Prion-like aggregation of mutant p53 in cancer

Opinion

Prion-like aggregation of mutant p53 in cancer Jerson L. Silva1,2, Claudia V. De Moura Gallo2,3, Danielly C.F. Costa1,2, and Luciana P. Rangel1,2,4 1

Instituto de Bioquı´mica Me´dica Leopoldo de Meis, Universidade Federal do Rio de Janeiro, Rio de Janeiro 21941-902, Brazil Instituto Nacional de Cieˆncia e Tecnologia (INCT) de Biologia Estrutural e Bioimagem, Universidade Federal do Rio de Janeiro, Rio de Janeiro 21941-902, Brazil 3 Departamento de Gene´tica, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil 4 Faculdade de Farmaćia, Universidade Federal do Rio de Janeiro, Rio de Janeiro 21941-902, Brazil 2

p53 is a master regulatory protein that participates in cellular processes such as apoptosis, DNA repair, and cell cycle control. p53 functions as a homotetrameric tumor suppressor, and is lost in more than 50% of human cancers. Recent studies have suggested that the formation of mutant p53 aggregates is associated with loss-offunction (LoF), dominant-negative (DN), and gain-offunction (GoF) effects. We propose that these phenomena can be explained by a prion-like behavior of mutant p53. We discuss the shared properties of cancer and neurodegenerative diseases and how the prion-like properties of p53 aggregates offer potential targets for drug development. Structure and function of p53, the guardian of the genome Elucidation of the molecular mechanisms of human diseases has revealed surprising connections, such as similarities between cancer and neurodegenerative disease. For some time, we have known that missense mutations in TP53 resulting in amino acid substitutions in the tumor suppressor protein p53 play a role in the pathogenesis and prognosis of more than 50% of malignant tumors. The importance of the tumor suppressor activity of wild type p53 is so great that it has been referred to as the ‘guardian of the genome’ [1]. It is a homotetrameric protein with a structural organization characteristic of transcription factors, including (i) an acidic N terminus that contains the transcriptional activation domains TAD1 (residues 1–42) and TAD2 (residues 43–63) as well as a proline-rich domain (PoliP; residues 64–92); (ii) a central, sequence-specific, well-conserved DNA-binding domain (DBD) (residues 102–306); and (iii) a C-terminal region that contains a nuclear localization signaling domain (NLS), an oligomerization domain (OD), and a negative regulation domain (Neg; residues 307–393) (Figure 1) [2–4]. The flexible nature of p53 allows it to interact with other proteins to exert its biological activities [4–6]. As a master regulatory protein, p53 controls different cellular processes in Corresponding author: Silva, J.L. ([email protected], [email protected]). Keywords: p53; protein aggregation; cancer; prion; amyloid; prion-like proteins. 0968-0004/ ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tibs.2014.04.001

260

Trends in Biochemical Sciences, June 2014, Vol. 39, No. 6

response to stress, such as apoptosis, DNA repair, cell cycle arrest, senescence, and metabolism. The complete genome sequencing of malignant tumors by research programs such as the Pan-Cancer Initiative of The Cancer Genome Atlas (TCGA) has confirmed that TP53 is the most frequently mutated gene in cancer cells [7]. In contrast to other tumor suppressor genes, TP53 is susceptible to a large spectrum of point mutations that confer structural diversity to the mutant protein. Interestingly, some amino acid residues, known as hot spots, are more frequently affected by mutational events than others, producing anomalous proteins that can have LoF, DN, and GoF effects [2,8,9] (Figure 1). Substitutions in p53 that have been frequently identified in human cancers include: R175H, which causes a major change to the structure of p53; and R248Q, R273H, R248W, R273C, and R282W, which cause changes to the p53–DNA contact sites [10]. According to the International Agency for Research on Cancer (IARC) TP53 Database (release 17, November 2013), more than 28 000 somatic mutations in p53 have been identified in human cancers, including more than 2300 p53 mutants with functional properties. Several proteins interact with the different domains of wild type (WT) p53 and p53 variants, including its paralogs p63 and p73 [2]. The altered functions of p53 variants may be related to the induction of ‘new’ target genes, in addition to the association of p53 with different types of proteins, potentially increasing carcinogenesis and tumor aggressiveness [11]. For example, the mutant R175H was one of the first reported examples of GoF activity, leading to activation of the multidrug resistance gene 1 (MDR1) promoter, which is not a target of WT p53 [12]. These interactions of mutant p53 result in malignant and more aggressive phenotypes, with clinicopathological consequences. Here, we discuss the evidence for prion-like and aggregation properties of p53, which appear to explain, at the molecular level, LoF, DN, and GoF effects in tumor pathogenesis. There is strong proof for the intracellular aggregation of mutant p53 in tumors. We argue that the accumulation of these aggregates (oligomers and fibrils) not only convert the wild type p53 into the misfolded conformation but also other tumor suppressors. The possibility that this prion-like conversion is further transmitted to

