Aptamer and N-Degron Ensemble (AptaGron) as a Target Protein Degradation Strategy Authors: Mohammad Faysal Al Mazid, Olha Shkel, Eunteg Ryu, Jiwon Kim, Kyung Ho Shin, Yun Kyung Kim, Hyun Suk Lim, Jun-Seok Lee Published: 2024 ABSTRACT AbstractTarget protein degradation (TPD) is a promising strategy for catalytic downregulation of target proteins through various cellular proteolytic pathways. Despite numerous reports on novel TPD mechanisms, the discovery of target-specific ligands remains a major challenge. Unlike small-molecule ligands, aptamers offer significant advantages, owing to their SELEX-based systematic screening method. To fully utilize aptamers for TPD, we designed an aptamer and N-degron ensemble system (AptaGron) that circumvents the need for synthetic conjugations between aptamers and proteolysis-recruiting units. In our AptaGron system, a peptide nucleic acid containing an N-degron peptide and a sequence complementary to the aptamer was designed. Using this system, we successfully degraded three target proteins, tau, nucleolin, and eukaryotic initiation factor 4E (eIF4E), which lack specific small-molecule ligands. Our results highlight the potential of the AptaGron approach as a robust platform for targeted protein degradation. Introduction Target protein degradation (TPD) has emerged as a promising strategy for the next-generation therapeutics, particularly for targeting “undruggable” proteins.1 Considering over 85% of the proteome is still undruggable because of the lack of well-defined ligand-binding pocket structures of the target protein,2 TPD has great potential because it does not limit the scope of the target proteins. To date, a series of TPD strategies have been developed, inspired by the pioneering concept of proteolysis-targeting chimeras (PROTAC),1b which utilize the proximal promiscuity of ubiquitin E3 ligase and the subsequent proteasomal degradation process. More recently, this field has expanded to include research on molecular glues3 and other TPD strategies that leverage lysosomal degradations, such as lysosomal targeting chimera (LYTAC),4 antibody-based PROTAC (AbTAC), covalent nanobody-based PROTAC (GlueTAC),5 and autophagy targeting chimera (AUTAC and AUTOTAC).6 Despite the variety of TPD platforms providing versatile options in terms of the degradation mechanism, the major hurdle for TPD remains the target protein engagement step. To achieve target protein binding, current TPD strategies have utilized the following: (i) known small molecule ligands; however, many proteins lack reported ligands; (ii) antibodies, which can only be applied to cell surface proteins owing to their lack of permeability; or (iii) intrinsic interaction partners based on the molecular glue structures; however, discovering neo-substrates is a highly challenging task and still relies on the serendipitous discovery at the moment. On the other hand, aptamers have a unique advantage of selective target-binding aptamers can be systematically discovered by taking advantage of SELEX for any targets.7 Additionally, the cellular delivery of oligonucleotides in the cytosol has been well-established by utilizing transfection agents, such as lipofectamine, or liposomes, as successfully demonstrated for the mRNA based COVID vaccine.8 Nevertheless, aptamers have been utilized, to a limited extent, in TPD studies for (i) cancer cell engagement with PROTAC molecules,9 (ii) extracellular or membrane protein targeting of LYTAC,10 or (iii) creation of functional units of PROTAC through the chemical conjugation of aptamers with ligands for VHL E3 ligase or cereblon.11 Although previous studies have presented the potential of aptamers as target protein-binding units in TPD, each degrader molecule must be individually conjugated between the aptamer and the E3 ligase-binding ligand. Given that the current challenge in degrader discovery lies in the precise optimization of the “target ligand-linker-proteolysis unit” scheme, developing simple and robust methods for derivatizing and generating degrader molecules is crucial. In this regard, we envisioned that a bifunctional ensemble consisting of an aptamer for target-selective binding and a functional motif for proximal proteolysis induction has great potential as a general TPD platform. Herein, we report the first ensemble system consisting of an aptamer and a N-degron peptide. To eliminate the hurdle of generating synthetic degrader molecules, we designed an ensemble system that hybridizes base pairing between the aptamer and the N-degron peptide linked to a peptide nucleic acid (PNA), which is a pseudopeptide mimicking nucleic acid (Figure 1).12 This design benefits from both the straightforward hybridization of aptamers and PNA, and its efficient degradation via the cellular N-end rule pathway. The N-end rule pathway is a proteolytic mechanism responsible for the removal