Transcriptional control of the proliferation cluster by the tumor suppressor p53

Abstract

When genome-wide expression profiles of tumors are compared to those of normal tissues, the most recurring transcriptional pattern is characterized by an increased expression of cell-cycle and proliferation-associated genes, collectively referred to as the ‘proliferation cluster’. Tumors with increased expression of the proliferation cluster are frequently associated with augmented proliferation rate, chromosomal instability and metastasis as well as with poor prognosis. Recent in vitro and in vivo data establish a link between the tumor suppressor p53 and the proliferation cluster, implicating loss of p53 activity as the major event responsible for elevated expression of the proliferation cluster in tumors. Moreover, a complex regulatory network, which links p53 with the transcription factors that govern the expression of the proliferation clustergenes, is being gradually elucidated.

---

Ran Brosh

Ran Brosh is a PhD student in the lab of Prof. Varda Rotter at the Weizmann Institute of Science, Israel. He earned his B.Sc from Tel Aviv University (2003) and his M.Sc from the Weizmann Institute of Science (2005). Ran’s scientific interests are the transcriptional activities of the p53 tumor suppressor, with a focus on the involvement of p53 in the regulation of cell-cycle-related protein-coding genes and microRNAs.

Varda Rotter

Varda Rotter serves as the Head of the Department of Molecular Cell Biology at the Weizmann Institute of Science, Israel. She is a molecular cancer researcher, focusing her studies, for the last 30 years, on the tumor suppressor p53. She was trained as an immunologist, and spent a post-doc tenure with David Baltimore in the Cancer Center at the Massachusetts Institute of Technology (1979–1981). Since 1981, she has been leading an active group of young scientists and students at the Weizmann Institute of Science. (http://www.weizmann.ac.il/mcb/Varda/)

---

The proliferation cluster

Tumor gene expression signatures can be employed for the development of genomic tests with the potential to improve predictions for the clinical outcome of cancer.¹ Whole-genome expression profiles of tumors provide more information than histopathological examination and other classical biomarkers, enabling the classification of tumors into distinct classes with different prognostic characteristics, and with varying responses to therapeutic drugs. Thus, expression signatures represent a powerful tool for personalized medicine.² Moreover, expression profiling can promote our understanding of tumor etiology by identifying specific signatures associated with activation of oncogenes or inactivation of tumor suppressors, as well as to facilitate the discovery of novel molecular targets for cancer therapy.

When utilizing microarray expression profiling to compare tumors to their corresponding normal tissues, the most recurring transcriptional signature observed is the upregulation of cell-cycle and proliferation-related genes.³ In fact, a similar group of proliferation-related genes, commonly referred to as the ‘proliferation cluster’ (PC) or the ‘proliferation signature’ is repeatedly found to be upregulated not only in the majority of cancer types, but also within sets of tumor samples and cancer cell lines. In those cases, increased expression of the PC frequently correlates with increased cellular proliferation rate^4,5 and mitotic index,^4,6,7 augmented chromosomal instability,⁸ high histological grade,^9,10 poor differentiation status,^11,12 metastasis,^7,13–15 p53 mutations,^16–18 and, importantly, poor clinical outcome.^{15–17,19–22}

Although the specific composition of genes in the numerous identified PCs varies between studies, some common features characterize all of the identified clusters; the hallmark of which is that most genes participate in the regulation of cell-cycle progression and proliferation-related processes.³ Despite the heterogeneity of the different clusters, some genes appear to populate the majority of identified PCs. To exemplify this, we analyzed 5 PCs from different cancer types, including breast,^6,16,17 cervical,²³ and thyroid,¹² as well as 2 clusters from meta-analyses of multiple cancer types, which identified transcriptional signatures associated with poor differentiation¹¹ and chromosomal instability,⁸ and 2 clusters associated with increased proliferation in cancer cell lines.^5,24 We specifically chose a heterogeneous group of clusters that were derived both from in vitro and in vivo data, including many types of malignancies, and they were associated not only with increased proliferation rate but also with other phenotypes, which are considered hallmarks of tumorigenesis. By intersection of the gene lists from the 9 clusters, several trends were revealed. The total number of genes in all 9 clusters is ∼600, of which one sixth appears in more than one cluster. As summarized in Table 1, 46 genes appear in 4 or more clusters, 22 genes appear in 5 or more clusters, and 10 genes appear in 6 or more clusters. The gene encoding topoisomerase IIα (TOP2A) appears in all 9 PCs analyzed. Thus, a common set of genes seems to represent a ‘core proliferation cluster’ (core PC).

Table 1 The core proliferation cluster (core PC)

Gene Symbol (Gene Name)	PC score^a	Cell cycle phase ^b	E2F target gene^c	NF-Y target gene^d
a PC (proliferation cluster) score represents the number of analyzed proliferation clusters in which the corresponding gene appears (out of a total of 9 PCs described in the following studies^5,6,8,11,12,16,17,23,24) b This column describes the cell-cycle phase in which the corresponding gene peaks, as reported by Whitfield et al.^25— represents genes that do not show cell-cycle-periodicity, N.D stands for genes that were not found in the dataset of Whitfield et al.^25 c ‘Yes’ indicates that the corresponding gene was reported to be activated by E2F. d ‘Yes’ indicates that the corresponding gene was reported to be activated by NF-Y.
TOP2A (topoisomerase (DNA) II alpha 170kDa)	9	G2	Yes	Yes
CCNA2 (cyclin A)	8	G2	Yes	Yes
CDC2 (cell division cycle 2), also known as Cdk1	8	G2	Yes	Yes
MAD2L1 (MAD2 mitotic arrest deficient-like 1)	8	G2	Yes	Yes
AURKA (aurora kinase A)	7	G2/M	Yes	Yes
CDC20 (cell division cycle 20 homolog)	6	G2/M	Yes	Yes
CENPF (centromereproteinF, 350/400ka (mitosin))	6	G2/M	No	Yes
PTTG1 (pituitary tumor-transforming 1)	6	M/G1	Yes	Yes
UBE2C (ubiquitin-conjugating enzyme E2C)	6	G2	No	No
FOXM1 (forkhead box M1)	6	G2/M	No	No
CCNB1 (cyclin B1)	5	G2/M	Yes	Yes
CCNB2 (cyclin B2)	5	G2/M	Yes	Yes
CDC6 (cell division cycle 6 homolog)	5	G1/S	Yes	Yes
CENPA (centromereprotein A)	5	G2/M	Yes	No
EZH2 (enhancer of zeste homolog 2)	5	S	Yes	No
NEK2 (NIMA (never in mitosisgene a)-related kinase 2)	5	G2/M	Yes	No
TTK (TTKprotein kinase)	5	G2/M	Yes	Yes
UBE2S (ubiquitin-conjugating enzyme E2S)	6	G2	No	No
RFC4 (replication factor C (activator 1) 4, 37kDa)	4	S	Yes	No
GGH (gamma-glutamyl hydrolase)	5	—	Yes	No
KIF23 (kinesin family member 23)	5	G2	Yes	No
PLK1 (polo-like kinase 1 (Drosophila))	5	G2/M	Yes	Yes
BIRC5 (baculoviral IAP repeat-containing 5)	4	G2/M	No	Yes
BUB1 (budding uninhibited by benzimidazoles 1 homolog)	4	G2/M	Yes	Yes
CDC25C (cell division cycle 25 homologC)	4	G2	Yes	Yes
CDCA8 (cell division cycle associated 8)	4	N.D.	No	Yes
CENPE (centromereprotein E, 312kDa)	4	G2/M	Yes	No
CKS2 (CDC28 protein kinase regulatory subunit 2)	4	G2/M	Yes	No
DLGAP5 (discs, large homolog-associated protein 5)	4	G2/M	No	Yes
ESPL1 (extra spindle pole bodieshomolog 1)	4	G2	No	Yes
FEN1 (flap structure-specific endonuclease 1)	4	S	Yes	No
H2AFX (H2A histone family, member X)	4	G2	Yes	Yes
KIF14 (kinesin family member 14)	4	G2/M	Yes	No
KIF2C (kinesin family member 2C)	4	G2/M	Yes	Yes
KIFC1 (kinesin family member C1)	4	G2	Yes	No
TRIP13 (thyroid hormone receptor interactor 13)	4	G2/M	No	No
MELK (maternal embryonic leucine zipper kinase)	4	G2	Yes	No
PBK (PDZ binding kinase)	4	M/G1	Yes	No
PCNA (proliferating cell nuclear antigen)	4	G1/S	Yes	Yes
PRC1 (protein regulator of cytokinesis 1)	4	M/G1	Yes	Yes
RRM2 (ribonucleotide reductase M2 polypeptide)	4	S	Yes	Yes
TPX2 (TPX2, microtubule-associated, homolog)	4	G2/M	No	No
MYBL2 (v-myb myeloblastosis viral oncogene homolog like 2)	4	—	Yes	No
ATAD2 (ATPase family, AAA domain containing 2)	4	N.D.	No	No
CEP55 (centrosomal protein55kDa)	4	N.D.	No	Yes
ZWINT (ZW10 interactor)	4	S	Yes	No

---

In a study that analyzed whole-genome expression profiles during human cell-cycle progression,²⁵ approximately 850 genes (∼3% of total genes analyzed) were identified as expressed in a cell-cycle-periodic (CCP) fashion, and were assigned to a specific cell-cycle phase. When compared to various cancer-derived PCs, it appears that approximately 50–60% of the PC genes are expressed in a cell-cycle-periodic fashion.²⁵ Interestingly, 98% of the core PC genes (Table 1) belong to the group of periodically-expressed genes, the majority of which peak during the G₂ and/or M phases of the cell-cycle. Thus, the core PC is comprised almost exclusively of cell-cycle genes.

