Evidence of a Role for Antizyme and Antizyme Inhibitor As Regulators of Human Cancer

Published OnlineFirst August 17, 2011; DOI: 10.1158/1541-7786.MCR-11-0178

Molecular Cancer Review Research

Evidence of a Role for Antizyme and Antizyme Inhibitor as Regulators of Human Cancer

Rachelle R. Olsen1,2 and Bruce R. Zetter1,2

Abstract Antizyme and its endogenous antizyme inhibitor have recently emerged as prominent regulators of cell growth, transformation, centrosome duplication, and tumorigenesis. Antizyme was originally isolated as a negative modulator of the enzyme ornithine decarboxylase (ODC), an essential component of the polyamine biosynthetic pathway. Antizyme binds ODC and facilitates proteasomal ODC degradation. Antizyme also facilitates degradation of a set of cell cycle regulatory proteins, including cyclin D1, Smad1, and Aurora A kinase, as well as Mps1, a protein that regulates centrosome duplication. Antizyme has been reported to function as a tumor suppressor and to negatively regulate tumor cell proliferation and transformation. Antizyme inhibitor binds to antizyme and suppresses its known functions, leading to increased polyamine synthesis, increased cell proliferation, and increased transformation and tumorigenesis. Gene array studies show antizyme inhibitor to be amplified in cancers of the ovary, breast, and prostate. In this review, we summarize the current literature on the role of antizyme and antizyme inhibitor in cancer, discuss how the ratio of antizyme to antizyme inhibitor can influence tumor growth, and suggest strategies to target this axis for tumor prevention and treatment. Mol Cancer Res; 9(10); 1285–93. 2011 AACR.

Introduction length antizyme protein can be either 29.5 kDa or 24 kDa in size. Antizyme decreases polyamine levels through 3 mecha- Regulation of the polyamine pathway by antizyme nisms, as summarized in Fig. 1. First, antizyme disrupts inhibitor and antizyme active ODC dimers, resulting in decreased polyamine syn- Antizyme inhibitor and antizyme are critically important thesis. Second, antizyme targets ODC for degradation by the for maintaining polyamine homeostasis within the cell. Poly- 26S proteasome in a reaction that does not require ubiquitin. amines are multivalent, organic cations, and adequate poly- ODC is already one of the most rapidly degraded proteins in amine levels are necessary for optimal cell growth. Within the mammalian cells, with a half-life of only 1 to 2 hours in the cell, polyamines function in diverse processes, including reg- absence of antizyme. When antizyme is present, however, ulating chromatin condensation, stabilizing the double helical this half-life decreases to only a few minutes (3). Third, structure of DNA, regulating cell differentiation, and regulat- antizyme inhibits polyamine uptake from the microenvi- ing translation through a unique posttranslational modifica- ronment through a mechanism that has yet to be completely tion known as hypusination (1). Polyamines are produced characterized. Thus, an increase in antizyme protein is from the amino acid ornithine in a rate-limiting reaction associated with lower levels of ODC, decreased polyamine catalyzed by the enzyme ornithine decarboxylase (ODC), as synthesis, and reduced rates of cell growth. shown in Fig. 1. ODC is only functional as a homodimeric Antizyme activity is further regulated by the protein complex, and individual ODC monomers have no enzymatic antizyme inhibitor, which is highly homologous to activity (2). ODC monomers bind to each other with relatively ODC, retaining 47.4% identity and 63.5% similarity at low affinity, and under physiologic conditions, ODC dimers the amino acid level between human antizyme inhibitor and are in rapid equilibrium with inactive ODC monomers. human ODC. Although antizyme inhibitor was originally ODC is negatively regulated by antizyme. High intracel- thought to be a derivative of ODC (4), subsequent cloning lular polyamine levels induce a þ1-frameshift that bypasses and sequencing showed that antizyme inhibitor is a distinct a stop codon and results in production of full-length anti- protein (5, 6). Antizyme inhibitor acts as a positive regulator zyme protein. Depending on the start codon used, full- of the polyamine pathway by binding to antizyme and preventing antizyme-mediated ODC degradation (Fig. 1). High levels of antizyme inhibitor correlate with increased Authors' Affiliations: 1Vascular Biology Program, Department of ODC protein, increased polyamine synthesis, and greater Surgery, Children's Hospital Boston, and 2Harvard Medical School, Boston, Massachusetts cell proliferation. Whether antizyme inhibitor functions as a Corresponding Author: Bruce Zetter, 1 Blackfan Circle, Karp 11.125, monomer or as a dimer has been examined in some detail. Boston, MA 02115. Phone: 617-919-2320; Fax: 617-730-0268; E-mail: Of the 8 residues involved in dimer formation in ODC, [email protected] 7 (87.5%) are also conserved in the antizyme inhibitor doi: 10.1158/1541-7786.MCR-11-0178 sequence, including a critical glycine residue (G387) nec- 2011 American Association for Cancer Research. essary for ODC dimerization (5, 7). A comparison of the

