The p53 circuit board

Biochimica et Biophysica Acta 1825 (2012) 229ndash244

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage wwwe lsev ie r com locate bbacan

Review

The p53 circuit board

Kelly D Sullivan Corrie L Gallant-Behm Ryan E Henry Jean-Luc Fraikin Joaquiacuten M Espinosa Howard Hughes Medical Institute amp Department of Molecular Cellular and Developmental Biology The University of Colorado at Boulder Boulder CO 80309-0347 USA

Corresponding author Tel +1 303 492 2857E-mail address joaquinespinosacoloradoedu (JM

0304-419X$ ndash see front matter copy 2012 Elsevier BV Aldoi101016jbbcan201201004

a b s t r a c t

a r t i c l e i n f o

Article historyReceived 6 December 2011Received in revised form 27 January 2012Accepted 28 January 2012Available online 7 February 2012

KeywordsGene networkPUMAp21ApoptosisCell cycle arrestPersonalized medicine

The p53 tumor suppressor is embedded in a large gene network controlling diverse cellular and organismalphenotypes Multiple signaling pathways converge onto p53 activation mostly by relieving the inhibitory ef-fects of its repressors MDM2 and MDM4 In turn signals originating from increased p53 activity diverge intodistinct effector pathways to deliver a specific cellular response to the activating stimuli Much attention hasbeen devoted to dissecting how the various input pathways trigger p53 activation and how the activity of thep53 protein itself can be modulated by a plethora of co-factors and post-translational modifications In thisreview we will focus instead on the multiple configurations of the effector pathways We will discuss howp53-generated signals are transmitted amplified resisted and eventually integrated by downstream genecircuits operating at the transcriptional post-transcriptional and post-translational levels We will also dis-cuss how context-dependent variations in these gene circuits define the cellular response to p53 activationand how they may impact the clinical efficacy of p53-based targeted therapies

copy 2012 Elsevier BV All rights reserved

Contents

1 Introduction 2292 Layered regulation of p53-derived signals the p21 circuit 232

21 Transcriptional co-regulation of p21 23222 Post-transcriptional co-regulation of p21 23223 Post-translational co-regulation of p21 233

3 Assembling circuits into integrated circuits and circuit boards the minimal case of p21 and 14-3-3σ versus PUMA and NOXA 23331 Gene-specific transcriptional regulation of p21 14-3-3σ PUMA and NOXA 23332 Gene-specific regulation of p21 14-3-3σ PUMA and NOXA at the post-transcriptional and post-translational levels 234

4 Context-dependent configurations of the p53 circuit board 23441 Cell type-specific configurations of the p53 circuit board 23442 Stimulus-specific configurations of the p53 circuit board 23543 Cell type- and stimulus-specific action of the p53 family members p63 and p73 235

431 Cousins in different places 235432 Cousins with different lifestyles 236

5 Clinical relevance of understanding p53 as a circuit board 23751 p53-based targeted therapies promise and obstacles 23752 The impact of cell type- and stimulus-specific assemblies of the p53 circuit board on the efficacy of p53-based therapies 23853 Personalized p53 medicines the urgent need for combinatorial therapies functional genomics and molecular diagnostics 238

6 Final remarks 240References 240

Espinosa)

l rights reserved

1 Introduction

As molecular biology evolved from the study of single moleculesto the study of large biological systems the meaning of the wordp53 displayed a parallel increase in sophistication When discoveredmore than 30 years ago p53 was first a protein in a gel [1ndash3] Early

Fig 1 The p53 circuit board Activation of p53 by diverse stimuli such as oncogene hyperactivation DNA damage and nutrient deprivation results in increased expression of nu-merous genes controlling different cellular outcomes such as cell cycle arrest senescence autophagy and apoptosis

230 KD Sullivan et al Biochimica et Biophysica Acta 1825 (2012) 229ndash244

on p53 was characterized as an oncogene a confounding issue de-rived from the study of the mutant p53 variants expressed in tumors[4ndash9] Years later the wild type version of the p53 gene was unequiv-ocally identified as a tumor suppressor [10ndash12] Eventually the p53protein was characterized as a DNA-binding transcription factor[13ndash17] With the discovery of MDM2 a repressor of p53 [18ndash20]and p21 (CDKN1A) a p53 target gene and effector [2122] a p53pathway was born With the identification of numerous upstreamregulators and hundreds of target genes with different functions[23ndash26] one pathway became many (Fig 1) Today the notion of ap53 pathway is obsolete p53 is a gene network An enormousamount of knowledge has been generated while studying the p53gene the p53 protein and the various p53 pathways What can welearn from studying p53 as a gene network

In this review we will abandon all p53-centric views and considerp53 as mere ldquosignal generatorrdquo within a vast gene network We willdeliberately neglect all information about the multiple events that af-fect the p53 molecule itself such as dozens of post-translational mod-ifications and hundreds of binding partners that regulate its activityWe refer those readers interested in the p53 molecule to many excel-lent p53-centered reviews [27ndash31] Instead in the next pages we willassume for the sake of argument that the p53 protein is one and thesame in every context and that its activity as a DNA binding proteinand transcriptional regulator is invariant across scenarios Havingblinded ourselves to the outstanding complexity of the molecularevents surrounding p53 activation we will focus our attention on

Fig 2 The p21 circuit The p53 target gene p21 is co-regulated by various factors acting atpotentially affect p53p21-dependent cell cycle arrest In both A and B from top to bottom trtranscriptional regulators into a single output functional mRNA In this context p53 is justgrator at this step which synthesizes p21 mRNA the input for the next level of regulation Frtranslation are combined via the ribosome signal integrator to achieve protein synthesis Finbilizing signals resolved by the proteasome In A the regulatory layers are displayed as carelectrical circuit comprising three signal integrators At the transcriptional level activatingintegrator (+) and are summed internally by an adding network (dashed blue box) inhibittivating and total inhibitory signals determines output amplitude in this case p21 mRNA levery in this example p21 mRNA levels become positive inputs for p21 protein production i

the downstream networks of genes that modulate p53-derived sig-nals Throughout this reviewwe will employ an electrical engineeringmetaphor and it is important for us to define at this point howwe willuse the terms circuit integrated circuit and circuit board

In our metaphor p53 is a switch turned to the lsquoONrsquo position by p53activating stimuli thus transmitting signals to downstream circuitryWe define a circuit as the collection of factors co-regulating a givenp53 target gene at the transcriptional post-transcriptional and post-translational levels In Section 2 we illustrate this for p21 a directtranscriptional target of p53 that mediates cell cycle arrest (Fig 2)[212232] An integrated circuit is defined as a group of p53 targetgenes that work together to drive a given cellular response For exam-ple the p21 and 14-3-3σ (SFN) circuits are part of an integrated cir-cuit that mediates p53-dependent cell cycle arrest whereas PUMA(p53-upregulated modulator of apoptosis BBC3) and NOXA (PMA-induced protein 1 PMAIP1) are two among many circuits mediatingp53-dependent apoptosis Finally a circuit board represents the as-sembly of several diverse integrated circuits which ultimately coa-lesces all of the various signals into a single cellular outcome(Fig 3) In Section 3 we assemble a minimal circuit board composedof p21 14-3-3σ PUMA and NOXA which includes their variousgene-specific co-regulators acting at the transcriptional post-transcriptional and post-translational levels In Section 4 we de-scribe how p53 circuit boards can be assembled into cell type- andstimulus-specific configurations that define context-dependent p53responses We discuss at this point the impact of the p53 family

the transcriptional post-transcriptional and post-translational levels all of which cananscriptional control of p21 gene expression merges the action of positive and negativeone among many regulators feeding into RNA polymerase II (RNAPII) the signal inte-om there signals from factors that positively and negatively influence RNA stability andally the levels of active p21 protein in the cell are fine-tuned by stabilizing and desta-toons of DNA RNA and protein molecules In B the same processes are depicted as ansignals including p53 (green) are applied to the positive terminals of the RNAPII signalory signals (red negative terminals) are combined similarly The difference in total ac-el Integrator output can subsequently serve as input to other signal processing machin-n the ribosome

231KD Sullivan et al Biochimica et Biophysica Acta 1825 (2012) 229ndash244

Fig 3 Integrated circuits of cell cycle arrest and apoptosis comprise a minimal p53 circuit board Following a similar schema to Fig 2 co-regulators of individual p53-target genesacting at the transcriptional post-transcriptional and post-translational levels are integrated into circuits Individual genes contributing to a particular cellular outcome followingp53 activation are then combined into integrated circuits represented here by the cell cycle arrest and apoptosis integrated circuits composed of p2114-3-3σ and PUMANOXArespectively The two integrated circuits are then assembled into the p53 circuit board which ultimately consolidates all positive and negative signals to define one cell fate Positiveregulators of gene activity are denoted by solid boxes and dashed boxes indicate negative regulators


members p63 and p73 on these differential assemblies Finally inSection 5 we discuss the biomedical importance of a gene network(ie circuit board) approach with an emphasis on a new generationof p53-based targeted therapies whose clinical worth is limited bythe highly pleiotropic character of p53

2 Layered regulation of p53-derived signals the p21 circuit

p21 is a keymediator of p53-dependent cell cycle arrest by acting asa potent inhibitor of cyclin-dependent kinases [212232] however p53is only one among dozens of regulators of p21 activity in the cell p21 isexquisitely regulated at various stages from transcription to proteindegradation [33] and these regulatory events tune p53-dependentcell cycle arrest Here we will assemble a small fraction of the knownregulators of p21 into a circuit (Fig 2)

21 Transcriptional co-regulation of p21

p53 is but one of many well documented transcriptional regula-tors of the p21 locus [33] Positive and negative regulation by theseother factors dictate the extent to which p53 transactivates p21 Forexample MIZ1 (MYC interacting zinc-finger protein-1) [34] and SP1(specificity protein 1) [3536] have been shown to regulate p21 inp53-dependent andor -independent manners Recently ZAC1(zinc-finger protein which regulates apoptosis and cell cycle arrest)was shown to synergistically transactivate p21 through a direct inter-action with SP1 and through a functional interaction with p53 [37]demonstrating the hierarchical action of three p21 transactivatorsIn contrast p53-dependent p21 induction is subject to myriad

antagonistic factors For example MYC represses p21 activation in re-sponse to multiple forms of DNA damage via its interaction with MIZ1[3438] ZBTB4 (zinc-finger and BTB domain containing 4) is a mem-ber of the POZ domain-containing family of transcriptional repressorsthat can also inhibit p21 transcription through interaction with MIZ1[39] Other POZ repressors including FBI-1 ZBTB2 and ZBTB5 alsoregulate p21 transactivation through direct competition with p53andor SP1 [40ndash42] These examples demonstrate the potential forregulatory complexity that exists at just one step of gene regulationIntegration of the effects from many transcription factors is requiredto fine-tune the amount of p21 mRNA produced which then func-tions as the input for the next step in the circuit regulation of RNAstability and translational control

22 Post-transcriptional co-regulation of p21

After transcription splicing and 3prime end formation mature mRNAsare regulated by numerous factors that influence their stability local-ization and access to ribosomes for active translation Viewed as a co-ordinated regulatory module these events consolidate multiplesignals into a single output functional protein (Fig 2) IncreasedRNA stability plays a major role in facilitating p21 protein upregula-tion in response to DNA damage via the binding of HuR (ELAVL1) toan AU-rich element (ARE) in the p21 3prime untranslated region (UTR)[4344] RNPC1 (RBM38 RNA binding motif 38) is a direct transcrip-tional target of p53 which also binds to an ARE in the p21 3prime UTR tostabilize it under DNA damaging conditions thus creating a link be-tween two adjacent levels of the circuit [45] In fact HuR andRNPC1 act cooperatively to bind the 3prime UTR and increase the stability


of the p21 mRNA [46] In contrast AUF1 (AU-rich binding protein 1hnRNP D) competes with HuR to regulate p21 mRNA stability antag-onistically In an intriguing set of experiments Lal et al demonstratedthat while AUF1 and HuR can bind to adjacent sites in the p21 3prime UTRto promote nuclear export in the absence of HuR AUF1 shifts itsbinding position to promote destabilization of the mRNA [47] In neu-ronal cells hnRNP K acts in concert with HuB (ELAVL2) to block trans-lation of p21 mRNA an interesting finding in light of the fact thathnRNP K also positively regulates p53-dependent transcription ofp21 [4849] Therefore hnRNP K is an example of a factor that can reg-ulate expression of p53 target genes at multiple sites within the cir-cuit Elements in the 5prime coding region of the p21 mRNA are alsosubject to regulation as CUGBP1 and calreticulin (CALR) competefor binding to CG-rich elements to increase or decrease translationrespectively in an interplay shown to affect differentiation and senes-cence [5051] Thus post-transcriptional regulation by competing RNAbinding factors significantly influences p21 protein levels Interestinglyp53 activates the expression of p21 both directly at the transcriptionallevel and indirectly by regulating factors that increase mRNA stabilityThe ultimate signal integrator of post-transcriptional events is the ribo-some which synthesizes protein that can in turn be post-translationallyregulated

23 Post-translational co-regulation of p21

p21 is subject to a variety of post-translational modifications thatalter its activity subcellular localization andor stability (Fig 2) Thep21 protein can be phosphorylated on at least seven SerThr residueswith variable effects on localization and stability For example AKTPKB can phosphorylate p21 on Thr145 or Ser146 This is an intriguingcase because AKT phosphorylation at Thr145 was shown to increasep21 stability and block its association with PCNA promoting cell cyclearrest whereas the Ser146 modification activates a Cyclin D1CDK4complex to drive proliferation [52] Additionally phosphorylation ofThr145 by the oncogenic PIM1 kinase was shown to alter p21 localiza-tion sequestering it in the cytoplasm where it is not competent toblock cell cycle progression [53] Furthermore p21 protein stabilitycan be altered both positively and negatively by p38αJNK1 phosphor-ylation of Ser130 and GSK3β phosphorylation of Ser114 respectively[5455] These examples represent only a small subset of the kinasesand phosphorylation sites that regulate p21 but they surely serve todemonstrate how a single type of post-translational modification canmodulate p21 activity well after p53-dependent transactivation

