Selective degradation of the p53‐R175H oncogenic hotspot mutant by an RNA aptamer‐based PROTAC

To the Editor: TP53 encodes the tumour suppressor protein p53, a master regulator of genomic integrity and cell survival, and is the most frequently mutated gene.1 p53 mutant proteins are stabilised and can acquire dominant-negative or oncogenic gain-of-function activities, thereby promoting malignant transformation, metastasis and chemoresistance.2, 3 Proteolysis targeting chimeras (PROTACs) that hijack cellular ubiquitin-proteasome machinery for targeted protein degradation have shown considerable promise in targeting previously undruggable proteins.4, 5 However, PROTACs targeting p53 mutants have not yet been reported, probably due to difficulty in identifying a suitable binder for these mutants. Some recent studies have revealed that aptamers can be exploited as ligands in place of standard peptides or small molecules for PROTACs.6-8 Here, using an aptamer-based strategy, we developed the first selective hotspot p53 mutant PROTAC. p53-R175H is the most common p53 hotspot mutation.9 In this study, we used an RNA aptamer10 (hereafter named p53m-RA) that selectively targets p53-R175H as a binder for PROTAC development. First, we used streptavidin pulldown assays to confirm its binding specificity. Both p53m-RA and N3-p53m-RA competitively abolished p53-R175H pulldown from cell lysates with N3-p53m-RA-biotin. By contrast, there was almost no pulldown of wild-type p53 (p53-WT) (Figure 1A). Structural analysis revealed that p53-R175H displays a much narrower binding area for the DNA major groove than the wild-type p53 (Figure 1B), but the newly formed groove between L2 and L3 facilitates the binding of p53m-RA to p53-R175H (Figure 1C). Moreover, the 5′ end of p53m-RA was far away from the ligand binding sites (Figure 1D). Therefore, an alkynylated CRBN ligand (CRBNL), thalidomide-O-amido-propargyl (Supporting Information Schemes), was connected to the 5′ end of N3-p53m-RA via a click reaction to generate the p53-R175H degrader, dp53m-RA (Figure 1E). Native polyacrylamide gel electrophoresis (PAGE) demonstrated a slight difference in dp53m-RA migration relative to N3-p53m-RA, indicating successful conjugation (Figure 2A). Simulation analysis indicated that dp53m-RA should actively bind p53-R175H and CRBN simultaneously (Figure 1D). Indeed, dp53m-RA retained its ability to compete with biotin-p53m-RA for binding to p53-R175H, while CRBNL did not (Figure 2B), suggesting that the degrader could form a complex with CRBN and p53-R175H. We next investigated its effects on p53-R175H degradation. In p53-R175H- or p53-WT-expressing H1299 cells (p53-null), dp53m-RA treatment destructed p53-R175H, but spared p53-WT in a dose-dependent manner (Figure 2C). Time-course experiments revealed that p53-R175H was significantly degraded in 12 h (Figure 2D). To substantiate these findings, we treated SKBR3 (p53-R175H), A549 (p53-WT), and H1975 (p53-R273H) cells with dp53m-RA. As expected, dp53m-RA treatment selectively degraded p53-R175H, but not p53-WT or p53-R273H (Figure 2E). The half-maximal degradation concentrations (DC50) for H1299-p53-R175H and SKBR3 cells were 1.06 and 1.28 µM, respectively (Figure 2F). When the effects of dp53m-RA were examined on other hotspot mutations, only p53-R175H was significantly degraded (Figure 2G). The expression levels of p53 downstream effectors, including MDM2, BAX, CDKN1A and PUMA, were elevated by dp53m-RA treatment in H1299-p53-R175H and SKBR3 cells but not in H1299-p53-WT and A549 cells (Figure 2H). Furthermore, pretreatment of p53m-RA compromised dp53m-RA-induced p53-R175H degradation (Figure 2I). Moreover, the proteasome inhibitor MG132 completely blocked dp53m-RA-induced p53-R175H degradation (Figure 2J), and dp53m-RA treatment increased polyubiquitination of p53-R175H over p53-WT (Figure 2K). These findings indicate that dp53m-RA is a selective p53-R175H degrader. We next examined the biological consequences of dp53m-RA-mediated p53-R175H degradation in cancer cells. dp53m-RA treatment dramatically inhibited the proliferation of p53-R175H-expressing H1299 cells and SKBR3 cells. By contrast, dp53m-RA treatment did not affect the proliferation of p53-WT-expressing H1299 cells and A549 cells (Figure 3A,B). dp53m-RA treatment consistently suppressed colony formation of p53-R175H-expressing cells, but not p53-WT-expressing cells (Figure 3C,D). However, p53m-RA pretreatment did not diminish the anti-tumour activities of dp53m-RA (Figure S1A,B), probably due to the anti-tumour activities of p53m-RA per se at higher concentration. One advantage of dp53m-RA over p53m-RA is that it significantly lowers drug concentration because of its catalytic nature. Moreover, Transwell migration and wound-healing assays revealed that dp53m-RA significantly attenuated the migration of p53-R175H-expressing cells, but not p53-WT-expressing cells (Figure 3E–H). These results indicate that dp53m-RA may have therapeutic potential in patients harbouring p53-R175H mutant cancers. In summary, by harnessing an RNA aptamer specifically targeting p53-R175H, the most common TP53 mutation with dominant-negative and oncogenic gain-of-function activities, as a binder in PROTAC design, we report the first selective mutant p53 degrader, dp53m-RA. dp53m-RA degraded p53-R175H but not wild-type p53 or other p53 mutants in a ubiquitin-proteasome-dependent manner. Importantly, dp53m-RA inhibited the proliferation and migration of cancer cells specifically harbouring the p53-R175H mutation (Figure 3I), suggesting its therapeutic potential for precision medicine. This work was supported by grants from the National Natural Science Foundation of China (32070708, 32270892, 81702678). The authors declare they have no conflicts of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.