Transactivation domain

DNA-binding domain

Telramerization DMA-binding domain Regulatory domain

Fig.15.1 A diagram of the structure of p53. Evolutionarily conserved regions are indicated as I-V. The functional domains are listed as: transactivation domain, DNA-binding domain, tetramerization domain, and DNA-binding regulatory domain. Phosphorylation, Acetylation, and Neddylation residues are indicated.

Ubiquitin ligases are divided into two classes depending on whether they contain a HECT domain or a RING domain (Pickart, 2004). Mdm2 falls within the family of RING E3 ligases possessing the characteristic CX2CX(9-39)CX(1-3)HX(2-3)C/HX2CX(4-48)CX2C motif at its extreme C-terminus (Joazeiro and Weissman, 2000). Its predominant substrate remains p53, though Mdm2 has been shown to be capable of ubiquitinating substrates such as (3-arrestin, PCAF, and the insulin-like growth factor 1 receptor (Girnita et al., 2003; Jin et al., 2004; Shenoy et al., 2001). Mdm2 can also ubiquitinate itself, and while this may be the predominant mechanism for maintaining its short half-life, it is clearly not the only answer. Mdm2 constructs expressed in cells that are either missing their RING domain (Mdm2 1-440) or contain a point mutation in the active site (Mdm2 C464A) are still capable of being degraded, suggesting other mechanisms for Mdm2 regulation are at play (Brooks and Gu, unpublished data).

In addition to Mdm2, other E3 ligases have been shown to impart specificity toward p53 and promote its proteasome-mediated degradation (Fig. 15.2). Pirh2, a RING-H2 domain containing protein, interacts with p53 and promotes Mdm2-independent p53 ubiquitination and degradation (Leng et al, 2003). Reminiscent of Mdm2, Pirh2 is a p53 transcriptional target gene and participates in a similar autoregulatory negative feedback loop. Another E3 ligase, COP1, has also been described recently as a direct ubiquitin ligase for p53. Using an epitope-tagging strategy, Dornan et al. purified a mammalian COP1 protein complex from the U20S cell line and found p53 as a strong interacting cellular factor (Dornan et al., 2004). COP1 is also a p53-inducible gene, and can ubiquitinate and degrade p53. Further, COP1 depletion by siRNA enhances p53-mediated G1 arrest and can sensitize cells to ionizing radiation. Together, Mdm2, COP1, and Pirh2 represent an army of E3 ligases the cell can call upon to regulate and maintain p53 levels (Fig. 15.1). They suggest that both Mdm2-dependent and independent mechanisms are used cooperatively by the cell for tight p53 regulation. It is yet uncertain exactly how these proteins are specifically regulated and under what situations they may be differentially activated. However, the redundancy of three ubiquitin ligases for p53 emphasizes the importance of keeping this tumor suppressor under tight lock and key.

Fig.15.2 A model for p53 regulation with both Mdm2-dependent and Mdm2-independent mechanisms. Mdm2, Pirh2, and COP1 all function as ubiquitin E3 ligases for p53 leading to its 26S proteasome dependent regulation. The protein level of Mdm2 is important for p53 mono- and polyubiquitination. When the level of Mdm2 is high it induces polyubiquitination of p53 and causes its efficient degradation. When the level is low, p53 is monoubiquitinated and moves into the cytoplasm where it can potentially be modified further.

Fig.15.2 A model for p53 regulation with both Mdm2-dependent and Mdm2-independent mechanisms. Mdm2, Pirh2, and COP1 all function as ubiquitin E3 ligases for p53 leading to its 26S proteasome dependent regulation. The protein level of Mdm2 is important for p53 mono- and polyubiquitination. When the level of Mdm2 is high it induces polyubiquitination of p53 and causes its efficient degradation. When the level is low, p53 is monoubiquitinated and moves into the cytoplasm where it can potentially be modified further.

B: Deubiquitination

The strong effect that ubiquitination has on p53 regulation suggests that its reversal may have an equal and opposite effect on protein stability. The first deubiquitinase specific for p53, HAUSP (Herpesvirus-associated ubiquitin-specific protease), was recently shown to have a profound effect on p53 stability (Li et al., 2002). HAUSP is capable of removing ubiquitin moieties for p53 both in vitro and in vivo and can stabilize p53 when overexpressesd. HAUSP was originally identified as a cellular protein that associated with the Herpesvirus protein ICPO and also been shown to associate with the Ebstein-Barr virus protein EBNA1 (Everett et al., 1999; Holowaty et al., 2003). The intimate relationship between HAUSP and p53 may be an attractive target for both viruses and oncogenes to exploit as a way of inhibiting p53's role in cell cycle arrest.