The final layer of regulation in this circuit comes at the level of pro-teasomal degradation Ubiquitin-mediated degradation of p21 occurs atdifferent phases of the cell cycle directed by different E3 ligases includ-ing SCFSKP2 APCCCdc20 and CRL4CDT2 [56ndash58] It has also been reportedthat p21 can be degraded by the proteasome independent of its ubiqui-tination status [5960] p21 can be stabilized without interceding post-translational modifications through the formation of complexes withCyclin D1 mediated by RAS overexpression or by binding to WISP39HSP90 both of which prevent the association of p21 with the protea-some [6162] This last layer of regulation can amplify or dampen thesignal initiated by active p53 to determine the level and activity ofp21 in the cell as well as its final effect on the cell cycle

3 Assembling circuits into integrated circuits and circuit boardsthe minimal case of p21 and 14-3-3σ versus PUMA and NOXA

As illustrated for p21 the regulation of individual p53 target genes isa complex process analogous to a circuit Determination of the ultimatebiological outcome following p53 activation owes to the convergence ofthese regulatory processes on numerous target genes The groups ofgenes that work together to affect a given phenotypic outcome areakin to an integrated circuit where numerous circuits operate in paral-lel In turn integrated circuits assemble into circuit boards governing

the overall cellular response to a specific stimulus In the next few par-agraphs we will assemble a minimal circuit board composed of themanifold gene-specific transcriptional post-transcriptional and post-translational regulators affecting four p53 target genes p21 14-3-3σPUMA and NOXA (Fig 3)

31 Gene-specific transcriptional regulation of p21 14-3-3σ PUMA andNOXA

Cell cycle arrest in response to p53 activation is mediated by twomain proteins p21 which arrests cells mainly in the G1 phase ofthe cell cycle and 14-3-3σ which arrests cells at the G2M transition[63] 14-3-3σ is a direct transcriptional target of p53 that binds to themitotic phosphatase CDC25 and sequesters it in the cytoplasm thuspreventing it from activating CDK1-cyclin B complexes [6364] p21and 14-3-3σ work coordinately to enforce cell cycle arrest in re-sponse to various DNA damaging agents and protect from apoptosis[6566] Relatively little is known about additional transcriptional cor-egulators of 14-3-3σ however BRCA1 has been shown to increase14-3-3σ mRNA levels in a p53-dependent fashion [67] In mouse EScells BRCA1 is required for full expression of 14-3-3σ a findingwhich translated to several p53 wild-type human cancer cell lineswhere overexpression of BRCA1 resulted in upregulation of 14-3-3σ[67] Negative regulation of 14-3-3σ transcription by promoter DNAhypermethylation occurs in many normal tissues and is widespreadin human cancers but the mechanisms by which it is established re-main largely elusive [68ndash70] Of note the IKKα kinase shields 14-3-3σ from silencing in keratinocytes by binding to the promoter andblocking the action of the histone- and DNA-methyl-transferasesSUV39H1 and DNMT3A respectively [71]

The apoptotic module of the p53 response involves numerous fac-tors including the BCL2-homology domain 3 (BH3)-only proteinsPUMA and NOXA which are both p53 target genes that inhibit pro-survival BCL2 family members such as BCL2 BCL2L1 (BCL-xL) andMCL1 thus activating the intrinsic apoptotic pathway [72ndash75]PUMA is also known to directly activate BAX the pore-forming pro-tein required for permeabilization of the outer mitochondrial mem-brane [76] Each of these genes is subject to additional regulation atthe transcriptional level PUMA for example can be directly activatedby the FOXO3 (Forkhead box O3a) transcription factor and may be re-pressed by MYC [7778] FOXO3-dependent induction of PUMA syner-gizes with transactivation-independent p53 activity to induceapoptosis [7879] PUMA is also induced by the transcription factorIRF1 as well as by NF-κB in response to TNF-α (tumor necrosisfactor-α) [8081] In some cell types the p53 target SLUG (SNAI2) di-rectly represses PUMA transcription in response to ionizing radiation[82] Finally we have demonstrated that PUMA is subject to a novelmode of gene-specific transcriptional repression whereby the insula-tor protein CTCF (CCCTC-binding factor) establishes a repressivechromatin boundary within the coding region of the gene [83]

A large cohort of factors similarly regulates NOXA transcriptionNOXA is transactivated under hypoxic conditions by HIF1A (Hyp-oxia inducible factor-1 alpha) independently of p53 [84] Protea-some inhibition and cytotoxic stimuli both result in NOXAupregulation mediated by MYC and FOXO1 (Forkhead box O1) re-spectively [8586] Conversely transcriptional repression of NOXAhas been demonstrated in several instances including silencing bythe polycomb group protein BMI1 and by glucocorticoids in acutelymphoblastic leukemia (ALL) cells [8788] E2F1 is a particularly in-teresting component in the apoptotic integrated circuit as it can bindto and activate the promoters of PUMA NOXA and several otherBH3-only genes [89] The integration of these complex and diverseregulatory signals generates a set of mature mRNAs that serve asinput for the post-transcriptional and post-translational stages ofthe circuit board


32 Gene-specific regulation of p21 14-3-3σ PUMA and NOXA at thepost-transcriptional and post-translational levels

In addition to p21 the cell cycle arrest integrated circuit is modu-lated by post-transcriptional regulators of 14-3-3σ (Fig 3) 14-3-3σprotein levels are regulated by several E3-ubiquitin ligases EFP (es-trogen-responsive finger protein TRIM25) a protein commonly over-expressed in breast cancers specifically targets 14-3-3σ forproteolysis and consistent with this function knockdown of EFP re-duces tumor growth in vivo [90] CARP2 (caspase 810 associatedRING protein 2) has also been implicated in proteasomal degradationof 14-3-3σ although the exact mechanism of CARP2 action remainsto be elucidated [91]

The apoptotic integrated circuit includes post-transcriptionalregulators of PUMA and NOXA A number of microRNAs includingmiR-125b -221222 and -296 have been shown to inhibit the pro-duction of PUMA protein in various cell types miR-125b promotestumor growth in vivo by targeting several pro-apoptotic mRNAs in-cluding the 3prime UTR of the PUMA transcript [92] miR-221222 havebeen shown to down-regulate PUMA expression in both glioblasto-ma and epithelial cancers [9394] The IKK1IKK2NEMO kinase com-plex was found to trigger phosphorylation of PUMA protein at Ser10

to promote its proteasomal degradation a mechanism that can beinitiated by inputs that compete with p53 activation such as cyto-kine signaling [9596] PUMA activity is also regulated at the levelof subcellular localization In glioblastoma cells EGFR (epidermalgrowth factor receptor) binds directly to PUMA and sequesters it inthe cytoplasm preventing its activity at the mitochondria [97] Reg-ulation of NOXA protein translation is less well understood althoughcytoplasmic sequestration has been reported as a mechanism to re-duce its activity at the mitochondrial membrane [98] In hematopoi-etic cells NOXA is constitutively expressed and phosphorylated onSer13 by CDK5 resulting in its cytoplasmic localization and inhibi-tion Glucose deprivation removes this mark activating the apopto-tic potential of NOXA [98] NOXA protein is also subject toproteasomal degradation In particular the Kruppel-like tumor sup-pressor KLF6-SV1 (splice variant 1) has been shown to promoteNOXA turnover using a mechanism that involves the E3-ubiquitin li-gase activity of MDM2 thereby promoting cancer progression andchemoresistance in vivo [99]

Clearly the examples highlighted so far are a minimal representa-tion of the p53 circuit board If more p53 target genes were includedin the analysis the complexity would become staggering Instead wewill leave the circuit board in this minimal form and explore how itsdifferential assembly in different contexts modulates the cellular re-sponse to p53 activation

4 Context-dependent configurations of the p53 circuit board

Context is the hallmark of biological processes The biological re-sponse observed upon activation of any node within a gene networkis affected by the contextual connectivity of said node The ONndashOFFswitch of a food blender operates in basically the same way as theONndashOFF switch of a bedside lamp The outcome of flipping the switchto the ON position is not defined by the switch itself but rather by thedownstream circuitry and machinery Such context-dependent out-comes are readily apparent in the p53 network and can be reasonablyresolved into two major types cell type-specific and stimulus-specificThe same p53 activating stimulus triggers starkly different responsesacross cell types A single cell type undergoes very different p53-dependent outcomes in response to distinct p53-activating stimuliIn this section examples of context-specific assemblies of the p53 cir-cuit board will be discussed with careful attention paid to how p53-dependent and p53-independent events are integrated to produce spe-cific outcomes

41 Cell type-specific configurations of the p53 circuit board

The two hundred or so differentiated cell types in the adult humanbody are genetically identical yet the fraction of their genomes that isexpressed is clearly different Although p53 is ubiquitously expressedacross tissues its surrounding network is massively different fromone cell type to another As epigenetic mechanisms silence much ofthe genome in a cell type-specific fashion the availability of p53 tar-get genes and their co-regulators varies immensely and so does thecellular response to p53 activation

A clear example of this pleiotropy is provided by elegant mousestudies where the wild type p53 gene was replaced by p53ERTAM a4-hydroxy-tamoxifen (4-OHT)-dependent variant of p53 [100101]Tissues and cells derived from these knock-in mice can be toggled be-tween p53-deficient and p53-proficient states by systemic adminis-tration and withdrawal of 4-OHT In the presence of MDM2addition of 4-OHT does not suffice to trigger a p53 response whichrequires additional activating stimuli such as DNA damage or onco-gene hyperactivation [101] However in p53ERTAM Mdm2minusminus mice4-OHT elicits full p53 activation [100] In this scenario activation ofp53ERTAM triggers efficient apoptosis in all classically radiosensitivetissues such as bone marrow thymus spleen white pulp and smalland large intestines which leads to their rapid atrophy [100] In con-trast classically radio-insensitive tissues like lung heart brain liverand kidney remain phenotypically normal In the case of testis al-though apoptosis is not observed atrophy is still obvious which isseemingly caused by a profound proliferation arrest Importantlythe authors confirmed that p21 and Puma mRNAs are effectively in-duced in all tissues regardless of outcome Thus although the switchis flipped to the ON position in every tissue and specific circuits withinthe circuit board are available and activated in every cell type tested (iep21 and PUMA) the final output is markedly different in each cell typewhich can only be explained by as yet undefined differences in the restof the circuit board

Although p21 is a key mediator of p53-dependent cell cycle arrestits loss does not completely abrogate the ability of p53 to halt prolif-eration which demonstrates the existence of redundant and cooper-ative pathways [102103] 14-3-3σ GADD45A and REPRIMO (RPRM)are additional p53 target genes which have been proposed to collab-orate with p21 mainly to deliver G2 arrest in specific cell types[6366104105] Another mediator of proliferation arrest is the p53-inducible microRNA miR-34a which is a direct transcriptional targetof p53 miR-34a over-expression induces arrest of various cell typesin the presence or absence of p21 whereas reducing miR-34a activitycompromises the arrest response [106ndash109] Is the cell cycle arrest in-tegrated circuit assembled in a cell type-specific fashion Indeed 14-3-3σ was first identified as an epithelial cell antigen exclusivelyexpressed in epithelia and it is silenced via DNA methylation inmany normal tissues [68] In the context of cancer 14-3-3σ REPRIMOand miR-34a are common targets of aberrant silencing via DNAmeth-ylation such that their availability varies across cancer cell types[6970110ndash117] Furthermore enhanced turnover of the p21 mRNAand impaired processing of the primary transcript for miR-34a pre-vent their accumulation upon p53 activation in some cell types [118]

Cell type-specific configurations of the apoptotic circuit are also ob-vious p53 transactivates genes in various apoptotic pathways includ-ing components of the intrinsic (mitochondrial) pathway such asPUMA NOXA BID BAX andAPAF1 [727375119ndash121] aswell asmem-bers of the extrinsic (death receptor) pathway such as FAS DR4 andDR5Killer [122ndash124] The specific contribution of these target genesto p53-dependent apoptosis varies greatly across tissues For examplein the thymus anddeveloping central nervous systemofmice apoptosisinduced by ionizing radiation requires p53 and PUMA but not BAX[125] In mice where either Puma or Noxa was disrupted both factorscontributed to DNA damage-induced apoptosis in fibroblasts but onlyloss of PUMA protected lymphocytes from cell death [126] Careful


analysis of double knockout Pumaminusminus Dr5minusminusmice demonstrates anunexpected interdependency of the intrinsic and extrinsic pathways forthe execution of p53-dependent apoptosis in some but not all tissues[127] In this comparative study apoptosis in vivo following sub-lethal whole-body IR is almost exclusively p53-dependent in the bonemarrow spleen thymus and GI tract Both Dr5 and Puma contributedsignificantly to cell death in the spleen and thymus yet Puma is themain contributor to cell death in the GI-tract and bone marrow

Expectedly pro-apoptotic factors tend to be lost or inactivatedduring tumor development including various p53 targets whichthen leads to multiple possible assemblies of the p53 apoptotic inte-grated circuit across cancer cell types For example PUMA is often si-lenced in lymphomas independently of p53 mutations [128] In smallcell lung carcinomas various combinations of p53 target genes in theextrinsic pathway are often silenced via DNA methylation includingFAS DR4 and caspase 8 [129] A recent genomics study of somaticcopy number alterations in 680 tumors representing 17 major cancertypes revealed multiple possible assemblies of the apoptotic circuit asdefined by deletions of PUMA and BOK and copy number gain forMCL1 and BCL-xL [130]

Taken together these observations reveal a great diversity in theassembly of the p53 circuit board across both normal and cancerouscell types