Detailed examination of the genes that recur in the various PCs reveal that they encode proteins which regulate key processes of cell-cycle progression, including cyclins and cyclin-dependent kinases (CCNB1, CCNB2, CCNA2, CCNF, CDC2), centromereproteins (CENPE, CENPF, CENPA), spindle checkpoint components and mitotic proteins (CDC20, MAD2L1, BUB1, BUB1B, PTTG1, ESPL1, PLK1, PRC1, ZWINT, AURKA, AURKB, ESPL1, CEP55), microtubule-dependent motors (KIF14, KIF23, KIF2C, KIFC1), DNA replication and repair factors (TOP2A, PCNA, RFC4, FEN1, RAD21), minichromosome maintenance proteins (MCM2-7), histones (H2AFX, H2AFZ), cell-cycle-associated ubiquitin conjugating enzymes (UBE2C, UBE2S), transcriptional regulators (E2F1, EZH2, FOXM1, MYBL2) and other cell-cycle-regulating proteins (CDC6, CDCA8, CDC25C, CKS2, MKI67, TTK).

The increased expression of PC genes in tumors may occur for various reasons. The most trivial one is that cancer samples contain a higher percentage of cycling cells than normal tissues, which contrarily contain many cells that are ‘resting’ in the G₀ phase. Similarly, more aggressive, less-differentiated, high-grade tumors may also contain more cycling cells than less aggressive, differentiated, low-grade tumors.³ This also coincides well with the positive correlation observed between the PC expression and proliferation rate in cell lines.^5,24 Indeed, the two most commonly used markers to evaluate tumor proliferation rate, Ki-67 and PCNA, are frequent members of PCs. Moreover, overexpression and knockdown studies have demonstrated the ability of numerous PC genes to promote the rate of proliferation and tumor growth.^26–33

However, it seems that upregulation of the PC is not merely an effect of increased proliferation, but rather the relationship between the PC and cell-cycle-dependent expression is more complicated. For example, numerous cell-cycle-periodic genes are not part of the PC; although it seems that genes that display a steep and clear oscillation during the cell-cycle show an increased tendency to populate PCs.²⁵ Furthermore, several clusters identified in cancer samples correlate with additional phenotypes that do not necessarily stem from an increased fraction of cycling cells. For example, Rhodes et al. identified a 69-gene signature upregulated in undifferentiated tumors,¹¹ while Carter et al. identified a 70-gene signature associated with chromosomal instability.⁸ Both these signatures contain a large number of cell-cycle-related genes, and are very similar to ‘classical’ PCs. The relationship between PC expression and genomic instability is probably due to the fact that a large fraction of the PC genes participate in the spindle checkpoint, which is crucial for proper chromosome segregation, and is the major barrier against the acquisition of gross chromosomal aberrations.³⁴ In fact, it was shown that several PC proteins can directly increase chromosomal instability,^32,35 grant resistance to microtubule-disrupting and other chemotherapeutic drugs,^33,36–39 maintain cell survival,^12,27,31,40 and drive tumorigenesis.³² Accordingly, several chemotherapeutic drugs commonly used in the clinic specifically target proteins encoded by core PC genes, including Hydroxyurea that targets ribonucleotide reductase, Doxorubicin and Etoposide that target topoisomerase II, and Flavopiridol that targets CDK1 (reviewed by Whitfield et al.³).

Transcriptional regulation of the proliferation cluster

As most PC genes are expressed in a cell-cycle-periodic fashion, it is not surprising that the transcription factors that govern their expression are key regulators of cell-cycle progression; namely, the E2F family members and NF-Y.²⁵ Evidence for regulation of specific PC genes by E2Fs and NF-Y is widespread, and as listed in Table 1, most of the core PC genes are known transcriptional targets of either E2F or NF-Y (or both). Interestingly, E2F1 is a frequent member of PCs, which coincides well with the fact that it regulates its own transcription⁴¹ and is regulated by NF-Y.^42,43

More systematic supporting evidence for the regulation of the PC by E2Fs and NF-Y is also available. For example, analysis of a PC derived from an in vitro malignant transformation process^44,45 revealed that the promoters of the PC genes are highly enriched with binding sites of E2F and NF-Y, with the vast majority of promoters harboring at least one of these binding sites.⁴⁶ Similarly, enrichment of E2F and NF-Ybinding sites was also found in a PC derived from comparison of highly malignant thyroid tumors with well-differentiated thyroid tumors and normal thyroid tissues.¹² Additionally, in both the above mentioned examples,^12,46 the promoters of the PC genes were found to be enriched not only with E2F and NF-Ybinding sites, but also with two additional regulatory motifs, namely, the cell-cycle-dependent element (CDE) and the cell-cycle genes homology region (CHR). As inferred from their names, both motifs are very much associated with transcriptional regulation of cell-cycle genes and often appear in close proximity in the promoters of cell-cycle genes; however, their cognate transcription factors are as yet unknown.⁴⁷

Extensive computational analysis of the promoter architecture of the PC genes revealed several interesting trends.⁴⁶ First, the presence of NF-Y or E2F motifs alone or together is not sufficient for a gene to display a characteristic PC expression pattern. However, the presence of both CHR and CDE motifs, in conjunction with an NF-Ybinding site drives a gene towards a PC expression pattern (in the presence or absence of E2F binding site). In the absence of a CHR motif, an E2F binding site is necessary for a characteristic PC expression pattern. The same type of analysis was also employed to study cell-cycle-periodic expression in the data published by Whitfield et al.²⁵ Here too, a CHR motif and an NF-Ybinding site were the most important elements for cell-cycle-periodic expression, whereas the CDE motif enhanced this pattern.⁴⁶ The low capacity of the E2F binding site to drive genes towards a characteristic PC or cell-cycle-periodic expression patterns seems to contradict the abundant data linking E2Fs with transcriptional regulation of cell-cycle and proliferation genes.⁴⁸ However, as the E2F family represents a complex network of transcription activators and repressors,⁴⁸ the presence of an E2F motif in a certain promoter is perhaps not sufficient to infer its transcriptional effect, and the entire promoter architecture should be analyzed to this end. Accordingly, the induction of the core PC geneCDC2 during the G₂ phase of the cell-cycle was shown to be dependent on distinct E2F binding sites.⁴⁹ A positive E2F binding site was found to bind E2F activators in an NF-Y-dependent manner, whereas a negative site binds E2F repressors in a CHR-dependent manner. In summary, it seems that the unique expression pattern of PC genes during the cell-cycle and during carcinogenesis is driven by alteration in the activities of the E2Fs, and to a greater extent, the NF-Ytranscription factor, as well as in the status of the unknown factors that regulate transcriptionvia the CHR and CDE motifs. Supporting this notion, it is well known that during tumor development, the activities of E2Fs are frequently augmented by varying mechanisms, including mutations in pRb, silencing of p16^INK4a expression, and amplification of cyclin D or CDK4.⁵⁰ Although the activity of NF-Y is important for cellular proliferation,^43,51 and accumulation of NF-Y increases proliferation rate,⁵² this pivotal cell-cycle regulator is currently not considered a target for oncogenic events. However, as we convey in the following sections, it is highly plausible that during tumor development the activities of NF-Y and E2F are augmented as a result of inactivation of the p53 tumor suppressor gene.

Introducing p53

p53 is a tumor suppressor which acts to suppress the growth of potentially malignant cells by inducing a variety of cellular processes, including cell-cycle arrest, apoptosis, DNA-repair, senescence and differentiation. Accordingly, p53 function is almost always compromised during tumorigenesis, usually as a result of somatic mutations, which occur in approximately half of all human cancers and constitute a cornerstone in tumorigenesis.^53,54 p53 protein levels and activities are kept low in unstressed cellsviaubiquitin-mediated degradation. However, upon cellular insults associated with cancer initiation, such as DNA-damage and oncogene activation, p53 undergoes a series of post-translational modifications resulting in its accumulation and activation.⁵⁵ Following activation, p53 acts primarily as a tetrameric transcription factor that regulates a large number of genes. Activation of genetranscription is considered the main mechanism by which p53 exerts its tumor suppressive functions,⁵⁶ and accordingly, most tumor-derived p53 mutant proteins are defective in DNA binding and transactivation of p53 target genes.⁵⁷ However, transactivation-independent activities of p53 are also important for its function, and include direct transcriptional repression,^58,59protein–protein interactions with other transcription factors and nuclear proteins, as well as cytoplasmatic activities which regulate apoptosis, autophagy and other processes.⁶⁰

Inhibition of cell-cycle progression and induction of cellular senescence are considered major activities through which p53 suppresses tumor development.^61–63 The mechanisms by which p53 inhibits cell-cycle progression and proliferation are diverse and include both activation and repression of genetranscription.^64,65 Transcriptional activation of the cell-cycle-inhibitory genesGADD45, 14-3-3σ, and in particular, p21 (CDKN1A), likely constitute the main mechanisms through which p53 inhibits proliferation.^66–68 While induction of p21 by p53 primarily mediates cell-cycle arrest at the G₁ phase (The G₁/S checkpoint), induction of GADD45, 14-3-3σ and additional less-characterized p53 targets is important to arrest cells at the G₂ phase (G₂/M checkpoint).^64,65 The complex interaction network that links p53, its transcriptional targets, cell-cycle progression, and the expression of the PC will be elaborated in the following sections.

p53 and the proliferation cluster—the in vitro connection

Ample data implicate p53 as the major regulator of the proliferation cluster; from molecular studies describing the ability of p53 to modulate the expression of individual PC genes, through in vitro and in vivo experimental systems in which manipulation of p53 activity resulted in a global alteration of the PC expression, to clinical data that reveal a strong correlation between p53 mutational status and the expression of the PC in human tumors. When combined, these data set forth the notion that p53 represses the expression of the PC.