Opinion

Trends in Biochemical Sciences June 2014, Vol. 39, No. 6

248

Hotspot mutaons relave frequency and localizaon

273

175 245 249

282

N

C 1

TAD1

TAD2

PoliP

DBD

NLS

OD

Neg

393 Ti BS

Figure 1. The domain structure of the human p53 protein. From the amino-terminal to carboxy-terminal regions: transactivation domain 1 (TAD1); transactivation domain 2 (TAD2); proline-rich domain (PoliP); DNA-binding domain (DBD); nuclear localization signaling domain (NLS); oligomerization domain (OD); and negative-regulation domain (Neg). Adapted from Surget et al. [3]. The bars above the p53 diagram indicate the relative frequencies of missense mutations at the residues based on version R17 (November 2013) of the International Agency for Research on Cancer (IARC) tumor suppressor protein p53 (TP53) Database (http://www-p53.iarc.fr/) [10]. The sites of hot spot mutations are indicated by the corresponding residue numbers.

other cells, through the aggregates, is an open question that is also debated below. p53 aggregation: does it have pathological implications? The achievement of a functional structure by a protein is driven by thermodynamics and kinetics. For some proteins, particularly those with low molecular weights, the state of lower free energy is reached rapidly and without requiring cellular assistance. However, most proteins require chaperones to ensure that the protein is present in the correct folding energy landscape at the proper time and place. The competing misfolding landscape results in a more rugged and complex structure, which leads to multiple protein states, including oligomeric aggregates, amyloid protofibrils, and fibrils (Figure 2). According to some authors, under certain conditions, most proteins will assume the deformed amyloid state [13]. In the past 20 years, we have learned that many neurodegenerative diseases are caused by the misfolding and aggregation of proteins and peptides, including Alzheimer’s disease, Parkinson’s disease, amyotrophic lateral sclerosis (ALS), familial amyloid polyneuropathy, and prion diseases, among others (Box 1). Most of these diseases are favored by advanced age, which is associated with the failure of some of the protein homeostasis systems, such as chaperones and protein degradation controls. The pathological state is the result of one or a combination of the following events: a lack of the functional protein; a toxic GoF effect of the aggregates; toxic GoF effects of stable intermediates in the aggregation pathway, usually referred to as ‘oligomers’; and GoF due to co-aggregation with other cellular proteins. Among the diseases caused by aggregation are transmissible spongiform encephalopathies (TSEs), which are related to mammalian prion proteins (PrPs) and have a crucial characteristic: the ability to transmit the phenotype of the misfolded state of a PrP. Although prion diseases are rare, they share several features that are typical of many neurodegenerative diseases [14]. Recent studies have provided evidence that the prion concept is more general and may encompass several neurodegenerative diseases [15–18] (Box 1). Key proteins involved in these diseases, such as amyloid-beta, tau, alpha-synuclein, superoxide dismutase 1 (SOD1), and TAR DNA-binding

protein 43 (TDP43), may act as prions, also referred to as prionoids [19], as demonstrated by their transmissibility in both animals and mammalian cell cultures [15–18]. Prionoid-related degenerative diseases are all based on the prion-like conversion of a correctly folded protein into its misfolded version, resulting in a toxic GoF that leads to cell death. Remarkably, malignant tumors featuring mutations in the p53 tumor suppressor protein may share a mechanism of propagation with neurodegenerative diseases [20,21], even given the proliferative nature of cancer. The aggregation of p53 into different types of structures, including amyloid oligomers and fibrils, has been demonstrated [20–26]. For instance, amyloid aggregates of mutant p53 have been detected in breast cancer tissues [20,27] and malignant skin tumors [28]. Although the three functional domains of p53 each have the potential to form amyloidlike aggregates [23–25], the part of the protein with the highest propensity to form amyloid oligomers and fibrils is the DNA-binding domain, p53C, where most p53 mutations related to cancer development are found [8]. It has been demonstrated that, depending on the type of p53 mutation in breast cancer biopsies, there were less or more amyloid oligomers [27]; a higher degree of aggregation seemed to correlate with more invasive tumors. Also, p53 mutations were associated with a high level of p53 immunostaining in aggregates containing both mutant and WT p53 in prostate cancer [29]. Mutant p53 may exert a DN regulatory effect on WT p53 in which amyloid-like mutant p53 converts WT p53 into an aggregated species, resulting in the complete loss of its tumor suppressor function (see next section); thus, hot spot mutations can make p53 prion-like [20]. In addition to inhibiting WT p53 function through aggregation, mutant p53 might co-aggregate important proteins in the cellular network, particularly transcription and translation factors, similar to artificial b-sheet proteins that were designed to form amyloid oligomers and fibrils in a recent study [30]. Some of the factors affected by the artificial b-sheet proteins also appear to be affected when different mutant p53 proteins are present [31]. Olzscha et al. have suggested that the mechanism might be broadly conserved among amyloids; that the main target of amyloid aggregation is a metastable subproteome [30]. As will be discussed in the next section, p53 has an intrinsic tendency to form 261

