Introduction
Id (Inhibitor of differentiation) proteins (Id1, Id2, Id3 and Id4) are dominant negative regulators of basic helix loop helix transcription factors such as TCF3[
1,
2]. Apart from blocking the general bHLH-DNA (E-box response element) interactions, the Id1, 2 and 3 proteins also interact with several non-bHLH proteins such as CASK, ELK1, 3 and 4, GATA4, caveolin, CDK2, PAX2, 5 and 8, Rb and related pocket proteins and ADD1 ([
1,
2] and public databases). Currently, the non-bHLH interaction partners for Id4 are not known. Id proteins can thus control many cellular processes such as cell growth, differentiation, and apoptosis[
3], through specific bHLH and non-bHLH interactions.
Id proteins in general, promote proliferation and inhibit differentiation with few exceptions such as Id2 and Id4 that can also promote differentiation in some organ systems. Id4 promotes differentiation of osteoblasts[
4], adipocytes[
5], neurons[
6], but inhibits oligodendroglial differentiation[
7] by blocking the transcriptional activity of bHLH protein Olig1/2.
Majority of studies have demonstrated tumor suppressor activity of Id4 which is largely based on the evidence that it is epigenetically silencing in cancers such as leukemia[
8], breast[
9,
10], colorectal[
11] mouse and human CLL (chronic lymphocytic leukemia[
12]) and gastric cancer[
13]. High Id4 expression is observed in bladder[
14] and rat mammary gland carcinomas,[
15], whereas chromosomal translocation of Id4 (t(6;14)(p22;q32)) was found in B-cell acute lymphoblastic leukemia[
16] and B-cell precursor acute lymphoblastic leukemia (BCP-ALL)[
17], suggesting that it may also have tumor promoter activity.
Decreased Id4 expression with increasing grade of prostate cancer is also associated with Id4 promoter hyper-methylation[
18,
19]. The prostate cancer cell line DU145 also lacks Id4 expression due to promoter hypermethylation whereas LNCaP cells express Id4[
20]. Interestingly, DU145 cells also harbor mutant p53 with extended half-life, a property associated with mutated forms of p53[
21]. The p53 mutants (P223L and V274F) in DU145 cells are rare but located within the DNA binding domain (DBD 94-292) known to abrogate p53 activity[
22,
23]. The V274F mutation in DU145 cells is next to R273H/C/L/P, a DNA contact and one of the most highly mutated amino acid in p53[
23]. Both these amino acids (274°F and 273H) are within the conserved region of p53 beta strand S10 whereas 223 L lies in the outer loop[
24]. Studies have shown that some but not all p53 mutations maintain transactivation potential for some promoters (e.g. CDKN1a) but not others (e.g. BAX, PUMA and Pig3)[
25]. Likewise, the mutant p53 in DU145 also lacks the ability to trans-activate CDKN1A[
21]. We have shown that ectopic expression of Id4 in DU145 cells triggers apoptosis and CDKN1A dependent cell cycle arrest[
20]. CDKN1A being a prototype p53 transcriptional target prompted us to investigate whether Id4 promoted mutant p53 transcriptional activity in DU145 cells. The results presented in this study demonstrate that Id4 can promote the binding of mutant p53 to its response element on the p21 promoter and other p53 responsive apoptotic target genes such as BAX and PUMA. At the mechanistic level we demonstrate that Id4 recruits acetyl transferase CBP/p300 to promote acetylation of p53. Thus, mutant p53 in DU145 may retain conformational flexibility which upon post-translational modification could achieve wild type activity. Studies reported earlier have indeed shown that PCAF dependent acetylation can restore wild type activity of certain p53 mutants (G245A and R175H)[
26]. Since more than one third of prostate cancers harbor mutant p53[
27,
28] and majority of prostate cancers also lack Id4[
18,
19]; hence physiological mechanisms involved in the transition of mutant p53 to wild type activity are of clinical relevance.
Discussion
In this study we provide evidence that Id4 regulates p53 at two different levels: transcriptional regulation of wt-p53 in LNCaP cells and restoration of the biological activity of mutant p53 in DU145 cells. In this study, we focused on investigating the mechanism by which Id4 restores the biological activity of mutant p53, clearly an area of high interest given that mutant p53 is observed in one third of prostate cancer[
27,
28] and more than 50% of all cancers[
48]. The down-regulation of wt-p53 protein expression in the absence of Id4 in LNCaP cells (LNCaP-Id4) is a significant observation that was not addressed in this study. We speculate that Id4 could interact and modify the transcriptional regulators of p53 expression which remains to be investigated.