In the early 1990s, when the basic features of p53 tumor suppressive activity had been unveiled, several studies identified its ability to repress the expression of cellular and viral genes.^58,69 Following this discovery, the transcriptional repression of many PC genes by p53 was demonstrated. Among these were highly frequent members of PCs, such as topoisomerase IIα,⁷⁰cyclin A2,^71,72CDC2,^73,74 cyclin B1⁷⁴ and B2,⁷²PRC1,⁷⁵CDC25C,⁷⁶CDC20,^46,77,78EZH2,²⁸TTK,⁷⁹BIRC5⁸⁰etc. With the development of cDNA microarrays, studies that exhibited the effect of p53 on the entire transcriptome emerged, resulting in the identification of new p53 repression targets and in the appreciation of p53’s ability to coordinately repress an array of proliferation-related genes.^{45,46,81–83} For example, in a study that analyzed the transcriptional effect of p53 null prostate cancer cells upon adenoviral delivery of p53,⁸¹ 111 genes were repressed by more than two fold, 41% of which were cell-cycle regulators (p-value < 5 × 10⁻²⁸), as defined by gene ontology annotations. Approximately half of the p53-repressed cell-cycle genes are part of the core PC (as appears in Table 1). This suppression was also validated in wild-type p53 expressing prostate cancer cells, which displayed robust p53-dependent transcriptional repression of many of the cell-cycle genes simultaneously with DNA damage-induced cell-cycle arrest.⁸¹

Another support for the regulation of the PC by p53 was provided by Milyavsky et al., who identified a 168-gene PC that was negatively regulated by p53 in human primary fibroblasts.^45,46 Specifically, these primary cells were part of an in vitro malignant transformation model, in which fibroblasts were gradually transformed by immortalization, p53 inactivation, H-Ras oncogene introduction and additional spontaneous processes which occurred along their prolonged culturing. Among several transcriptional programs that accompanied the transformation process, a gradual increase in the expression of the PC was identified and was found to correlate both with increased proliferation rate and with defects in the mitotic checkpoint.⁴⁵ The major genetic determinant that governed the expression of this PC was the status of p53: the cluster displayed the lowest expression in senescent cells that harbor an active p53, and the highest expression in fully transformed cells with an inactivated p53. This PC highly resembles those found in human tumors, e.g. it shares approximately 30–40% of its genes with PCs derived from cervical and breast cancers,^23,84 and is highly enriched with cell-cycle genes.

The last example derives from an elegant cellular system developed by Sur et al.,⁸² in which a homologous recombination (‘knock-in’) strategy was utilized to manipulate the p53 status in four different human colorectal cancer cell lines. From each original line, derivatives with different p53 status were created, including p53^+/+, p53^+/− and p53^−/− derivatives from HCT116, RKO and SW480 lines, as well as mutant p53-expressing derivatives from HCT116, RKO and DLD-1 lines. These isogenic derivatives were than exposed to γ-irradiation, which induces DNA double stand breaks and p53-dependent cell-cycle arrest,⁸⁵ and analyzed for their genome-wide expression profiles. Using strict criteria to identify genes that were commonly repressed by wild-type p53 in all cell lines by more than two fold, 35 genes were identified, the majority of which participate in cell-cycle regulation (particularly in the G₂/M phases),⁸² and one third are part of the core PC (Table 1).

To further exemplify the similarity between cancer-derived transcriptional signatures and signatures of p53-mediated transcriptional repression, we analyzed the behavior of different cancer-derived PCs as a function of p53 activity (Table 2). Specifically, we collected five different cancer-derived PCs, including two clusters that are associated with increased tumor grade in breast cancer,^6,17 one cluster which is upregulated in cervical cancer compared to normal cervix, and appears to correlate with an unfavorable outcome,²³ one meta-signature of undifferentiated tumors,¹¹ and one meta-signature of chromosomal instability.⁸ Additionally, a PC which derived from ∼60 different cancer cell lines and is repressed upon DNA damage was also included,²⁴ as well the core PC (Table 1). As shown in Table 2, when tested for functional enrichment of gene annotations (using DAVID⁸⁶), all of these clusters displayed striking enrichment of the ‘cell-cycle’ annotation. In the next step we tested whether the genes that comprise each of the clusters are repressed by p53 in the two in vitro systems described above, namely, the primary human fibroblast transformation model, where wild-type p53 was inactivated at several stages along the transformation process (Milyavsky et al.^44–46), and the isogenic sets of colorectal cancer cell lines differing in their p53 status and exposed to γ-irradiation (Sur et al.⁸²). Strikingly, the percentage of genes repressed by p53 (tested with a univariate paired t-test, p < 0.05) among the different cancer-derived PCs ranges between 60–85% according to the data of Sur et al., and 40–80% according to the data of Milyavsky et al. The higher percentages of p53-repressed genes in the Sur et al. data probably stem from the fact that in this system, p53 was activated by γ-irradiation, in contrast to the system described by Milyavsky et al., p53 was not activated by exogenous drugs. The core PC showed an even higher percentage of repression by p53, reaching 95%. Thus, the majority of the genes that populate cancer-derived PCs are repressed by p53 in normal and cancer cells. Moreover, nearly all genes that recur in different PCs (the core PC) are repressed by p53. Next, we analyzed the functional enrichment for the cell-cycle annotation among the genes that are repressed or not repressed by p53. This analysis revealed that the p53-repressed genes are highly enriched for the cell-cycle annotation, while the genes that are not repressed are either not or only marginally enriched for the cell-cycle annotation. This suggests that among PC members, only (or almost only) the cell-cycle-related genes are repressed by p53.

Table 2 Cancer-derived proliferation clusters are repressed by p53

Study		Core PC	Langerød et al.^17	Perou et al.^6	Rosty et al.^23	Rhodes et al.^11	Carter et al.^8	Amundson et al.^24
a Repression by p53 was determined using the BRB-Array tools^137 using univariate paired t-test with a threshold of 0.05, on all the clustergenes identified in Sur et al.^82 or Milyavsky et al.^45 data. b Cell-cycle enrichment was determined according to the DAVID database of gene annotations.^86 c Motif enrichment was determined as described in Tabach et al.^46
Sample type		Mixed cancers and cell lines	Primary breast cancers	Primary breast cancers	Normal cervix, cervical cancer, and cervical cell lines	Meta-analysis of multiple cancer types	Meta-analysis of multiple cancer types	NCI60 panel of multiple cancer cell lines
Comparison		Genes that appear in many different PCs	High-grade mutant-p53 vs. low-grade W.T. p53 tumors	High-grade vs. low grade-tumors	Tumors and cell lines vs. normal	Undifferentiated vs. Well-differentiated tumors	Chromosomal instability signature	Genes repressed by ionizing radiation
# of known genes in cluster		40	98	69	123	67	70	39
	Cell-cycle enrichment^b	4.6 × 10^−27	8.1 × 10^−24	3.9 × 10^−13	1.2 × 10^−30	6 × 10^−12	3.8 × 10^−18	3.3 × 10^−36
Behavior in Sur et al.
% repressed by p53^a		95	71	63	81	58	77	84
	Cell-cycle enrichment^b	3.5 × 10^−28	5.1 × 10^−28	1.2 × 10^−20	5.5 × 10^−33	2.2 × 10^−16	8.2 × 10^−21	1.8 × 10^−30
	Motif enrichment^c	NF-Y, CHR	NF-Y, CHR	NF-Y, CHR, E2F	NF-Y, CHR, CDE	NF-Y, CHR	NF-Y, CHR	NF-Y, CHR
% not repressed by p53^a		5	29	37	19	42	23	16
	Cell-cycle enrichment^b	Not enriched	Not enriched	1.0 × 10^−2	5.6 × 10^−2	Not enriched	Not enriched	Not enriched
	Motif enrichment^c	None	None	None	None	None	None	None
Behavior in Milyavsky et al.
% repressed by p53^a		89	61	54	70	40	60	78
	Cell-cycle enrichment^b	2.0 × 10^−16	2.1 × 10^−13	3.7 × 10^−8	1.1 × 10^−20	3.4 × 10^−12	1.0 × 10^−9	4.0 × 10^−19
% not repressed by p53^a		11	39	46	30	60	40	22
	Cell-cycle enrichment^b	Not enriched	3.5 × 10^−2	1.4 × 10^−5	9.6 × 10^−6	8.9 × 10^−4	1.2 × 10^−2	3.0 × 10^−2

---

p53 and the proliferation cluster—the clinical connection

The final line of evidence linking p53 to the PC derives from human clinical data. As described above, p53 is inactivated during tumorigenesis in most human cancers.⁵⁴ This inactivation is frequently achieved viamissense mutations that result in the expression of a full length mutant p53 protein, readily detected in human tumors.^53,87,88 It is well documented that p53 inactivation results in an increased proliferation rate in cell lines as well as in tumors, where p53 mutations are repeatedly associated with increased mitotic index.⁸⁹ Since this index is estimated according to the percentage of cells that are stained immuno-positive for Ki-67, which is encoded by a frequent PC member (MKI67); it points to the potential of wild-type p53 to repress PC genes in human tumors, and demonstrates that this transcriptional repression is manifested also at the protein level. Additional negative correlations exist between wild-type p53 presence and expression of core PC genes and proteins, such as topoisomerase IIα,⁹⁰ cyclin A,^91,92cyclin B1,^92,93EZH2,⁹⁴CENPA,⁹⁵ aurora kinases A and B,^96,97BIRC5⁹⁸etc.

