{"paper_id":"343e415e-36cb-4ac0-bf65-1048f3ba89a6","body_text":"Discovery of Alkenyl Oxindole as a Novel PROTAC Moiety for \nTargeted Protein Degradation via CRL4DCAF11 Recruitment  \nYing Wanga,†, Tianzi Weib,†, Man Zhaoa,†, Aima Huangc, Fan Sunc, Lu Chena, Risheng Linb, Yubao Xiea, \nMing Zhanga, Shiyu Xuc, Zhihui Sunc, Liang Hongc,*, Rui Wanga,d,*, Ruilin Tianb,*, Guofeng Lia,* \naSchool of Pharmacy, Shenzhen University Medical School, Shenzhen University, Shenzhen 518055, \nChina \nbKey University Laboratory of Metabolism and Health of Guangdong, Department of Medical \nNeuroscience, School of Medicine, Southern University of Science and Technology, Shenzhen 518055, \nChina \ncGuangdong Key Laboratory of Chiral Molecule and Drug Discovery, School of Pharmaceutical Sciences, \nSun Yat-sen University, Guangzhou 510006, China \ndInstitute of Materia Medica and Research Unit of Peptide Science, Chinese Academy of Medical \nSciences & Peking Union Medical College, Beijing 100050, China \n†These authors contributed equally to this work \n*Corresponding authors. \nE-mail addresses: liguofeng@szu.edu.cn (Guofeng Li); tianrl@sustech.edu.cn (Ruilin Tian); \nwangrui@lzu.edu.cn (Rui Wang); hongliang@sysu.edu.cn (Liang Hong) \n  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nAbstract \nAlkenyl oxindoles have been characterized as autophagosome-tethering compounds (ATTECs), which can \ntarget mutant huntingtin protein (mHTT) for lysosomal degradation. In order to expand the application of \nalkenyl oxindoles for targeted protein degradation, we designed and synthesized a series of \nhetero-bifunctional compounds by conjugating different alkenyl oxindoles with the BRD4 inhibitor JQ1. \nThrough structure-activity relationship study, we successfully developed JQ1-alkenyl oxindole conjugates \nthat potently degrade BRD4. Unexpectedly, we found that these molecules degrade BRD4 through the \nubiquitin-proteasome system, rather than the autophagy-lysosomal pathway. Using pooled CRISPR \ninterference (CRISPRi) screening, we revealed that JQ1-alkenyl oxindole conjugates recruit the E3 ubiquitin \nligase complex CRL4\nDCAF11 for substrate degradation. Furthermore, we validated the most potent \nhetero-bifunctional molecule HL435 as a promising drug-like lead compound to exert antitumor activity \nboth in vitro and  in vivo . Our research provides new employable PROTAC moieties for targeted protein \ndegradation, providing new possibilities for drug discovery. \nKEY WORDS: PROTAC; DCAF11; Alkenyl oxindole; BRD4; Antitumor  \n1. Introduction \nTargeted protein degradation (TPD) has emerged as a promising approach for drug discovery. It uses  \nmultispecific small molecules to selectively recognize target proteins, facilitating their degradation via cell's \nintrinsic protein degradation pathways\n1. Compared with traditional inhibitors, TPD drugs possess the unique \nability to not only inhibit protein activity but also facilitate the degradation of target proteins. This dual \nfunctionality empowers TPD drugs to elicit stronger therapeutic effects and holds promise for targeting \nproteins that were previously deemed \"undruggable\"\n2-4. Currently, TPD strategies primarily utilize two \nmajor degradation pathways: the ubiquitin-proteasome system and the lysosomal degradation pathway 5-7. \nAccording to mechanism of action, the major TPD strategies include proteolysis targeting chimeras \n(PROTACs), lysosome targeting chimeras (L YTACs) and autophagy targeting chimeras (AUTACs) 8-11. \nAmong them, PROTAC technology is the most extensively studied and has achieved significant \nbreakthroughs. It has been successfully applied to degrade more than 100 target proteins, including those \npreviously considered \"undruggable\"12. Moreover, more than 20 PROTACs are currently undergoing clinical \ntrials since 201913-16, indicating that PROTAC technology is a promising therapeutic strategy.  \nPROTACs are heterobifunctional molecules consisting of two ligand domains joined by a chemical linker. \nOne ligand domain binds to the protein target of interest, and the other ligand recruits an E3 ubiquitin ligase. \nBy engaging both the target protein and E3 ligase simultaneously, PROTACs facilitate the \npolyubiquitination and proteasomal degradation of the target protein17. While over 600 E3 ligases have been \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nidentified in the human genome 18, only a small fraction (< 3%) have been successfully recruited by \nPROTACs19. The most commonly utilized E3 ligases include CRBN, VHL, MDM2 and IAPs. More recently, \nKEAP120, 21, RNF11422, 23, DCAF1524 and DCAF1625 have expanded the toolbox of accessible E3 domains. \nHowever, the vast majority (> 90%) of reported PROTAC molecules continue to rely primarily on just two \nligases: CRBN and VHL 12. This narrow E3 diversity poses a major challenge, as the development and \ntargeting potential of PROTAC degraders is constrained by the limited pool of recruited E3s. Therefore, \nbroadening the range of E3 ligases that can be engaged by small molecule ligands could unlock new avenues \nto potentially degrade a wider range of protein targets, as well as circumvent the acquired drug resistance \nthat caused by mutations in certain E3 ligases\n26, 27.  \nThe alkenyl oxindole framework is commonly found in synthetic or natural compounds that exhibit a \nwide range of biological activities and have attracted research interests from pharmacologists and chemists28. \nMany alkenyl oxindoles have been developed as lead compounds or marketed drugs against tumors, such as \nsunitinib29-31. Recently, two alkenyl oxindoles ( 10O5 and AN1) were found to act as molecular glues that \ntether mHTT to LC3, leading to the autophagy-lysosomal degradation of mHTT 32. This prompted us to test \nwhether this strategy can be expanded to degrade other substrates. To this end, we synthesized a series of \nheterobifunctional molecules by linking JQ1 with different alkenyl oxindoles, followed by assessing their \ntargeted degradation activity. This led to the identification of HL435, a highly potent alkenyl oxindole-based \nBRD4 degrader. However, when we investigated the protein degradation mechanism of HL435, we found \nthat it degraded BRD4 through the ubiquitin-proteasome system rather than the autophagy-lysosomal \npathway. Based on this unexpected finding, we hypothesized that alkenyl oxindoles may act as novel E3 \nligase ligands. To verify our hypothesis, we performed a pooled CRISPR interference (CRISPRi) screen, \nfrom which we revealed that the E3 ligase complex CRL4\nDCAF11 is in charge of HL435-induced proteasomal \ndegradation of BRD4. We further validated the anti-tumor efficacy of HL435 both in vitro and in vivo . \nOverall, we discovered that alkenyl oxindoles can act as recruitment moiety for CRL4 DCAF11 and developed \nalkenyl oxindole-based PROTAC molecules with high degradation efficiency and anti-tumor effects, \nexpanding the toolbox of E3 ligases available for