42 Stimulus-specific configurations of the p53 circuit board

In addition to the cell type-specific configurations of the p53 net-work a given cell type can also adopt alternative p53-dependent cellfates in a stimulus-specific manner There are disappointingly fewpublications wherein distinct stimuli that elicit different p53-dependent responses have been compared in a single cell type in anattempt to understand how alternative cell fates are established

p21 has provided an excellent paradigm to understand stimulus-specific regulation within the p53 network Not all p53-activatingagents cause p21 upregulation p53 activation by Nutlin-3 5-fluorouracil (5-FU) doxorubicin daunorubicin and ionizing radiation(IR) results in effective accumulation of the p21 protein in diverse can-cer cell types whereas p53 activation byUltraviolet Light C (UVC) or hy-droxyurea (HU) does not [131ndash135] Interestingly all these agentsproduce equivalent accumulation of p53 effective binding of p53 tothe enhancers in the p21 promoter and p53-dependent recruitment ofspecific p53 cofactors such as histone acetyl-transferases and subunitsof the core Mediator complex [131ndash134] However in the case of UVCafter a transient wave of early transcription the transcriptional appara-tus at the p21 promoter is partially disassembled as defined by loss ofi) the CDK8-module of the Mediator complex ii) the general transcrip-tion factors TFIIF and TFIIB iii) transcription elongation factors and iv)elongating RNA polymerase II (RNAPII) [132133] Interestingly theTAF1 protein a subunit of the TFIID general transcription factorknown to downregulate p21 expression is strongly recruited to thep21 promoter after UVC [132ndash134] In the case of HU p21 inactivationoccurs not at early transcriptional events but rather at late elongationsteps [131135] During the S-phase checkpoint triggered byHU p53 ac-tivation leads to efficient recruitment and assembly of the transcrip-tional apparatus and RNAPII escape from the p21 promoter butRNAPII fails to complete elongation throughout the p21 coding regionInterestingly this block in late elongation is relieved by genetic or phar-macological ablation of the stress-induced kinase CHK1 [131] Thusparallel signals created by UVC and HU can impact the transcriptionalmachinery acting at the p21 promoter and intragenic region respec-tively to block the activating signal generated by p53 at the enhancersImportantly these effects are gene-specific as other p53 target genesdo not display such stimulus-specific regulation [131ndash135]

The p53 apoptotic circuit can also be assembled in a stimulus-specific fashion In response to DNA damage mouse embryonic fibro-blasts (MEFs) undergo p53-dependent G1 arrest However when

MEFs are transformed with the adenoviral E1A oncoprotein thesame stimulus leads to p53-dependent apoptosis Using this cellularsystem and a subtractive cloning strategy the Jacks group identifiedPERP as an ldquoapoptosis-specific generdquo that is expressed in a p53-dependent manner at much higher levels in the apoptotic setting[136] PERP is a membrane protein whose overexpression is sufficientto cause cell death in various cell types [137138] Strikingly severalother p53 targets in the arrest and apoptosis integrated circuitswere equally activated in each scenario including p21 Bax and Dr5Curiously the p53 target gene IGFBP3 is expressed only in arrestingcells In order to understand how the proliferative signals generatedby E1A resulted in differential expression of PERP the authorsshowed that E2F1 overexpression activates PERP but only in thepresence of p53 Thus oncogenic transformation creates an extra sig-nal that enables the activation of an additional apoptotic p53 target

43 Cell type- and stimulus-specific action of the p53 family membersp63 and p73

The p53 family of transcription factors includes p63 and p73[139ndash141] which share with p53 an N-terminal transactivation do-main (TA) a highly conserved DNA binding domain (DBD) and anoligomerization domain (OD) (reviewed in [142] Fig 4) Howeverp63 and p73 diverge from p53 at their C-termini which contain asterile active motif (SAM) domain and a transcription inhibition do-main (TID) [143144] p63 and p73 are subject to alternative pro-moter usage and alternative splicing which creates myriadisoforms varying in their N-termini and C-termini respectively Ofnote several p53 isoforms have also been identified whose func-tions remain enigmatic We direct the reader to an excellent reviewby Marcel et al on the current state of knowledge related to the p53isoforms [145]

As the DNA binding domains of both p63 and p73 exhibit highidentity with the p53 DBD [140] the consensus binding sites for allthree factors are virtually indistinguishable from one another[146ndash148] However the fact that the various isoforms contain differ-ent transcription activation and repression domains creates a uniqueopportunity to positively or negatively regulate gene expression in acombinatorial fashion In general the TAp63 and TAp73 isoformsare positive transcriptional regulators within the p53 network andthe ΔNp63 and ΔNp73 isoforms have been described mostly as nega-tive regulators Unlike p53 the p63 and p73 isoforms are not ubiqui-tously expressed and show instead exquisite tissue-specific patternsof expression which then contributes to cell type-specific assembliesof the p53 circuit board

431 Cousins in different placesThe TA andΔN isoforms of p63 clearly affect the p53 circuit board in

a cell type-specific manner TAp63 has been shown to transactivate thecell cycle arrest genes p21 and 14-3-3σ [149ndash151] as well as the pro-apoptotic genes PUMA and NOXA [149150152] (Fig 5) ConverselyΔNp63α has been reported as a transcriptional repressor of p21 14-3-3σ PUMA and NOXA [152ndash157] TAp63 is highly expressed in germcells of the ovary and testis [158] and is essential for DNA damage-induced oocyte death [159] thereby protecting the germline genomein a p53-independent manner Furthermore TAp63 is essential forRAS-induced senescence of fibroblasts and therefore may serve as ap53-independent tumor suppressor in mesenchymal tissues [160] Incontrast ΔNp63 is a potent pro-proliferative factor that is highlyexpressed in the basal cells of all stratified epithelia including skin cer-vix vaginal epithelium urothelium and prostate [140] The role ofΔNp63 in epithelial stem cell maintenance and proliferation is wellestablished due to a number of insightful studies performed usingp63 transgenic mice Mice lacking all p63 isoforms (p63minusminus) exhibitprofound developmental defects lacking all stratified epithelial tissueshair follicles teeth mammary lachrymal and salivary glands and limbs

Fig 4 The p53 family of transcription factors Schematic of the gene architecture and well characterized isoforms of p53 and the closely related factors p63 and p73 Arrows rep-resent alternative promoters and boxes represent exons (black segments are untranslated regions) TAD transactivation domain TA2 second transactivation domain PY prolinerich domain DBD DNA binding domain NLS nuclear localization signal OD oligomerization domain SAM sterile-alpha motif TID trans-inhibitory domain


[139161] Genetic complementation of p63minusminus mice with eitherΔNp63 or TAp63 demonstrate that there are unique roles for the p63isoforms in regulating epithelial development ΔNp63 is required forepidermal stem cell proliferation whereas TAp63 may contribute tothe differentiation of suprabasal keratinocytes and formation of thema-ture pluristratified epithelium [162] Thus while TAp63 feeds activatingsignals into the p53 circuit board in germ cells mesenchymal tissue andsuprabasal keratinocytes triggering differentiation cell cycle arrest se-nescence andor apoptosis depending on the contextΔNp63 attenuatessignalingwithin the circuit board to allow the continued proliferation ofstem cells in the epithelium

Somewhat less is known about the cell type-specific action of thep73 isoforms Like p53 and TAp63 TAp73 can increase the transcriptionof p21 14-3-3σ PUMA and NOXA [152156160163ndash165] thereby acti-vating the p53 circuit board in response to DNA damage and cell stressConversely ΔNp73 has been shown to repress the transcription of p21PUMA and NOXA [166ndash168] and may directly bind to and inactivatep53 TAp63 andor TAp73 [169] thereby serving to dampen signalingin the circuit board Initial characterization of the p73minusminus mouse de-termined that the loss of all p73 isoforms precipitated profound neu-rological defects [170] Furthermore two independently generatedΔNp73 knockout mouse strains show signs of neurodegenerationand brain atrophy [171172] indicating that ΔNp73 plays a crucialrole in regulating neuronal survival Indeed TAp73 and ΔNp73 havebeen shown to antagonize one another in neural tissue TAp73 stimu-lates neuronal apoptosis through the p53-independent activation of

PUMA expression a phenomenon which may be suppressed by theexogenous expression of ΔNp73 [152] ΔNp73 can also inhibit thepro-apoptotic effects of NGF (nerve growth factor) withdrawal there-by mediating neuronal survival through p53-dependent and-independent mechanisms [173] As ΔNp73 is the predominant iso-form expressed in sympathetic neurons [173] and in numerous partsof the brain [172] this demonstrates a cell type-specific role forΔNp73 in promoting neuronal survival In comparison studies usingthe TAp73minusminus mouse demonstrate its role as a bona fide tumor sup-pressor and in maintaining genomic stability in a broad range of tis-sues [174] Thus tissue-specific expression of the TA and ΔNisoforms of p73 allows for the exquisite regulation of their functionsin the p53 network in a cell type-specific manner

432 Cousins with different lifestylesp63 and p73 isoforms are regulated by signaling pathwayswhich do

not affect p53 directly thus creating additional regulatory diversity forthe stimulus-specific assembly of the p53 circuit board Unlike for p53MDM2 does not appear to function as an E3 ubiquitin ligase for p63 orp73 Instead the E3 ubiquitin ligase ITCH has been shown to targetthe TA and ΔN isoforms of both p63 and p73 for proteasomal degrada-tion [175176] ITCH itselfmay be positivelymodulated byphosphoryla-tion of Ser199 Thr222 and Ser232 by JNK1 [177] and negativelymodulated by phosphorylation of Tyr371 by the Src kinase FYN [178]or by associationwith the competitive inhibitor N4BP1 (Nedd4-bindingpartner 1) [179] thereby creating unique opportunities for stimulus-

Fig 5 The impact of p63 and p73 isoforms on the p53 circuit board This wiring diagram illustrates the ability of various p53 family members to co-regulate both positively andnegatively canonical p53 target genes involved in cell cycle arrest and apoptosis Different isoforms of p63 and p73 are expressed in diverse cell types thus providing cell type-specific regulatory capacity Furthermore p63 and p73 isoforms are positively and negatively regulated by diverse factors that do not affect p53 directly (gray boxes on top)thus providing additional opportunities for stimulus-specific regulation within the circuit board


specific regulation of p63 and p73 levels in the cell Other E3 ubiquitinligases targeting p63 and p73 for degradation are SCFFbw7 [180] andPIR2 [181] respectively which add other entry points for regulation atthe protein turnover level On the other hand CABLES1 (CDK 5 andAbl enzyme substrate 1) has been shown to protect TAp63 andΔNp63 from proteasomal degradation and CABLES1 expression is re-quired for maximal stabilization of TAp63 and subsequent inductionof apoptosis in female germ cells following DNA damage [182] Similar-ly NQO1 (NAD(P)H quinone oxidoreductase) and the PML (promyelo-cytic leukemia) protein have been shown to protect TAp63 and TAΔNp73 from proteasomal degradation respectively [183184] Finallyphosphorylation of TAp73 by PLK1 (polo-like kinase 1) [185] or variousCDKs [186] may inhibit p73s transactivation abilities These examplesprovide mechanisms by which specific factors outside of the p53 net-work may modulate the expression stability and activation of thep53 family members in a cell type-andor stimulus-specific mannerthereby providing additional avenues for input into the regulation ofthe p53 circuit board

5 Clinical relevance of understanding p53 as a circuit board

51 p53-based targeted therapies promise and obstacles

As the paradigm for cancer therapy shifts from blunt cocktails ofgenotoxic agents antimetabolites and mitotic poisons to biologicallytargeted therapies the development of non-genotoxic p53-based ther-apies is gaining momentum The validity of p53 as a therapeutic targethas been elegantly proven in animal models where specific activationof p53 in tumors leads to their demise via apoptosis or senescence[187188] Virtually every tumor expresses p53 In about half of thesetumors p53 has been inactivated by a single point mutation that dis-rupts its tumor suppression function and in some cases confers uponit oncogenic properties (reviewed in [189]) Since mutant p53 can not

transactivateMDM2 these tumors express large amounts of the formerIn the other half of tumors wild type p53 is expressed at low levels andkept in check by hyperactivation of its repressorsMDM2 andorMDM4Therefore two p53-based targeted therapies can be envisioned i) smallmolecules that convert (even if partially) mutant p53 back to the wildtype form and ii) small molecule inhibitors of MDM2 andor MDM4Approximately 11 million patients worldwide would benefit fromeach type of therapy [190] It is hard to envision a more widely usefultype of targeted therapymdash perhaps only inhibitors of the RAS oncogenewould match this potential Currently both classes of p53-based thera-pies are being tested in the clinic

The first molecules targeting p53 activity with any degree of spec-ificity were described in the late 1990s and early 2000s CP-31398 wasdiscovered by screening small molecules for the ability to restore con-formational stability to p53 using antibodies specific for the active anddenatured forms of p53 CP-31398 was shown to rescue the ability ofmutant p53 to transactivate a p53 reporter and inhibit tumor growthin vivo [191] A subsequent screen for compounds to reactivate mutantp53 identified PRIMA-1 (p53 reactivation and induction of massive ap-optosis) [192] This small molecule was found to restore p53 activity incells with mutant p53 and suppress tumor growth in a mouse xeno-graft model The first small molecule inhibitors of the MDM2ndashp53 in-teraction Nutlins were described in 2004 [193] Nutlin-3 mimics threehydrophobic residues on p53 required forMDM2 binding thusworkingas a competitive inhibitor of the interaction [193] Additional drugs inthis group include another MDM2 inhibitor MI-219 and RITA whichbinds instead to p53 to disrupt the p53ndashMDM2 interaction [194195]Nutlin-3 was shown to bind MDM2with nanomolar efficiency activatep53 and suppress tumor growth all without inducing the cytotoxic sideeffects associated with traditional chemotherapeutics [193] Unfortu-nately it quickly became clear that in a majority of cancer cell linesthe effects of Nutlin-3 treatment were cytostatic rather than apoptotic[196] limiting its therapeutic efficacy