Opinion


Misfolded p53

Gibbs free energy

Oligomers

Nave p53 Fibrils

Folding landscape Misfolding/aggregaon landscape

Ti BS

Figure 2. Folding and misfolding landscapes of p53 aggregation. Inter- and intramolecular contacts promote the formation of funneling conformations, leading either to the properly folded, native state or to the misfolded, aggregated forms of p53, which include oligomers and amyloid structures. Adapted from Silva et al. [21].

protein assemblies, most likely because it has evolved to make protein–protein interactions. Effects of DN p53 mutations on oligomerization and fibrillar formation A protein misfolding disease (PMD) can also be caused or accentuated when a mutation leads to protein misfolding, inducing the loss of WT function [32]. One example of this phenomenon is epidermolysis bullosa simplex, which is caused by mutations in genes encoding some keratin proteins [32–35]. However, single amino acid substitutions in the core domain of p53 have long been known to produce DN, in addition to LoF effects [36]. As discussed above, somatic mutations in p53 are the most important of the common genetic changes in cancer. p53 ordinarily forms tetramers, therefore one obvious mechanism for the DN effects of p53 mutants is the neutralization of the functional state of the WT form by heterotetramerization. In fact, the formation of heterotetramers (mutant and WT p53) has been considered the simplest explanation of the DN effect [36,37]. However, it is unlikely that heterotetramerization can account alone for the DN effect because the dissociation of tetramers was shown to be very slow [38]. An alternative mechanism is the formation of higher order aggregates containing mutant and WT p53. Several years ago, it was reported that the p53 core domain exhibited a propensity to undergo amyloid aggregation under certain conditions [23]; from these and later data, it was proposed that the decreased stability and increased propensity of mutant p53 to aggregate might contribute to its DN effect [21,23,39]. The electron microscopy analysis of in vitro aggregation of p53C reveals a mixture of oligomers 262

and fibrils [20,23]. The propensity of other p53 domains to aggregate [24,25] further suggests that single amino acid substitutions in p53 can lead to the misfolding of the WT conformer. In silico analyses of aggregation-prone sequences of p53 have revealed that p53 contains a large number of hot spots for aggregation, similar to mammalian prion proteins and other proteins involved in prion-like diseases [40]. The paralogous proteins p63 and p73 also exhibit a high propensity to form aggregates [40,41]. As suggested by the pioneering study by Milner and Medcalf [36], the dominant-negative effect of p53 may be reminiscent of a prion. The prion-like behavior of mutant p53 is supported by experiments showing that a seed of a mixture of R248Q p53C amyloid oligomers and fibrils accelerated the aggregation of WT p53C [20]. Whereas oligomers are formed by a limited number of protein units and more difficult to characterize, the fibrils are orderly arranged cross b-sheet structures. More recently, it was clearly shown that aggregates of recombinant p53C are internalized by cells and co-aggregate with endogenous p53 protein [42], demonstrating the prion-like conversion of cellular p53 into an aggregated conformation induced by misfolded p53C taken by the cell. However, the ability of a cell to release mutant p53 aggregates in the extracellular milieu and to spread the aggregation phenomenon to other cells has yet to be proven. p53 aggregates can be toxic to particular cell lines, which could support a mechanism in which a misfolded conformation is spread from one cell to another [20,23,28]. One might argue that if the aggregate is toxic it will kill the cell, which would compromise rather than support the growth of malignant tumors. However, this toxicity could

Opinion


Box 1. Amyloids and prion-like diseases In the early 1980s, Stanley Prusiner developed the prion concept to explain the nature of the etiologic agent of TSEs [66]. TSEs are classified as progressive neurodegenerative diseases that are inevitably fatal. Their agent of transmission has long been debated because no coding DNA could be detected. Prusiner proposed that the etiologic agent might be a protein in a misfolded conformation that was able to promote the conformational change of a native protein into a misfolded one [66,67]. Through this mechanism, the disease would be able to self-propagate and infect other individuals, even without a classical transmission agent. Several different origins have been described for TSEs, including genetic mutations, which lead to a less stable protein that is more prone to misfolding; the ingestion of contaminated beef; and sporadic forms of unidentified origin. The presence of a cofactor that could lower the energy barrier between the folded and misfolded conformations has been consid-

(A)

ered [55–57,63,68]. The proposed mechanisms of prion conversion to the misfolded, pathogenic conformation are described in Figure I. Recent developments in this field suggest that other PMDs, such as ALS, Huntington’s, Parkinson’s, and Alzheimer’s diseases, might also be due to prion-like isoforms [15,16,69]. The self-propagation of amyloids of the different causative proteins of these diseases could induce the misfolding of their native counterparts in cells, tissues, and animal models [15,16,70,71]. Although these pathogenic protein forms are self-propagating within different cells and tissues of an individual and even within individuals, their infectivity is yet to be confirmed because their natural acquisition from the environment or an infected host has not been described. In addition, reports of coaggregation between different proteins involved in neurodegenerative diseases (such as beta-amyloid, alpha-synuclein, and tau) [30,72– 74] have revealed that this complex story is far from complete.

