( H) ChIP analysis of p23 binding to the COX-2 promoter region by PCR. ( G) Nuclear p23 was analyzed by Western blotting. ( F) The cellular localization of p23 was assessed by confocal microscopy, and the relative fluorescence units were also calculated. ( D and E) Analysis of p23 succinylation in lung cancer specimens and A549-derived (D) or H322-derived (E) xenograft tumor tissues. ( A to C) p23 succinylation in 293T cells transiently expressing Flag-p23 (A) in A549 and H322/OEp23 cells treated with or without succinate (B) and (C) was determined by p23 IP and Western blotting using an antibody against succinylated lysine. P values were calculated using two-tailed unpaired Student’s t test, and differences with P values <0.05 were statistically significant (** P < 0.01 and *** P < 0.001). Dual-luciferase reporter assays were performed.
( M) 293T cells were cotransfected COX-2-reporter, pRL-TK, and p23 or each of its mutants. The streptavidin-biotin pulldown was analyzed for p23. ( L) The nuclear extract from 293T cells transiently transfected with p23 or its mutants as indicated were incubated with biotin-conjugated CPDF. The hydrogen bonds are depicted as red dotted lines. The nucleotides and the key residues of p23 responsible for its binding are shown as stick models. ( J) Dynamic HDX analysis for p23 binding to COX-2 promoter DNA. ( I) The binding affinity of p23 to the COX-2 promoter or its mutants was determined by EMSA. Dual-luciferase reporter analysis was performed. ( H) 293T cells were cotransfected with p23 and pRL-TK, and further with COX-2 reporter or each of its mutants. ( G) DNA binding motif analysis for p23 using the DREME Network. ( F) Anti-p23 antibody was used for supershift analysis in P1, P2, and P6. The COX-2 sequence fragments bound to p23 are in red. P1 to P7: the seven segments of the COX-2 promoter. ( E) EMSA was performed to detect the binding affinity of p23 to the COX-2 promoter. The ChIP signal was normalized and quantified. ( D) ChIP analysis with anti-p23 antibody for its binding to the COX-2 promoter in A549 cells. ( C) The biotin-labeled COX-2 promoter DNA was incubated with nuclear extract to analyze p23 binding to the COX-2 promoter by Western blotting. ( A and B) Analysis of p23 functions in activating COX-2 transcription with dual-luciferase assay using the optimal COX-2 promoter region (A) or its truncate mutants (B). P values were calculated using two-tailed unpaired Student’s t test, and differences with P < 0.05 were statistically significant (* P < 0.05, ** P < 0.01, and *** P < 0.001). ( N and O) The expression of COX-2, p23, and PCNA in H322-derived tumor tissues from (M) was analyzed by Western blotting. The H322 colony formation (J) and tumor growth in nude mice were assessed: tumor appearance (K), volumes (L), and weights (M). P23 and COX-2 expressions were confirmed by Western blotting (I). ( I to M) Gain of p23 functions promotes H322 colony formation and tumor growth through COX-2. ( G and H) IHC and Western blotting analysis of p23 and COX-2 expression in lung cancer specimens (represented by 2) and corresponding adjacent normal tissues (represented by 1). ( F) The RT-qPCR analysis of selected DEGs and correlation with RNA-seq data. ( E) The KEGG pathways of common DEGs shared by p23-KD groups. ( D) Kaplan-Meier analysis of the overall 5-year survival rates of patients with low, intermediate, and high nucleus p23 expression ( P = 0.015). ( C) Nuclear p23 expression in lung cancer specimens by subcellular fractionation. ( B) P23 levels in lysates from lung cancer specimens (represented by 2) and corresponding adjacent normal tissues (represented by 1) were analyzed by Western blotting. ( A) IHC staining for p23 expression in lung cancer tissue sections and representative images magnified with a ×4 or ×40 objective is shown. Therefore, our study defines p23 as a succinate-activated transcription factor in tumor progression and provides a rationale for inhibiting p23 succinylation as an anticancer chemotherapy. M16 inhibited p23 succinylation and nuclear translocation, attenuated COX-2 transcription in a p23-dependent manner, and markedly suppressed tumor growth. We then identified M16 as a potent p23 succinylation inhibitor from 1.6 million compounds through a combined virtual and biological screening.
Intratumor succinate promotes p23 succinylation at K7, K33, and K79, which drives its nuclear translocation for COX-2 transcription and consequently fascinates tumor growth. Here, we found that p23 is a previously unidentified transcription factor of COX-2, and its nuclear localization predicts the poor clinical outcomes. The molecular nature underlying how this HSP90-independent p23 function is achieved remains as a biological mystery. P23, historically known as a heat shock protein 90 (HSP90) co-chaperone, exerts some of its critical functions in an HSP90-independent manner, particularly when it translocates into the nucleus.