If we agree that p53 is a mere signal generator upstream of a vastgene circuit controlling starkly diverse cellular responses it is thenobvious to anticipate that p53-based targeted therapeutics will in-duce tumor regression in a small fraction of patients It is fair to as-sume that these successful cases will be due to rapid commitmentto apoptosis or establishment of senescence within the tumors Cellcycle arrest and autophagy are deemed the least preferred outcomesfrom a therapeutic perspective as they are reversible and would leadonly to a temporary stall in tumor growth for as long as the therapy isadministered Furthermore it should be noted that the effects ofNutlin-3 are rapidly reversible Within half an hour of washingNutlin-3 from cell cultures p53 levels drop drastically to low basallevels most likely due to the fact that MDM2 accumulated duringthe treatment [197] In contrast genotoxic p53 activating agents pro-duce long lasting signaling which may explain their potency in theclinic Some unavoidable questions then arise What molecular mech-anisms define cell type-specific responses to Nutlin-3 How does thep53 circuit board react to genotoxic versus non-genotoxic p53 acti-vating agents Can drugs like Nutlin-3 be used in the clinic alone orin combination to elicit potent cell death in tumors but without thesystemic long lasting side effects of genotoxic agents

52 The impact of cell type- and stimulus-specific assemblies of the p53circuit board on the efficacy of p53-based therapies

While studying the mechanisms of cell fate choice in response toNutlin-3 we uncovered combinatorial assemblies of the cell cycle ar-rest integrated circuit across cancer cell types which impact the deci-sion to undergo arrest or commit to apoptosis (Fig 6A) [118] UponNutlin-3-treatment some cancer cell lines undergo reversible cellcycle arrest without apoptosis (eg HCT116 colorectal cancer A549lung cancer) others arrest at first and then commit to apoptosis(eg SJSA osteosarcoma LNCaP prostate cancer) and a few selecttypes undergo apoptosis without signs of arrest (eg BV173 chronicmyelogenous leukemia) Although p53 and several apoptotic targetsare equally activated in all cell lines we observe a clear correlationbetween the expression of p21 14-3-3σ and miR-34a and the out-come adopted HCT116 and A549 cells express high levels of thethree arrest genes SJSA and LNCaP cells express high levels of p21mid-levels of miR-34a and no 14-3-3σ and BV173 cells expressnone of these three p53 targets Mechanistic studies showed thatthe observed differences are due to a combination of cell type-specific degradation of p21 mRNA 14-3-3σ promoter DNA methyla-tion and impaired processing of the primary miR-34a transcript[118] Importantly we showed that attenuation of miR-34a functionin HCT116 p21minusminus 14-3-3σminusminus cells leads to a significant increasein apoptosis upon Nutlin-3 treatment relative to HCT116 cells expres-sing all three genes suggesting that the concerted action of p21 14-3-3σ and miR-34a protects cells from p53-dependent apoptosis andthat their expression level may determine at least partially thechoice of p53 response

In a comparative study of Nutlin-3 versus genotoxic agents weidentified stimulus-specific assemblies of the apoptotic circuit thatdefine the p53 response (Fig 6B) Interestingly we found that thesame cell lines that adopt a cell cycle arrest response upon Nutlin-3treatment can effectively undergo p53-dependent apoptosis whentreated with the antimetabolite 5-FU Using this paradigm we inves-tigated the contribution of ~20 components of the p53 network to cellfate choice [198] Strikingly we found that arresting cells display ef-fective transactivation of PUMA concurrent with translocation ofBAX to the mitochondria However these cells fail to release cyto-chrome C into the cytosol and activate caspases which is explainedby the failure of BAX to oligomerize at the mitochondria On theother hand cells undergoing p53-dependent apoptosis accumulatep21 14-3-3σ and other genes involved in cell cycle arrest but theyfail to arrest and show instead p53-dependent activation of caspases

Genetic dissection of different branches of the apoptotic integratedcircuit revealed that the key stimulus-specific events are i) p53-dependent activation of caspase 8 ii) caspase 8-dependent activationof the BH3-only protein BID and iii) BID-dependent activation andoligomerization of poised BAX Interestingly we found that thedeath receptor DR4 is strongly induced only in 5-FU-treated cellsvia a combination of p53-dependent transactivation and p53-independent mRNA stabilization In fact DR4 is required for activa-tion of caspase 8 BID and BAX Thus parallel signals generated by5-FU complement p53 action to tip the balance toward apoptosis

53 Personalized p53 medicines the urgent need for combinatorialtherapies functional genomics and molecular diagnostics

Given the stochastic nature of the mutations driving cancer pro-gression it is safe to assume that even across a unique tumor typethe p53 circuit board will adopt a large number of possible configura-tions with the consequent variability in the efficacy of p53-based tar-geted therapies One way of circumventing these shortcomings is tocombine Nutlin-3 with other agents that would tip the balance to-ward a rapid apoptotic response Many recent efforts have been de-voted to such combinatorial strategies using both traditional andtargeted therapeutics For example Nutlin-3 promotes apoptosis inconcert with ionizing radiation in otherwise radioresistant lung can-cer cells [199] Similarly Nutlin-3 exhibits synergy with genotoxicdrugs in diverse cancers ranging from chronic lymphocytic leukemiato hepatocellular carcinoma [200201] These potential treatmentsrepresent progress but remain a rather blunt approach that merelycombines DNA damage with high levels of p53 induction Of noteNutlin-3-induced cell cycle arrest protects from the killing effects ofmitotic poisons such as paclitaxel and taxol an observation that ledto the hypothesis that Nutlin-3 could be used for cyclotherapy pro-tecting healthy dividing cells from the harmful effects of certaindrugs meant to target cancer cells [202203] Nutlin-3 is also more ef-fective when combined with more specific drugs and small mole-cules For example several groups have reported that CDK inhibitorspotentiate the apoptotic effects of Nutlin treatment [204205] andsimilar results have been reported for MAPK inhibition in AML cells[206] In a more directed approach Nutlin-3 was combined withABT-737 a small molecule that sequesters pro-survival BCL2 familymembers in studies that demonstrated striking synergy betweenthe drugs in AML and breast cancer cells [207208] A recent studyshowed that inhibition of Hsp90 by 17AAG strongly activates p53-mediated apoptosis in response to Nutlin-3 treatment both in vitroand in vivo [209] Multiple studies have also shown that Nutlin-3 in-creases susceptibility to TRAIL via upregulation of death receptors[210211] These examples represent only a fraction of the drug com-binations tested with Nutlin but all carry the same challenges as sin-gle drug strategies which combinatorial therapies will be effective onany given tumor Certainly tumors where TRAIL receptors have beensilenced via DNAmethylation will not respond to a [Nutlin-3+TRAIL]combination [129] Thus there is an increasing need for both a largemenu of possible combinatorial strategies and the diagnostic toolsto predict which strategy will be most effective for a given tumor

With the advent of functional genomics it is now possible to inter-rogate the entire genome for pathways displaying synthetic lethalitywith targeted therapeutics In fact RNAi based screens for sensitizersto several targeted therapies approved for clinical use have identifiedadditional drug targets that increase the efficacy of the drugs[212213] These screens could be adapted to identify novel combina-torial strategies that improve the efficacy of p53-based targeted ther-apies These ldquosynthetic lethal with p53 activation pathwaysrdquo (SLPAPs)may behave as such in some but not all cancer types We can envisionthe existence of ldquocorerdquo and ldquofacultativerdquo SLPAPs depending on theiruniversality However core SLPAPs may also function as such innon-cancerous tissues which would limit their applicability due to

Fig 6 Context-dependent configurations of the p53 circuit board define the efficacy of p53 based therapies A An example of cell type-specific p53 responses is provided by non-genotoxic p53 activation by Nutlin-3 across cancer cell lines In BV173 cells (left) the cell cycle arrest integrated circuit is impaired by p21 mRNA decay 14-3-3σ promoter meth-ylation and impaired processing of miR-34a In contrast these three cell cycle arrest genes are effectively activated in HCT116 cells (right) where they function coordinately toestablish a cell cycle arrest response even though potent apoptotic genes such as PUMA have also been induced B Stimulus-specific assembly of the p53 circuit in response top53 activation by Nutlin-3 versus 5-FU in HCT116 cells Both Nutlin-3 and 5-FU strongly activate genes involved in both cell cycle arrest and apoptosis however only 5-FU treat-ment results in p53-independent stabilization of DR4 mRNA and concomitant upregulation of DR4 protein levels which is required for caspase 8 activation and proteolytic activa-tion of BID into tBID Activation of the DR4tBID axis by 5-FU drives the apoptotic response by promoting oligomerization of poised BAX at the mitochondria



side effects of the combinatorial therapy Upon identification of thesepathways the challenge will move then to defining which patientswill benefit from which combinatorial therapies highlighting theneed for molecular diagnostics Obviously Nutlin-3 and otherMDM2 inhibitors have no effects in p53 mutant cells [202] Thusp53 mutational status should be the first biomarker to be analyzedbefore deciding on a therapeutic strategy However great variabilityis to be expected across wild type p53 tumors as some may undergocell cycle arrest and others apoptosis and no biomarker is availableyet to distinguish these subclasses It is reasonable to predict that tu-mors with impaired arrest integrated circuits (as defined by loss ofp21 14-3-3σ andor miR-34a) may respond better to these drugsHowever loss of just one of these arrest factors does little to changethe response to Nutlin-3 [118214] Thus extensive gene expressionprofiling of tumors displaying variable responses to the drug seemnecessary to identify a more powerful gene signature Recently syn-thetic lethal screens have been proposed to accelerate biomarker dis-covery [215] The hypothesis is that many synthetic lethal geneswhich protect ldquoresistantrdquo cell lines from the drug will be expressedat lower levels in the ldquosensitiverdquo cells Thus cross-referencing geneprofiling and synthetic lethality datasets will provide the gene signa-ture with highest prognostic value However whereas gene profilingcan be easily performed in tumor samples RNAi screens are techni-cally amenable only to established cell cultures and it is unclear ifsynthetic lethality observed in vitro can be extrapolated to the clinicalsetting Technical difficulties aside the potential of these approachesis undeniable and they set clear research goals for the field

6 Final remarks

We ask the reader to picture the following scenario not too far intothe future During an annual check up the reader is found to carry anon-resectable tumor Pathologists report that it is a tumor type forwhich no single agent therapy has proven useful The pathology reportindicates that the tumor iswild type for p53MDM2positive andMDM4negative The oncologist orders two assays i) a genome wide shRNAscreen for synthetic lethality with an approved MDM2 inhibitor to beperformed on a growing explant derived from a fresh tumor biopsyand ii) a global RNA and protein profile of gene expression for thetumor tissues and a few select normal tissues Two weeks later the re-sults arrive as a list of possible synthetic lethal drug combinationsand digital blueprints of the p53 circuit boards in the tumor and normaltissues A week later treatment begins with sequential combinations ofan MDM2 inhibitor with other targeted therapies predicted to be syn-thetically lethal only in tumor cells All drugs are administered at lowdoses and for short periods of time Within days the tumor regressesand the reader is considered cancer free Science fiction or rational op-timism

ldquoThe rung of a ladder was never meant to rest upon but only to hold amans foot long enough to enable him to put the other somewhathigherrdquo~Thomas H Huxley

References

[1] AB DeLeo G Jay E Appella GC Dubois LW Law LJ Old Detection of atransformation-related antigen in chemically induced sarcomas and other trans-formed cells of the mouse Proc Natl Acad Sci U S A 76 (1979) 2420ndash2424

[2] DI Linzer AJ Levine Characterization of a 54K Dalton cellular SV40 tumor an-tigen present in SV40-transformed cells and uninfected embryonal carcinomacells Cell 17 (1979) 43ndash52

[3] DP Lane LV Crawford T antigen is bound to a host protein in SV40-transformed cells Nature 278 (1979) 261ndash263

[4] D Eliyahu A Raz P Gruss D Givol M Oren Participation of p53 cellular tu-mour antigen in transformation of normal embryonic cells Nature 312 (1984)646ndash649

[5] JR Jenkins K Rudge GA Currie Cellular immortalization by a cDNA cloneencoding the transformation-associated phosphoprotein p53 Nature 312(1984) 651ndash654

[6] DP Lane Cell immortalization and transformation by the p53 gene Nature 312(1984) 596ndash597

[7] SJ Baker ER Fearon JM Nigro SR Hamilton AC Preisinger JM Jessup PvanTuinen DH Ledbetter DF Barker Y Nakamura R White B VogelsteinChromosome 17 deletions and p53 gene mutations in colorectal carcinomasScience 244 (1989) 217ndash221

[8] SJ Baker AC Preisinger JM Jessup C Paraskeva S Markowitz JK Willson SHamilton B Vogelstein p53 gene mutations occur in combination with 17p al-lelic deletions as late events in colorectal tumorigenesis Cancer Res 50 (1990)7717ndash7722

[9] PW Hinds CA Finlay RS Quartin SJ Baker ER Fearon B Vogelstein AJLevine Mutant p53 DNA clones from human colon carcinomas cooperate withras in transforming primary rat cells a comparison of the ldquohot spotrdquo mutantphenotypes Cell Growth Differ 1 (1990) 571ndash580

[10] LA Donehower M Harvey BL Slagle MJ McArthur CA Montgomery Jr JSButel A Bradley Mice deficient for p53 are developmentally normal but suscep-tible to spontaneous tumours Nature 356 (1992) 215ndash221

[11] SJ Baker S Markowitz ER Fearon JK Willson B Vogelstein Suppression ofhuman colorectal carcinoma cell growth by wild-type p53 Science 249 (1990)912ndash915

[12] CA Finlay PW Hinds AJ Levine The p53 proto-oncogene can act as a suppres-sor of transformation Cell 57 (1989) 1083ndash1093