Misfolded protein Cofactor

Nave protein

(B)

Fragmentaon/ seed formaon

Misfolded protein template

Smaller aggregates/ oligomers

Nave protein Nave protein

Larger aggregates

Oligomers causing neurodegeneraon Ti BS

Figure I. Mechanisms for prion-like effects. (A) A misfolded form of a protein might originate with the help of a cofactor, which would be able to lower the energetic barrier between the two conformations (native and misfolded). (B) The misfolded protein is able to act as a template for the conversion of other native protein units. This leads to the formation of small aggregates, which are fragmented into smaller units that are thought to be the causative agents of neurodegeneration and may act as seed to form larger aggregates and amyloid fibers.

also lead to the release of the misfolded protein, allowing it to interact with other cells and rapidly propagate to other cells in the tissue (this is schematically shown in Figure 3). Although this hypothesis has not been confirmed, spreading of the prion conformation of p53 to distant locations would have alarming implications by making it a true prion. Do GoF p53 mutations cause p53 to form amyloid oligomers with other proteins? In addition to providing an explanation for the DN effects of oncogenic p53 mutants, prion-like behavior may also explain the GoF properties of several of the mutations,

which usually result in increased metastatic potential [8]. Mutant p53 aggregation not only induces WT p53 aggregation [20] but also the aggregation of its paralogs p63 and p73 [41]. In addition, there is evidence that aggregation may occur in the presence of the chaperone heat shock protein 70 (Hsp70) [43] and acetyltransferase p300 [44]. In the past decade, many studies have demonstrated that p53 mutations not only lead to the loss of WT p53 function but also elicit a myriad of effects. Several p53 mutations produce oncogenic properties, leading to tissue invasion, rapid proliferation, and metastasis [8,31]. The R248Q mutant, which has a higher propensity to aggregate 263

Opinion


Mutant p53 cancer cell

WT p53 normal cell

(B) Aggregaon–prone mutant p53 – DN, LOF

(A) Funconal p53 at the nucleolus

(D) Aggregate/oligomer is toxic to cells

? (C) GOF of oligomers

(E) Spread to other cells

(F) cancer progression Ti BS

Figure 3. The role of mutant p53 aggregation on cancer proliferation. (A) Normal cells contain wild type (WT) p53, which is functional at the nucleolus, able to control the cell cycle, and preserve cell integrity. (B) When a cell expresses an aggregation-prone mutant p53, due to genomic or cancer cell-specific mutation events, p53 is inactivated. This leads to loss-of-function (LoF) effects, because p53 is no longer active in the nucleolus. p53 aggregation might lead to two novel situations: (C) a differential activity of oligomeric forms of mutant p53, which might interact with different proteins such as p63 and p73 [41,75–77], or interact in nonphysiological ways with WT p53 binding partners such as mouse double minute 2 homolog (MDM2) and heat shock protein 70 (HSP70) [43]. These events might lead to the interaction with new binding sites in DNA and explain gain-of-function (GOF) effects, which increase cancer aggressiveness and progression (F). (D) Aggregates are toxic to cells [20,28]. This might lead to cell death and disruption, with the release of aggregates and oligomers in the extracellular medium (E). p53 aggregates have been shown to be captured by cells through macropinocytosis [42]. Consequently, the spread of the prion-like phenomenon to neighboring cells might also lead to cancer progression.

and can seed the aggregation of WT p53, was recently shown to lead to more aggressive tumors in mice compared with another hot spot mutant, G245S, and also p53-null alleles [45,46]. Interestingly, both mutants present GoF activities [45]. The authors also correlated these data with Li-Fraumeni syndrome carriers bearing R248 missense mutations, who present with earlier tumor onset and higher tumor numbers per at-risk person than Li-Fraumeni carriers with G245 missense and null mutations [45]. Interestingly, Xu and coworkers also found that the R248Q mutant was prone to co-aggregation with p63 and p73 [41]. As noted by Muller and Vousden in a recent review [8], the mechanism of GoF of p53 mutants is far from being completely elucidated. They proposed that at least four different types of mechanisms underlie the GoF activity of p53 mutants. Combinations of two or more of these mechanisms would most likely be required to produce GoF. Among the four types of mechanisms, one of the most important may be the prion-like aggregation of mutant p53 with WT p53, p63, p73, and other transcription factors. The co-localization of p53 with small amyloid oligomers in breast cancer tissues [20,27] and in basal cell carcinomas 264

[28] supports the co-aggregation hypothesis for the GoF of some p53 mutants. Interestingly, biopsies of breast cancer patients harboring the R175H substitution contained few or no aggregates [27], most likely because this mutant has much lower stability than WT p53 and other p53 mutants, thereby limiting the formation of stable oligomers. It is not clear why a complex mixture of amyloid oligomers and fibrils was formed in the in vitro experiments with mutant p53, whereas only oligomers were detected in breast cancer cells [20,27]. In the case of neurodegenerative diseases, there is strong evidence that the effects of oligomers on cellular homeostasis are greater than those of fibrils [13]. One might assume that the toxicity of these oligomers is usually limited because the cell has sufficient time to develop a mechanism to counterbalance their effects. However, with time and age, it appears that the oligomers prevail due to their ability to cause cellular death and to spread to other cells in a prion-like fashion. Whereas these mechanisms have strong experimental evidence in some neurodegenerative diseases, further data will be needed to test whether these oligomers can pass from cell to cell and carry the prion-like phenotype in cancer.

