissues.Based on previous research, the 7-position-substituted benzoxazoles attracted our attention [20].Although Zhou et al. successfully converted the 7-position-substituted benzoxazoles compound 92 into ERβ selective probe through radiobromination, there was no further cell tests and in vivo PET imaging to investigate the performance of the probe [28].In our study, before the radiosynthesis, a competitive radiometric binding assay was performed using 3 H-estradiol, revealing a 12.46-fold higher selectivity of PVBO for ERβ over ERα.This value is lower than the reported value for ERB-041 (255.5-foldhigher selectivity for ERβ) [29].To gain insight into the binding nature of PVBO to ERβ, molecular docking simulations were performed.The docking score of PVBO against ERβ is worse than that of ERB-041, -8. 25 Kcal/mol vs. -8.30Kcal/mol, suggesting that the chemical modi cation of 7-position-substituted of benzoxazoles decreases the ERβ selectivity.Although PVBO adopts a similar binding mode to ERB-041 against ERβ regarding the ERB-041 part, the extended side chain of PVBO abolishes the hydrogen bond between Arg-346 in the ERβ and phenolic hydroxyl in the PVBO.Meanwhile, no signi cant cytotoxicity was found at a concentration as high as 100 µΜ, inspiring us to carry further radiolabeling and cell uptake studies. 18F-PVBO was successfully prepared by two-step radiosynthetic method based on the precursor 12. 18 F-PVBO meets the requirements of PET/CT imaging as its high radiochemical purity, speci c activity, high stability and low cytotoxicity.The uptake of 18 F-PVBO in DU145 cells increased with extending incubation times due to their ERβ expression and could be signi cantly blocked by ERB-041, demonstrating the ERβ binding speci city of 18 F-PVBO.
Inspired by the speci c ERβ binding ability of 18 F-PVBO, we further explored its PET imaging performance in DU145 tumor-bearing nude mice models.In vivo PET imaging successfully visualized ERβ positive DU145 xenograft tumors in mice models.The uptake of 18 F-PVBO gradually accumulated in the tumor during dynamic scanning.Co-administration of PVBO or ERB-041 reduced tumor uptake at 2h PI by approximated 50% and 75%, respectively, demonstrating the ERβ targeting speci city of 18 F-PVBO. 18F-FES possessed 2.5-fold higher selectivity for ERα and was considered to provide accurate information about ER expression [30].In our study, at 2h PI, a 1.3-fold higher T/M ratio was observed of 18 F-PVBO than 18 F-FES, indicating that 18 F-PVBO provides better tumor visualization ability for monitoring ERβ-positive tumors.Collectively, the in vitro and in vivo results showed that 18 F-PVBO can be used to image ERβ positive tumors.
To further clarify the distribution of 18 F-PVBO, a biodistribution study was performed using BALC/c mice.
After injection of 18 F-PVBO, the radioactive signal in the blood reached a high level at 30 min and then decreased rapidly at 60 and 120 min.The uptake in the bone was detected 30 min PI and reached a moderate level at 60 min PI, suggesting that 18 F-PVBO was de uorinated in vivo [31], which may be related to the liver metabolism [30,32].Interestingly, it has been found that the bone uptake of 18 F-labeled probes in mice is much higher than in primates [33].Whether 18 F-PVBO will be seriously de uorinated in primates needs further investigation.
Our study meets the urgent needs of noninvasive ERβ detection in living systems.Signi cant uptake and slow clearance were observed in the heart, which con rmed its future application in ERβ-related heart disease.The imaging data also showed that 18 F-PVBO passed the blood-brain barrier, implying 18 F-PVBO may provide an advantage for the diagnosis of ERβ-related intracranial diseases.Because of the lack of in vivo ERβ quanti cation method, research on the gradual change of ERβ in different disease models is complicated. 18F-PVBO has the potential to change such situation.

ConclusionA
novel 18 F-labeled small molecule, 18 F-PVBO, has been developed for PET imaging of ERβ expression.The probe shows 12.46-fold stronger binding to ERβ over ERα, with high radiochemical stability and low cytotoxicity. 18F-PVBO demonstrates the feasibility of noninvasively imaging ERβ positive tumors by small-animal PET.It may provide a new strategy for visualizing ERβ in vivo.Molecular imaging with PET/CT using ERβ-speci c PET probes is expected to help understand the roles of ERβ in various diseases and facilitate the detection of ERβ in clinical setting.Declarations Funding This was partially supported by the National Key Research and Development Program of China 2021YFC2100300 (JM), the National Natural Science Foundation of China under Grant No. 82071976 (HC) and 81901799 (CQ), Shanghai Pujiang Program (19PJ1411100) (HC), and Shanghai Municipal Science and Technology Major Project (HC, ZC), National Key Research and Development Plan Digital Diagnosis and Treatment Equipment Research and Development Project (2017YFC0113300) (XL) and Pudong Hospital, Fudan University, college project (YJYJRC202108/ YJYJRC202101/Zdzk2020-14) (XL).The authors wish to thank Haibing Zhou (School of Pharmaceutical sciences, Wuhan university) to provide ERβ protein.We thank Sheng Liang (Department of nuclear medicine, Xinhua hospital, Shanghai jiao tong university) and Shaoli Song (Department of nuclear medicine, Shanghai cancer center, Fudan University) for technical assistance in radiosynthesis of PET probes.

Figure 1 Synthetic scheme of PVBO Figure 2 Figure 3 Figure 5 Representative
Figure 1