The effect of p53 mutations on the expression of proliferation-related genes as a set was observed in several clinical studies that analyzed whole-genome expression as well as the mutational status of p53.^16–18,22 For example, Miller et al., analyzed 251 primary breast tumors and identified a PC that was highly correlated with p53 mutations.¹⁸ In fact, of the top 30 genes that were most correlated with p53 mutations, 15 are members of the core PC. Similarly, Langerød et al. analyzed 80 primary breast tumors, and identified a large set of genes positively associated with p53 mutations.¹⁷ This gene set was highly enriched (p-value 4.7 × 10⁻¹⁴) with genes belonging to the cell-cycle annotation. Out of the 10 genes that were most highly correlated with p53 mutations, 8 are part of the core PC.

To further exemplify this notion we analyzed the expression of the PC derived from the data of Langerød et al.¹⁷ This PC is comprised of genes upregulated in tumors with a high histological grade (Fig. 1). As the figure depicts, a positive correlation also exists between the expression of this PC and p53 mutations. Additionally, many genes in this PC are part of the core PC, and most of these core PC genes are expressed in a cell-cycle-periodic manner (see bars on the right). The positive correlation between the increased expression of the PC and p53 mutations does not attest to a causative relationship between these two variables. Therefore, we plotted the expression of the genes from this breast cancer PC according to the data of Sur et al.⁸² These data includes expression profiles of isogenic cell lines differing only in their p53 status, and hence, can be utilized to test whether a particular gene can be transcriptionally repressed by p53. As demonstrated in Fig. 2, the vast majority of genes displayed clear downregulation in cells that contain wild-type p53 compared to p53 null or mutant cells. Another trend identified in this analysis indicates that p53-repressed genes tend to be included in the core PC and to display cell-cycle-periodicity. Thus, p53 represses a large proportion of the PC genes in tumors, and in particular, genes that are associated with the process of the cell-cycle.

Fig. 1 A breast cancer proliferation cluster. The figure depicts a cluster of ∼100 genes according to their expression in primary human breast tumors, as described by Langerød et al. (ref. 17). The cluster is upregulated in tumors with high histological grade (tumor grade bar on top). A correlation with p53 mutational status is also apparent (p53 status bar on top), as the cluster shows increased expression in samples with a mutant p53 (analyzed by sequencing as described by Langerød et al.). The core PC bar on the right indicates genes that appear in 3 (orange) or at least 4 (red) out of 9 analyzed proliferation clusters (see Table 1). The CCP bar indicates genes that are expressed in a cell-cycle-periodic fashion according to Whitfield et al. (ref. 25). Note that most of the core PC genes are expressed in a cell-cycle-periodic fashion. Analyses were performed using BRB-ArrayTools developed by Dr Richard Simon and BRB-ArrayTools Development Team.

Fig. 2 A breast cancer-derived proliferation cluster is repressed by p53 in human colorectal cell lines. The figure depicts the expression patterns of ∼100 genes that constitute a proliferation cluster from Langerød et al. (ref. 17, see Fig. 1) according to their expression in the data from Sur et al. (ref. 82). The vast majority of genes (denoted by a red dendogram on the left) display increased expression in cell lines with an inactivated p53 (indicated as −/− for p53 null of MUT for mutant p53-expressing cells in the p53 genotype bar on top). The expression level of p21 is also displayed (on top) and is negatively correlated with the expression of most of the displayed genes. The core PC bar on the right indicates genes that appear in 3 (orange) or at least 4 (red) out of 9 analyzed proliferation clusters (see Table 1). The CCP bar indicates genes that are expressed in a cell-cycle-periodic fashion according to Whitfield et al. (Ref. 25). Note that most of the genes that display p53-dependent repression are part of the core PC and are expressed in a cell-cycle-periodic fashion. Analyses were performed using BRB-ArrayTools developed by Dr Richard Simon and BRB-ArrayTools Development Team.

A similar analysis was performed by Troester et al.,¹⁶ which utilized p53-specific small interference RNA (siRNA) to identify a common gene set that was regulated by p53 in 4 different breast cancer cell lines. This set was then intersected with another set that correlated with p53 mutations in primary breast tumors, yielding a 52-gene signature shared between the in vitro and in vivo data. In agreement with our analysis, the majority of the genes that were upregulated in mutant p53 tumors and in p53 siRNAcells were cell-cycle-related, and more than half are core PC genes.

In summary, it appears that inactivation of the p53 pathway is the major oncogenic event leading to the upregulation of the PC in human tumors, and in parallel, to increased proliferation rate and aggressiveness of these tumors. The regulatory network linking the upstream p53 tumor suppressor with the expression of the downstream PC genes will be described in the following section.

p53 and the proliferation cluster—the underlying mechanisms

The vast majority of the PC promoters do not contain a consensus p53 binding site, and besides few exceptions,^76,80,99 their p53-mediated repression does not involve p53 binding to their promoters. As discussed above, the transcriptional regulation of most PC genes, and particularly, the cell-cycle-periodic ones, is governed by E2F, NF-Y, CHR and CDE regulatory motifs. Therefore, it is not surprising that these motifs were found to mediate the p53-dependent repression of many individual PC genes. For example, in reporter assays, NF-Ybinding sites were found important for the repression of topoisomerase IIα,⁷⁰CDC2,⁷³ cyclin B1¹⁰⁰ and B2,^100,101CDC25C¹⁰⁰ and CDC20;⁴⁶CDE and/or CHR motifs were required for repression of CDC2,⁷⁴CDC25C⁷⁶ and CDC20;^77,99 and E2F binding sites were important for repression of CDC2⁷⁴ and MCM7.⁸³Table 2 demonstrates this concept in a more global manner; i.e., when the analyzed PC genes are segregated into p53-repressed and non-repressed groups (according to Sur et al.), promoter enrichment for NF-Y, E2F, CHR and CDE motifs is observed among the p53-repressed promoters, and not among the non-repressed promoters. The involvement of E2Fs and NF-Y in p53-mediated suppression of proliferation is supported by studies which demonstrated their requirement for p53-dependent senescence,^102,103 a process involving the transcriptional repression of the PC.^45,46 An additional support for the notion that the p53-mediated repression of PC genes is indirect is derived from an experiment that demonstrated the inability of a transcriptionally-inactive p53 mutant to exert transcriptional repression^28,46,70,104 and induce cell-cycle arrest.¹⁰⁵

The fact that the p53-dependent repression of PC genes requires p53 transactivation activity and is mediated by E2Fs and NF-Y leads to the conclusion that one (or more) of p53’s target genes mediate(s) its transcriptional repression, and probably inhibit(s) E2F and/or NF-Y activities. Indeed, induction of p21 (CDKN1A) by p53 was shown to be necessary for repression of many core PC genes, including TOP2A, CDC2, CDC20, CDC25C, NEK2, MAD2L1, PRC1, EZH2, cyclins A2, B1 and B2, BIRC5 and CENPF.^{28,46,74,77,83,104} Additionally, p21 was shown to be sufficient for repressing the transcription of TOP2A, CDC2, CDC25C, BIRC5, CDC20 and cyclin A2 and B1.^74,77,104 Array data may also highlight p21 importance, as its expression was found to negatively correlate with the expression of PC genes.⁴⁶ This trend can be easily detected in Fig. 2 (see p21 expression bar on the top). The involvement of p21 in the repression of PC genes coincides well with its molecular function, as its major effect is mediated via inhibition of cyclin dependent kinases (CDKs). p21 binds and inhibits CDK2/4, thus preventing the phosphorylation of pRb and the release of E2Fs from pRb-mediated inhibition.⁶⁴ While this inhibitory effect mediates G₁ arrest, p21 can also induce a G₂ arrest by inhibiting the activity of CDK1 (encoded by the core PC geneCDC2).^64,65,106 Furthermore, p21 can also negatively regulate NF-Y activity since CDK2-dependent phosphorylation is necessary for NF-Y function.¹⁰⁷ Thus, induction of p21 by p53, and the subsequent inactivation of CDK2 can lead to inhibition of both E2F and NF-Y activities. An alternative mechanism for the p53-dependent repression of NF-Y target genes was suggested by Imbriano et al., who claimed that upon DNA damage, p53 physically binds NF-Y and is recruited onto NF-Y target promoters, where it exerts transcriptional repression by recruiting histone deacetylases (HDACs).¹⁰⁸ The involvement of histone deacetylation is supported by experiments that demonstrate the ability of Trichostatin A (TSA), an HDACinhibitor, to interfere with p53-mediated transcriptional repression.^99,109,110