Opinion In malignant tumors, oligomers of mutant p53 might interact reversibly with WT p53, p63, p73, and other proteins, similar to the artificial b-sheets transfected into cells by Olzscha et al. [30]. Hetero-oligomers of mutant p53 with other proteins provide a challenging theme for structural biology and cell biology, particularly by correlating a particular structure with an observed GoF. Two recent fascinating studies have shown how prion-like polymerization of the adaptor protein ASC (apoptosis-associated speck-like protein containing a C-terminal caspaserecruitment domain) participates in inflammasome assembly signaled by microbes and leading to the release of inflammatory cytokines [47,48]. Likewise, the formation of aggregated structures of p53 with other cellular transcription factors may act as a malignant signal. For example, overwhelming GoF effects result from the interaction of some mutants of p53 with p63 [49,50], which leads to either the increased expression of genes targeted by p63 or the binding of the p53–p63 complex to uncommon DNA sequences. We do not know the structural details of these interactions. Recently, Kirilyuk et al. have shed some light on the structure and site of co-aggregation of p53 with other proteins [44]. They determined that the prion-like domain (PSPD) of the acetyltransferase p300 directed p53 localization into aggregates containing ubiquitin and the 20S subunit of the proteasome [44]. Again, only by unveiling these structures will we be able to understand the modified functions of p53 in cancerous cells. Prion-like transmission of phenotypes to other cells The p53 core domain has a greater tendency to form aggregates than the globular domain of the prion protein PrP [40]. Milner and Medcalf [36] demonstrated that the co-translation of mutant and WT p53 caused WT p53 to assume the mutant conformation. Roucou’s group recently demonstrated in vitro transmissibility of p53 [42]. They found that WT p53 aggregates could penetrate HeLa and NIH3T3 cells via macropinocytosis and induce the aggregation of intracellular p53. According to our proposed prion hypothesis of the DN and GoF effects of mutant p53, the aggregates induce toxic effects in some cells, which results in the release of aggregates to the extracellular environment. Aggregates can thereby be transmitted to contiguous cells, which take up these oligomers, resulting in immortalization in some cells and toxic effects in others (Figure 3). There is mounting evidence of the prion-like transmission of a misfolded phenotype in several neurodegenerative diseases; the prion-like protein could be a new therapeutic target [51]. Although the evidence of intercellular prion transmission in cancer is not as substantive as in the neurodegenerative diseases, there is a possibility of targeting the same mechanisms of intracellular action of mutant p53. In the case of Parkinson’s and other Lewy body diseases, experimental data have revealed that a-synuclein is released from neuronal cells by exocytosis, contributing to the pathological features of these diseases [52]. However, there have been few studies of the secretion of p53 and its absorption from the extracellular environment. In some K-Ras-mutated tumors, p53 is secreted via exocytosis with Snail [53]; this


is followed by endocytosis by neighboring cells, although part of p53 is digested by extracellular proteases. K-Rasmutated cells can take up p53 through caveolin-1-mediated endocytosis [54]. The susceptibility of soluble p53 aggregates to proteases would limit their cell-to-cell transmissibility; thus, the transmission of soluble and amyloid aggregates of p53 requires additional investigation using in vivo and in vitro models. Another relevant question is related to the actions of cellular cofactors and crowding (Box 1). Prion diseases appear to require a cofactor for pathogenicity [55–57]. Thus, similar to PrP aggregation, lipids and polyanions may be involved in p53 aggregation. Recently, Nieva and coworkers demonstrated that cholesterol secosterol aldehydes, which are associated with chronic inflammation, induce p53 amyloid aggregation [58]. Concluding remarks and future perspectives Prions are the most challenging entities in biology and medicine. It appears that the mechanism in which corrupted proteins initiate a sequence of reactions that leads to aggregation, which is typical of prions, is involved not only in neurodegenerative diseases and classical amyloidosis but also in tumors bearing p53 mutations. A prion-like seeding mechanism would provide a mechanistic explanation for both the DN and GoF effects of p53 mutations. Moreover, the co-aggregation of mutant p53 with other tumor suppressors, such as p63 and p73 [41], could occur in a cross-seeding fashion, similar to the cross-seed fibrillization of tau protein induced by aggregates of a-synuclein [59], especially because both p63 and p73 have a high propensity to form amyloid aggregates [40]. As it was also found in the case of sequestration of metastable proteins by artificial b-sheet proteins [30], other proteins can be incorporated into aggregates of mutant p53, without necessarily participating in the beta-sheet structures inside the polymer. The inhibition of cell proliferation due to the DN or GoF effects of the prion-like behavior of mutant p53 appears to be a novel therapeutic target. Researchers have sought compounds with the ability to inhibit p53 aggregation by binding to the cavity that results from the Y220C substitution [26,60]. However, because accumulation and aggregation of mutant p53 depend on both the mutation and the cellular environment, each case must be cautiously investigated. For example, in the case of Y220C substitution, both the apo-protein (without Zn) and the aggregated holoY220C caused acceleration in the aggregation of the holoprotein, implying co-aggregation [61]. Another potential strategy is the use of nucleic acid aptamers and glycosaminoglycans to prevent aggregation and prion-like conversion in p53-mutant cancers [9,39], because similar approaches have shown positive results in TSEs and other neurodegenerative diseases [62–65]. The prion-like properties of nucleation, templating, multiplication, and spreading should be considered as promising targets for the development of new anticancer therapies. Acknowledgments We thank Martha M. Sorenson for carefully reading the manuscript and providing helpful suggestions. Our laboratories were supported by grants 265