[13] S Fields SK Jang Presence of a potent transcription activating sequence in thep53 protein Science 249 (1990) 1046ndash1049

[14] L Raycroft HY Wu G Lozano Transcriptional activation by wild-type but nottransforming mutants of the p53 anti-oncogene Science 249 (1990)1049ndash1051

[15] SE Kern KW Kinzler A Bruskin D Jarosz P Friedman C Prives B VogelsteinIdentification of p53 as a sequence-specific DNA-binding protein Science 252(1991) 1708ndash1711

[16] G Farmer J Bargonetti H Zhu P Friedman R Prywes C Prives Wild-type p53activates transcription in vitro Nature 358 (1992) 83ndash86

[17] GP Zambetti J Bargonetti K Walker C Prives AJ Levine Wild-type p53 me-diates positive regulation of gene expression through a specific DNA sequenceelement Genes Dev 6 (1992) 1143ndash1152

[18] J Momand GP Zambetti DC Olson D George AJ Levine The mdm-2 onco-gene product forms a complex with the p53 protein and inhibits p53-mediated transactivation Cell 69 (1992) 1237ndash1245

[19] FS Leach T Tokino P Meltzer M Burrell JD Oliner S Smith DE Hill DSidransky KW Kinzler B Vogelstein p53 mutation and MDM2 amplificationin human soft tissue sarcomas Cancer Res 53 (1993) 2231ndash2234

[20] MH Kubbutat SN Jones KH Vousden Regulation of p53 stability by Mdm2Nature 387 (1997) 299ndash303

[21] WS el-Deiry T Tokino VE Velculescu DB Levy R Parsons JM Trent D LinWE Mercer KW Kinzler B Vogelstein WAF1 a potential mediator of p53tumor suppression Cell 75 (1993) 817ndash825

[22] JW Harper GR Adami N Wei K Keyomarsi SJ Elledge The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinasesCell 75 (1993) 805ndash816

[23] T Tokino S Thiagalingam WS el-Deiry T Waldman KW Kinzler BVogelstein p53 tagged sites from human genomic DNA Hum Mol Genet 3(1994) 1537ndash1542

[24] J Yu L Zhang PM Hwang C Rago KW Kinzler B Vogelstein Identificationand classification of p53-regulated genes Proc Natl Acad Sci U S A 96(1999) 14517ndash14522

[25] R Zhao K Gish M Murphy Y Yin D Notterman WH Hoffman E Tom DHMack AJ Levine Analysis of p53-regulated gene expression patterns using oli-gonucleotide arrays Genes Dev 14 (2000) 981ndash993

[26] T Riley E Sontag P Chen A Levine Transcriptional control of human p53-regulated genes Nat Rev Mol Cell Biol 9 (2008) 402ndash412

[27] KH Vousden C Prives Blinded by the light the growing complexity of p53 Cell137 (2009) 413ndash431

[28] KH Vousden X Lu Live or let die the cells response to p53 Nat Rev Cancer 2(2002) 594ndash604

[29] B Vogelstein D Lane AJ Levine Surfing the p53 network Nature 408 (2000)307ndash310

[30] MV Poyurovsky C Prives Unleashing the power of p53 lessons from mice andmen Genes Dev 20 (2006) 125ndash131

[31] AJ Levine M Oren The first 30 years of p53 growing ever more complex NatRev Cancer 9 (2009) 749ndash758

[32] WS el-Deiry JW Harper PM OConnor VE Velculescu CE Canman JJackman JA Pietenpol M Burrell DE Hill Y Wang et al WAF1CIP1 is in-duced in p53-mediated G1 arrest and apoptosis Cancer Res 54 (1994)1169ndash1174

[33] YS Jung Y Qian X Chen Examination of the expanding pathways for the reg-ulation of p21 expression and activity Cell Signal 22 (2010) 1003ndash1012

[34] S Herold M Wanzel V Beuger C Frohme D Beul T Hillukkala J Syvaoja HPSaluz F Haenel M Eilers Negative regulation of the mammalian UV responseby Myc through association with Miz-1 Mol Cell 10 (2002) 509ndash521

[35] G Koutsodontis I Tentes P Papakosta A Moustakas D Kardassis Sp1 plays acritical role in the transcriptional activation of the human cyclin-dependent ki-nase inhibitor p21(WAF1Cip1) gene by the p53 tumor suppressor protein JBiol Chem 276 (2001) 29116ndash29125


[36] K Nakano T Mizuno Y Sowa T Orita T Yoshino Y Okuyama T Fujita NOhtani-Fujita Y Matsukawa T Tokino H Yamagishi T Oka H Nomura TSakai Butyrate activates the WAF1Cip1 gene promoter through Sp1 sites in ap53-negative human colon cancer cell line J Biol Chem 272 (1997)22199ndash22206

[37] PY Liu TY Hsieh ST Liu YL Chang WS Lin WM Wang SM Huang Zac1an Sp1-like protein regulates human p21(WAF1Cip1) gene expression inHeLa cells Exp Cell Res 317 (2011) 2925ndash2937

[38] J Seoane HV Le J Massague Myc suppression of the p21(Cip1) Cdk inhibitor influ-ences theoutcomeof thep53 response toDNAdamage Nature 419 (2002) 729ndash734

[39] A Weber J Marquardt D Elzi N Forster S Starke A Glaum D Yamada PADefossez J Delrow RN Eisenman H Christiansen M Eilers Zbtb4 repressestranscription of P21CIP1 and controls the cellular response to p53 activationEMBO J 27 (2008) 1563ndash1574

[40] DI Koh WI Choi BN Jeon CE Lee CO Yun MW Hur A novel POK familytranscription factor ZBTB5 represses transcription of p21CIP1 gene J BiolChem 284 (2009) 19856ndash19866

[41] BN Jeon WI Choi MY Yu AR Yoon MH Kim CO Yun MW Hur ZBTB2 anovel master regulator of the p53 pathway J Biol Chem 284 (2009)17935ndash17946

[42] WI Choi BN Jeon CO Yun PH Kim SE Kim KY Choi SH Kim MW HurProto-oncogene FBI-1 represses transcription of p21CIP1 by inhibition of tran-scription activation by p53 and Sp1 J Biol Chem 284 (2009) 12633ndash12644

[43] M Gorospe X Wang NJ Holbrook p53-dependent elevation of p21Waf1 ex-pression by UV light is mediated through mRNA stabilization and involves avanadate-sensitive regulatory system Mol Cell Biol 18 (1998) 1400ndash1407

[44] W Wang H Furneaux H Cheng MC Caldwell D Hutter Y Liu N Holbrook MGorospe HuR regulates p21 mRNA stabilization by UV light Mol Cell Biol 20(2000) 760ndash769

[45] L Shu W Yan X Chen RNPC1 an RNA-binding protein and a target of the p53family is required for maintaining the stability of the basal and stress-inducedp21 transcript Genes Dev 20 (2006) 2961ndash2972

[46] SJ Cho J Zhang X Chen RNPC1 modulates the RNA-binding activity of and co-operates with HuR to regulate p21 mRNA stability Nucleic Acids Res 38 (2010)2256ndash2267

[47] A Lal K Mazan-Mamczarz T Kawai X Yang JL Martindale M Gorospe Con-current versus individual binding of HuR and AUF1 to common labile targetmRNAs EMBO J 23 (2004) 3092ndash3102

[48] A Moumen P Masterson MJ OConnor SP Jackson hnRNP K an HDM2 targetand transcriptional coactivator of p53 in response to DNA damage Cell 123(2005) 1065ndash1078

[49] M Yano HJ Okano H Okano Involvement of Hu and heterogeneous nuclear ri-bonucleoprotein K in neuronal differentiation through p21 mRNA post-transcriptional regulation J Biol Chem 280 (2005) 12690ndash12699

[50] NA Timchenko P Iakova ZJ Cai JR Smith LT Timchenko Molecular basis forimpaired muscle differentiation in myotonic dystrophy Mol Cell Biol 21(2001) 6927ndash6938

[51] P Iakova GL Wang L Timchenko M Michalak OM Pereira-Smith JR SmithNA Timchenko Competition of CUGBP1 and calreticulin for the regulation ofp21 translation determines cell fate EMBO J 23 (2004) 406ndash417

[52] Y Li D Dowbenko LA Lasky AKTPKB phosphorylation of p21CipWAF1 en-hances protein stability of p21CipWAF1 and promotes cell survival J BiolChem 277 (2002) 11352ndash11361

[53] Z Wang N Bhattacharya PF Mixter W Wei J Sedivy NS Magnuson Phos-phorylation of the cell cycle inhibitor p21Cip1WAF1 by Pim-1 kinase BiochimBiophys Acta 1593 (2002) 45ndash55

[54] GY Kim SE Mercer DZ Ewton Z Yan K Jin E Friedman The stress-activatedprotein kinases p38 alpha and JNK1 stabilize p21(Cip1) by phosphorylation JBiol Chem 277 (2002) 29792ndash29802

[55] JY Lee SJ Yu YG Park J Kim J Sohn Glycogen synthase kinase 3beta phos-phorylates p21WAF1CIP1 for proteasomal degradation after UV irradiationMol Cell Biol 27 (2007) 3187ndash3198

[56] G Bornstein J Bloom D Sitry-Shevah K Nakayama M Pagano A HershkoRole of the SCFSkp2 ubiquitin ligase in the degradation of p21Cip1 in S phaseJ Biol Chem 278 (2003) 25752ndash25757

[57] V Amador S Ge PG Santamaria D Guardavaccaro M Pagano APCC(Cdc20)controls the ubiquitin-mediated degradation of p21 in prometaphase Mol Cell27 (2007) 462ndash473

[58] Y Kim NG Starostina ET Kipreos The CRL4Cdt2 ubiquitin ligase targets thedegradation of p21Cip1 to control replication licensing Genes Dev 22 (2008)2507ndash2519

[59] C Cayrol B Ducommun Interaction with cyclin-dependent kinases and PCNAmodulates proteasome-dependent degradation of p21 Oncogene 17 (1998)2437ndash2444

[60] R Touitou J Richardson S Bose M Nakanishi J Rivett MJ Allday A degrada-tion signal located in the C-terminus of p21WAF1CIP1 is a binding site for theC8 alpha-subunit of the 20S proteasome EMBO J 20 (2001) 2367ndash2375

[61] ML Coleman CJ Marshall MF Olson Ras promotes p21(Waf1Cip1) proteinstability via a cyclin D1-imposed block in proteasome-mediated degradationEMBO J 22 (2003) 2036ndash2046

[62] T Jascur H Brickner I Salles-Passador V Barbier A El Khissiin B Smith RFotedar A Fotedar Regulation of p21(WAF1CIP1) stability by WISp39 aHsp90 binding TPR protein Mol Cell 17 (2005) 237ndash249

[63] H Hermeking C Lengauer K Polyak TC He L Zhang S Thiagalingam KWKinzler B Vogelstein 14-3-3 sigma is a p53-regulated inhibitor of G2M pro-gression Mol Cell 1 (1997) 3ndash11

[64] CY Peng PR Graves RS Thoma Z Wu AS Shaw H Piwnica-Worms Mitoticand G2 checkpoint control regulation of 14-3-3 protein binding by phosphory-lation of Cdc25C on serine-216 Science 277 (1997) 1501ndash1505

[65] TA Chan H Hermeking C Lengauer KW Kinzler B Vogelstein 14-3-3Sigma isrequired to prevent mitotic catastrophe after DNA damage Nature 401 (1999)616ndash620

[66] TA Chan PM Hwang H Hermeking KW Kinzler B Vogelstein Cooperativeeffects of genes controlling the G(2)M checkpoint Genes Dev 14 (2000)1584ndash1588

[67] O Aprelikova AJ Pace B Fang BH Koller ET Liu BRCA1 is a selective co-activator of 14-3-3 sigma gene transcription in mouse embryonic stem cells JBiol Chem 276 (2001) 25647ndash25650

[68] K Bhatia AK Siraj A Hussain R Bu MI Gutierrez The tumor suppressor gene14-3-3 sigma is commonly methylated in normal and malignant lymphoid cellsCancer Epidemiol Biomarkers Prev 12 (2003) 165ndash169

[69] AT Ferguson E Evron CB Umbricht TK Pandita TA Chan H Hermeking JRMarks AR Lambers PA Futreal MR Stampfer S Sukumar High frequency ofhypermethylation at the 14-3-3 sigma locus leads to gene silencing in breastcancer Proc Natl Acad Sci U S A 97 (2000) 6049ndash6054

[70] D Lodygin H Hermeking Epigenetic silencing of 14-3-3sigma in cancer SeminCancer Biol 16 (2006) 214ndash224

[71] F Zhu X Xia B Liu J Shen Y Hu M Person Y Hu IKKalpha shields 14-3-3sigma a G(2)M cell cycle checkpoint gene from hypermethylation prevent-ing its silencing Mol Cell 27 (2007) 214ndash227

[72] J Yu L Zhang PM Hwang KW Kinzler B Vogelstein PUMA induces the rapidapoptosis of colorectal cancer cells Mol Cell 7 (2001) 673ndash682

[73] K Nakano KH Vousden PUMA a novel proapoptotic gene is induced by p53Mol Cell 7 (2001) 683ndash694

[74] RJ Youle A Strasser The BCL-2 protein family opposing activities that mediatecell death Nat Rev Mol Cell Biol 9 (2008) 47ndash59

[75] E Oda R Ohki H Murasawa J Nemoto T Shibue T Yamashita T Tokino TTaniguchi N Tanaka Noxa a BH3-only member of the Bcl-2 family andcandidate mediator of p53-induced apoptosis Science 288 (2000)1053ndash1058

[76] H Kim HC Tu D Ren O Takeuchi JR Jeffers GP Zambetti JJ Hsieh EHCheng Stepwise activation of BAX and BAK by tBID BIM and PUMA initiates mi-tochondrial apoptosis Mol Cell 36 (2009) 487ndash499