Several studies demonstrated that p53 can transcriptionally repress PC genes in p21 null cells,^76,99 perhaps implying that additional p53 target genes may mediate its suppressive effects on cell-cycle progression and on the expression of the PC genes. For example, GADD45 and 14-3-3σ are two p53 targets involved primarily in G₂/M cell-cycle arrest. While GADD45 promotes the dissociation of CDK1 (CDC2) from its activating cyclinB; 14-3-3σ sequesters CDK1-cyclin B complexes in the cytoplasm.⁶⁴ Moreover, a less characterized p53 target, Reprimo, is also capable of inducing a G₂/M cell-cycle arrest.^65,111 Although the direct molecular pathways linking these p53 targets with E2F and NF-Y are not fully characterized, their mere ability to induce cell-cycle arrest is likely sufficient to indirectly inhibit the activity of E2F and NF-Y, as both these factors are activated in a cell-cycle-periodic manner.^50,112 Finally, a recently identified target of p53, microRNA-34a,¹¹³ was shown to inhibit several E2F family members and to induce cell-cycle arrest and downregulation of PC genes.¹¹⁴ The complex network linking p53 and the transcriptional regulation of the PC genes is depicted in Fig. 3.

Fig. 3 Schematic model for the regulatory network linking p53 and the proliferation cluster. Arrows and bar-headed lines correspond to activation and inhibition, respectively. Dashed lines represent poorly characterized interactions. Gray round or elliptical shapes are assigned to proteins that inhibit proliferation, while black rectangular shapes are assigned to proteins that promote proliferation. Note that the CHR and CDE motifs are not included in this model, despite their important roles in the transcription regulation of the PC, since the proteins that bind these motifs are as yet uncharacterized.

Since p53 suppresses the expression of the PC via inhibition of E2F function; other tumor suppressors that regulate E2F may also exert such activity. Two such candidates are p16^INK4a, which, similarly to p21, functions as a cyclin-dependent kinase inhibitor, as well as retinoblastoma (pRb). Loss of p16^INK4a or pRb activities may indeed account for increased expression of the PC in cancer as they are both negative regulators of E2F activity and are frequently mutated, deleted, silenced or virally-inactivated during tumorigenesis.^115,116 Accordingly, several studies have demonstrated correlations between loss of p16^INK4 or pRb activities and increased expression of the PC.^15,46

Mutant p53 gain-of-function—activation of the proliferation cluster expression

Almost all p53 mutations observed in human cancers abrogate its ability to bind and activate wild-type p53 target genes.⁵⁷ Moreover, most missense mutations grant the mutant protein a dominant negative activity, with the accepted mechanism involving hetero-oligomerization with the wild-type protein.^117–119 However, many mutant p53 forms not only lose their wild-type function and acquire dominant negative activities, but also gain new oncogenic properties.^88,119 Such properties are often detected when mutant p53 proteins are ectopically expressed in p53 null cells, or silenced in mutant p53 expressing cells. They include a variety of activities that can enhance the oncogenic potential of cells, such as enhancement of colony formation ability, cooperation with H-Ras in cellular transformation, attenuation of apoptosis and resistance to chemotherapeutic drugs.⁸⁸ This notion, termed the ‘gain-of-function hypothesis’, is evident clinically as an association of p53 mutations with poor prognosis and drug response in several malignancies.⁸⁹ Thus, the increased expression of the PC in mutant p53-harboring tumors may stem not only from loss of wild-type p53 repressive activity but also from the gain-of-function properties of the mutant proteins. Indeed, recent studies demonstrated the ability of several common p53 mutant forms to enhance the activity of NF-Y¹²⁰ and E2F1,¹²¹ and to transcriptionally activate PC genes such as cyclin A2, cyclin B1, CDC2 and CDC25C.¹²⁰ For example, Di Agostino et al.¹²⁰ have recently demonstrated that similarly to wild-type p53, two highly common forms of mutant p53 (p53^R175H and p53^R273H) bind NF-Y and are recruited onto NF-Y responsive elements within NF-Y target gene promoters. However, in contrast to wild-type p53, which binds the histone deacetylase HDAC1, mutant p53 proteins bind the histone acetyltransferase p300 and recruit it to NF-Y target genes promoters, resulting in their transcriptional activation. Thus mutant p53 proteins can not only interfere with the ability of wild-type p53 to inhibit NF-Y, but can also actively enhance the transcription of its target genes. Accordingly, in addition to their ability to interfere with wild-type p53-mediated cell-cycle arrest, many mutant p53 proteins can also actively promote proliferation.^88,122,123

microRNAs as new members of the proliferation cluster

We recently analyzed the effect of p53 on the expression of microRNAs, and identified a group of 15 microRNAs that are transcriptionally repressed by p53 in human senescent fibroblasts.¹²⁴ These 15 microRNAs are organized in the genome in three polycistrons (miR-17-92, miR-106a-92 and miR-106b-25), each containing 3–6 microRNAs. There are several indications that these 15 microRNAs represent novel bona fide members of the PC. First, as many PC members, they are overexpressed in a variety of cancer types.^125–128 Moreover, the miR-106b-25 polycistron resides within an intron of the protein-codinggeneMCM7, which is a frequent member of PCs and is frequently overexpressed in cancers.^129–131 Second, the three polycistrons are directly activated by E2F,^{124,128,132–134} and their p53-dependent repression is mediated via inhibition of E2F activity.¹²⁴ Third, almost all validated targets of these microRNAs are key regulators of cell-cycle progression, including E2Fs, pocket proteins, and CDKinhibitors.^{124,125,132–136} Consequently, overexpression of these microRNAs stimulates proliferation and accelerates tumor growth.^{124,128,132,135} Finally, these microRNAs are upregulated in primary breast tumors that harbor mutant p53 compared to wild-type p53 tumors.¹²⁴ In summary, it seems that the PC definition can now be expanded to include additional transcript types such as microRNAs, the repression of which by p53 is plausibly important for its tumor suppressive function.

Concluding remarks

Numerous studies have demonstrated the prognostic significance of the level of expression of individual PC genes in various malignancies, with the common trend being positive correlation between their expression level and poor clinical outcome. It is not surprising therefore, that the assessment of the entire PC expression has a significant prognostic value, usually higher than the assessment of individual genes.³

Increased expression of many PC genes can directly promote proliferation, hamper cell-cycle checkpoints, predispose cells to genomic instability and drive tumorigenesis. Moreover, transcriptional repression of the PC genes by p53 likely represents an important mechanism by which p53 inhibits proliferation and maintains genomic stability. It is unclear though, whether the poor prognosis and drug response that are associated with increased PC expression stem from the activity of the PC genes or perhaps from their associated p53 status, which is considered to be a key player in the regulation of the cancer cell phenotype and drug sensitivity. Regardless of the mechanism, measuring the expression of the PC represents a powerful method for the assessment of p53 status in tumors; a method shown to outperform sequencing-based detection of p53 mutations in predicting prognosis.¹⁶ Finally, it is our opinion that the commonly ignored ability of p53 to transcriptionally repress gene expression, and in particular, the expression of proliferation-related genes, is central for its tumor-suppressive function.

Acknowledgements

We are grateful to Dr Michael Milyavsky, Dr Igor Shats and Dr Yuval Tabach, who pioneered the study on p53 and the proliferation cluster in our research group. VR is the incumbent of the Norman and Helen Asher Professorial Chair for Cancer Research at the Weizmann Institute. The research was supported by a Center of Excellence grant from the Flight Attendant Medical Research Institute (FAMRI).