Opinion from Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq awards and INCT Program), Fundaçaõ Carlos Chagas Filho de Amparo a` Pesquisa do Estado do Rio de Janeiro (FAPERJ), Ministerio da Saude (Decit Program) and Coordenaçaõ de Aperfeiçoamento de Pessoal de Nı´vel Superior (CAPES).

References 1 Lane, D.P. and Crawford, L.V. (1979) T antigen is bound to a host protein in SY40-transformed cells. Nature 278, 261–263 2 Freed-Pastor, W.A. and Prives, C. (2012) Mutant p53: one name, many proteins. Genes Dev. 26, 1268–1286 3 Surget, S. et al. (2014) Uncovering the role of p53 splice variants in human malignancy: a clinical perspective. Onco Targets Ther. 2014, 57–68 4 Joerger, A.C. and Fersht, A.R. (2008) Structural biology of the tumor suppressor p53. Annu. Rev. Biochem. 77, 557–582 5 Okorokov, A.L. et al. (2006) The structure of p53 tumour suppressor protein reveals the basis for its functional plasticity. EMBO J. 25, 5191–5200 6 Lavin, M.F. and Gueven, N. (2006) The complexity of p53 stabilization and activation. Cell Death Differ. 13, 941–950 7 Kandoth, C. et al. (2013) Mutational landscape and significance across 12 major cancer types. Nature 502, 333–339 8 Muller, P.A.J. and Vousden, K.H. (2013) p53 mutations and cancer. Nat. Cell Biol. 15, 2–8 9 Silva, J.L. et al. (2013) Expanding the prion concept to cancer biology: dominant-negative effect of aggregates of mutant p53 tumour suppressor. Biosci. Rep. 33, 593–603 10 Petitjean, A. et al. (2007) Impact of mutant p53 functional properties on TP53 mutation patterns and tumor phenotype: lessons from recent developments in the IARC TP53 Database. Hum. Mutat. 28, 622–629 11 Brosh, R. et al. (2009) When mutants gain new powers: news from the mutant p53 field. Nat. Rev. Cancer 9, 701–713 12 Chin, K.V. et al. (1992) Modulation of activity of the promoter of the human MDR1 gene by Ras and p53. Science 255, 459–462 13 Chiti, F. and Dobson, C.M. (2006) Protein misfolding, functional amyloid, and human disease. Annu. Rev. Biochem. 75, 333–366 14 Prusiner, S.B. (1998) Prions. Proc. Natl. Acad. Sci. U.S.A 95, 13363– 13383 15 Soto, C. (2012) Transmissible proteins: expanding the prion heresy. Cell 149, 968–977 16 Prusiner, S.B. (2012) Cell biology. A unifying role for prions in neurodegenerative diseases. Science 336, 1511–1513 17 Polymenidou, M. and Cleveland, D.W. (2011) The seeds of neurodegeneration: prion-like spreading in ALS. Cell 147, 498–508 18 Irwin, D.J. et al. (2013) Evaluation of potential infectivity of Alzheimer and Parkinson disease proteins in recipients of cadaver-derived human growth hormone. JAMA Neurol. 70, 462–468 19 Aguzzi, A. and Rajendran, L. (2009) The transcellular spread of cytosolic amyloids, prions, and prionoids. Neuron 64, 783–790 20 Ano Bom, A.P. et al. (2012) Mutant p53 aggregates into prion-like amyloid oligomers and fibrils: implications for cancer. J. Biol. Chem. 287, 28152–28162 21 Silva, J.L. et al. (2010) Ligand binding and hydration in protein misfolding: insights from studies of prion and p53 tumor suppressor proteins. Acc. Chem. Res. 43, 271–279 22 Bullock, A.N. et al. (1997) Thermodynamic stability of wild-type and mutant p53 core domain. Proc. Natl. Acad. Sci. U.S.A. 94, 14338–14342 23 Ishimaru, D. et al. (2003) Fibrillar aggregates of the tumor suppressor p53 core domain. Biochemistry 42, 9022–9027 24 Lee, A.S. et al. (2003) Reversible amyloid formation by the p53 tetramerization domain and a cancer-associated mutant. J. Mol. Biol. 327, 699–709 25 Rigacci, S. et al. (2008) The (1-63) region of the p53 transactivation domain aggregates in vitro into cytotoxic amyloid assemblies. Biophys. J. 94, 3635–3646 26 Wilcken, R. et al. (2012) Kinetic mechanism of p53 oncogenic mutant aggregation and its inhibition. Proc. Natl. Acad. Sci. U.S.A. 109, 13584–13589 27 Levy, C.B. et al. (2011) Co-localization of mutant p53 and amyloid-like protein aggregates in breast tumors. Int. J. Biochem. Cell Biol. 43, 60–64