The core domain (aa 98-303) of p53 is inherently unstable. Point mutations in this domain promote instability and unfolding, leading to decreased or completely abrogated transcriptional activity[
49]. Both the alleles of p53 in DU145 cells (p223L and V274F) carry mutations within this core domain resulting in increased expression of mutant p53[
22] with predominantly denatured conformation. The attenuated transactivation potential of p53 P223L and V274F mutants is also observed when over-expressed in p53 null PC3 cells[
50]. Hence the mutants in DU145 cells are excellent models to understand the mechanisms involved in promoting its function in context of Id4 which is epigenetically silenced in DU145 cells.
In our studies we clearly show high mutant p53 expression in DU145 cells with attenuated transactivation potential and DNA binding activity as compared to LNCaP cells with wt-p53. Multiple lines of evidence support the gain of transactivation potential of mutant p53 in DU145 cell over-expressing Id4: First, mutant p53 in DU145 + Id4 cells promotes p53 dependent luciferase reporter activity, second, mutant p53 gains DNA binding activity as demonstrated by EMSA and direct DNA binding followed by detection and quantitation of binding with p53 specific antibody and thirdly, site specific binding to the respective p53 binding sites on BAX, PUMA, p21 and MDM2 P2 promoters. Studies have also shown that virtually all tumor derived p53 mutants are unable to activate BAX transcription but some retain the ability to activate p21 transcription[
25]. However, our results suggest the p53 mutations in DU145 are incapable of trans-activating not only p21 but BAX as well due to lack of promoter binding. The decrease in the expression of mutant p53 in DU145 + Id4 cells as compared to DU145 could also suggest that mutant p53 responds to the regulatory network required to maintain its normal physiological (compared to LNCaP cells) levels that needs to be investigated. The post-translation modifications within p53 (discussed below) can promote its function at multiple levels by attenuating its interaction with MDM2, recruitment to p53 responsive promoters and favoring nuclear retention as observed in DU145 + Id4 cells.
The discrepancy between p21 expression at the transcript and protein level was also observed in LNCaP-Id4 cells. The amount of p53 bound to the respective response element and RNA pol II, especially on the p21 promoter is not the sole determinant of transcriptional repression[
39] as seen in LNCaP-Id4 cells, in which p21 transcript abundance is not significantly different from LNCaP cells. A significant decrease in p21 protein expression in LNCaP-Id4 cells could be due to increased proteolysis. Increased MDM2 expression in LNCaP-Id4 could facilitate the binding of p21 with the proteosomal C8-subunit[
51] in a ubiquitin independent manner. Alternatively, loss of Id4 may promote proteolysis of p21 through ubiquitin dependent mechanisms involving e.g. Skp1/cullin/F-box (SCF) complexes that remain to be investigated (reviewed in[
52]).
Acetylation at lysine residues has emerged as a critical post-translational modification of p53 for its function
in vivo such as growth arrest, DNA binding, stability and co-activator recruitment ([
45,
46] and reviewed in[
53]). The global de-acetylation of p53 and specifically at K320 and K373 in LNCaP-Id4 cells provide strong evidence that acetylation is a major modification required to maintain wild type p53 activity. Our results on mutant p53 acetylation, global and K320/ 373 specific in DU145 + Id4 are particularly novel and provide direct evidence that mutant p53 activity can be restored by acetylation. The increased K320 acetylation of DU145 p53 mutants is most likely also mediated by PCAF but we did not directly investigate this mechanism. However, a significant observation made in this study was co-elution CBP/P300 with wt-(LNCaP) and mutant p53 (DU145 + Id4) and increased K373 acetylation in an Id4 dependent manner. Moreover, co-elution of Id4 as part of this complex with p53 antibody and co-elution of p53 with Id4 antibody suggest that Id4 can recruit CBP/P300 on wt-and mutant p53 to promote acetylation. Alternatively, CBP/p300 could recruit Id4 to promote large macromolecular assembly on p53 that could result in its acetylation and increased biological activity. Thus certain p53 mutations with some degree of conformational flexibility, upon co-factor recruitment such as Id4 and CBP/p300 could gain biological activity that is similar to wt-p53.