PROTAC drug development. \n2. Results \n2.1. Compounds development and structure-activity relationship studies on alkenyl oxindole-based    \nhetero-bifunctional degraders \nTo explore the potential of alkenyl oxindole for target protein degradation, we designed and synthesized a \nseries of hetero-bifunctional molecules by connecting JQ1 with different alkenyl oxindoles using various \nlinkers, followed by examining their ability to degrade BRD4 (Table 1 and Figure S1). Firstly, JQ1 and the \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nreported alkenyl oxindole ( 10O5) were connected directly with saturated or unsaturated alkane chains of \ndifferent lengths to afford compounds H1-H4. However, they had little ability to degrade BRD4. When PEG \nlinker was used to replace the alkane chain ( H5), the degradation of BRD4 was observed at a concentration \nof 1.0 μ M. Furthermore, the direct connection of linker and 10O5 through amide bond ( H6) significantly \nimproved the degradation ability. Therefore, we conducted a subsequent structural-activity study on the \nalkenyl oxindole moiety using the linker of compound H6 . Subsequently, we examined the degradation \nactivity of JQ1-alkenyl oxindole conjugates constructed from different alkenyl oxindole derivatives, \nincluding the addition of electron-poor substituents on the phenyl group or the benzo moiety of oxindole \ncore (Table 1, H7 - H28). The results showed that enhanced degradation activity can be achieved by \nsubstituting the alkenyl oxindole with trifluoromethyl group (Table 1, R = 6-CF 3). These modifications led \nto the development of HL435 ( H27), an excellent BRD4 degrader with a degradation efficiency > 99% at \n1.0 μ M.  \nTable 1: Target Degradation Efficiency and Antiproliferation Activities of Compounds \n \nCompoun\nd Linker Alkenyl Oxindoles \nBRD4 \nDegradation \nRate (%)a \nIC50 (μ M)b \n0.1 \nμ M 1.0 μ M MCF\n-7 \nMDA-MB-23\n1 \nH1  \n \nR = 5-I <10 <10 116.8\n0 2.17 \nH2  R = 5-I <10 <10 147.\n40 36.54 \nH3  R = H <10 <10 6.66 1.96 \nH4c \n R = H <10 <10 21.2\n4 1.37 \nH5 \n \nR = H <10 38 2.66 1.58 \nH6 \n \nR = H 25 49 0.49 0.48 \nH7 \n \n \nR = I 22 50 0.46 0.31 \nH8 R = 5-F 35 86 0.77 0.81 \nH9 R = \n5-Cl 44 78 0.87 0.81 \nH10 R = \n5-Br 30 71 0.95 0.82 \nH11 R = 52 96 0.19 0.50 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n6-CF3 \nH12 \n \nR = 5-F - 24 3.39 1.89 \nH13 R = \n5-Cl - 30 3.05 1.71 \nH14 R = \n5-Br - 30 2.92 2.18 \nH15 R = \n6-CF3 - 55 1.37 2.04 \nH16 \n \nR = 5-F - 47 1.91 1.86 \nH17 R = \n5-Cl - 38 2.73 1.67 \nH18 R = \n5-Br <10 34 2.08 2.06 \nH29 R = \n6-CF3 <10 59 0.72 0.83 \nH20 \n \nR = 5-F 12 62 2.64 1.34 \nH21 R = \n5-Cl 28 32 1.99 1.35 \nH22 R = \n5-Br <10 55 2.75 1.91 \nH23 R = \n6-CF3 35 88 1.67 1.07 \nH24 \n \nR = 5-F 28 89 2.26 1.18 \nH25 R = \n5-Cl 52 94 1.77 0.98 \nH26 R = \n5-Br 53 95 2.28 0.80 \nH27\n（ HL435）  \nR = \n6-CF3 87 >99 0.38 0.21 \nH28 \n \nR = \n6-CF3 27 60 4.02 1.45 \nD27\n（ HL389）   \nN/A N/A N/A ND ND 11.39 9.66 \nJQ1 N/A N/A N/A ND ND 9.96 0.90 \naBRD4 degradation rate was relative quantification result of Figure S1 (WB). bIC50 values against cells proliferation were \naverages from triplicate measurements, determinated by CCK8 assay at 48 h (JQ1 at 72 h). cThe Ar motif of H4 was \n4-iodobenzaldehyde. \n2.2. HL435 potently degrades BRD4 through the ubiquitin-proteasome pathway \nTo evaluate the efficacy of HL435 (Figure 1A) in depleting BRD4, we conducted a \nconcentration-dependent study in human breast cancer cells (Figure 1B and S3A). The maximum \ndegradation efficiency (D max) of HL435 was > 99%, with DC 50 values of 11.9 and 21.9 nM in \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nMDA-MB-231 and MCF-7 cells, respectively (Figure 1C). Kinetics study of BRD4 degradation showed that \ndegradation was observed just 1 h after HL435 treatment (Figure 1D and S3B), with a half-life of 1.38 and \n1.31 h in MDA-MB-231 and MCF-7 cells (Figure 1E), respectively. Meanwhile, HL435 was demonstrated \nto efficiently degrade BRD4 in a concentration-dependent manner in multiple cell lines (Figure S2). \nAlthough the efficacies varied slightly among different cell lines, the D\nmax all > 99%.  \nTo validate the mechanism of BRD4 depletion induced by HL435, we first assessed the mRNA levels of \nBRD4. The mRNA level of BRD4 in the HL435 treated group was not lower than that of control group in \nbreast cancer cells, indicating that HL435 did not affect BRD4 at the transcriptional level but rather at the \nprotein level (Figure 1F). Alkenyl oxindoles were found to bind both LC3 and mHTT for inducing \nautophagy degradation of mHTT32, so we explored whether autophagy-lysosomal inhibitors could rescue the \ndegradation of BRD4 induced by JQ1- alkenyl oxindole-conjugated compounds. Surprisingly, both CQ and \nbafilomycin failed to rescue the degradation of BRD4 induced by HL435, H1, H6 or H7 (10f) in different \ncell lines (Figure 1G, 1H, S4B, S4D). We next treated WT-, ATG5KO -, and ATG4BKO-Hela cells with \nHL435 and found that the absence of LC3 and autophagosomes did not affect the efficiency of HL435 in \ndegrading BRD4 (Figure 1I). Similar results were obtained for H1 (Figure S4C). These results suggested \nthat the degradation of BRD4 by JQ1- alkenyl oxindole-conjugated compounds was independent of the \nautophagy-lysosomal pathway, thus, we suspected that it was perhaps mediated by ubiquitin-proteasome \nsystem. To validate this hypothesis, E1 ubiquitin-activating enzyme inhibitor PYR-41 or proteasome \ninhibitors MG132 or PS-341 was employed to co-treatment with our compounds. As expected, pre-treatment \nwith PYR-41, MG132 and PS-341 all successfully rescued the degradation of BRD4 induced by HL435, H6 \nor H7 (10f) (Figure 1J, 1K, S4D, S4E). Pre-treatment with NEDD8 activating E1 enzyme (NAE1) inhibitor \nMLN4924 also block the degradation of BRD4 by HL435 (Figure 1L), indicating that the degradation \nrequired the activation of Cullin-RING E3 ligase (CRL). When the degradation process was blocked by \nproteasome inhibitor MG132, HL435 increased the ubiquitination level of BRD4 (Figure 1M). Finally, we \nvalidated that both JQ1 and structurally modified alkenyl oxindole HL389, whether used alone or in \ncombination, cannot deplete BRD4 in MDA-MB-231 cells. Moreover, an excess of JQ1 could competitively \nblock the degradation of BRD4 induced by HL435 (Figure 1N, 1O). All these results suggested that \nJQ1-alkenyl oxindole-conjugated compounds including HL435 degrade BRD4 through the \nubiquitin-proteasome pathway rather than the autophagy-lysosomal pathway, and the degradation process \ndepends on compounds simultaneously interacting with the substrate protein and the ubiquitin-proteasome \nsystem. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nFigure 1. HL435 potently degrades BRD4 through the ubiquitin-proteasome pathway. A, Design and developm en\nHL435 as a potent BRD4 degrader. B, Representative WB results, Cells were treated with gradient concentrations of HL\nfor 12 h. C , The DC 50 values of HL435 to degrade BRD4 in MDA-MB-231 and MCF- 7 cells, from relative quan tita\nanalysis. D, Representative WB results, MDA-MB- 231cells were treated with HL435 at 0.5 μ M for gradient ti me\nRelative quantitative analysis for the time-dependent degradation of BRD4. F, The relative mRNA levels of BRD4, \nwere treated with HL435 at 1.0 μ M for 12 h, GAPDH used as control. G, Representative WB results, cells were co- tre\nwith HL435 and chloroquine (CQ) or bafilomycin (Baf) for 6 h, CQ or Baf was pre-treated for 2 h. H , Relative quantita\nanalysis of BRD4 from G. I, Representative WB results (n=3). WT-Hela, ATG5KO-Hela or ATG4BKO-Hela cel ls w\ntreated with indicated concentration of HL435 for 6 h. J , WB results for BRD4 degradation, cells were pre- treated \nPYR-41, MG132 or PS-341 for 2 h, followed by HL435 treatment for 6 h. KRelative quantitative analysis of BRD4 fro\nL, WB results, HL435 and MLN4924 were co-treated at indicated concentration for 6 h, MLN4924 was pre- treated for\nent o f \nHL435 \ntitative \nme. E, \n, cells \ntreated \ntitative \nls w ere \nd with \nfrom J. \nfor 2 h. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nfollowed by treatment with MG132 for 8 h and HL435 for 6 h. Cell lysates were Immunoprecipitated with anti-Flag \nmagnetic beads and immunoblotted for ubiquitination level of BRD4. N, WB, JQ1, HL389 or HL435 was treated for 6 h. O, \nRelative quantitative analysis BRD4 from N. Data were presented as mean ± SEM. Statistical significance was determined \nby One-way analysis of variance (ANOV A) or Mann Whitney test (F). *p < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; \nns, no statistical significance. \n2.3. A focused CRISPRi screen identified CRL4 DCAF11 complex potentially responsible for \nHL435-induced proteasomal degradation activity  \nTo identify the E3 ligase mediating HL435-induced degradation of BRD4, we determined to c onduct a \npooled CRISPR interference (CRISPRi) screen. We first constructed a dual-fluorescence reporter, containing \nthe BRD4 bromodomain 1 (BD1) fused to mScarlet, followed by a P2A self-cleaving fragment and an \nenhanced green fluorescent protein (EGFP) for normalization (Figure 2A). This reporter was stably \ntransduced into HEK293T cells constitutively expressing the CRISPRi machinery (dCas9-BFP-KRAB) from \nthe CL YBL safe harbor locus\n33. Upon treatment with HL435, the relative BD1 intensity was significantly \nreduced, which could be fully restored by MG132 (Figure 2B-D), confirming the sensitivity of the reporter. \nNext, we designed a focused sgRNA library targeting all known human E1, E2, and E3 enzymes, consisting \nof 5071 sgRNAs against 993 genes with 5 sgRNAs per gene, and more than 100 non-targeting control \nsgRNAs. Using this library, we performed a fluorescence activated cell sorting (FACS)-based CRISPRi \nscreen in the BD1 reporter cells based on relative BD1-mScarlet signal (BD1-mScarlet intensity normalized \nto EGFP intensity). In the HL435-treated group, knockdown of any components mediating HL435-induced \ndegradation activity would result in an increased relative BD1-mScarlet signal (Figure 2E), which would not \nincrease in the DMSO-treated group. As shown in Figure 2F and 2H, components of the CRL4\nDCAF11 \ncomplex, including the E3 ligase scaffold Cullin-4B (CUL4B), the RING-finger protein RING-box1 (RBX1), \nthe adaptor Damage-specific DNA binding protein 1( DDB1) and the substrate receptor DDB1 and CUL4 \nassociated factor 11  (DCAF11) were among the top positive hits, whose knockdown increased \nBD1-mScarlet signal in the HL435-treated group. In addition, NAE1 and ubiquitin like modifier activating \nenzyme 3 ( UBA3), which are responsible for CRL neddylation and activation, were also strong hits in the \nscreen (Figure 2F). Importantly, these genes showed no phenotype or only weak phenotype in the DMSO \ngroup. These data indicated that the CRL4\nDCAF11 complex is specifically involved in HL435-induced \nsubstrate degradation (Figure 2F, G and Supplementary Table 1).  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n \nFigure 2. CRISPRi screen identified the CRL4 DCAF11 complex as a potential target mediating HL435- induced proteaso\ndegradation of BRD4. A, Design of BRD4 BD1 dual fluorescent reporter. BRD4 BD1 domain was fused with m Sca\nfollowed by a P2A self-cleaving fragment and an EGFP. B, Validation of BD1 reporter response to HL435 tre atm\nRepresentative fluorescent microscope fields for BD1 reporter levels in HEK29T cells treated with DMSO, HL435 (5 0 n\nor HL435 (50 nM) and MG132 (5 μ M) for 24 hours. Bar = 200 μ m. C, Validation of BD1 reporter response to HL\nasomal \nScarlet, \natment. \n0 nM), \nHL435 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\ntreatment. Relative BD1-mScarlet signal in HEK293T cells was determined by the ratio of mScarlet and EGFP as measured \nby flow cytometry. The cells were treated with DMSO, HL435 (50 nM), or HL435 (50 nM) and MG132 (5 μ M) for 24 \nhours. D, Quantification of the relative BD1-mScarlet signal in the indicated groups was shown in the bar graph (mean ± \ns.d., n = 3 biological replicates). E,  CRISPRi screen strategy. CRISPRi HEK293T cells harboring BD1 reporter were \ntransduced with an sgRNA library targeting all human E1, E2, and E3 enzymes. Cells were treated with DMSO or HL435 \n(50 nM, 24 hours) and 5 million cells were taken as “input”. Cells with top 25% relative BD1-mScarlet signal (BD1\nhigh) \nwere sorted via FACS. The frequencies of BD1 high and input cells expressing each sgRNA were determined by \nnext-generation sequencing, and were compared to determine sgRNAs enriched or depleted in the BD1 high population. The \nscreens were performed in duplicates. F, Screening results analyzed by the MAGeCK-iNC pipeline were shown for \nHL435-treated and DMSO-treated groups. A positive phenotype indicates the corresponding sgRNA was enriched in the \nBD1\nhigh population and vice versa. Dots in red, blue, grey and orange represent positive hits, negative hits, negative control \nand other genes, respectively. Genes encoding components of the CRL4 DCAF11 complex and the NEDD8-activating enzyme \nwere highlighted. G, Scatter plot comparing gene scores for the screens under HL435 and DMSO treatment. Genes \nencoding components of the CRL4 DCAF11 complex and the NEDD8-activating enzyme were highlighted. H, Predicted \nstructure of CRL4DCAF11 complex using Alphafold. Statistical significance was determined by One-way ANOVA. ****P < \n0.0001. \nTo validate the requirement of CRL4 DCAF11 complex for the degradation activity of HL435, we \nindividually cloned two separate sgRNAs targeting each of the following genes: DCAF11, DDB1, CUL4B, \nRBX1, NAE1 and UBA3, followed by evaluating their effects on HL435-induced BD1 reporter degradation \n(Figure 3A). As expected, knocking down any of these genes blocked the BD1 reporter degradation upon \nHL435 treatment (Figure 3B-D). Additionally, we confirmed the requirement of DCAF11 in HL435-induced \ndegradation of endogenous BRD4 (Figure 3E and F). Furthermore, we detected interaction between BRD4 \nBD1 and DCAF11 only in the presence of HL435, suggesting the formation of a ternary complex between \nBRD4, HL435 and DCAF11(Figure 3G). Taken together, these data indicate that the CRL4\nDCAF11 complex is \nresponsible for HL435-induced proteasomal degradation of BRD4. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nFigure 3. HL435 recruits CRL4 DCAF11 complex to induce proteasomal degradation of BRD4. A, Validation of knoc kd\nefficiency of different hit genes in CRISPRi HEK293T reporter cells by RT- qPCR (mean ± s.d., n = 3 technical repl ica\nB, Representative fluorescent microscope fields for BD1 reporter levels in the reporter cells expressing sgRNAs ta rge\nindividual hit genes after DMSO or HL435 treatment (50 nM, 24 hours). Bar = 200 μ m. C, Relative BD1-mScarlet si\nwas quantified by flow cytometry for hit gene knockdown in BD1 reporter cells after DMSO or HL435 treatment ( 50 \n24 hours). D, Quantification of the relative BD1 intensity in the indicated groups was shown in the bar graph ( mean ± s.\n= 3 biological replicates). E, Validation of knockdown efficiency of two sgRNAs targeting DCAF11 in CRISPRi HEK2\nreporter cells by RT-qPCR (mean ± s.d., n = 3 technical replicates). F , Western blot showing endogenous protein le ve\nBRD4 and α -tubulin in CRISPRi HEK293T cells expressing sg RNAs targeting DCAF11 after DMSO or HL435 treatm\n \nkdown \nicates). \nrgeting \nt signal \n50 nM, \n s.d., n \nK293T \nvels of \natment \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n(50 nM, 24 hours). G , Co-immunoprecipitation analysis of HA-DCAF11 with Flag-BD1 in the absence or presence of \nHL435 (5 μ M, 6 hours) in HEK293T cells. \n2.4. HL435 potently inhibits proliferation and induced apoptosis of tumor cells in vitro \nBRD4 has been identified as a potential therapeutic target for tumors owing to its contribution to tumor \npathogenesis34, 35. As shown in Table 1,  most of JQ1- alkenyl oxindole-conjugated compounds exhibited \nexcellent anti-proliferation abilities in breast cancer cell lines MCF-7 and MDA-MB-231, with HL435 \nperformed the best overall. The anti-proliferation activities of different compounds were positively \ncorrelated with their BRD4 degradation abilities, suggesting that BRD4 degradation contributed \nsubstantially to the anti-proliferation activity of breast cancer. Notably, the half-maximal inhibitory \nconcentrations (IC\n50) of HL435 against 22RV1 (prostate cancer) was as low as 8.7 nM (Figure 4A), \nsignificantly superior to JQ1 (IC 50 =157 nM). To further explore the biological effects of HL435 on -breast \ncancer, we performed flow cytometric analysis and Western blotting to assess the influence on cell cycle and \napoptosis in MCF-7 and MDA-MB-231 cells. HL435 acted similarly to JQ1 at low concentration, blocking \nthe cell cycle at G0/G1 phase, while higher concentrations of HL435 arrested cell cycle at G2/M phase \n(Figure 4B, 4C, S5A and S5B). Consistent with the results of flow cytometric analysis, immunoblotting \nresults showed that HL435 treatment up-regulated P53 and P21 levels and down-regulated the levels of \nCyclin D1 and Cyclin B1 (Figure 4E, S5C), which contributed to block the G1/S and G2/M transitions. The \nability of apoptosis induction by HL435 in breast cancer cells was > 20-fold more potent than JQ1 (Figure \n4D, S6A). Treatment with HL435 at 1.0µM for 36 h in MDA-MB-231 cells led to an apoptotic rate of 55.9 ± \n1.9%, and the levels of cleaved caspase-9 and PARP1 were profoundly increased accordingly (Figure 4F, \nS6B). Oncogenes c-Myc is the key downstream signal used to evaluate the function of BRD4\n36. As \ndisplayed in Figure 4E, S5C and S7A, both mRNA and protein expression levels of c-Myc were \nsignificantly downregulated in breast cancer cells treated with HL435. These data indicated that HL435 not \nonly exhibits excellent anti-proliferative capacity against multiple tumor cell lines, but also effectively \narrest\ns  the cell cycle and induce s  apoptosis in breast cancer cells, validating the stronger therapeutic \nefficacy of degraders compared to inhibitors. \n2.5. HL435 suppresses tumor growth in vivo \nTo evaluate the anti-tumor ability of HL435 in vivo, a mouse xenograft tumor model of MDA-MB-231 \ncells was employed. Mice bearing xenograft tumor were treated daily with HL435 (20 mg/kg) or vehicle (10% \nDMSO+90% corn oil) 6 days per week. After 27 days of treatment, the HL435 treatment group attenuated \ntumor progression, with a tumor growth inhibition rate (TGI) of 54.34% (Figure 4H) and a 51.12% \nreduction in tumor weight (Figure 4I) compared to vehicle group. Meanwhile, the body weight of \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nHL435-treated group was comparable to that of the vehicle group, and no obvious toxicity or adverse eff\nwere observed throughout the experiment period (Figure 4J), indicating that HL435 was tolerated w\nThese data further validated the anti-tumor efficacy of HL435 in vivo, providing a promising drug- like l\ncompound for anticancer drug development. \nFigure 4. HL435 exhibited excellent anti-tumor efficacy in vitro and in vivo. A , IC50 values of HL435 against m ul\ntumor cell lines, determinated by the CCK8 assay after 48 h treatment. B, Representative flow cytom etry analysis resul\neffects \n well. \ne lead \n \nultiple \nsults of \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nanalysis of cell cycle for B. D , Representative flow cytometry analysis results and quantitative statistical analysis of \napoptosis, MDA-MB-231 cells were treated with indicated compounds for 36 h before stained with an 7AAD/APC \nApoptosis Detection kit. E, Representative Western blotting results of c-Myc and cycle relevant proteins in MCF-7 cells \n(n=3). F, Representative Western blotting results of apoptosis relevant proteins in MDA-MB-231 cells (n=3). G, Picture of \nstripped xenograft tumors at the end of experiment (day 45). NOD-SICD mice bearing the MDA-MB-231 xenograft were \ndaily administered with vehicle (10% DMSO+90% corn oil, i.p.) or HL435 (20 mg/kg, i.p.) 6 days per week for 27 days. H, \nGrowth curve of xenograft tumors after treatment. I, weights of stripped xenograft tumors at day 45. J, Body weight curves \nduring treatment. Data was presented as mean ± SEM (n\n≥ 3). One-way ANOVA or Unpaired t test was employed to \ndetermine statistical significance. **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, no statistical significance. \n3. Discussion and Conclusions \nAlkenyl oxindoles have been characterized as molecular glues that tether mHTT to LC3, enabling the \nlysosomal degradation of mHTT 32. We initially sought to expand the versatility of this approach by \nconjugating alkenyl oxindoles to other substrate binding moieties, generating bifunctional molecules that \nmay facilitate the degradation of various substrate proteins through the autophagy-lysosomal