[77] H You M Pellegrini K Tsuchihara K Yamamoto G Hacker M Erlacher AVillunger TW Mak FOXO3a-dependent regulation of Puma in response to cyto-kinegrowth factor withdrawal J Exp Med 203 (2006) 1657ndash1663

[78] S Amente J Zhang M Lubrano Lavadera L Lania EV Avvedimento B MajelloMyc and PI3KAKT signaling cooperatively repress FOXO3a-dependent PUMAand GADD45a gene expression Nucleic Acids Res 39 (2011) 9498ndash9507

[79] H You K Yamamoto TW Mak Regulation of transactivation-independentproapoptotic activity of p53 by FOXO3a Proc Natl Acad Sci U S A 103(2006) 9051ndash9056

[80] P Wang W Qiu C Dudgeon H Liu C Huang GP Zambetti J Yu L ZhangPUMA is directly activated by NF-kappaB and contributes to TNF-alpha-induced apoptosis Cell Death Differ 16 (2009) 1192ndash1202

[81] J Gao M Senthil B Ren J Yan Q Xing J Yu L Zhang JH Yim IRF-1 transcrip-tionally upregulates PUMA which mediates the mitochondrial apoptotic path-way in IRF-1-induced apoptosis in cancer cells Cell Death Differ 17 (2010)699ndash709

[82] W-S Wu S Heinrichs D Xu SP Garrison GP Zambetti JM Adams AT LookSlug antagonizes p53-mediated apoptosis of hematopoietic progenitors byrepressing puma Cell 123 (2005) 641ndash653

[83] NP Gomes JM Espinosa Gene-specific repression of the p53 target gene PUMAvia intragenic CTCF-Cohesin binding Genes Dev 24 (2010) 1022ndash1034

[84] JY Kim HJ Ahn JH Ryu K Suk JH Park BH3-only protein Noxa is a mediatorof hypoxic cell death induced by hypoxia-inducible factor 1alpha J Exp Med199 (2004) 113ndash124

[85] MA Nikiforov M Riblett WH Tang V Gratchouck D Zhuang Y Fernandez MVerhaegen S Varambally AM Chinnaiyan AJ Jakubowiak MS SoengasTumor cell-selective regulation of NOXA by c-MYC in response to proteasomeinhibition Proc Natl Acad Sci U S A 104 (2007) 19488ndash19493

[86] K Valis L Prochazka E Boura J Chladova T Obsil J Rohlena J Truksa LFDong SJ Ralph J Neuzil HippoMst1 stimulates transcription of the proapopto-tic mediator NOXA in a FoxO1-dependent manner Cancer Res 71 (2011)946ndash954

[87] M Yamashita M Kuwahara A Suzuki K Hirahara R Shinnaksu H HosokawaA Hasegawa S Motohashi A Iwama T Nakayama Bmi1 regulates memoryCD4 T cell survival via repression of the Noxa gene J Exp Med 205 (2008)1109ndash1120

[88] C Ploner J Rainer S Lobenwein S Geley R Kofler Repression of the BH3-onlymolecule PMAIP1Noxa impairs glucocorticoid sensitivity of acute lymphoblas-tic leukemia cells Apoptosis 14 (2009) 821ndash828

[89] T Hershko D Ginsberg Up-regulation of Bcl-2 homology 3 (BH3)-only proteinsby E2F1 mediates apoptosis J Biol Chem 279 (2004) 8627ndash8634

[90] T Urano T Saito T Tsukui M Fujita T Hosoi M Muramatsu Y Ouchi S InoueEfp targets 14-3-3 sigma for proteolysis and promotes breast tumour growthNature 417 (2002) 871ndash875

[91] W Yang DT Dicker J Chen WS El-Deiry CARPs enhance p53 turnover bydegrading 14-3-3sigma and stabilizing MDM2 Cell Cycle 7 (2008) 670ndash682

[92] XB Shi L Xue AH Ma CG Tepper HJ Kung RW White miR-125b promotesgrowth of prostate cancer xenograft tumor through targeting pro-apoptoticgenes Prostate 71 (2011) 538ndash549


[93] CZ Zhang JX Zhang AL Zhang ZD Shi L Han ZF Jia WD Yang GX WangT Jiang YP You PY Pu JQ Cheng CS Kang MiR-221 and miR-222 targetPUMA to induce cell survival in glioblastoma Mol Cancer 9 (2010) 229

[94] C Zhang J Zhang A Zhang Y Wang L Han Y You P Pu C Kang PUMA is anovel target of miR-221222 in human epithelial cancers Int J Oncol 37(2010) 1621ndash1626

[95] M Fricker J OPrey AM Tolkovsky KM Ryan Phosphorylation of Puma mod-ulates its apoptotic function by regulating protein stability Cell Death Dis 1(2010) e59

[96] JJ Sandow AM Jabbour MR Condina CP Daunt FC Stomski BD Green CDRiffkin P Hoffmann MA Guthridge J Silke AF Lopez PG Ekert Cytokine re-ceptor signaling activates an IKK-dependent phosphorylation of PUMA to pre-vent cell death Cell Death Differ (2011) doi 101038cdd2011131 [Onlineonly]

[97] H Zhu X Cao F Ali-Osman S Keir HW Lo EGFR and EGFRvIII interact withPUMA to inhibit mitochondrial translocalization of PUMA and PUMA-mediatedapoptosis independent of EGFR kinase activity Cancer Lett 294 (2010)101ndash110

[98] XH Lowman MA McDonnell A Kosloske OA Odumade C Jenness CBKarim R Jemmerson A Kelekar The proapoptotic function of Noxa in humanleukemia cells is regulated by the kinase Cdk5 and by glucose Mol Cell 40(2010) 823ndash833

[99] A Difeo F Huang J Sangodkar EA Terzo D Leake G Narla JA MartignettiKLF6-SV1 is a novel antiapoptotic protein that targets the BH3-only proteinNOXA for degradation and whose inhibition extends survival in an ovarian can-cer model Cancer Res 69 (2009) 4733ndash4741

[100] I Ringshausen CC OShea AJ Finch LB Swigart GI Evan Mdm2 is criticallyand continuously required to suppress lethal p53 activity in vivo Cancer Cell10 (2006) 501ndash514

[101] MA Christophorou D Martin-Zanca L Soucek ER Lawlor L Brown-SwigartEW Verschuren GI Evan Temporal dissection of p53 function in vitro and invivo Nat Genet 37 (2005) 718ndash726

[102] C Deng P Zhang JW Harper SJ Elledge P Leder Mice lacking p21CIP1WAF1undergo normal development but are defective in G1 checkpoint control Cell82 (1995) 675ndash684

[103] J Brugarolas C Chandrasekaran JI Gordon D Beach T Jacks GJ Hannon Ra-diation-induced cell cycle arrest compromised by p21 deficiency Nature 377(1995) 552ndash557

[104] MB Kastan Q Zhan WS el-Deiry F Carrier T Jacks WV Walsh BS PlunkettB Vogelstein AJ Fornace Jr A mammalian cell cycle checkpoint pathway utiliz-ing p53 and GADD45 is defective in ataxiandashtelangiectasia Cell 71 (1992)587ndash597

[105] R Ohki J Nemoto H Murasawa E Oda J Inazawa N Tanaka T Taniguchi Rep-rimo a new candidate mediator of the p53-mediated cell cycle arrest at the G2phase J Biol Chem 275 (2000) 22627ndash22630

[106] L He X He LP Lim E de Stanchina Z Xuan Y Liang W Xue L Zender JMagnus D Ridzon AL Jackson PS Linsley C Chen SW Lowe MA ClearyGJ Hannon A microRNA component of the p53 tumour suppressor networkNature 447 (2007) 1130ndash1134

[107] V Tarasov P Jung B Verdoodt D Lodygin A Epanchintsev A Menssen GMeister H Hermeking Differential regulation of microRNAs by p53 revealedby massively parallel sequencing miR-34a is a p53 target that induces apoptosisand G(1)-arrest Cell Cycle 6 (2007)

[108] H Tazawa N Tsuchiya M Izumiya H Nakagama Tumor-suppressivemiR-34a in-duces senescence-like growth arrest through modulation of the E2F pathway inhuman colon cancer cells Proc Natl Acad Sci U S A 104 (2007) 15472ndash15477

[109] K Kumamoto EA Spillare K Fujita I Horikawa T Yamashita E Appella MNagashima S Takenoshita J Yokota CC Harris Nutlin-3a activates p53 toboth down-regulate inhibitor of growth 2 and up-regulate mir-34a mir-34band mir-34c expression and induce senescence Cancer Res 68 (2008)3193ndash3203

[110] R Henrique C Jeronimo MO Hoque AL Carvalho J Oliveira MR Teixeira CLopes D Sidransky Frequent 14-3-3 sigma promoter methylation in benign andmalignant prostate lesions DNA Cell Biol 24 (2005) 264ndash269

[111] E Kunze M Wendt T Schlott Promoter hypermethylation of the 14-3-3 sigmaSYK and CAGE-1 genes is related to the various phenotypes of urinary bladdercarcinomas and associated with progression of transitional cell carcinomasInt J Mol Med 18 (2006) 547ndash557

[112] JP Hamilton F Sato Z Jin BD Greenwald T Ito Y Mori BC Paun T Kan YCheng S Wang J Yang JM Abraham SJ Meltzer Reprimo methylation is a po-tential biomarker of Barretts-associated esophageal neoplastic progressionClin Cancer Res 12 (2006) 6637ndash6642

[113] N Sato N Fukushima H Matsubayashi CA Iacobuzio-Donahue CJ Yeo MGoggins Aberrant methylation of Reprimo correlates with genetic instabilityand predicts poor prognosis in pancreatic ductal adenocarcinoma Cancer 107(2006) 251ndash257

[114] TS Wong DL Kwong JS Sham WI Wei AP Yuen Methylation status of Rep-rimo in head and neck carcinomas Int J Cancer 117 (2005) 697

[115] M Suzuki H Shigematsu T Takahashi N Shivapurkar UG Sathyanarayana TIizasa T Fujisawa AF Gazdar Aberrant methylation of Reprimo in lung cancerLung Cancer 47 (2005) 309ndash314

[116] T Takahashi M Suzuki H Shigematsu N Shivapurkar C Echebiri M Nomura VStastny M Augustus CW Wu II Wistuba SJ Meltzer AF Gazdar Aberrantmethylation of Reprimo in humanmalignancies Int J Cancer 115 (2005) 503ndash510

[117] M Vogt J Munding M Gruner ST Liffers B Verdoodt J Hauk L SteinstraesserA Tannapfel H Hermeking Frequent concomitant inactivation of miR-34a and

miR-34bc by CpG methylation in colorectal pancreatic mammary ovarianurothelial and renal cell carcinomas and soft tissue sarcomas Virchows Arch458 (2011) 313ndash322

[118] R Paris RE Henry SJ Stephens M McBryde JM Espinosa Multiple p53-independent gene silencing mechanisms define the cellular response to p53 ac-tivation Cell Cycle 7 (2008) 2427ndash2433

[119] T Miyashita JC Reed Tumor suppressor p53 is a direct transcriptional activatorof the human bax gene Cell 80 (1995) 293ndash299

[120] JK Sax P Fei ME Murphy E Bernhard SJ Korsmeyer WS El-Deiry BID reg-ulation by p53 contributes to chemosensitivity Nat Cell Biol 4 (2002) 842ndash849

[121] MC Moroni ES Hickman E Lazzerini Denchi G Caprara E Colli F Cecconi HMuller K Helin Apaf-1 is a transcriptional target for E2F and p53 Nat Cell Biol3 (2001) 552ndash558

[122] M Muller S Wilder D Bannasch D Israeli K Lehlbach M Li-Weber SLFriedman PR Galle W Stremmel M Oren PH Krammer p53 activates theCD95 (APO-1Fas) gene in response to DNA damage by anticancer drugs JExp Med 188 (1998) 2033ndash2045

[123] X Liu P Yue FR Khuri SY Sun p53 upregulates death receptor 4 expressionthrough an intronic p53 binding site Cancer Res 64 (2004) 5078ndash5083

[124] GS Wu TF Burns ER McDonald III W Jiang R Meng ID Krantz G Kao DDGan JY Zhou R Muschel SR Hamilton NB Spinner S Markowitz G Wu WSel-Deiry KILLERDR5 is a DNA damage-inducible p53-regulated death receptorgene Nat Genet 17 (1997) 141ndash143

[125] JR Jeffers E Parganas Y Lee C Yang J Wang J Brennan KH MacLean J HanT Chittenden JN Ihle PJ McKinnon JL Cleveland GP Zambetti Puma is anessential mediator of p53-dependent and -independent apoptotic pathwaysCancer Cell 4 (2003) 321ndash328

[126] A Villunger EM Michalak L Coultas F Mullauer G Bock MJ AusserlechnerJM Adams A Strasser p53- and drug-induced apoptotic responses mediatedby BH3-only proteins puma and noxa Science 302 (2003) 1036ndash1038

[127] K Kuribayashi N Finnberg JR Jeffers GP Zambetti WS El-Deiry The relativecontribution of pro-apoptotic p53-target genes in the triggering of apoptosis fol-lowing DNA damage in vitro and in vivo Cell Cycle 10 (2011) 2380ndash2389

[128] SP Garrison JR Jeffers C Yang JA Nilsson MA Hall JE Rehg W Yue J Yu LZhang M Onciu JT Sample JL Cleveland GP Zambetti Selection againstPUMA gene expression in Myc-driven B-cell lymphomagenesis Mol Cell Biol28 (2008) 5391ndash5402

[129] S Hopkins-Donaldson A Ziegler S Kurtz C Bigosch D Kandioler C Ludwig UZangemeister-Wittke R Stahel Silencing of death receptor and caspase-8 ex-pression in small cell lung carcinoma cell lines and tumors by DNA methylationCell Death Differ 10 (2003) 356ndash364