The role of HAUSP in p53 regulation became more obscure with the discovery that removal of the gene in somatic HCT116 cells caused profound p53 stabilization. This result was counterintuitive to what was expected, and a detailed biochemical analsysis of this pathway revealed a provocative model. Two independent systems, namely the HeLa cell line and transient partial reduction of HAUSP gene expression by siRNA, both show p53 destabilization, indicating that HAUSP does in part function as a stabilizing effector of p53. However, full reduction or removal of HAUSP expression leads to destabilized Mdm2 and the subsequent stabilization of p53. HAUSP was also shown to act as a deubiquitinase for Mdm2, and this intricate interplay between p53, Mdm2, and HAUSP yields a cellular system that is delicately responsive to changes in HAUSP level (Brooks and Gu, 2004). The physiology underlying this pathway remains to be elucidated. One possibility is that HAUSP is differentially regulated depending on the cellular signaling involved, so that at times it preferentially stabilizes p53 (i.e. DNA damage) and at others it stabilizes Mdm2 (i.e. normal cellular conditions). It is also quite likely that there is a substrate balance in this pathway and that HAUSP does not exhibit an all-or-nothing response. Nevertheless, future studies into HAUSP regulation may yield more insight into the p53-Mdm2 pathway as well as some viral commandeering strategies.

C: Neddylation

The Nedd8 protein is in the ubiquitin-like family of proteins and uses El activating and E2 transfer enzymes in a similar fashion as ubiquitin. The Nedd8 modifying pathway is important for yeast growth and viability

(Lammer et al., 1998; Osaka et al., 1998). In mammalian systems, it has been shown to be an important modification for the cullin family of proteins. In the case of the SCF complexes, neddylation of cullins seems to regulate their activity. Neddylation can promote the recruitment of an E2 ubiquitin-conjugating enzyme to the SCF complex (Kawakami et al., 2001; Wu et al., 2002). It has also been shown to promote the disassociation of the SCF ligase and cullin inhibitor pl20 Cand'(Liu et al., 2002). In this context, a component of the SCF complex, Rocl/Rbxl/Hrtl, has been proposed to possess E3 Nedd8 ligase activity toward the complex (Gray et al., 2002; Kamura et al., 1999; Morimoto et al., 2003).

Neddylation was recently shown to occur on p53 as well, and this modification inhibited its transactivation activity (Xirodimas et al., 2004). Mdm2 acts as a specific E3 Nedd8 ligase for p53 and can undergo self-neddylation as well. Lysines 370, 372, and 373 seem to be important for the Nedd8 conjugation pathway as p53 KR mutants of this region are no longer inhibited in their transcriptional activity. Together, the data suggests yet another layer of regulation of p53 as a transcription factor and an increase in complexity of Mdm2 function. Further work on this newly evolving aspect of p53 regulation will certainly yield more answers in the future.

D: Phosphorylation

Several residues at both the amino terminus and the carboxyl terminus of p53 are phosphorylated or dephosphorylated in response to genotoxic stress (Appella and Anderson, 2001). Several of these phosphorylation events occur within the N-terminal Mdm2 binding domain of p53 and abrogate the p53-Mdm2 interaction. The DNA-damage induced kinases Chkl and Chk2 both phosphorylate Ser20 after DNA damage while ATM and ATR both phosphorylate Serl5 and Ser37 (Bode and Dong, 2004). Other kinases including DNA-PK, CK1, CAK, and p38 also phosphorylate p53 at specific residues within this region with unclear roles in the DNA damage response. Despite these findings, it has been difficult to show strong phenotypic changes when these sites are mutated, suggesting a more complex and yet undetermined physiologic role for these events. It is also possible that specific combinations of phosphorylation events are differentially mediating specific p53 responses or regulating effector interactions.

The role of phosphorylation sites outside of the Mdm2 binding domain is less clear. A protein complex consisting of CK2, hSptl6, and SSRP1 specifically phosphorylates Ser392 upon UV irradiation and enhances p53 transactivation activity on the p21 promoter (Keller et al., 2001). JNK also phosphorylates p53 at residue Thr81 and stabilizes the protein (Buschmann et al., 2001). However, when not active, JNK seems to promote p53 degradation by the 26S proteasome through an ill-defined mechanism. In addition, CKII phosphorylates Ser392 and PK-C phosphorylates Ser371, 376, and 378 (Pluquet and Hainaut, 2001). These modifications have been proposed to enhance DNA binding as they occur within the often alluded to C-terminal regulatory domain (see acetylation below).