pathway. As a \nproof-of-principle, we generated a series of JQ1-alkenyl oxindole conjugates, from which we indeed \nidentified molecules that potently degrade the target BRD4. However, we discovered that BRD4 degradation \ninduced by JQ1-alkenyl oxindole conjugates does not occur via the autophagy-lysosomal pathway, but \nthrough the ubiquitin-proteasome pathway. We speculated that the oxidized indole structure may recruit E3 \nubiquitin ligase for its degradation activity. To determine the responsible E3 ubiquitin ligase, we conducted a \npooled CRISPRi screen, from which we identified the CRL4\nDCAF11 complex as a potential target for \nmediating the degradation activity of JQ1-alkenyl oxindole conjugates. We showed that alkenyl oxindoles \ncan recruit DCAF11, thus acting as a novel PROTAC moiety for targeted protein degradation. Previously, \nthe Cravatt\n37 and Gray38 groups respectively revealed that DCAF11 serves as an E3 ligase that can support \nprotein degradation triggered by electrophilic PROTACs. Very recently, while we were preparing this \nmanuscript for reviewing, similar findings were reported by Waldmann and Winter et al 39. As alkenyl \noxindole possesses Michael acceptor properties, they confirmed that it recruits DCAF11 through a covalent \nmodification approach, potentially engaging with cysteine residues.  \nPrevious structure-activity studies on PROTACs have mainly focused on the effects of different linker \nlengths on degradation activity, with few studies on the structure-activity of the E3 ligase ligand part. In this \nstudy, we found that the ability to degrade BRD4 were significantly improved through structural \noptimization of E3 ligase ligands, and an excellent BRD4 degrader HL435 was identified, whose JQ1 \nmoiety was conjugated with the trifluoromethyl-substituted alkenyl oxindole via PEG chain. The D\nmax of \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nHL435 > 99%, with DC50 values of 11.9 nM in MDA-MB-231 cells. To explore the druggability potential of \nhetero-bifunctional compounds conjugated with alkenyl oxindoles, we evaluated their anti-tumor abilities \nboth in vitro and in vivo. Most of hetero-bifunctional compounds exhibited more excellent anti-proliferation \nabilities against breast cancer cells than JQ1 or alkenyl oxindoles, supporting the more excellent therapeutic \npotential of degraders compared to inhibitors. Consistent with the degradation efficiency of BRD4, HL435 \nshowed the best anti-proliferative activity overall, with an IC50 as low as 8.7 nM against prostate cancer cells \n22RV1. In addition to outstanding antiproliferative abilities against multiple tumor cells, HL435 can \neffectively arrest the cell cycle and induce apoptosis in breast cancer cells by blocking BRD4 downstream \nsignaling pathway in a concentration-dependent manner. Finally, the anti-tumor efficacy of HL435 in vivo \nwas validated in a mouse TNBC xenograft model, with good tolerability. These data suggested that HL435, \na compound composed of structure modified alkenyl oxindole and BRD4 inhibitor JQ1, was a promising \ndrug-like lead compound for anticancer drug development. \nAlthough there are more than 600 E3 ubiquitin ligases in human cells, the ligand molecules currently \navailable to recruit E3 ubiquitin ligases only cover less than 3%\n18, 19. In addition, with the emergence of E3 \nubiquitin ligase resistance, PROTACs based on the same E3 ubiquitin ligase ligand may be ineffective 40-42. \nTherefore, the development of new E3 ubiquitin ligase ligands can not only solve the limitations of existing \nligands, but also be a major way to expand the scope of PROTACs therapeutic targets and provide better \ntreatment opportunities\n26. Moreover, as DCAF11 is localized in the nucleus, the identification of ligands \ncapable of recruiting DCAF11 provides new possibilities for targeting the degradation of nuclear proteins.  \nIn summary, we discovered alkenyl oxindole as a novel PROTAC moiety for targeted protein degradation \nvia CRL4 DCAF11 recruitment. We also developed JQ1-alkenyl oxindole-conjugated bifunctional molecules \nwith high BRD4 degradation efficiencies in multiple cell lines and proved their anti-cancer effect both in \nvitro and in vivo. Our study expands the E3 toolbox available for PROTACs, which will potentially broaden \nthe spectrum of degradable proteins and improve the efficiency of target degradation, providing new \npossibilities for drug discovery. \n4. Experimental Section \n4.1. Cell lines culture and plasmid transfection  \nHCT116, MCF-7, 22RV1, A549, K562, THP-1, Hela and HEK293T cell lines were previously obtained \nfrom ATCC and cryopreserved in our laboratory. MDA-MB-231 was newly purchased from Procell (Wuhan, \nChina). ATG4BKO-Hela, ATG5KO-Hela, ATG4BKO-HCT116 were kindly gifts from Professor Li (Sun \nYat-sen University). The CRISPRi HEK293T cell line was established by integrating dCas9-BFP-KRAB \ncassette into the CL YBL safe harbor locus via homology-directed repair (HDR) as described previously\n33. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nAll cell lines were mycoplasma-free. HA-Ub and Flag-BRD4 plasmids were purchased from Miaoling \nBiology (Wuhan, China). Cell lines were cultured in RPMI-1640 medium (Gibco, Thermo Fisher Scientific, \nWaltham, MA, USA) or DMEM (Gibco) supplemented with 1% penicillin-streptomycin (Gibco) and 10% \nfetal bovine serum (FBS, sigma), the cultures were maintained in a CO\n2 incubator at 37 °C with 5% (v/v) \nCO2. For transient transfection, HEK293T cells were seeded into 6-well plates and cultured to about 50% \ndensity, then co-transfected with HA-Ub and Flag-BRD4 plasmids for 32 h by Hieff Trans TM Liposomal \nTransfection Reagent (Y easen, China). \n4.2. DNA constructs  \nThe coding sequence of BRD4 BD1 domain ( amino acid N44-E168) was amplified from full-length of \nBRD4 cDNA in a pcDNA3.1-BRD4-3Flag with the forward primer \n(5’-ATGACGA TGACAAGACTAGTaaccccccgcccccagagacctcca-3’) and reverse primer \n(5’-ccgccttcttctgtgggtagctcatttatt-3’) and it was fused with mScarlet-P2A-EGFP from pRT117 vector into the \npLVX-3Flag-Hygro vector with the forward primer \n(5’-ctacccacagaagaaGGCGGTGGCTCGGTGAGCAA-3’) and reverse primer \n(5’-AGGGGCGGGATCCGCGGCCGCttactagtcggttcaactctaggtg-3’) by Hieff Clone\n® Universal One Step \nCloning Kit (YEASEN, 10922ES20). The coding sequence of DCAF11 (Youbio, L11006) was inserted into \na pcDNA3.1-3HA vector with the forward primer (5’- \nTACCTGACTACGCTGGTACCatgggatcgcggaacagcagcag-3’) and reverse primer (5’- \nGATATCTGCAGAATTCctactggggtgaggaaaaggg-3’) by Hieff Clone\n® Universal One Step Cloning Kit \n(YEASEN, 10922ES20). \n4.3. CRISPRi screen  \nThe overall CRIPSRi screen process was performed as described previously 33, 43, 44. In brief, HEK293T \ncells harboring the BRD4 BD1 domain dual fluorescence reporter were infected with the sgRNA library \ntargeting all human E1, E2, and E3 enzymes as described previously. The MOI value was controlled under \n0.3 when library was transduced to the cells. The cells were selected by 2 μ g/mL puromycin for 2 days. \nAfter expansion and puromycin selection, the cells were treated with DMSO and HL435 respectively. 