[130] R Beroukhim CH Mermel D Porter G Wei S Raychaudhuri J Donovan JBarretina JS Boehm J Dobson M Urashima KT Mc Henry RM PinchbackAH Ligon YJ Cho L Haery H Greulich M Reich W Winckler MSLawrence BA Weir KE Tanaka DY Chiang AJ Bass A Loo C Hoffman JPrensner T Liefeld Q Gao D Yecies S Signoretti E Maher FJ Kaye HSasaki JE Tepper JA Fletcher J Tabernero J Baselga MS Tsao FDemichelis MA Rubin PA Janne MJ Daly C Nucera RL Levine BL EbertS Gabriel AK Rustgi CR Antonescu M Ladanyi A Letai LA Garraway MLoda DG Beer LD True A Okamoto SL Pomeroy S Singer TR Golub ESLander G Getz WR Sellers M Meyerson The landscape of somatic copy-number alteration across human cancers Nature 463 (2010) 899ndash905

[131] R Beckerman AJ Donner M Mattia MJ Peart JL Manley JM Espinosa CPrives A role for Chk1 in blocking transcriptional elongation of p21 RNA duringthe S-phase checkpoint Genes Dev 23 (2009) 1364ndash1377

[132] AJ Donner JM Hoover SA Szostek JM Espinosa Stimulus-specific transcrip-tional regulation within the p53 network Cell Cycle 6 (2007)

[133] AJ Donner S Szostek JM Hoover JM Espinosa CDK8 is a stimulus-specificpositive coregulator of p53 target genes Mol Cell 27 (2007) 121ndash133

[134] JM Espinosa RE Verdun BM Emerson p53 functions through stress- andpromoter-specific recruitment of transcription initiation components beforeand after DNA damage Mol Cell 12 (2003) 1015ndash1027

[135] M Mattia V Gottifredi K McKinney C Prives p53-Dependent p21 mRNA elon-gation is impaired when DNA replication is stalled Mol Cell Biol 27 (2007)1309ndash1320

[136] LD Attardi EE Reczek C Cosmas EG Demicco ME McCurrach SW Lowe TJacks PERP an apoptosis-associated target of p53 is a novel member of thePMP-22gas3 family Genes Dev 14 (2000) 704ndash718

[137] EE Reczek ER Flores AS Tsay LD Attardi T Jacks Multiple response ele-ments and differential p53 binding control Perp expression during apoptosisMol Cancer Res 1 (2003) 1048ndash1057

[138] RA Ihrie E Reczek JS Horner L Khachatrian J Sage T Jacks LD Attardi Perpis a mediator of p53-dependent apoptosis in diverse cell types Curr Biol 13(2003) 1985ndash1990

[139] A Yang R Schweitzer D Sun M Kaghad N Walker RT Bronson C Tabin ASharpe D Caput C Crum F McKeon p63 is essential for regenerative prolifer-ation in limb craniofacial and epithelial development Nature 398 (1999)714ndash718

[140] A Yang M Kaghad Y Wang E Gillett MD Fleming V Dotsch NC Andrews DCaput F McKeon p63 a p53 homolog at 3q27ndash29 encodes multiple productswith transactivating death-inducing and dominant-negative activities MolCell 2 (1998) 305ndash316

[141] M Kaghad H Bonnet A Yang L Creancier JC Biscan A Valent A Minty PChalon JM Lelias X Dumont P Ferrara F McKeon D Caput Monoallelicallyexpressed gene related to p53 at 1p36 a region frequently deleted in neuroblas-toma and other human cancers Cell 90 (1997) 809ndash819


[142] F Murray-Zmijewski DP Lane JC Bourdon p53p63p73 isoforms an orches-tra of isoforms to harmonise cell differentiation and response to stress CellDeath Differ 13 (2006) 962ndash972

[143] CD Thanos JU Bowie p53 family members p63 and p73 are SAM domain-containing proteins Protein Sci 8 (1999) 1708ndash1710

[144] P Ghioni F Bolognese PH Duijf H Van Bokhoven R Mantovani L GuerriniComplex transcriptional effects of p63 isoforms identification of novel activa-tion and repression domains Mol Cell Biol 22 (2002) 8659ndash8668

[145] V Marcel ML Dichtel-Danjoy C Sagne H Hafsi D Ma S Ortiz-Cuaran MOlivier J Hall B Mollereau P Hainaut JC Bourdon Biological functions ofp53 isoforms through evolution lessons from animal and cellular models CellDeath Differ 18 (2011) 1815ndash1824

[146] K Ortt S Sinha Derivation of the consensus DNA-binding sequence for p63 re-veals unique requirements that are distinct from p53 FEBS Lett 580 (2006)4544ndash4550

[147] CA Perez J Ott DJ Mays JA Pietenpol p63 consensus DNA-binding site iden-tification analysis and application into a p63MH algorithm Oncogene 26 (2007)7363ndash7370

[148] EN Kouwenhoven SJ van Heeringen JJ Tena M Oti BE Dutilh ME AlonsoE de la Calle-Mustienes L Smeenk T Rinne L Parsaulian E Bolat R Jurgele-naite MA Huynen A Hoischen JA Veltman HG Brunner T Roscioli EOates M Wilson M Manzanares JL Gomez-Skarmeta HG Stunnenberg MLohrum H van Bokhoven H Zhou Genome-wide profiling of p63 DNA-binding sites identifies an element that regulates gene expression during limbdevelopment in the 7q21 SHFM1 locus PLoS Genet 6 (2010) e1001065

[149] A Yang Z Zhu P Kapranov F McKeon GM Church TR Gingeras K Struhl Re-lationships between p63 binding DNA sequence transcription activity and bio-logical function in human cells Mol Cell 24 (2006) 593ndash602

[150] B Trink M Osada E Ratovitski D Sidransky p63 transcriptional regulation ofepithelial integrity and cancer Cell Cycle 6 (2007) 240ndash245

[151] A Petitjean C Ruptier V Tribollet A Hautefeuille F Chardon C Cavard APuisieux P Hainaut C Caron de Fromentel Properties of the six isoforms ofp63 p53-like regulation in response to genotoxic stress and cross talk with Del-taNp73 Carcinogenesis 29 (2008) 273ndash281

[152] HM Mundt W Stremmel G Melino PH Krammer T Schilling M MullerDominant negative (DeltaN) p63alpha induces drug resistance in hepatocellularcarcinoma by interference with apoptosis signaling pathways Biochem Bio-phys Res Commun 396 (2010) 335ndash341

[153] MD Westfall DJ Mays JC Sniezek JA Pietenpol The Delta Np63 alpha phos-phoprotein binds the p21 and 14-3-3 sigma promoters in vivo and has tran-scriptional repressor activity that is reduced by HayndashWells syndrome-derivedmutations Mol Cell Biol 23 (2003) 2264ndash2276

[154] M LeBoeuf A Terrell S Trivedi S Sinha JA Epstein EN Olson EE MorriseySE Millar Hdac1 and Hdac2 act redundantly to control p63 and p53 functionsin epidermal progenitor cells Dev Cell 19 (2010) 807ndash818

[155] H Lu X Yang P Duggal CT Allen B Yan J Cohen L Nottingham RA RomanoS Sinha KE King WC Weinberg Z Chen C Van Waes TNF-alpha promotes c-RELDeltaNp63alpha interaction and TAp73 dissociation from key genes thatmediate growth arrest and apoptosis in head and neck cancer Cancer Res 71(2011) 6867ndash6877

[156] JW Rocco CO Leong N Kuperwasser MP DeYoung LW Ellisen p63 medi-ates survival in squamous cell carcinoma by suppression of p73-dependent ap-optosis Cancer Cell 9 (2006) 45ndash56

[157] MR Ramsey L He N Forster B Ory LW Ellisen Physical association of HDAC1and HDAC2 with p63 mediates transcriptional repression and tumor mainte-nance in squamous cell carcinoma Cancer Res 71 (2011) 4373ndash4379

[158] T Kurita GR Cunha SJ Robboy AA Mills RT Medina Differential expressionof p63 isoforms in female reproductive organs Mech Dev 122 (2005)1043ndash1055

[159] EK Suh A Yang A Kettenbach C Bamberger AH Michaelis Z Zhu JA ElvinRT Bronson CP Crum F McKeon p63 protects the female germ line duringmeiotic arrest Nature 444 (2006) 624ndash628

[160] X Guo WM Keyes C Papazoglu J Zuber W Li SW Lowe H Vogel AA MillsTAp63 induces senescence and suppresses tumorigenesis in vivo Nat Cell Biol11 (2009) 1451ndash1457

[161] AA Mills B Zheng XJ Wang H Vogel DR Roop A Bradley p63 is a p53 ho-mologue required for limb and epidermal morphogenesis Nature 398 (1999)708ndash713

[162] E Candi A Rufini A Terrinoni D Dinsdale M Ranalli A Paradisi V DeLaurenzi LG Spagnoli MV Catani S Ramadan RA Knight G Melino Differen-tial roles of p63 isoforms in epidermal development selective genetic comple-mentation in p63 null mice Cell Death Differ 13 (2006) 1037ndash1047

[163] Y Yamamura WL Lee MX Goh Y Ito Role of TAp73alpha in induction of ap-optosis by transforming growth factor-beta in gastric cancer cells FEBS Lett 582(2008) 2663ndash2667

[164] Y Shi H Takenobu K Kurata Y Yamaguchi R Yanagisawa M Ohira K KoikeA Nakagawara LL Jiang T Kamijo HDM2 impairs Noxa transcription and af-fects apoptotic cell death in a p53p73-dependent manner in neuroblastomaEur J Cancer 46 (2010) 2324ndash2334

[165] M Sang Y Li T Ozaki S Ono K Ando H Yamamoto T Koda C Geng ANakagawara p73-Dependent induction of 14-3-3sigma increases the chemo-sensitivity of drug-resistant human breast cancers Biochem Biophys Res Com-mun 347 (2006) 327ndash333

[166] GMelino F BernassolaM Ranalli K YeeWX ZongM Corazzari RA Knight DRGreen C Thompson KH Vousden p73 induces apoptosis via PUMA transactiva-tion and Bax mitochondrial translocation J Biol Chem 279 (2004) 8076ndash8083

[167] M Fricker S Papadia GE Hardingham AM Tolkovsky Implication of TAp73 inthe p53-independent pathway of Puma induction and Puma-dependent apopto-sis in primary cortical neurons J Neurochem 114 (2010) 772ndash783

[168] NN Kartasheva C Lenz-Bauer O Hartmann H Schafer M Eilers MDobbelstein DeltaNp73 can modulate the expression of various genes in ap53-independent fashion Oncogene 22 (2003) 8246ndash8254

[169] WM Chan WY Siu A Lau RY Poon How many mutant p53 molecules areneeded to inactivate a tetramer Mol Cell Biol 24 (2004) 3536ndash3551

[170] A Yang N Walker R Bronson M Kaghad M Oosterwegel J Bonnin C VagnerH Bonnet P Dikkes A Sharpe F McKeon D Caput p73-Deficient mice haveneurological pheromonal and inflammatory defects but lack spontaneous tu-mours Nature 404 (2000) 99ndash103

[171] MT Wilhelm A Rufini MK Wetzel K Tsuchihara S Inoue R Tomasini AItie-Youten A Wakeham M Arsenian-Henriksson G Melino DR KaplanFD Miller TW Mak Isoform-specific p73 knockout mice reveal a novel rolefor delta Np73 in the DNA damage response pathway Genes Dev 24 (2010)549ndash560

[172] F Tissir A Ravni Y Achouri D Riethmacher G Meyer AM Goffinet DeltaNp73regulates neuronal survival in vivo Proc Natl Acad Sci U S A 106 (2009)16871ndash16876

[173] AF Lee DK Ho P Zanassi GS Walsh DR Kaplan FD Miller Evidence thatDeltaNp73 promotes neuronal survival by p53-dependent and p53-independent mechanisms J Neurosci 24 (2004) 9174ndash9184

[174] R Tomasini K Tsuchihara M Wilhelm M Fujitani A Rufini CC Cheung FKhan A Itie-Youten A Wakeham MS Tsao JL Iovanna J Squire I JurisicaD Kaplan G Melino A Jurisicova TW Mak TAp73 knockout shows genomicinstability with infertility and tumor suppressor functions Genes Dev 22(2008) 2677ndash2691

[175] M Rossi RI Aqeilan M Neale E Candi P Salomoni RA Knight CM Croce GMelino The E3 ubiquitin ligase Itch controls the protein stability of p63 ProcNatl Acad Sci U S A 103 (2006) 12753ndash12758

[176] M Rossi V De Laurenzi E Munarriz DR Green YC Liu KH Vousden GCesareni G Melino The ubiquitinndashprotein ligase Itch regulates p73 stabilityEMBO J 24 (2005) 836ndash848

[177] E Gallagher M Gao YC Liu M Karin Activation of the E3 ubiquitin ligase Itchthrough a phosphorylation-induced conformational change Proc Natl AcadSci U S A 103 (2006) 1717ndash1722

[178] C Yang W Zhou MS Jeon D Demydenko Y Harada H Zhou YC Liu Negativeregulation of the E3 ubiquitin ligase itch via Fyn-mediated tyrosine phosphory-lation Mol Cell 21 (2006) 135ndash141

[179] A Oberst M Malatesta RI Aqeilan M Rossi P Salomoni R Murillas P SharmaMR Kuehn M Oren CM Croce F Bernassola G Melino The Nedd4-bindingpartner 1 (N4BP1) protein is an inhibitor of the E3 ligase Itch Proc Natl AcadSci U S A 104 (2007) 11280ndash11285

[180] F Galli M Rossi Y DAlessandra M De Simone T Lopardo Y Haupt OAlsheich-Bartok S Anzi E Shaulian V Calabro G La Mantia L GuerriniMDM2 and Fbw7 cooperate to induce p63 protein degradation following DNAdamage and cell differentiation J Cell Sci 123 (2010) 2423ndash2433