References

1. C. Sotiriou and L. Pusztai, N. Engl. J. Med., 2009, 360, 790–800.
2. A. Potti and J. R. Nevins, Curr. Opin. Genet. Dev., 2008, 18, 62–67.
3. M. L. Whitfield, L. K. George, G. D. Grant and C. M. Perou, Nat. Rev. Cancer, 2006, 6, 99–106.
4. C. M. Perou, S. S. Jeffrey, M. van de Rijn, C. A. Rees, M. B. Eisen, D. T. Ross, A. Pergamenschikov, C. F. Williams, S. X. Zhu, J. C. Lee, D. Lashkari, D. Shalon, P. O. Brown and D. Botstein, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 9212–9217.
5. D. T. Ross, U. Scherf, M. B. Eisen, C. M. Perou, C. Rees, P. Spellman, V. Iyer, S. S. Jeffrey, M. Van de Rijn, M. Waltham, A. Pergamenschikov, J. C. Lee, D. Lashkari, D. Shalon, T. G. Myers, J. N. Weinstein, D. Botstein and P. O. Brown, Nat. Genet., 2000, 24, 227–235.
6. C. M. Perou, T. Sorlie, M. B. Eisen, M. van de Rijn, S. S. Jeffrey, C. A. Rees, J. R. Pollack, D. T. Ross, H. Johnsen, L. A. Akslen, O. Fluge, A. Pergamenschikov, C. Williams, S. X. Zhu, P. E. Lonning, A. L. Borresen-Dale, P. O. Brown and D. Botstein, Nature, 2000, 406, 747–752.
7. E. LaTulippe, J. Satagopan, A. Smith, H. Scher, P. Scardino, V. Reuter and W. L. Gerald, Cancer Res., 2002, 62, 4499–4506.
8. S. L. Carter, A. C. Eklund, I. S. Kohane, L. N. Harris and Z. Szallasi, Nat. Genet., 2006, 38, 1043–1048.
9. D. S. Rickman, M. P. Bobek, D. E. Misek, R. Kuick, M. Blaivas, D. M. Kurnit, J. Taylor and S. M. Hanash, Cancer Res., 2001, 61, 6885–6891.
10. T. Bonome, J. Y. Lee, D. C. Park, M. Radonovich, C. Pise-Masison, J. Brady, G. J. Gardner, K. Hao, W. H. Wong, J. C. Barrett, K. H. Lu, A. K. Sood, D. M. Gershenson, S. C. Mok and M. J. Birrer, Cancer Res., 2005, 65, 10602–10612.
11. D. R. Rhodes, J. Yu, K. Shanker, N. Deshpande, R. Varambally, D. Ghosh, T. Barrette, A. Pandey and A. M. Chinnaiyan, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 9309–9314.
12. G. Salvatore, T. C. Nappi, P. Salerno, Y. Jiang, C. Garbi, C. Ugolini, P. Miccoli, F. Basolo, M. D. Castellone, A. M. Cirafici, R. M. Melillo, A. Fusco, M. L. Bittner and M. Santoro, Cancer Res., 2007, 67, 10148–10158.
13. C. H. Chung, J. S. Parker, G. Karaca, J. Wu, W. K. Funkhouser, D. Moore, D. Butterfoss, D. Xiang, A. Zanation, X. Yin, W. W. Shockley, M. C. Weissler, L. G. Dressler, C. G. Shores, W. G. Yarbrough and C. M. Perou, Cancer Cell, 2004, 5, 489–500.
14. J. Lapointe, C. Li, J. P. Higgins, M. van de Rijn, E. Bair, K. Montgomery, M. Ferrari, L. Egevad, W. Rayford, U. Bergerheim, P. Ekman, A. M. DeMarzo, R. Tibshirani, D. Botstein, P. O. Brown, J. D. Brooks and J. R. Pollack, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 811–816.
15. K. K. Deeb, A. M. Michalowska, C. Y. Yoon, S. M. Krummey, M. J. Hoenerhoff, C. Kavanaugh, M. C. Li, F. J. Demayo, I. Linnoila, C. X. Deng, E. Y. Lee, D. Medina, J. H. Shih and J. E. Green, Cancer Res., 2007, 67, 8065–8080.
16. M. A. Troester, J. I. Herschkowitz, D. S. Oh, X. He, K. A. Hoadley, C. S. Barbier and C. M. Perou, BMC Cancer, 2006, 6, 276.
17. A. Langerød, H. Zhao, O. Borgan, J. M. Nesland, I. R. Bukholm, T. Ikdahl, R. Kåresen, A. L. Børresen-Dale and S. S. Jeffrey, Breast Cancer Res., 2007, 9, R30.
18. L. D. Miller, J. Smeds, J. George, V. B. Vega, L. Vergara, A. Ploner, Y. Pawitan, P. Hall, S. Klaar, E. T. Liu and J. Bergh, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 13550–13555.
19. H. Dai, L. van't Veer, J. Lamb, Y. D. He, M. Mao, B. M. Fine, R. Bernards, M. van de Vijver, P. Deutsch, A. Sachs, R. Stoughton and S. Friend, Cancer Res., 2005, 65, 4059–4066.
20. S. Paik, S. Shak, G. Tang, C. Kim, J. Baker, M. Cronin, F. L. Baehner, M. G. Walker, D. Watson, T. Park, W. Hiller, E. R. Fisher, D. L. Wickerham, J. Bryant and N. Wolmark, N. Engl. J. Med., 2004, 351, 2817–2826.
21. A. Rosenwald, G. Wright, A. Wiestner, W. C. Chan, J. M. Connors, E. Campo, R. D. Gascoyne, T. M. Grogan, H. K. Muller-Hermelink, E. B. Smeland, M. Chiorazzi, J. M. Giltnane, E. M. Hurt, H. Zhao, L. Averett, S. Henrickson, L. Yang, J. Powell, W. H. Wilson, E. S. Jaffe, R. Simon, R. D. Klausner, E. Montserrat, F. Bosch, T. C. Greiner, D. D. Weisenburger, W. G. Sanger, B. J. Dave, J. C. Lynch, J. Vose, J. O. Armitage, R. I. Fisher, T. P. Miller, M. LeBlanc, G. Ott, S. Kvaloy, H. Holte, J. Delabie and L. M. Staudt, Cancer Cell, 2003, 3, 185–197.
22. T. Sorlie, C. M. Perou, R. Tibshirani, T. Aas, S. Geisler, H. Johnsen, T. Hastie, M. B. Eisen, M. van de Rijn, S. S. Jeffrey, T. Thorsen, H. Quist, J. C. Matese, P. O. Brown, D. Botstein, P. Eystein Lonning and A. L. Borresen-Dale, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 10869–10874.
23. C. Rosty, M. Sheffer, D. Tsafrir, N. Stransky, I. Tsafrir, M. Peter, P. de Cremoux, A. de La Rochefordiere, R. Salmon, T. Dorval, J. P. Thiery, J. Couturier, F. Radvanyi, E. Domany and X. Sastre-Garau, Oncogene, 2005, 24, 7094–7104.
24. S. A. Amundson, K. T. Do, L. C. Vinikoor, R. A. Lee, C. A. Koch-Paiz, J. Ahn, M. Reimers, Y. Chen, D. A. Scudiero, J. N. Weinstein, J. M. Trent, M. L. Bittner, P. S. Meltzer and A. J. Fornace. Jr., Cancer Res., 2008, 68, 415–424.
25. M. L. Whitfield, G. Sherlock, A. J. Saldanha, J. I. Murray, C. A. Ball, K. E. Alexander, J. C. Matese, C. M. Perou, M. M. Hurt, P. O. Brown and D. Botstein, Mol. Biol. Cell, 2002, 13, 1977–2000.
26. K. Yoshida and I. Inoue, FEBS Lett., 2003, 553, 213–217.
27. L. Michel, E. Diaz-Rodriguez, G. Narayan, E. Hernando, V. V. Murty and R. Benezra, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 4459–4464.
28. X. Tang, M. Milyavsky, I. Shats, N. Erez, N. Goldfinger and V. Rotter, Oncogene, 2004, 23, 5759–5769.
29. K. A. Honeycutt, Z. Chen, M. I. Koster, M. Miers, J. Nuchtern, J. Hicks, D. R. Roop and J. M. Shohet, Oncogene, 2006, 25, 4027–4032.
30. T. V. Kalin, I. C. Wang, T. J. Ackerson, M. L. Major, C. J. Detrisac, V. V. Kalinichenko, A. Lyubimov and R. H. Costa, Cancer Res., 2006, 66, 1712–1720.
31. X. X. Wang, R. Liu, S. Q. Jin, F. Y. Fan and Q. M. Zhan, Cell Res., 2006, 16, 356–366.
32. R. Sotillo, E. Hernando, E. Diaz-Rodriguez, J. Teruya-Feldstein, C. Cordon-Cardo, S. W. Lowe and R. Benezra, Cancer Cell, 2007, 11, 9–23.
33. I. Androic, A. Kramer, R. Yan, F. Rodel, R. Gatje, M. Kaufmann, K. Strebhardt and J. Yuan, BMC Cancer, 2008, 8, 391.
34. R. Bharadwaj and H. Yu, Oncogene, 2004, 23, 2016–2027.
35. G. Mondal, S. Sengupta, C. K. Panda, S. M. Gollin, W. S. Saunders and S. Roychoudhury, Carcinogenesis, 2007, 28, 81–92.