266


28 Lasagna-Reeves, C.A. et al. (2013) Dual role of p53 amyloid formation in cancer; loss of function and gain of toxicity. Biochem. Biophys. Res. Commun. 430, 963–968 29 Kluth, M. et al. (2014) Clinical significance of different types of p53 gene alteration in surgically treated prostate cancer. Int. J. Cancer http:// dx.doi.org/10.1002/ijc.28784 30 Olzscha, H. et al. (2011) Amyloid-like aggregates sequester numerous metastable proteins with essential cellular functions. Cell 144, 67–68 31 Bisio, A. et al. (2014) TP53 mutants in the tower of babel of cancer progression. Hum. Mutat. http://dx.doi.org/10.1002/humu.22514 32 Valastyan, J.S. and Lindquist, S. (2014) Mechanisms of protein-folding diseases at a glance. Dis. Model Mech. 7, 9–14 33 Chamcheu, J.C. et al. (2011) Keratin gene mutations in disorders of human skin and its appendages. Arch. Biochem. Biophys. 508, 123–137 34 Russell, D. et al. (2004) Mechanical stress induces profound remodelling of keratin filaments and cell junctions in epidermolysis bullosa simplex keratinocytes. J. Cell Sci. 117, 5233–5243 35 Werner, N.S. et al. (2004) Epidermolysis bullosa simplex-type mutations alter the dynamics of the keratin cytoskeleton and reveal a contribution of actin to the transport of keratin subunits. Mol. Biol. Cell 15, 990–1002 36 Milner, J. and Medcalf, E.A. (1991) Cotranslation of activated mutant p53 with wild type drives the wild-type p53 protein into the mutant conformation. Cell 65, 765–774 37 Chan, W.M. et al. (2004) How many mutant p53 molecules are needed to inactivate a tetramer? Mol. Cell. Biol. 24, 3536–3551 38 Natan, E. et al. (2009) Ultraslow oligomerization equilibria of p53 and its implications. Proc. Natl. Acad. Sci. U.S.A. 106, 14327–14332 39 Ishimaru, D. et al. (2009) Cognate DNA stabilizes the tumor suppressor p53 and prevents misfolding and aggregation. Biochemistry 48, 6126– 6135 40 Rangel, L.P. et al. (2014) The aggregation of mutant p53 produces prion-like properties in cancer. Prion http://dx.doi.org/10.4161/ pri.27776 41 Xu, J. et al. (2011) Gain of function of mutant p53 by coaggregation with multiple tumor suppressors. Nat. Chem. Biol. 7, 285–295 42 Forget, K.J. et al. (2013) p53 aggregates penetrate cells and induce the co-aggregation of intracellular p53. PLoS ONE 8, http://dx.doi.org/ 10.1371/journal.pone.0069242 43 Wiech, M. et al. (2012) Molecular mechanism of mutant p53 stabilization: the role of HSP70 and MDM2. PLoS ONE 7, http:// dx.doi.org/10.1371/journal.pone.0051426 44 Kirilyuk, A. et al. (2012) An intrinsically disordered region of the acetyltransferase p300 with similarity to prion-like domains plays a role in aggregation. PLoS ONE 7, http://dx.doi.org/10.1371/ journal.pone.0048243 45 Hanel, W. et al. (2013) Two hot spot mutant p53 mouse models display differential gain of function in tumorigenesis. Cell Death Differ. 20, 898–909 46 Mello, S.S. and Attardi, L.D. (2013) Not all p53 gain-of-function mutants are created equal. Cell Death Differ. 20, 855–857 47 Lu, A. et al. (2014) Unified polymerization mechanism for the assembly of ASC-dependent inflammasomes. Cell 156, 1193–1206 48 Cai, X. et al. (2014) Prion-like polymerization underlies signal transduction in antiviral immune defense and inflammasome activation. Cell 156, 1207–1222 49 Neilsen, P.M. et al. (2011) Mutant p53 uses p63 as a molecular chaperone to alter gene expression and induce a pro-invasive secretome. Oncotarget 2, 1203–1217 50 Martynova, E. et al. (2012) Gain-of-function p53 mutants have widespread genomic locations partially overlapping with p63. Oncotarget 3, 132–143 51 Jucker, M. and Walker, L.C. (2013) Self-propagation of pathogenic protein aggregates in neurodegenerative diseases. Nature 501, 45–51 52 Lee, H.J. et al. (2014) Extracellular a-sinuclein – a novel and crucial fator in Lewy body diseases. Nat. Rev. Neurol. 10, 92–98 53 Lee, S.H. et al. (2009) p53, secreted by K-Ras-Snail pathway, is endocyted by K-Ras mutated cells; implication of target-specific drug delivery and early diagnostic marker. Oncogene 28, 2005–2014 54 Lee, S.H. et al. (2013) Extracellular p53 fragment re-enters K-Ras mutated cells through the caveolin-1 dependent early endosomal system. Oncotarget 4, 2523–2531