The missing mechanistic link between phosphorylation status and p53 transactivation was characterized with the finding of the peptidyl-prolyl isomerase Pinl (Zacchi et al., 2002; Zheng et al., 2002). This protein specifically binds to phosphorylated Ser/ Thr-Pro motifs on p53 and induces a conformational change in the protein thereby enhancing its transactivation activity. Importantly, cells deficient in Pinl show a marked decrease in p53 transactivation of some pro-apoptotic genes, indicating a role for Pinl in linking p53 phosphorylation with transactivation. Pinl also interacts with the family member p73 and induces a similar conformational change as with p53 (Mantovani et al., 2004). Here, c-Abl-dependent phosphorylation of p73 enhances the Pinl-p73 interaction and this in turn enhances p300-mediated acetylation of p73. Acetylated p73 is then able to bind to and transactivate pro-apoptotic genes such as bax, p53AIP, and pig3. Pinl therefore may represent a common translator of phosphorylation status for the p53 family of transcription factors. It also might prove to be a more general mechanism for linking phosphorylation with transactivation function particularly in the case of p53.

E: Acetylation

The covalent linkage of an acetyl group to lysine residues located on histone tails, the enzymatic process of acetylation (Jenuwein and Allis, 2001), is soundly believed to be involved with transcriptional regulation as it was first shown decades ago to correlate with an increase in transcriptional activity (Kouzarides, 2000). The significance of histone acetylation in transcriptional regulation seems indisputable; however, the precise role of this event is still not completely understood. Histones are not the only proteins that can be acetylated however, and the discovery that acetylation of a transcription factor could increase its transactivation activity has opened up an entirely new field of research.


CBP/p300, a protein possessing histone acetyl-transferase (HAT) activity, acts as a coactivator of p53 and augments its transcriptional activity as well as its biological function in vivo (Avantaggiati et al., 1997; Gu et al., 1997; Lill et al., 1997). The observation that p53 could have functional synergism with CBP/p300, together with the intrinsic HAT activity of CBP/p300, led to the discovery of a novel transcriptional factor acetyl-transferase (FAT) function of CBP/p300 on p53 (Gu and Roeder, 1997). The significance of CBP/p300, and arguably its interaction with p53, is emphasized by the presence of p300 mutations in several types of tumors (Goodman and Smolik, 2000). Additionally, mutations of CBP in human Rubeinstein-Taybi syndrome as well as CBP knockout mice lead to a higher risk of tumorigenesis.

Many transcription factors have been demonstrated as bona fide substrates for acetyl-transferases such as GATA-1, MyoD, HMG-1, E2F-1, ACTR, EKLF and Smad7 (Sterner and Berger, 2000). The functional consequences of acetylation are diverse and include increased DNA binding, enhancement of stability, and changes in protein-protein interactions. p53 is specifically acetylated at multiple lysine residues (Lys370, 371, 372, 381, 382) of the C-terminal regulatory domain by CBP/p300, and to a lesser extent Lys320 by PCAF (Appella and Anderson, 2001). The acetylation levels of p53 are significantly enhanced in vivo in response to almost every type of stress, well correlated with its activation and stabilization induced by stress (Ito et al., 2001). These acetylation sites of p53 are essential for its ubiquitination and subsequent degradation by Mdm2. Acetylation may even have a more direct role in p53 stabilization, as it was shown that C-terminal acetylation of p53 can inhibit Mdm2-mediated ubiquitination and prolong the half-life of p53. They may also have a significant impact on proteinprotein interactions between p53 and transcriptional co-activators such as CBP/p300 and PCAF. Indeed, p53 acetylation has been shown to be critically important for the efficient recruitment of these complexes to promoter regions and the activation of p53 target genes in vivo (Barlev et al., 2001).

Numerous studies indicate that the C-terminus of p53 acts as a critical regulator of p53 and negatively modulates its transcriptional activation. Deletion of the C-terminus, injection of antibodies specific for the C-terminus (PAb421), single-strand DNA, protein-protein interactions such as HMG-1, and post-translational modifications at this region all induce profound p53 transactivation abilities (Hupp et al., 1992; Jayaraman and Prives, 1995; Jayaraman et al., 1998). Consistent with this model, the acetylation of p53 can dramatically stimulate its sequence-specific DNA binding activity in vitro, possibly as a result of an acetylation-induced conformational change (Espinosa and Emerson, 2001; Gu and Roeder, 1997; Liu et al., 1999; Sakaguchi et al., 1998). The model has also been confirmed in vitro and in vivo using purified acetylated p53, which augments its site-specific DNA binding to both short and long DNA fragments.