36. K. Taniguchi, N. Momiyama, M. Ueda, R. Matsuyama, R. Mori, Y. Fujii, Y. Ichikawa, I. Endo, S. Togo and H. Shimada, Anticancer Res., 2008, 28, 1559–1563.
37. C. Swanton, I. Tomlinson and J. Downward, Cell Cycle, 2006, 5, 818–823.
38. Y. Tao, P. Zhang, F. Girdler, V. Frascogna, M. Castedo, J. Bourhis, G. Kroemer and E. Deutsch, Oncogene, 2008, 27, 3244–3255.
39. D. J. Burgess, J. Doles, L. Zender, W. Xue, B. Ma, W. R. McCombie, G. J. Hannon, S. W. Lowe and M. T. Hemann, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 9053–9058.
40. A. A. Dar, A. Belkhiri, J. Ecsedy, A. Zaika and W. El-Rifai, Cancer Res., 2008, 68, 8998–9004.
41. D. G. Johnson, K. Ohtani and J. R. Nevins, Genes Dev., 1994, 8, 1514–1525.
42. W. Wang, L. Dong, B. Saville and S. Safe, Mol. Endocrinol., 1999, 13, 1373–1387.
43. Q. Hu and S. N. Maity, J. Biol. Chem., 2000, 275, 4435–4444.
44. M. Milyavsky, I. Shats, N. Erez, X. Tang, S. Senderovich, A. Meerson, Y. Tabach, N. Goldfinger, D. Ginsberg, C. C. Harris and V. Rotter, Cancer Res., 2003, 63, 7147–7157.
45. M. Milyavsky, Y. Tabach, I. Shats, N. Erez, Y. Cohen, X. Tang, M. Kalis, I. Kogan, Y. Buganim, N. Goldfinger, D. Ginsberg, C. C. Harris, E. Domany and V. Rotter, Cancer Res., 2005, 65, 4530–4543.
46. Y. Tabach, M. Milyavsky, I. Shats, R. Brosh, O. Zuk, A. Yitzhaky, R. Mantovani, E. Domany, V. Rotter and Y. Pilpel, Mol. Syst. Biol., 2005, 1, 2005.0022.
47. J. Zwicker, F. C. Lucibello, L. A. Wolfraim, C. Gross, M. Truss, K. Engeland and R. Muller, EMBO J., 1995, 14, 4514–4522.
48. A. P. Bracken, M. Ciro, A. Cocito and K. Helin, Trends Biochem. Sci., 2004, 29, 409–417.
49. W. Zhu, P. H. Giangrande and J. R. Nevins, EMBO J., 2004, 23, 4615–4626.
50. J. R. Nevins, Hum. Mol. Genet., 2001, 10, 699–703.
51. Q. Hu, J. F. Lu, R. Luo, S. Sen and S. N. Maity, Nucleic Acids Res., 2006, 34, 6272–6285.
52. I. Manni, G. Caretti, S. Artuso, A. Gurtner, V. Emiliozzi, A. Sacchi, R. Mantovani and G. Piaggio, Mol. Biol. Cell, 2008, 19, 5203–5213.
53. M. Hollstein, D. Sidransky, B. Vogelstein and C. C. Harris, Science, 1991, 253, 49–53.
54. B. Vogelstein, D. Lane and A. J. Levine, Nature, 2000, 408, 307–310.
55. M. Oren, Cell Death Differ., 2003, 10, 431–442.
56. M. Nistér, M. Tang, X.-Q. Zhang, C. Yin, M. Beeche, X. Hu, G. Enblad, T. van Dyke and G. M. Wahl, Oncogene, 2005, 24, 3563–3573.
57. S. Kato, S. Y. Han, W. Liu, K. Otsuka, H. Shibata, R. Kanamaru and C. Ishioka, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 8424–8429.
58. J. Ho and S. Benchimol, Cell Death Differ., 2003, 10, 404–408.
59. V. V. Grinkevich, F. Nikulenkov, Y. Shi, M. Enge, W. Bao, A. Maljukova, A. Gluch, A. Kel, O. Sangfelt and G. Selivanova, Cancer Cell, 2009, 15, 441–453.
60. D. R. Green and G. Kroemer, Nature, 2009, 458, 1127–1130.
61. A. Ventura, D. G. Kirsch, M. E. McLaughlin, D. A. Tuveson, J. Grimm, L. Lintault, J. Newman, E. E. Reczek, R. Weissleder and T. Jacks, Nature, 2007.
62. W. Xue, L. Zender, C. Miething, R. A. Dickins, E. Hernando, V. Krizhanovsky, C. Cordon-Cardo and S. W. Lowe, Nature, 2007.
63. D. Schwartz and V. Rotter, Semin. Cancer Biol., 1998, 8, 325–336.
64. L. E. Giono and J. J. Manfredi, J. Cell. Physiol., 2006, 209, 13–20.
65. W. R. Taylor and G. R. Stark, Oncogene, 2001, 20, 1803–1815.
66. M. B. Kastan, Q. Zhan, W. S. el-Deiry, F. Carrier, T. Jacks, W. V. Walsh, B. S. Plunkett, B. Vogelstein and A. J. Fornace, Jr, Cell, 1992, 71, 587–597.
67. H. Hermeking, C. Lengauer, K. Polyak, T. C. He, L. Zhang, S. Thiagalingam, K. W. Kinzler and B. Vogelstein, Mol. Cell, 1997, 1, 3–11.
68. W. S. el-Deiry, T. Tokino, V. E. Velculescu, D. B. Levy, R. Parsons, J. M. Trent, D. Lin, W. E. Mercer, K. W. Kinzler and B. Vogelstein, Cell, 1993, 75, 817–825.
69. D. Ginsberg, F. Mechta, M. Yaniv and M. Oren, Proc. Natl. Acad. Sci. U. S. A., 1991, 88, 9979–9983.
70. Q. Wang, G. P. Zambetti and D. P. Suttle, Mol. Cell. Biol., 1997, 17, 389–397.
71. M. Yamamoto, M. Yoshida, K. Ono, T. Fujita, N. Ohtani-Fujita, T. Sakai and T. Nikaido, Exp. Cell Res., 1994, 210, 94–101.
72. C. Badie, J. Bourhis, J. Sobczak-Thépot, H. Haddada, M. Chiron, M. Janicot, F. Janot, T. Tursz and G. Vassal, Br. J. Cancer, 2000, 82, 642–650.
73. J. Yun, H. D. Chae, H. E. Choy, J. Chung, H. S. Yoo, M. H. Han and D. Y. Shin, J. Biol. Chem., 1999, 274, 29677–29682.
74. W. R. Taylor, A. H. Schonthal, J. Galante and G. R. Stark, J. Biol. Chem., 2001, 276, 1998–2006.
75. C. Li, M. Lin and J. Liu, Oncogene, 2004, 23, 9336–9347.
76. S. St. Clair, L. Giono, S. Varmeh-Ziaie, L. Resnick-Silverman, W. J. Liu, A. Padi, J. Dastidar, A. DaCosta, M. Mattia and J. J. Manfredi, Mol. Cell, 2004, 16, 725–736.
77. T. Kidokoro, C. Tanikawa, Y. Furukawa, T. Katagiri, Y. Nakamura and K. Matsuda, Oncogene, 2007.
78. T. Banerjee, S. Nath and S. Roychoudhury, Nucleic Acids Res., 2009, 37, 2688–2698.
79. M. R. Bhonde, M. L. Hanski, J. Budczies, M. Cao, B. Gillissen, D. Moorthy, F. Simonetta, H. Scherubl, M. Truss, C. Hagemeier, H. W. Mewes, P. T. Daniel, M. Zeitz and C. Hanski, J. Biol. Chem., 2006, 281, 8675–8685.
80. W. H. Hoffman, S. Biade, J. T. Zilfou, J. Chen and M. Murphy, J. Biol. Chem., 2002, 277, 3247–3257.
81. K. B. Spurgers, D. L. Gold, K. R. Coombes, N. L. Bohnenstiehl, B. Mullins, R. E. Meyn, C. J. Logothetis and T. J. McDonnell, J. Biol. Chem., 2006, 281, 25134–25142.
82. S. Sur, R. Pagliarini, F. Bunz, C. Rago, L. A. Diaz, Jr., K. W. Kinzler, B. Vogelstein and N. Papadopoulos, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 3964–3969.
83. M. J. Scian, E. H. Carchman, L. Mohanraj, K. E. Stagliano, M. A. Anderson, D. Deb, B. M. Crane, T. Kiyono, B. Windle, S. P. Deb and S. Deb, Oncogene, 2008, 27, 2583–2593.
84. L. J. van ‘t Veer, H. Dai, M. J. van de Vijver, Y. D. He, A. A. Hart, M. Mao, H. L. Peterse, K. van der Kooy, M. J. Marton, A. T. Witteveen, G. J. Schreiber, R. M. Kerkhoven, C. Roberts, P. S. Linsley, R. Bernards and S. H. Friend, Nature, 2002, 415, 530–536.
85. F. Bunz, A. Dutriaux, C. Lengauer, T. Waldman, S. Zhou, J. P. Brown, J. M. Sedivy, K. W. Kinzler and B. Vogelstein, Science, 1998, 282, 1497–1501.
86. G. Dennis, Jr, B. T. Sherman, D. A. Hosack, J. Yang, W. Gao, H. C. Lane and R. A. Lempicki, Genome Biol., 2003, 4, P3.
87. A. Petitjean, E. Mathe, S. Kato, C. Ishioka, S. V. Tavtigian, P. Hainaut and M. Olivier, Hum. Mutat., 2007, 28, 622–629.
88. L. Weisz, M. Oren and V. Rotter, Oncogene, 2007, 26, 2202–2211.