24 \nhours later, 5 million cells were taken as “input”, and the remaining cells were subsequently collected for \nFACS, where the cells were sorted into the top 25% based on the ratio of mScarlet and EGFP signal. For \neach sample, cells corresponding to at least 4,000-fold over the library coverage were sorted per replicate. \nSorted populations were collected and genomic DNA was isolated using DNAiso Reagent (Takara, 9770A). \nsgRNA cassettes were amplified by PCR and subjected to Next-generation sequencing (NGS) by NovaSeq \n6000 PE150. Sequencing results were analyzed using MAGeCK-iNC as previously described\n33. \n4.4. Cell viability assay  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nCells were seeded in 96-well plates at a density of 2000-4000 cells per well. After overnight incubation, \ncompounds were administered at the gradient concentrations for 2-3 days. Then, remove the old medium and \nadd 100 μ L fresh medium with 10% CCK8 reagent (Bimake; Selleck Chemicals; cat. no. B34304) for each \nwell. And the plate was incubated in a cell incubator at 37 °C for 1-3 h. The optical density (OD) value at \n450 nm, which stands for the vitality of the cells, was detected with a BioTek Synergy H1 microplate reader. \nIC50 values of compounds against cell lines were calculated from triplicate measurements. \n4.5. Co-Immunoprecipitation and Western blot analysis  \nCells were lysed in RIPA buffer (Beyotime, Haimen, China) supplemented with phosphatase inhibitors \n(Bimake; Selleck Chemicals, Houston, TX, USA) or protease inhibitor cocktail (Roche, Basel, Switzerland). \nThe protein in each sample was quantified by a BCA protein assay (ThermoFisher, Rockford, I L), and \nboiled with 5×loading buffer (LB) for 5 min. For immunoprecipitation, 1 mg protein in each sample was \nincubated with anti-Flag (P2115, Beyotime, China) or anti-HA magnetic beads (P2121, Beyotime, China) \novernight at 4°C. After washed away non-specifically bound proteins, anti-Flag magnetic beads were boiled \nwith 1× LB for transsexual washout. 8%, 10% and 12% SDS-PAGE gels or 3% Tris-Acetate Polyacrylamide \nGradient Gels were used to separate protein samples, and then transfered to a PVDF membrane (Millipore; \nMerck KGaA). 5% skim milk was used to block membranes at RT for 1 hour. Primary antibodies were \nblotted at 4°C overnight. The next day, the membranes were slowly flipped in secondary antibodies \nconjugated with horseradish peroxidase for 1 hour at RT. Images were captured by Tanon 5200 (Shanghai, \nChina). Image J was used to quantify the intensities of bands. The antibodies used in this paper were as \nbelow: Anti-\n/i2 -Tubulin (T6047) , anti-LC3B (L7543) and anti- β -Actin were purchased from Sigma (St. \nLouis, MO, USA); Anti-BRD4 (13440), anti-PARP1 (9542), anti-Caspase 9 (9505), anti-HA (3724) and \nanti-Cyclin D1(2922) were purchased from Cell Signaling Technology (Danvers, MA); Anti-Ubiquitin \n(sc-8017) was from Santa Cruz (Dallas, TX, USA). Anti-ATG4B (M134), anti-ATG5 (M153) and anti-Flag \nwere from MBL (Tokyo, Japan); Anti-Cyclin B1(55004-1-AP), anti-p53 (10442-1-AP), anti-p21 \n(10355-1-AP), anti-c-Myc (10828-1-AP), and anti-GAPDH (60004-1-Ig), anti-Vinculin (66305-1- Ig), Goat \nanti-Mouse IgG (H+L) (SA00001-1) and Goat anti-Rabbit IgG (H+L) (SA00001-2) were purchased from \nProteintech. \n4.6. Quantitative real-time polymerase chain reaction (qRT-PCR ) analysis  \nMDA-MB-231 or MCF-7 cells were plated in 12-well plates and treated with compounds for 12 h after \novernight incubation. Total RNA was extracted using TRIzol (Invitrogen). A High-Capacity cDNA Reverse \nTranscription kit (Thermo Fisher Scientific) was used to create cDNA from purified RNA. The real-time \nPCR was conducted on a real-time fluorescence quantitative PCR equipment (light-Cycler480II, Roche) \naccording to the protocol of SYBR Green qPCR Mix (Dongsheng Biotech, China). Results analyses were \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nperformed from three or four biological replicates, each data in biological replicate was triplicate. The \nexpression level of genes was calculated with the 2 -ΔΔ Ct technique, and GAPDH was used as an internal \nreference. The expression level of each gene in the DMSO group was normalized to 1, and that of treatment \ngroups were presented as fold-change relative to the DMSO group.  \nThe primer sequences used were as follows:  \nGAPDH-F: GAGTCAACGGA TTTGGTCGT, GAPDH-R: GACAAGCTTCCCGTTCTCAG;  \nBRD4-F: CTCCGCAGACA TGCTAGTGA, BRD4-R: GTAGGATGACTGGGCCTCTG;  \nc-MYC-F: CACCGAGTCGTAGTCGAGGT, c-MYC-R: GCTGCTTAGACGCTGGATTT;  \nP21-F: TGTCCGTCAGAACCCATGC, P21-R: AAAGTCGAAGTTCCATCGCTC; \nDCAF11-F: CAATGATCTGGGCTTCACTGAT, DCAF11-R: TCTTGGCAAGCAGACATGAAT; \nDDB1-F: ATGTCGTACAACTACGTGGTAAC, DDB1-R: CGAAGTAAAGTGTCCGGTCAC; \nNAE1-F: ACCTGTTCGAGGCACAA TTCC, NAE1-R: TCTTTGCTTTTTCACGGTAAACG; \nUBA3-F: CGATCTGGACCCTTCACACAC, UBA3-R: GCCAGCTCCAATGACTAGAAC; \nCUL4B-F: ACTCCTCCTTTACAACCCAGG, CUL4B-R: TCTTCGCATCAAACCCTACAAAC; \nRBX1-F: TTGTGGTTGATAACTGTGCCAT, RBX1-R: GACGCCTGGTTAGCTTGACAT; \nTubulin-F: ACCTTAACCGCCTTATTAGCCA, Tubulin-R: ACATTCAGGGCTCCATCAAATC. \n4.7. Cell cycle assay  \nThe influence of compounds on the cell cycle was detected by cell flow cytometry following instructions \nof the Cell Cycle Analysis Kit (C1052, Beyotime). In brief, seed cells into a 6-well plate at an appropriate \ndensity. After overnight incubation, compounds were administered at the respective concentration for 24 h. \nPre-cooled PBS was used to wash cells before and after centrifugation, followed by overnight fixation in 70% \nethanol at 4 °C. On the next day, ethanol was removed by centrifugation, then cells were dealt with RNase \nfor 30 min at 37 °C. Subsequently, they were stained with propidium iodide (PI) at room temperature for an \nadditional 30 min. If stored at 4 °C, the stained cells can be detected on a flow cytometer (BD FACSCalibur, \nBD Biosciences, USA) within 24 h and analyzed for cell cycle distribution using FlowJo software. \n4.8. Cell apoptosis assay \nThe effect of compounds on inducing apoptosis was detected using cell flow cytometry according to the \ninstructions of Annexin V APC/7-AAD apoptosis kit (AP105-100, liankebio, China). In brief, seed cells into \na 6-well plate at an appropriate density. On the next day, compounds were administered at the respective \nconcentrations for 36 hours. 500 \nμ L 1X binding buffer was used to resuspend the harvested cells, and then \nadd Annexin V-APC (5 μ L) and 7-AAD (10 μ L). Gently mix the solution and incubate it in the dark at RT \nfor 5 minutes. Finally, a flow cytometer (BD FACSCalibur, BD Biosciences, USA) was employed to detect \nthe cells as soon as possible. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n4.9. Animal experiments  \nAnimal experiments were performed at the Experimental Animal Center of Sun Yat-sen University (East \nCampus), and female NOD-SCID mice were purchased from Guangdong Yaokang Biotechnology Co., LTD. \nInject subcutaneously 5 million MDA-MB-231 cells on the right dorsal side of each mouse at the age of 6-7 \nweeks. When tumors size reached 60-70 mm 3 about half a month later, mice were randomly divided into 2 \ngroups. Mice were daily injected with vehicle (10% DMSO+90% corn oil, i.p.) or HL435 (20 mg/kg, i.p.) 6 \ndays per week for 27 days. V olume of xenograft tumor and body weight of each mouse were measured every \n2-4 days. V olume of xenograft tumor = length x width\n2/2. sacrifice all of mice at day 27 post treatment and \nxenograft tumors were excised for weight measurement. TGI (%) = [1 - (TV Treatment/Dx - TV Treatment/D1)/ \n(TVVehicle/Dx– TVVehicle/D1)] × 100%, X = days post treament. The animal experiments were conducted strictly \naccording to animal ethics guidelines and the protocol (No. SYSU-IACUC-2023-000327), approved by the \nInstitutional Animal Care and Use Committee (IACUC) of Sun Yat-sen University Cancer Center. \n4.10. Chemistry \nUnless otherwise stated, all solvents and the compounds without provided synthesis routes were \ncommercially purchased. All solvents were purified and dried according to standard methods before use. The \nspectra of \n1H nuclear magnetic resonance (NMR) was recorded on a Varian instrument (500 MHz or 400 \nMHz), and the tetramethylsilane signal or residual protio solvent signals was used as the internal standard. \n13C NMR was recorded on a V arian instrument (125 MHz or 100 MHz). Data for 1H NMR were recorded as \nfollows: chemical shift (δ , ppm), multiplicity (s = singlet, d = doublet, t = triplet, m = multiplet, q = quartet \nor unresolved, coupling constant (s) in Hz, integration). Data for 13C NMR were reported in terms of \nchemical shift (δ , ppm). The progress of the reaction was monitored by thin-layer chromatography (TLC) on \nglass plates coated with a fluorescent indicator (GF254). Flash column chromatography was performed on \nsilica gel (200-300 mesh). The ESI ionization sources were employed to obtain high resolution mass spectra \n(HRMS). The purity of final key products was confirmed by a Waters e2695 HPLC system equipped with an \nXBridge C18 (5 um, 4.6 x 250 mm) and eluted with methanol/water (97.5: 2.5) at a flow rate of 1.0 mL/min. \nThe yields indicated were from single step reactions. All compounds used in biological tests have been \nfurther purified by preparative liquid chromatography, and all of them showed > 95% purity using the HPLC \nmethods described above. \nAcknowledgements \nThe authors are grateful for financial support from the National Natural Science Foundation of China \n(22271317 to L.H., 22101306 to Ming Z., 32100766 and 82171416 to R.T.), the Medical Innovation and \nDevelopment Project of Lanzhou University (lzuyxcx-2022-156 to R.W.), the CAMS Innovation Fund for \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nMedical Sciences (CIFMS) (2019-I2M-5-074, 2021-I2M-1-026, 2021-I2M-3-001 and 2022-I2M-2-002 to \nR.W.), Guangdong Basic and Applied Basic Research Foundation (2023B1515020075 to R.T.), the Science, \nTechnology and Innovation Commission of Shenzhen Municipality (RCBS20210609103800006, \nJCYJ20220530112602006 and RCYX20221008092845052 to R.T.), the Lingang Laboratory Grant \n(LG-QS-202203-11 to R.T.), and the China Postdoctoral Science Foundation (2023M731523 to T.W.). \nConflict of interest \nAll authors declare no conflict of interest. \nAuthor contributions \nG.L., R.T., L.H., and R.W conceived and designed the project. Y .W., T.W., Man Z., A.H., F.S., L.C., R.L., \nY .X., Ming Z., S.X. and Z.S. performed the experimental work. Y .W., T.W. and Man Z. analyzed the results \nand wrote the manuscript. Contributions to the experimental work include: compounds development and \nstructure-activity relationship studies, A.H., Y .W., F.S., L.C., Man Z., Ming Z., S.X. and Z.S.; degradation \nmechanism verification and anti-tumor efficacy research in vivo and in vitro, Y .W., Man Z. and Y .X.; \nIdentification of recruiting E3 ligase by pooled CRISPRi screening and validation, T.W. and R.L. All authors \nedited and approved the manuscript. \nData Availability Statement \nData supporting the findings of this study is available in the supplementary information of this article. \nReferences \n(1) Deshaies, R. J. Multispecific drugs herald a new era of biopharmaceutical innovation. Nature 2020, \n580 (7803), 329-338.  \n(2) Winter, G. E.; Buckley, D. L.; Paulk, J.; Roberts, J. M.; Souza, A.; Dhe-Paganon, S.; Bradner, J. E. \nPhthalimide conjugation as a strategy for in vivo target protein degradation. Science 2015, 348 (6241), \n1376-1381.  \n(3) Wu, T.; Yoon, H.; Xiong, Y .; Dixon-Clarke, S. E.; Nowak, R. P .; Fischer, E. S. Targeted protein \ndegradation as a powerful research tool in basic biology and drug target discovery. Nat. Struct. Mol. Biol. \n2020, 27 (7), 605-614.  \n(4) Salami, J.; Crews, C. M. Waste disposal—An attractive strategy for cancer therapy. Science 2017, 355 \n(6330), 1163-1167.  \n(5) Ciechanover, A. Intracellular Protein Degradation: From a Vague Idea, through the Lysosome and the \nUbiquitin–Proteasome System, and onto Human Diseases and Drug Targeting. Angew. Chem. Int. Ed. 2005, \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n44 (37), 5944-5967.  \n(6) Dikic, I. Proteasomal and Autophagic Degradation Systems. Annu. Rev. Biochem. 2017, 86 (1), \n193-224. \n(7) Lu, K.; den Brave, F.; Jentsch, S. Pathway choice between proteasomal and autophagic degradation. \nAutophagy 2017, 13 (10), 1799-1800. \n(8) Nalawansha, D. A.; Crews, C. M. PROTACs: An Emerging Therapeutic Modality in Precision \nMedicine. Cell Chem. Bio. 2020, 27 (8), 998-1014.  \n(9) Ding, Y .; Fei, Y .; Lu, B. Emerging New Concepts of Degrader Technologies. Trends Pharmacol. Sci. \n2020, 41 (7), 464-474. \n(10) Takahashi, D.; Moriyama, J.; Nakamura, T.; Miki, E.; Takahashi, E.; Sato, A.; Akaike, T.; \nItto-Nakama, K.; Arimoto, H. AUTACs: Cargo-Specific Degraders Using Selective Autophagy. Mol. Cell \n2019, 76 (5), 797-810  \n(11) Banik, S. M.; Pedram, K.; Wisnovsky, S.; Ahn, G.; Riley, N. M.; Bertozzi, C. R. Lysosome-targeting \nchimaeras for degradation of extracellular proteins. Nature 2020, 584 (7820), 291-297.  \n(12) Cao, C.; He, M.; Wang, L.; He, Y .; Rao, Y . Chemistries of bifunctional PROTAC degraders. Chem. \nSoc. Rev. 2022, 51 (16), 7066-7114. \n(13) Mullard, A. Targeted protein degraders crowd into the clinic. Nat. Rev. Drug. Disc. 2021, 20 (4), \n247-250. \n(14) Li, J.; Chen, X.; Lu, A.; Liang, C. Targeted protein degradation in cancers: Orthodox PROTACs and \nbeyond. The Innovation 2023, 4 (3), 100413. \n(15) Kong, N. R.; Jones, L. H. Clinical Translation of Targeted Protein Degraders. Clin. Pharmacol. Ther. \n2023, 114 (3), 558-568.  \n(16) Mullard, A. First targeted protein degrader hits the clinic. Nat. Rev. Drug. Disc. 2019, 18 (4), \n237-239. \n(17) Burslem, G. M.; Crews, C. M. Proteolysis-Targeting Chimeras as Therapeutics and Tools for \nBiological Discovery. Cell 2020, 181 (1), 102-114. \n(18) Clague, M. J.; Heride, C.; Urbé, S. The demographics of the ubiquitin system. Trends Cell Biol. 2015, \n25 (7), 417-426.  \n(19) Mi, D.; Li, Y .; Gu, H.; Li, Y .; Chen, Y . Current advances of small molecule E3 ligands for \nproteolysis-targeting chimeras design. Eur. J. Med. Chem. 2023, 256, 115444. \n(20) Wei, J.; Meng, F.; Park, K.-S.; Yim, H.; Velez, J.; Kumar, P.; Wang, L.; Xie, L.; Chen, H.; Shen, Y .; et \nal. Harnessing the E3 Ligase KEAP1 for Targeted Protein Degradation. J. Am. Chem. Soc. 2021, 143 (37), \n15073-15083. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n(21) Du, G.; Jiang, J.; Henning, N. J.; Safaee, N.; Koide, E.; Nowak, R. P.; Donovan, K. A.; Yoon, H.; Y ou, \nI.; Yue, H.; et al. Exploring the target scope of KEAP1 E3 ligase-based PROTACs. Cell Chem. Bio. 2022, 29 \n(10), 1470-1481. \n(22) Tong, B.; Spradlin, J. N.; Novaes, L. F. T.; Zhang, E.; Hu, X.; Moeller, M.; Brittain, S. M.; McGregor, \nL. M.; McKenna, J. M.; Tallarico, J. A.; et al. A Nimbolide-Based Kinase Degrader Preferentially Degrades \nOncogenic BCR-ABL. ACS Chem. Biol. 2020, 15 (7), 1788-1794.  \n(23) Luo, M.; Spradlin, J. N.; Boike, L.; Tong, B.; Brittain, S. M.; McKenna, J. M.; Tallarico, J. A.; \nSchirle, M.; Maimone, T. J.; Nomura, D. K. Chemoproteomics-enabled discovery of covalent \nRNF114-based degraders that mimic natural product function. Cell Chem. Bio. 2021, 28 (4), 559-566.e515. \n(24) Li, L.; Mi, D.; Pei, H.; Duan, Q.; Wang, X.; Zhou, W.; Jin, J.; Li, D.; Liu, M.; Chen, Y . In vivo target \nprotein degradation induced by PROTACs based on E3 ligase DCAF15. Signal Transduct. Target Ther. 2020, \n5 (1), 129. \n(25) Zhang, X.; Crowley, V . M.; Wucherpfennig, T. G.; Dix, M. M.; Cravatt, B. F. Electrophilic PROTACs \nthat degrade nuclear proteins by engaging DCAF16. Nat. Chem. Biol. 2019, 15 (7), 737-746.  \n(26) Békés, M.; Langley, D. R.; Crews, C. M. PROTAC targeted protein degraders: the past is prologue. \nNat. Rev. Drug. Disc. 2022, 21 (3), 181-200.  \n(27) Lee, J.; Lee, Y .; Jung, Y . M.; Park, J. H.; Yoo, H. S.; Park, J. Discovery of E3 Ligase Ligands for \nTarget Protein Degradation. Molecules 2022, 27(19), 6515.  \n(28) Chaudhari, P.; Bari, S.; Surana, S.; Shirkhedkar, A.; Wakode, S.; Shelar, S.; Racharla, S.; Ugale, V .; \nGhodke, M. Logical synthetic strategies and structure-activity relationship of indolin-2-one hybrids as small \nmolecule anticancer agents: An overview. J. Mol. Struct. 2022, 1247, 131280.  \n(29) Andreani, A.; Granaiola, M.; Locatelli, A.; Morigi, R.; Rambaldi, M.; Varoli, L.; Vieceli Dalla Sega, \nF.; Prata, C.; Nguyen, T. L.; Bai, R.; et al. Cytotoxic activities of substituted \n3-(3,4,5-trimethoxybenzylidene)-1,3-dihydroindol-2-ones and studies on their mechanisms of action. Eur. J. \nMed. Chem. 2013, 64, 603-612.  \n(30) Hopkins, T. G.; Marples, M.; Stark, D. Sunitinib in the management of gastrointestinal stromal \ntumours (GISTs). Eur. J. Surg. Oncol. 2008, 34 (8), 844-850.  \n(31) Yousefian, M.; Ghodsi, R. Structure–activity relationship studies of indolin-2-one derivatives as \nvascular endothelial growth factor receptor inhibitors and anticancer agents. Arch. Pharm. 2020, 353 (12), \n2000022.  \n(32) Li, Z.; Wang, C.; Wang, Z.; Zhu, C.; Li, J.; Sha, T.; Ma, L.; Gao, C.; Yang, Y .; Sun, Y .; et al. \nAllele-selective lowering of mutant HTT protein by HTT–LC3 linker compounds. Nature 2019, 575 (7781), \n203-209.  \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\n(33) Tian, R.; Gachechiladze, M. A.; Ludwig, C. H.; Laurie, M. T.; Hong, J. Y .; Nathaniel, D.; Prabhu, A. \nV .; Fernandopulle, M. S.; Patel, R.; Abshari, M.; et al. CRISPR Interference-Based Platform for Multimodal \nGenetic Screens in Human iPSC-Derived Neurons. Neuron 2019, 104 (2), 239-255. \n(34) Duan, Y .; Guan, Y .; Qin, W.; Zhai, X.; Y u, B.; Liu, H. Targeting Brd4 for cancer therapy: inhibitors \nand degraders. Med. Chem. Comm. 2018, 9 (11), 1779-1802. \n(35) Stathis, A.; Bertoni, F. BET Proteins as Targets for Anticancer Treatment. Cancer Discov. 2018, 8 (1), \n24-36. \n(36) Lovén, J.; Hoke, Heather A.; Lin, Charles Y .; Lau, A.; Orlando, David A.; Vakoc, Christopher R.; \nBradner, James E.; Lee, Tong I.; Young, Richard A. Selective Inhibition of Tumor Oncogenes by Disruption \nof Super-Enhancers. Cell 2013, 153 (2), 320-334.  \n(37) Zhang, X.; Luukkonen, L. M.; Eissler, C. L.; Crowley, V . M.; Yamashita, Y .; Schafroth, M. A.; \nKikuchi, S.; Weinstein, D. S.; Symons, K. T.; Nordin, B. E.; et al. DCAF11 Supports Targeted Protein \nDegradation by Electrophilic Proteolysis-Targeting Chimeras. J. Am. Chem. Soc. 2021, 143 (13), 5141-5149.  \n(38) Sarott, R. C.; You, I.; Li, Y .-D.; Toenjes, S. T.; Donovan, K. A.; Seo, P.; Ordonez, M.; Byun, W. S.; \nHassan, M. M.; Wachter, F.; et al. Chemical Specification of E3 Ubiquitin Ligase Engagement by \nCysteine-Reactive Chemistry. J. Am. Chem. Soc. 2023, 145 (40), 21937-21944. \n(39) Xue, G.; Xie, J.; Hinterndorfer, M.; Cigler, M.; Dötsch, L.; Imrichova, H.; Lampe, P.; Cheng, X.; \nAdariani, S. R.; Winter, G. E.; et al. Discovery of a Drug-like, Natural Product-Inspired DCAF11 Ligand \nChemotype. Nat. Commun. 2023, 14 (1), 7908.  \n(40) Gooding, S.; Ansari-Pour, N.; Towfic, F.; Ortiz Estévez, M.; Chamberlain, P. P.; Tsai, K.-T.; Flynt, E.; \nHirst, M.; Rozelle, D.; Dhiman, P.; et al. Multiple cereblon genetic changes are associated with acquired \nresistance to lenalidomide or pomalidomide in multiple myeloma. Blood 2021, 137 (2), 232-237.  \n(41) Shirasaki, R.; Matthews, G. M.; Gandolfi, S.; de Matos Simoes, R.; Buckley, D. L.; Raja V ora, J.; \nSievers, Q. L.; Brüggenthies, J. B.; Dashevsky, O.; Poarch, H.; et al. Functional Genomics Identify Distinct \nand Overlapping Genes Mediating Resistance to Different Classes of Heterobifunctional Degraders of \nOncoproteins. Cell Rep. 2021, 34 (1), 108532.  \n(42) Zhang, L.; Riley-Gillis, B.; Vijay, P.; Shen, Y . Acquired Resistance to BET-PROTACs \n(Proteolysis-Targeting Chimeras) Caused by Genomic Alterations in Core Components of E3 Ligase \nComplexes. Mol. Cancer Ther. 2019, 18 (7), 1302-1311.  \n(43) Tian, R.; Abarientos, A.; Hong, J.; Hashemi, S. H.; Yan, R.; Drager, N.; Leng, K.; Nalls, M. A.; \nSingleton, A. B.; Xu, K.; et al. Genome-wide CRISPRi/a screens in human neurons link lysosomal failure to \nferroptosis. Nat. Neurosci. 2021, 24 (7), 1020-1034.  \n(44) Samelson, A. J.; Tran, Q. D.; Robinot, R.; Carrau, L.; Rezelj, V . V .; Kain, A. M.; Chen, M.; \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint \n\nRamadoss, G. N.; Guo, X.; Lim, S. A.; et al. BRD2 inhibition blocks SARS-CoV-2 infection by reducing \ntranscription of the host cell receptor ACE2. Nat. Cell Biol. 2022, 24 (1), 24-34.  \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted February 19, 2024. ; https://doi.org/10.1101/2024.02.15.580430doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}