of misfolded or redundant proteins by identifying those with destabilizing N-terminal amino acids (N-degrons).13 This pathway is widespread and operates in both prokaryotic and eukaryotic organisms. In the mammalian system, unmodified Arg acts as a primary N-degron that is recognized directly by the E3 N-recognin, which is a sensing component for N-degrons comprising ∼80-residue regions called the UBR domain. To date, type 1 degrons (positively charged residues: Arg, Lys, and His) are known to recognize several UBR proteins, including UBR1, UBR2, UBR4, and UBR5. Therefore, any of these degron-containing proteins are recognized by UBR proteins and processed for degradation. Our research group has previously reported the selective degradation of steroid receptor coactivator-1 (SRC-1) through the N-degron pathway using a tetrapeptide handle.14 Moreover, we demonstrated that the N-degron-dependent TPD agent showed universal target degradation regardless of the cell type, whereas conventional thalidomide-based PROTAC exhibited cell-type dependent degradation patterns due to the distinctive expression of the E3 ligase CRBN.1b Our aptamer N-degron ensemble system (AptaGron) in turn provides a robust platform for designing TPD agents for any target proteins for which aptamers are available. We illustrated this with three case studies involving tau, nucleolin, and elF4E. Schematic comparison of conventional PROTAC approach versus aptamer and N-degron ensemble (AptaGron) system. PROTAC design relies on conjugating a ligand for the target protein with an E3 ligase ligand. AptaGron transforms an aptamer with high affinity for the target protein into a degrader by hybridizing it with an N-degron-containing PNA through complementary sequences. Results and Discussion Design of Aptamer and N-Degron Ensemble (AptaGron) The N-degron pathway controls the lifetime of the protein, depending on the N-terminal amino acid. The N-terminal arginine is a type-1 N-degron that recognizes the UBR box domains of UBR1, UBR2, UBR4, and UBR5.15 The N-degron PNA was designed based on the previous arginine containing N-terminal peptide with additional linker RLAAAC and complementary sequence against the 5′-end linkers of the aptamer (Figure 2a). This peptide motif has been known as a UBR binder with a moderately good binding affinity (KD = 4.2) against the UBR box domain,16 making it the optimum choice. Because the ternary structure of UBR and arginine N-degron binding may require extra space for effective ubiquitination of target proteins, we designed four series of linkers diversifying the length (9 and 18 nt) and sequence compositions (AT-rich and GC-rich). In 2021, the crystal structure of the UBR1 initiation complex involved in N-degron-dependent ubiquitination was determined,17 revealing the spatial arrangement between the N-degron and ubiquitin (Figure S1). Specifically, the distance between the N-degron and ubiquitin-Lys48 was measured to be 51.5 Å (Å). We envisaged that an 18 nt linker (around 57.8 Å) would effectively span this distance. This length ensures adequate flexibility and spatial coverage to bridge the gap between the substrate and ubiquitin. Additionally, a shorter version of the linker was also tested to assess whether a more compact design could still maintain sufficient efficiency in proximity-induced ubiquitination. (a) Schematic of the AptaGron system. (b) Stability profile under phosphate-buffered saline (PBS) buffer (pH 7.4) for wtaptamertau, GC18ntaptamertau, and hybridized GC18ntAptaGrontau. (c) AptaGron stability under cell lysate exposures up to 48 h. (d) TPD of AptaGron against tau. Tau-GFP degradation was obtained using the ChemiDoc MP Imaging System (filterset: FITC channel). (e) Fluorescent live cell images of Tau-GFP expressed HEK293T cells upon administration of GC18ntAptaGrontau at serial incubation time. (f) Bar graph of quantitation values of fluorescent signals from the live cell images of Tau-GFP expressed HEK293T cell after treatment of GC18ntAptaGrontau. (g) Fluorescence signal from Tau-GFP showing degradations after administration of GC18ntAptaGrontau. (h) Quantitation of the fluorescence signal from PAGE images using ImageJ (*: p-value <0.05, **: p-value <0.005). We chose three proteins that are well characterized for disease associations, but currently, no selective small-molecule ligands are available for them. It should also be noted that the three target proteins have different subcellular localizations, with tau and elF4E localizing in the cytosol and nucleolin in the nucleus. The aptamers for the tau (wtaptamertau: 57 nt, KD = 19.5 ± 1.7 nM),18 nucleolin (wtaptamernuc: 26 nt),19 and elF4E (wtaptamerelF: 53 nt)20 were selected from the previously reported DNA aptamers (Table S1). All PNA and DNA aptamers were obtained from commercial vendors and used without additional processing. AptaGron Stability