89. M. Olivier, P. Hainaut and A. L. Borresen, in 25 Years of p53 Research, ed. P. Hainaut and K. G. Wiman, Springer, Dordrecht, 2005, pp. 321–338.
90. H. E. Miettinen, T. A. Jarvinen, U. Kellner, P. Kauraniemi, R. Parwaresch, I. Rantala, H. Kalimo, L. Paljarvi, J. Isola and H. Haapasalo, Neuropathol. Appl. Neurobiol., 2000, 26, 504–512.
91. Q. Chen, H. Zhou, W. Guo, L. P. Samaranayake, M. Zhou and B. Li, Cytobios, 2001, 106, 87–99.
92. M. Bai, J. Vlachonikolis, N. J. Agnantis, E. Tsanou, S. Dimou, C. Nicolaides, S. Stefanaki, N. Pavlidis and P. Kanavaros, Mod. Pathol., 2001, 14, 1105–1113.
93. K. Aaltonen, R. M. Amini, P. Heikkila, K. Aittomaki, A. Tamminen, H. Nevanlinna and C. Blomqvist, Br. J. Cancer, 2009, 100, 1055–1060.
94. K. Kidani, M. Osaki, T. Tamura, K. Yamaga, K. Shomori, K. Ryoke and H. Ito, Oral Oncol., 2009, 45, 39–46.
95. Y. M. Li, X. H. Liu, X. Z. Cao, L. Wang and M. H. Zhu, Zhonghua Bing Li Xue Za Zhi, 2007, 36, 175–178.
96. Y. C. Shen, F. C. Hu, Y. M. Jeng, Y. T. Chang, Z. Z. Lin, M. C. Chang, C. Hsu and A. L. Cheng, Cancer Epidemiol. Biomark. Prev., 2009, 18, 417–423.
97. Y. M. Jeng, S. Y. Peng, C. Y. Lin and H. C. Hsu, Clin. Cancer Res., 2004, 10, 2065–2071.
98. Z. Khan, R. P. Tiwari, R. Mulherkar, N. K. Sah, G. B. Prasad, B. R. Shrivastava and P. S. Bisen, Head Neck, 2009.
99. T. Banerjee, S. Nath and S. Roychoudhury, Nucleic Acids Res., 2009, 37, 2688–2698.
100. I. Manni, G. Mazzaro, A. Gurtner, R. Mantovani, U. Haugwitz, K. Krause, K. Engeland, A. Sacchi, S. Soddu and G. Piaggio, J. Biol. Chem., 2001, 276, 5570–5576.
101. K. Krause, M. Wasner, W. Reinhard, U. Haugwitz, C. L. Dohna, J. Mossner and K. Engeland, Nucleic Acids Res., 2000, 28, 4410–4418.
102. M. S. Jung, J. Yun, H. D. Chae, J. M. Kim, S. C. Kim, T. S. Choi and D. Y. Shin, Oncogene, 2001, 20, 5818–5825.
103. B. D. Rowland, S. G. Denissov, S. Douma, H. G. Stunnenberg, R. Bernards and D. S. Peeper, Cancer Cell, 2002, 2, 55–65.
104. K. Lohr, C. Moritz, A. Contente and M. Dobbelstein, J. Biol. Chem., 2003, 278, 32507–32516.
105. J. Zhu, J. Jiang, W. Zhou, K. Zhu and X. Chen, Oncogene, 1999, 18, 2149–2155.
106. C. J. Sherr, Science, 1996, 274, 1672–1677.
107. J. Yun, H. D. Chae, T. S. Choi, E. H. Kim, Y. J. Bang, J. Chung, K. S. Choi, R. Mantovani and D. Y. Shin, J. Biol. Chem., 2003, 278, 36966–36972.
108. C. Imbriano, A. Gurtner, F. Cocchiarella, S. Di Agostino, V. Basile, M. Gostissa, M. Dobbelstein, G. Del Sal, G. Piaggio and R. Mantovani, Mol. Cell. Biol., 2005, 25, 3737–3751.
109. M. Murphy, J. Ahn, K. K. Walker, W. H. Hoffman, R. M. Evans, A. J. Levine and D. L. George, Genes Dev., 1999, 13, 2490–2501.
110. I. Shats, M. Milyavsky, X. Tang, P. Stambolsky, N. Erez, R. Brosh, I. Kogan, I. Braunstein, M. Tzukerman, D. Ginsberg and V. Rotter, J. Biol. Chem., 2004, 279, 50976–50985.
111. R. Ohki, J. Nemoto, H. Murasawa, E. Oda, J. Inazawa, N. Tanaka and T. Taniguchi, J. Biol. Chem., 2000, 275, 22627–22630.
112. R. Mantovani, Gene, 1999, 239, 15–27.
113. N. Raver-Shapira, E. Marciano, E. Meiri, Y. Spector, N. Rosenfeld, N. Moskovits, Z. Bentwich and M. Oren, Mol Cell, 2007.
114. H. Tazawa, N. Tsuchiya, M. Izumiya and H. Nakagama, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 15472–15477.
115. W. H. Liggett, Jr. and D. Sidransky, J. Clin. Oncol., 1998, 16, 1197–1206.
116. D. W. Goodrich, Oncogene, 2006, 25, 5233–5243.
117. J. Milner, E. A. Medcalf and A. C. Cook, Mol. Cell. Biol., 1991, 11, 12–19.
118. J. Milner and E. A. Medcalf, Cell, 1991, 65, 765–774.
119. A. Sigal and V. Rotter, Cancer Res, 2000, 60, 6788–6793.
120. S. Di Agostino, S. Strano, V. Emiliozzi, V. Zerbini, M. Mottolese, A. Sacchi, G. Blandino and G. Piaggio, Cancer Cell, 2006, 10, 191–202.
121. G. Blandino, 4th International Workshop on Mutant p53, Akko, Israel, 2009.
122. G. Bossi, E. Lapi, S. Strano, C. Rinaldo, G. Blandino and A. Sacchi, Oncogene, 2006, 25, 304–309.
123. M. J. Scian, K. E. Stagliano, D. Deb, M. A. Ellis, E. H. Carchman, A. Das, K. Valerie, S. P. Deb and S. Deb, Oncogene, 2004, 23, 4430–4443.
124. R. Brosh, R. Shalgi, A. Liran, G. Landan, K. Korotayev, G. H. Nguyen, E. Enerly, H. Johnsen, Y. Buganim, H. Solomon, I. Goldstein, S. Madar, N. Goldfinger, A. L. Borresen-Dale, D. Ginsberg, C. C. Harris, Y. Pilpel, M. Oren and V. Rotter, Mol. Syst. Biol., 2008, 4, 229.
125. S. Volinia, G. A. Calin, C. G. Liu, S. Ambs, A. Cimmino, F. Petrocca, R. Visone, M. Iorio, C. Roldo, M. Ferracin, R. L. Prueitt, N. Yanaihara, G. Lanza, A. Scarpa, A. Vecchione, M. Negrini, C. C. Harris and C. M. Croce, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 2257–2261.
126. L. He, J. M. Thomson, M. T. Hemann, E. Hernando-Monge, D. Mu, S. Goodson, S. Powers, C. Cordon-Cardo, S. W. Lowe, G. J. Hannon and S. M. Hammond, Nature, 2005, 435, 828–833.
127. A. J. Schetter, S. Y. Leung, J. J. Sohn, K. A. Zanetti, E. D. Bowman, N. Yanaihara, S. T. Yuen, T. L. Chan, D. L. Kwong, G. K. Au, C. G. Liu, G. A. Calin, C. M. Croce and C. C. Harris, JAMA, J. Am. Med. Assoc., 2008, 299, 425–436.
128. F. Petrocca, R. Visone, M. R. Onelli, M. H. Shah, M. S. Nicoloso, I. de Martino, D. Iliopoulos, E. Pilozzi, C. G. Liu, M. Negrini, L. Cavazzini, S. Volinia, H. Alder, L. P. Ruco, G. Baldassarre, C. M. Croce and A. Vecchione, Cancer Cell, 2008, 13, 272–286.
129. B. Ren, G. Yu, G. C. Tseng, K. Cieply, T. Gavel, J. Nelson, G. Michalopoulos, Y. P. Yu and J. H. Luo, Oncogene, 2006, 25, 1090–1098.
130. S. S. Li, W. C. Xue, U. S. Khoo, H. Y. Ngan, K. Y. Chan, I. Y. Tam, P. M. Chiu, P. P. Ip, K. F. Tam and A. N. Cheung, Histopathology, 2005, 46, 307–313.
131. A. Facoetti, E. Ranza, E. Benericetti, M. Ceroni, F. Tedeschi and R. Nano, Anticancer Res., 2006, 26, 1071–1075.
132. K. A. O’Donnell, E. A. Wentzel, K. I. Zeller, C. V. Dang and J. T. Mendell, Nature, 2005, 435, 839–843.
133. Y. Sylvestre, V. De Guire, E. Querido, U. K. Mukhopadhyay, V. Bourdeau, F. Major, G. Ferbeyre and P. Chartrand, J. Biol. Chem., 2007, 282, 2135–2143.
134. K. Woods, J. M. Thomson and S. M. Hammond, J. Biol. Chem., 2007, 282, 2130–2134.
135. I. Ivanovska, A. S. Ball, R. L. Diaz, J. F. Magnus, M. Kibukawa, J. M. Schelter, S. V. Kobayashi, L. Lim, J. Burchard, A. L. Jackson, P. S. Linsley and M. A. Cleary, Mol. Cell. Biol., 2008, 28, 2167–2174.
136. Q. Wang, Y. C. Li, J. Wang, J. Kong, Y. Qi, R. J. Quigg and X. Li, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 2889–2894.
137. R. Simon, A. Lam, M. C. Li, M. Ngan, S. Menenzes and Y. Zhao, Cancer Inform., 2007, 3, 11–17.

---