Opinion 55 Silva, J.L. et al. (2008) Intriguing nucleic-acid-binding features of mammalian prion protein. Trends Biochem. Sci. 33, 132–140 56 Cordeiro, Y. et al. (2001) DNA converts cellular prion protein into the beta-sheet conformation and inhibits prion peptide aggregation. J. Biol. Chem. 276, 49400–49409 57 Deleault, N.R. et al. (2003) RNA molecules stimulate prion protein conversion. Nature 425, 717–720 58 Nieva, J. et al. (2011) Cholesterol secosterol aldehydes induce amyloidogenesis and dysfunction of wild-type tumor protein p53. Chem. Biol. 18, 920–927 59 Guo, J.L. et al. (2013) Distinct a-synuclein strains differentially promote tau inclusions in neurons. Cell 154, 103–117 60 Liu, X. et al. (2013) Small molecule induced reactivation of mutant p53 in cancer cells. Nucleic Acids Res. 41, 6034–6044 61 Wang, G. and Fersht, A.R. (2012) First-order rate-determining aggregation mechanism of p53 and its implications. Proc. Natl. Acad. Sci. U.S.A. 109, 13590–13595 62 Kocisko, D.A. et al. (2006) Potent antiscrapie activities of degenerate phosphorothioate oligonucleotides. Antimicrob. Agents Chemother. 50, 1034–1044 63 Vieira, T.C. et al. (2011) Heparin binding by murine recombinant prion protein leads to transient aggregation and formation of RNA-resistant species. J. Am. Chem. Soc. 133, 334–344 64 Holmes, B.B. et al. (2013) Heparan sulfate proteoglycans mediate internalization and propagation of specific proteopathic seeds. Proc. Natl. Acad. Sci. U.S.A. 110, http://dx.doi.org/10.1073/pnas. 1301440110 65 Vieira, T.C.R.G. et al. (2014) Heparin binding confers prion stability and impairs its aggregation. FASEB J. http://dx.doi.org/10.1096/fj.13246777 66 Prusiner, S.B. (1982) Novel proteinaceous infectious particles cause scrapie. Science 216, 136–144


67 Kocisko, D.A. et al. (1994) Cell-free formation of protease-resistant prion protein. Nature 370, 471–474 68 Caughey, B. and Baron, G.S. (2006) Prions and their partners in crime. Nature 443, 803–810 69 Marciniuk, K. et al. (2013) Evidence for prion-like mechanisms in several neurodegenerative diseases: potential implications for immunotherapy. Clin. Dev. Immunol. 2013, http://dx.doi.org/10.1155/ 2013/473706 70 Grad, L.I. et al. (2011) Intermolecular transmission of superoxide dismutase 1 misfolding in living cells. Proc. Natl. Acad. Sci. U.S.A. 108, 16398–16403 71 Li, J.Y. et al. (2008) Lewy bodies in grafted neurons in subjects with Parkinson’s disease suggest host-to-graft disease propagation. Nat. Med. 14, 501–503 72 Chaves, R.S. et al. (2010) Protein aggregation containing b-amyloid asynuclein and hyperphosphorylated t in cultured cells of hippocampus, substantia nigra and locus coeruleus after rotenone exposure. BMC Neurosci. 11, http://dx.doi.org/10.1186/1471-2202-11-144 73 Nu¨bling, G. et al. (2012) Synergistic influence of phosphorylation and metal ions on tau oligomer formation and coaggregation with asynuclein at the single molecule level. Mol. Neurodegener. 7, http:// dx.doi.org/10.1186/1750-1326-7-35 74 Takei, Y. et al. (2013) a-Synuclein coaggregation in familial amyotrophic lateral sclerosis with SOD1 gene mutation. Hum. Pathol. 44, 1171–1176 75 Gaiddon, C. et al. (2001) A subset of tumor-derived mutant forms of p53 down-regulate p63 and p73 through a direct interaction with the p53 core domain. Mol. Cell. Biol. 21, 1874–1887 76 Strano, S. et al. (2001) From p63 to p53 across p73. FEBS Lett. 490, 163–170 77 Strano, S. et al. (2002) Physical interaction with human tumor-derived p53 mutants inhibits p63 activities. J. Biol. Chem. 277, 18817–18826

267