An opposing model suggests that the C-terminus of p53 has a regulatory effect on short pieces of DNA but has no effect on longer DNA templates (Espinosa and Emerson, 2001). This is based on the observation that unpurified acetylated p53 significantly inhibits its activity in an in vitro chromatin assay (Espinosa and Emerson, 2001). However, this notion is flawed when considering the data set forth. First, Espinosa and Emerson show that acetylation on the C-terminus of p53 has no effect on DNA binding when compared to unacetylated p53 in an in vitro DNase footprinting assay. However, the enzyme used in the assay was subsequently supershifted by an antibody that specifically recognizes the C-terminus of p53, pAb421. If the enzyme used in this assay was indeed completely acetylated, then pAb421 would not recognize p53 and the protein would not be supershifted. Indeed, when purified acetylated p53 is used in a similar assay, there is a profound increase in the DNA binding ability of p53 (Luo et al, 2004). This holds true for the in vivo setting as well. The concept of the C-terminus of p53 acting as a positive regulator of transcriptional activation is also unlikely fit for the endogenous p21 promoter, since the p53 C-terminal mutant exhibits very strong cell growth repression and transactivation of this promoter in vivo (Crook et al., 1994; Hupp et al., 1992; Jayaraman et al., 1998; Marston et al., 1994). Taken together, a substantial amount of data obtained by several researchers still indicates that acetylation of the C-terminus of p53 is involved in transcriptional regulation, protein stability, and protein-protein interactions.

G: Deacetylases

Histone deacetylase complexes (HDACs) have emerged as notable components in regulating transcriptional activation as well. HDACs are often associated with corepressor complexes and can exert their repressive effects on both histone and non-histone proteins by removing acetyl groups (Smith, 2002). In contrast, much less is known about HDAC activity on p53 function and the general role of p53 deacetylation. Normal resting cells have a very low level of acetylated p53. Treatment of cells with the HDAC inhibitor trichostatin A (TSA) increases levels of acetylated p53 and led to the identification of the adaptor protein PID/MTA2, a component of the HDAC1 complex that can enhance HDAC 1-mediated deacetylation of p53 (Juan et al., 2000; Luo et al., 2000). Subsequent work has identified Sir2a (SIRT1), a TSA-resistant, NAD-dependent histone deacetylase that can both deacetylate p53 and attenuate its transcriptional activity (Luo et al., 2001; Vaziri et al., 2001). Sir2a co-localizes in PML nuclear bodies with p53 (Langley et al., 2002) and was structurally shown to undergo a conformational change when bound to acetylated p53 (Avalos et al., 2002). Further, PML and oncogenic Ras can upregulate acetylated p53 levels in primary fibroblasts (Pearson et al., 2000). The novelty of the Sir2a family of HDACs suggests an interesting link between nicotinamide (vitamin B3), cellular metabolism, and p53-mediated cellular responses to genotoxic stress. Transgenic mice harboring an N-terminus p53 deletion mutant exhibit an early-ageing phenotype (Tyner et al., 2002), and Sir2a is involved in gene silencing and extension of life span in yeast and C. elegans (Guarente, 2000). Taken together, Sir2a may provide a possible link between p53 and mammalian longevity.

The prominence of deacetylase activity on p53 certainly raises the defining question of its physiological purpose. One possibility is that deacetylation provides a quick acting mechanism to stop p53 function once transcriptional activation of target genes is no longer needed (Fig. 15.3). Targeted deacetylation has been shown to occur very quickly amidst a global equilibrium of genomic acetylation and deacetylation (Katan-Khaykovich and Struhl, 2002). Restoration of this steady-state level at p53 target genes is crucial for cellular homeostasis once DNA repair is complete. Deacetylation could also serve as an important step in MDM2-mediated p53 degradation.

Protein-protein Interactions p53 is highly dependent on its interaction with specific cellular proteins for rapid stabilization during times of cellular stress. These proteins tend to function by binding to either p53 (e.g. INGlb) or binding to Mdm2 (e.g. pRb, pl9ARF, and MdmX) (Hsieh et al., 1999; Kamijo et al., 1998; Leung et al., 2002; Sharp et al., 1999). Presumably, interactions in this regard block Mdm2-mediated ubiquitination of p53 and thereby stabilize the protein.

pl4ARF was first discovered as an alternative reading frame gene product from the lNK4a/ARF gene

Delicious Diabetic Recipes

Delicious Diabetic Recipes

This brilliant guide will teach you how to cook all those delicious recipes for people who have diabetes.

Get My Free Ebook

Post a comment