and Linker Optimization For the AptaGron system, we designed PNAs, consisting of N-degron and linker with (AXXAXXAXX)n motifs (n = 1 or 2). For hybridization, a complementary sequence linker was added to 5′-end of the tau aptamer. Initially, we examined the stability of AptaGron in PBS (pH 7.4) using wtaptamertau, an 18 nt GC-rich linker-conjugated aptamer (GC18ntaptamertau), and hybridized GC18ntAptaGrontau. None of the three forms showed significant degradation over a 48 h incubation period at 37 °C (Figure 2b). In HEK293T cell lysates, the AptaGron bands were detectable in the agarose gel with a half-life of 24 h (Figure 2c). Next, we investigated the degradation of tau by AptaGron, which contained different sequences and linker lengths. To monitor TPD, we used GFP-fusion tau (Tau-GFP), overexpressed in the HEK293T cells,21 and the degradation mediated by AptaGron was monitored using the GFP signal in fluorescence images from polyacrylamide gel electrophoresis (PAGE) (Figure 2d). A series of excess AptaGron (300 μM) was mixed with the cell lysate with or without Tau-GFP overexpression for 16 h, and the fluorescence gel pattern showed apparent downregulation of Tau-GFP. Notably, 18 nt long linkers were more effective than 9 nt long linkers, demonstrating that the special degree of freedom is a critical factor for the ubiquitination of target proteins through UBR proteins. This was supported by the case of AT9ntAptaGrontau, which showed negligible tau degradation, possibly due to its short length and lower affinity, owing to the fewer hydrogen bonding pairs. Another interesting observation was the shift of the fluorescent band to a higher molecular weight with multiple bands, indicating polyubiquitination resulting from the binding between Tau-GFP and AptaGron (Figure S2). To optimize incubation time and working concentrations, we screened 18 nt linker versions of the AptaGron across a range of 10–300 μM for 7.5 min to 16 h (Figure S3). The most effective conditions for degradation were observed using the GC-rich 18 nt linker version, GC18ntAptaGrontau. Tau degradation was observed at concentrations as low as 10 μM following 16 h incubation. Considering these experimental findings and the previously published UBR1 structure, which indicates that the distance between the N-degron and Lys48 residue of ubiquitin falls within the span of an 18 nt linker, we used an 18 nt linker system for the AptaGron platform and conducted the case studies to validate the robustness of this method for various target proteins. Degradation of Tau: A Case Study for a Neurodegenerative Disease Target Deposition of intraneuronal tau fibrils is a major hallmark of many neurodegenerative diseases including Alzheimer’s disease and frontotemporal dementia.22 As the accumulation of fibrillar tau induces neurotoxicity, therapeutics can be developed to eliminate tau fibrils. To date, a series of TPD studies have aimed to downregulate tau using a tau-binding peptide (YQQYQDATADEQG),23 or the 5H-pyrido [4,3-b]indole scaffold of PET tracer T807.24 Despite the promising results of these studies, the overall efficiency of TPD has been limited by the available ligands. Moreover, a significant breakthrough in the development of next-generation agents is unlikely unless entirely new ligands for the target proteins are identified. In this regard, the AptaGron platform offers unique advantages, as the adaptation of a new aptamer to the AptaGron is straightforward. Following the linker optimization, we assessed GC18ntAptaGrontau for tau degradation in the animal samples. It has been reported that TauP301L is one of the most prevalent pathogenic mutations associated with frontotemporal dementia, and our team previously developed a transgenic mice model expressing human TauP301L-BiFC.25 A brain proteome from a TauP301L-BiFC expressing mouse was prepared to verify whether our GC18ntAptaGrontau could effectively degrade tau from animal samples as a proof-of-concept study. Western blot analysis revealed significant degradation of both TauP301L mutant and mouse tau after 24 h at 37 °C (Figure S4). This indicates that the AptaGron can degrade both exogenous overexpressed and endogenous target proteins. Moreover, the successful degradation of the aggregation-inducing mutant form TauP301L implies its potential for therapeutic application. Next, we investigated the application of AptaGron in live cell conditions. Unlike in vitro assays, AptaGron cannot directly engage with tau in live cells because of the poor cell permeability of the negatively charged aptamer. After forming the AptaGron ensemble through the hybridization process, we used lipofectamine 3000 for cell delivery. To monitor tau expression, HEK293T cells expressing Tau-GFP were prepared, and live cell fluorescence images were obtained after transfection with GC18ntAptaGrontau. As shown in Figure 