[181] BS Sayan AL Yang F Conforti P Tucci MC Piro GJ Browne M Agostini SBernardini RA Knight TW Mak G Melino Differential control of TAp73 andDeltaNp73 protein stability by the ring finger ubiquitin ligase PIR2 Proc NatlAcad Sci U S A 107 (2010) 12877ndash12882

[182] N Wang L Guo BR Rueda JL Tilly Cables1 protects p63 from proteasomaldegradation to ensure deletion of cells after genotoxic stress EMBO Rep 11(2010) 633ndash639

[183] O Hershkovitz Rokah O Shpilberg G Granot NAD(P)H quinone oxidoreduc-tase protects TAp63gamma from proteasomal degradation and regulatesTAp63gamma-dependent growth arrest PLoS One 5 (2010) e11401

[184] F Bernassola P Salomoni A Oberst CJ Di Como M Pagano G Melino PPPandolfi Ubiquitin-dependent degradation of p73 is inhibited by PML J ExpMed 199 (2004) 1545ndash1557

[185] SM Soond SP Barry G Melino RA Knight DS Latchman A Stephanou p73-Mediated transcriptional activity is negatively regulated by polo-like kinase 1Cell Cycle 7 (2008) 1214ndash1223

[186] C Gaiddon M Lokshin I Gross D Levasseur Y Taya JP Loeffler C Prives Cyclin-dependent kinases phosphorylate p73 at threonine 86 in a cell cycle-dependentmanner and negatively regulate p73 J Biol Chem 278 (2003) 27421ndash27431

[187] W Xue L Zender C Miething RA Dickins E Hernando V Krizhanovsky CCordon-Cardo SW Lowe Senescence and tumour clearance is triggered byp53 restoration in murine liver carcinomas Nature 445 (2007) 656ndash660

[188] A Ventura DG Kirsch ME McLaughlin DA Tuveson J Grimm L Lintault JNewman EE Reczek R Weissleder T Jacks Restoration of p53 function leadsto tumour regression in vivo Nature 445 (2007) 661ndash665

[189] N Rivlin R Brosh M Oren V Rotter Mutations in the p53 tumor suppressorgene important milestones at the various steps of tumorigenesis Genes Cancer2 (2011) 466ndash474

[190] CJ Brown S Lain CS Verma AR Fersht DP Lane Awakening guardian angelsdrugging the p53 pathway Nat Rev Cancer 9 (2009) 862ndash873

[191] BA Foster HA Coffey MJ Morin F Rastinejad Pharmacological rescue of mu-tant p53 conformation and function Science 286 (1999) 2507ndash2510

[192] VJ Bykov N Issaeva A Shilov M Hultcrantz E Pugacheva P Chumakov JBergman KG Wiman G Selivanova Restoration of the tumor suppressor functionto mutant p53 by a low-molecular-weight compound Nat Med 8 (2002) 282ndash288

[193] LT Vassilev BT Vu B Graves D Carvajal F Podlaski Z Filipovic N Kong UKammlott C Lukacs C Klein N Fotouhi EA Liu In vivo activation of the p53pathway by small-molecule antagonists of MDM2 Science 303 (2004) 844ndash848


[194] S Shangary SWang Small-molecule inhibitors of theMDM2-p53 proteinndashproteininteraction to reactivate p53 function a novel approach for cancer therapy AnnuRev Pharmacol Toxicol 49 (2009) 223ndash241

[195] N Issaeva P Bozko M Enge M Protopopova LG Verhoef M Masucci APramanik G Selivanova Small molecule RITA binds to p53 blocks p53-HDM-2interaction and activates p53 function in tumors Nat Med 10 (2004)1321ndash1328

[196] C Tovar J Rosinski Z Filipovic B Higgins K Kolinsky H Hilton X Zhao BTVu W Qing K Packman O Myklebost DC Heimbrook LT Vassilev Small-molecule MDM2 antagonists reveal aberrant p53 signaling in cancer implica-tions for therapy Proc Natl Acad Sci U S A 103 (2006) 1888ndash1893

[197] NP Gomes JM Espinosa Disparate chromatin landscapes and kinetics of inac-tivation impact differential regulation of p53 target genes Cell Cycle 9 (2010)3428ndash3437

[198] RE Henry Z Andrysik R Paris MD Galbraith JM Espinosa A DR4tBID axisdrives the p53 apoptotic response by promoting oligomerization of poisedBAX EMBO J (2012) doi 101038emboj2011498

[199] C Cao ET Shinohara TK Subhawong L Geng K Woon Kim JM Albert DEHallahan B Lu Radiosensitization of lung cancer by nutlin an inhibitor of mu-rine double minute 2 Mol Cancer Ther 5 (2006) 411ndash417

[200] L Coll-Mulet D Iglesias-Serret AF Santidrian AM Cosialls M de Frias E Castano CCampas M Barragan AF de Sevilla A Domingo LT Vassilev G Pons J Gil MDM2antagonists activate p53 and synergize with genotoxic drugs in B-cell chronic lym-phocytic leukemia cells Blood 107 (2006) 4109ndash4114

[201] T Zheng J Wang X Song X Meng S Pan H Jiang L Liu Nutlin-3 cooperates withdoxorubicin to induce apoptosis of human hepatocellular carcinoma cells throughp53 or p73 signaling pathways J Cancer Res Clin Oncol 136 (2010) 1597ndash1604

[202] D Carvajal C Tovar H Yang BT Vu DC Heimbrook LT Vassilev Activation ofp53 by MDM2 antagonists can protect proliferating cells from mitotic inhibitorsCancer Res 65 (2005) 1918ndash1924

[203] SV Tokalov ND Abolmaali Protection of p53 wild type cells from taxol bynutlin-3 in the combined lung cancer treatment BMC Cancer 10 (2010) 57

[204] J Ribas J Boix L Meijer (R)-roscovitine (CYC202 Seliciclib) sensitizes SH-SY5Yneuroblastoma cells to nutlin-3-induced apoptosis Exp Cell Res 312 (2006)2394ndash2400

[205] CF Cheok A Dey DP Lane Cyclin-dependent kinase inhibitors sensitize tumorcells to nutlin-induced apoptosis a potent drug combination Mol Cancer Res 5(2007) 1133ndash1145

[206] K Kojima M Konopleva IJ Samudio V Ruvolo M Andreeff Mitogen-activatedprotein kinase kinase inhibition enhances nuclear proapoptotic function of p53in acute myelogenous leukemia cells Cancer Res 67 (2007) 3210ndash3219

[207] K Kojima M Konopleva IJ Samudio WD Schober WG Bornmann MAndreeff Concomitant inhibition of MDM2 and Bcl-2 protein function synergis-tically induce mitochondrial apoptosis in AML Cell Cycle 5 (2006) 2778ndash2786

[208] M Wade LW Rodewald JM Espinosa GM Wahl BH3 activation blocks Hdmxsuppression of apoptosis and cooperates with Nutlin to induce cell death CellCycle 7 (2008) 1973ndash1982

[209] AV Vaseva AR Yallowitz ND Marchenko S Xu UM Moll Blockade of Hsp90by 17AAG antagonizes MDMX and synergizes with Nutlin to induce p53-mediated apoptosis in solid tumors Cell Death Dis 2 (2011) e156

[210] T Hori T Kondo M Kanamori Y Tabuchi R Ogawa QL Zhao K Ahmed TYasuda S Seki K Suzuki T Kimura Nutlin-3 enhances tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis through up-regulation of death receptor 5 (DR5) in human sarcoma HOS cells and humancolon cancer HCT116 cells Cancer Lett 287 (2010) 98ndash108

[211] HY Tseng CC Jiang A Croft KH Tay RF Thorne F Yang H Liu P Hersey XDZhang Contrasting effects of nutlin-3 on TRAIL- and docetaxel-induced apopto-sis due to upregulation of TRAIL-R2 and Mcl-1 in human melanoma cells MolCancer Ther 9 (2010) 3363ndash3374

[212] AWWhitehurst BO Bodemann J Cardenas D Ferguson L Girard M PeytonJD Minna C Michnoff W Hao MG Roth XJ Xie MA White Synthetic lethalscreen identification of chemosensitizer loci in cancer cells Nature 446 (2007)815ndash819

[213] MA Gregory TL Phang P Neviani F Alvarez-Calderon CA Eide T OHare VZaberezhnyy RT Williams BJ Druker D Perrotti J Degregori WntCa2+NFATsignaling maintains survival of Ph+ leukemia cells upon inhibition of Bcr-AblCancer Cell 18 (2010) 74ndash87

[214] M Xia D Knezevic LT Vassilev p21 does not protect cancer cells from apopto-sis induced by nongenotoxic p53 activation Oncogene 30 (2011) 346ndash355

[215] R Bernards Its diagnostics stupid Cell 141 (2010) 13ndash17

Fig 1 The p53 circuit board Activation of p53 by diverse stimuli such as oncogene hyperactivation DNA damage and nutrient deprivation results in increased expression of nu-merous genes controlling different cellular outcomes such as cell cycle arrest senescence autophagy and apoptosis


on p53 was characterized as an oncogene a confounding issue de-rived from the study of the mutant p53 variants expressed in tumors[4ndash9] Years later the wild type version of the p53 gene was unequiv-ocally identified as a tumor suppressor [10ndash12] Eventually the p53protein was characterized as a DNA-binding transcription factor[13ndash17] With the discovery of MDM2 a repressor of p53 [18ndash20]and p21 (CDKN1A) a p53 target gene and effector [2122] a p53pathway was born With the identification of numerous upstreamregulators and hundreds of target genes with different functions[23ndash26] one pathway became many (Fig 1) Today the notion of ap53 pathway is obsolete p53 is a gene network An enormousamount of knowledge has been generated while studying the p53gene the p53 protein and the various p53 pathways What can welearn from studying p53 as a gene network

In this review we will abandon all p53-centric views and considerp53 as mere ldquosignal generatorrdquo within a vast gene network We willdeliberately neglect all information about the multiple events that af-fect the p53 molecule itself such as dozens of post-translational mod-ifications and hundreds of binding partners that regulate its activityWe refer those readers interested in the p53 molecule to many excel-lent p53-centered reviews [27ndash31] Instead in the next pages we willassume for the sake of argument that the p53 protein is one and thesame in every context and that its activity as a DNA binding proteinand transcriptional regulator is invariant across scenarios Havingblinded ourselves to the outstanding complexity of the molecularevents surrounding p53 activation we will focus our attention on

Fig 2 The p21 circuit The p53 target gene p21 is co-regulated by various factors acting atpotentially affect p53p21-dependent cell cycle arrest In both A and B from top to bottom trtranscriptional regulators into a single output functional mRNA In this context p53 is justgrator at this step which synthesizes p21 mRNA the input for the next level of regulation Frtranslation are combined via the ribosome signal integrator to achieve protein synthesis Finbilizing signals resolved by the proteasome In A the regulatory layers are displayed as carelectrical circuit comprising three signal integrators At the transcriptional level activatingintegrator (+) and are summed internally by an adding network (dashed blue box) inhibittivating and total inhibitory signals determines output amplitude in this case p21 mRNA levery in this example p21 mRNA levels become positive inputs for p21 protein production i

the downstream networks of genes that modulate p53-derived sig-nals Throughout this reviewwe will employ an electrical engineeringmetaphor and it is important for us to define at this point howwe willuse the terms circuit integrated circuit and circuit board

In our metaphor p53 is a switch turned to the lsquoONrsquo position by p53activating stimuli thus transmitting signals to downstream circuitryWe define a circuit as the collection of factors co-regulating a givenp53 target gene at the transcriptional post-transcriptional and post-translational levels In Section 2 we illustrate this for p21 a directtranscriptional target of p53 that mediates cell cycle arrest (Fig 2)[212232] An integrated circuit is defined as a group of p53 targetgenes that work together to drive a given cellular response For exam-ple the p21 and 14-3-3σ (SFN) circuits are part of an integrated cir-cuit that mediates p53-dependent cell cycle arrest whereas PUMA(p53-upregulated modulator of apoptosis BBC3) and NOXA (PMA-induced protein 1 PMAIP1) are two among many circuits mediatingp53-dependent apoptosis Finally a circuit board represents the as-sembly of several diverse integrated circuits which ultimately coa-lesces all of the various signals into a single cellular outcome(Fig 3) In Section 3 we assemble a minimal circuit board composedof p21 14-3-3σ PUMA and NOXA which includes their variousgene-specific co-regulators acting at the transcriptional post-transcriptional and post-translational levels In Section 4 we de-scribe how p53 circuit boards can be assembled into cell type- andstimulus-specific configurations that define context-dependent p53responses We discuss at this point the impact of the p53 family

the transcriptional post-transcriptional and post-translational levels all of which cananscriptional control of p21 gene expression merges the action of positive and negativeone among many regulators feeding into RNA polymerase II (RNAPII) the signal inte-om there signals from factors that positively and negatively influence RNA stability andally the levels of active p21 protein in the cell are fine-tuned by stabilizing and desta-toons of DNA RNA and protein molecules In B the same processes are depicted as ansignals including p53 (green) are applied to the positive terminals of the RNAPII signalory signals (red negative terminals) are combined similarly The difference in total ac-el Integrator output can subsequently serve as input to other signal processing machin-n the ribosome















































































6 Final remarks



References











































































































































































































































































































6 Final remarks



References










































































































































































































































































































6 Final remarks



References
































































































































































































































































































6 Final remarks



References




















































































































































































































































































6 Final remarks



References







































































































































































































































































6 Final remarks



References

























































































































































































































































6 Final remarks



References


















































































































































































































































6 Final remarks



References










































































































































































































































6 Final remarks



References

































































































































































































































6 Final remarks



References































































































































































































































6 Final remarks



References
























































































































































































































































































































































































































































































































































































































































Date post:	09-Dec-2023
Category:	Documents
Upload:	colorado
View:	0 times
Download:	0 times

The p53 circuit board

Documents