Acetylation at specific lysine residues can also promote specific p53 functional modifications: acetylation at K320 by PCAF results in increased cytoplasmic levels whereas CBP/P300 dependent acetylation of K370/372/373 leads to increased nuclear retention of p53[
46,
47]. In contrast, MDM2, a negative regulator of p53, actively suppresses p300/CBP-mediated p53 acetylation
in vivo and
in vitro[
54]. In this study we did not investigate the role of phosphorylation in regulating wt-or mut-p53 activity. K373 acetylation mimic p53Q373 undergoes hyper-phosphorylation and interacts more strongly with low affinity pro-apoptotic promoters such as BAX. In contrast, the p53Q320 interacts efficiently with the high-affinity p21 promoter[
46]. The ChIP data demonstrating high p53 binding on p21 promoter in DU145 + Id4 cells with increased p53 K320 acetylation may suggest increased phosphorylation that correlates well and further supports acetylation dependent increase in mutant p53 activity.
As such, low MDM2 levels observed in DU145 + Id4 cells as compared to DU145 could be one of the mechanism by which mutant p53 could gain its trans-activation potential together with increased acetylation. MDM2 binds to the N-terminal end of p53 to inhibit its trans-activation function partly by suppressing p300/CBP-mediated p53 acetylation[
54]. Acetylation also destabilizes p53-MDM2 interaction and enables p53 mediated response including recruitment to respective promoters and apoptosis[
38]. Studies in DU145 and LNCaP cells using nutlin, a disruptor of p53-MDM2 interaction, suggested that blocking MDM2 interaction or decreasing its cellular levels may be sufficient to promote wt-p53 activity (LNCaP cells) but is not sufficient for promoting mutant p53 transcriptional activity in DU145 cells[
21].
In a recent study[
55], Id4 expression was shown to be regulated by mutant p53 in an E2F1 dependent manner in breast cancer cell lines SKBR3 (p53 R175H) and MDA-MB-231 (p53 R280K). Both these cell lines were also shown to express Id4[
55]. Meta-analysis on clinical samples revealed that mutant p53 breast cancer tumors under-express Id4 suggesting an inverse correlation[
56] as seen in DU145 cells. Based on our results, we speculate that in the study by Fontemaggi et al.,[
55] Id4 could restore functional conformation of mut-p53, by acetylation in breast cancer cell lines leading to increased transcriptional activity. The mut-p53 in SKBR3 cells can be restored to functional conformation by Zinc[
57] further suggesting that mut-p53 retains the flexibility to undergo functional conformation to mimic wild type p53 activity.
Conclusions
We provide evidence that mutant p53 in DU145 cells retains the ability to undergo acetylation in the presence of Id4. Id4, a transcriptional regulator, may promote the p53 acetylation by recruiting CBP/p300 and/or PCAF, independent of p53 mutations. Acetylated p53 in turn acquires transcriptional activity through structural changes that could possibly involve destabilization of p53-MDM2 interaction, and subsequent recruitment to p53 responsive genes and promote apoptosis. The global acetylation promoted by Id4 suggests that additional lysines such as K120 and K164, known to increase binding to specific p53 responsive genes such as PUMA could also be involved, but remains to be investigated. Whether Id4 promotes the activity of p53 mutants found only in DU145 cells or it has the ability to promote transactivation potential of other well-known p53 hot-spot mutants is an obvious next step that needs to be investigated. Nevertheless, the acetylation mechanism is nearly universal in nature and suggests that Id4 could promote the biological activity of other mutants, however whether such mutants retains sufficient structural flexibility following acetylation remains to be determined. Our results also suggest that Id4 regulates the activity of wild type p53, a significant observation that requires further validation in other cell types.
Acknowledgements
The work was supported by NIH/NCI CA128914 (JC) and in part by NIH/NCRR/RCMI G12RR03062. The authors wish to thank Prof. Deborah Core (PhD), Dept. of English, Eastern Kentucky University, Richmond, KY for critical review of the manuscript.
Support
The work was supported by NIH/NCI CA128914 (JC) and in part by NIH/NCRR/RCMI G12RR03062.
Competing interests
The authors declare that there is no competing financial interest in relation to the work described.
Financial competing interests: Authors declare no financial competing interests.
Authors’ contributions
AEK: developed the L-Id4 cell lines, ChIP experiments, Immuno-blot studies, qRT-PCR, First draft of the manuscript. DP: Apoptosis assays, Mitochondrial membrane potential, ChIP experiments, immuno-blots, luciferase assays. DJM: Immuno-blots, p53 DNA binding ELISA. PS: Immuno-precipitations and Gel shift assays. SG: Immuno-cytochemistry. JC: Conceived the study and final draft of the manuscript. All authors read and approved the final manuscript.