2e, the Tau-GFP signal gradually increased over time in control cells, indicating that more Tau-GFP proteins were expressed and formed fibrils. Cells treated with GC18ntAptaGrontau (10 μM) exhibited lower fluorescence intensity than the vehicle, demonstrating that AptaGron downregulated tau expression levels, and the effect was accumulated over the incubation time (Figure 2f). Both live cell fluorescence images and PAGE analysis consistently demonstrated tau degradation (Figure 2g,h). Although targeted downregulation exhibited a dose-dependent effect, the effective dose range of AptaGron was limited, owing to the cytotoxicity of the delivery agents. This limitation arises because a proportional amount of Lipofectamine is required for the effective delivery of AptaGron. However, our results showed that concentrations in the range of tens of micromolar ranges were effective in live cell experiments. Encouraged by this successful demonstration of tau degradation, we used the AptaGron system for other target proteins. Degradation of Nucleolin and elF4E: A Case Study for Targeted Cancer Therapy Cancer is the leading cause of death worldwide among individuals aged <70 years worldwide.26 Kinases have been extensively explored as drug targets owing to their crucial role as key mediators of cell growth signal transduction and the straightforward design of ligand derivatives based on the binding pocket structures.27 However, there is a significant demand for drug targets that operate through alternative mechanisms of actions. Nucleolin has emerged as a tumor-associated antigen that is up-regulated to a higher degree in tumors than in normal tissues.28 Additionally, eukaryotic initiation factor 4E (elF4E), which has a rate-limiting effect on net mRNA translation, is associated with metastasis.29 Degrader molecules that downregulate these proteins have great therapeutic significance; however, they are still in their early stages of development. For the previously reported aptamers (Table S1), we extended the 5′-end with 18 nt linkers. In the case of nucleolin, an AT-rich linker was chosen to minimize the complications of hybridization between the aptamer and PNA, because wtaptamernuc contains repeated GGT motifs. The AT18ntAptaGronnuc (10 μM) was prepared by thermally annealing the mixture of PNA and AT18ntaptamernuc. Then, the AptaGron complex was applied to the total cell lysate of MCF7 breast cancer cells, with serial incubation times to monitor its effect. The degradation pattern of nucleolin was monitored using Western blot analysis, and the band intensity was observed to decrease with the incubation time (Figure 3a). Cellular delivery was also tested using transfection agents, and the downregulated expression of nucleolin was detected after incubation for 12 h (Figure 3b,c) with a dose dependency (Figure S5a). As previously reported, the cellular localization of nucleolin is restricted to the nucleus. Our immunocytochemistry (ICC) images showed antibody staining signals mostly within the nucleus, and the intensity and number of green spots were reduced during incubation. (a) Western blot analysis of nucleolin degradation after administration of AT18ntAptaGronnuc (10 μM) for 40 h. (b) Immunocytochemistry images of MCF7 cells after treatment of AT18ntAptaGronnuc (10 μM). Blue color: nucleus was stained using Hoechst34580 (5 μM). Green color: nucleolin was imaged using nucleolin antibody (ab136649 from Abcam) and secondary antibody conjugated with Alexa488 (goat antimouse: A11001 from Invitrogen). (c) Bar graph of the fluorescence intensity profile obtained from immunocytochemistry. (d) Western blot analysis of tau, nucleolin, and eIF4E and with or without cotreatment of proteasome inhibitor, MG132. (e) Immunocytochemistry images of MCF7 cells after treatment of GC18ntAptaGroneIF4E (10 μM). Blue color: nucleus was stained using Hoechst34580 (5 μM). Red color: eIF4E was imaged using the eIF4E antibody (ab33768 from Abcam) and secondary antibody conjugated with Alexa594 (goat antirabbit: A110142 from Invitrogen). (f) Bar graph of fluorescence intensity from immunocytochemistry images. Next, we prepared AptaGron for eIF4E, consisting of 18 nt GC-rich linkers, using the same protocol. GC18ntAptaGroneIF4E (10 μM) showed a clear degradation of eIF4E in vitro (Figure S6). It should be noted that this degradation was inhibited by the addition of the proteasome inhibitor MG132. This suggested that the degradation of the target protein was a ubiquitin-proteolysis-dependent process (Figures 3d and S7). A similar downregulation pattern was also observed when GC18ntAptaGroneIF4E was delivered to live cells using a transfection agent, with effects apparent at 12 and 24 h (Figure 3e,f), along with dose-dependent responses (Figure S5b). Our ICC images showed the cellular localization of eIF4E in the cytosol, as previously reported, and degradation was consistently detected by using both Western blotting and ICC. Conclusions In summary, we designed an ensemble system consisting of an aptamer and an N-degron PNA for a simple and robust TPD strategy. Taking advantage of the complementary sequence hybridization, our AptaGron design can be applied to any target protein after SELEX-based aptamer discovery. We demonstrated the concept with three proteins as case studies. Serial optimization of the linker sequence and length showed that 18 nt of both AT and GC, depending on the aptamer, functioned effectively in our system. Most importantly, the degradation process is highly dependent on proteasomal activity. We confirmed the formation of the ternary complex “UBR1-AptaGron-target protein” through pull-down Western blot analysis, demonstrating that each target protein was recruited to UBR1 exclusively in the presence of AptaGron (Figure S8), and AptaGron does not alter the mRNA transcript level (Figure S9). Although there is a possibility of undesired off-target effects since the PNA linker sequence could interact with endogenus other genes (Table S2), it is not plausible to exchange the AptaGron to endogenous off-target, because the PNA:DNA hybridization binding in AptaGron has significantly high melting temperature, which makes it difficult to disrupt in physiological conditions (GC18ntTm = 96.3 °C and AT18ntTm = 54.6 °C).30 The overall TPD efficiency relies on the aptamer’s target binding affinity and optimized linkers to form the N-degron and UBR complex, followed by ubiquitination of the target protein. Therefore, further optimization of each component is required for enhanced therapeutic development. It is also worth noting that our AptaGron system demonstrates a particular advantage over traditional CRBN-ligand PROTACs in cells with low expression of the CRBN E3 ligase, such as prostate cancer-derived DU145 cells (Human Proteome Atlas: https://www.proteinatlas.org/ENSG00000113851-CRBN/cell+line). By employing the same sequence aptamer used for AptaGron, we designed a pomalidomide-conjugated aptamer through azide-DBCO coupling, and AptaGron exhibited distinct efficacy in CRBN low-expressing cells (Figures S10 and S11). Each platform offers distinct advantages for targeted protein degradation. When diseased cells highly express a specific E3 ligase with an available high-affinity ligand, PROTAC design enables selective degradation. In contrast, AptaGron provides a versatile alternative for cases where E3 ligase expression is low, suitable ligands are unavailable, or target protein ligands have yet to be identified. In this study, we focused on the proof-of-concept of the AptaGron system. However, there is still room for improvement in areas such as target selectivity, degradation efficiency, toxicity reduction, and cellular delivery enhancement. First, because the target selectivity and degradation efficiency are largely influenced by the aptamer, developing a high-affinity aptamer is crucial for optimizing the AptaGron system. Second, the N-degron warhead, the RLAA sequence we used, had a reasonable binding affinity (KD = 4.2 μM),16 emphasizing that a UBR binding motif with higher affinity could enhance overall ubiquitination efficiency. Third, one should carefully design the PNA linker to minimize the undesired off-target effect and examine the global selectivity profile. Finally, the current AptaGron design was not optimal for cellular uptake, despite the successful degradation of target proteins, including cytosolic target proteins in this study. The AptaGron system requires transfection agents for cellular uptake, thereby limiting its high-dose administration in cellular or animal models. To address these challenges, we are currently investigating other aptamers and developing effective cellular delivery methods. We foresee the potential of the AptaGron approach, particularly with the incorporation of cell-penetrating peptides, which could enable efficient delivery without transfection agents. This advancement could facilitate the development of the next version of AptaGron technology. Acknowledgments This study was supported by the Ministry of Science and ICT through the National Research Foundation of Korea (NRF-2020R1A2C2004422, RS-2023-00217996, RS-2023-00217701, RS-2024-00396224, and NRF-2022M3A9G8017710) and National Research Council of Science and Technology (NST) grant (CAP23011-400). Supporting Information Available The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acschembio.4c00536. Detailed information on the materials, experimental procedures, and characterization (PDF) Detailed information on the materials, experimental procedures, and characterization (PDF) The authors declare no competing financial interest. Supplementary Material cb4c00536_si_001.pdf References Associated Data Supplementary Materials cb4c00536_si_001.pdf