www.aacrjournals.org 1285

Olsen and Zetter

AZ AZI

ODC

ODC-monomer AZI (inactive) Figure 1. Antizyme inhibitor and antizyme regulate polyamine ODC AZ ODC biosynthesis through ODC. ODC-dimer ODC/antizyme Dimeric ODC catalyzes the first (active) ODC complex step in polyamine synthesis, converting ornithine into the polyamine putrescine. Antizyme disrupts ODC homodimers, 26S targets ODC for degradation by AZ Polyamine Proteasome the 26S proteasome in an biosynthesis ubiquitin-independent manner, and inhibits polyamine uptake. Activity of antizyme is further ODC regulated by antizyme inhibitor. Polyamine Frameshift: degradation Antizyme inhibitor can displace +1 Frameshift translation of uptake ODC from the ODC-antizyme complex, restoring polyamine AUGAUG UGA antizyme AZ biosynthesis by ODC. AZ, AZ ORF1 antizyme; AZI, antizyme inhibitor. AZ ORF2 UG(A)U Polyamine Cys transporter

structure and conserved amino acids in antizyme inhibitor members of the antizyme family. All antizyme proteins inhibit and ODC is shown in Fig. 2. Although early studies ODC activity (10). Antizyme-2, like antizyme-1, has a wide reported that antizyme inhibitor could form homodimers tissue distribution in vertebrates, though it is expressed at that lack ODC enzymatic activity (4, 8), more recent much lower levels than antizyme-1 in most tissues. Antizyme- evidence from size exclusion chromatography and crystal- 2 is more conserved evolutionarily than antizyme-1, suggest- lography indicates that antizyme inhibitor functions as a ing it may have an important physiologic role. monomer (9). Although antizyme-2 is structurally similar to antizyme- 1, antizyme-2 does not promote proteasomal degradation of Isoforms of antizyme and antizyme inhibitor ODC in vitro, even though it does inhibit both ODC The most predominant antizyme protein is antizyme-1, activity and polyamine uptake. In vivo, however, antizyme- although there are multiple antizyme genes and at least 4 2 promotes ODC degradation and inhibits both ODC activity and polyamine uptake (11). The difference in antizyme activity in vitro was subsequently mapped to 2 131 135 AZI N ++ ++ +++ C Asp amino acids in antizyme-2, replacing Arg and Ala in antizyme-1 (12). The physiologic role of antizyme-2 in facilitating protein degradation in vivo is not yet well AZ-binding Site understood, although it can promote ODC degradation (63% conserved) PEST Sequences in human embryonic kidney cells (13). Expression of antizyme-3 is testis specific and is restricted ODC N ++ ++ ++++ C to a late stage in sperm production. This highly restricted expression suggests that antizyme-3 is necessary to abruptly ODC Active Site alter polyamine levels during sperm morphogenesis (3, 14). + ODC Dimerization This idea is supported by reports that animals overexpressing PLP Binding Substrate Binding ODC in the testes have defects in spermatogenesis, perhaps because the high level of ODC overwhelms the levels of antizyme-3 (15). Antizyme-3, like antizyme-2, has the Figure 2. Comparison of conserved residues in antizyme inhibitor and ability to inhibit both ODC activity and polyamine uptake, ODC. Amino acids involved in formation of the ODC active site (*), ODC 0 but it does not target ODC for degradation (13). By yeast dimerization (þ), pyridoxal 5 -phosphate (PLP) binding (¤), and substrate binding (~) in both antizyme inhibitor and ODC proteins. two-hybrid screen, antizyme-3 was found to interact with AZ, antizyme; AZI, antizyme inhibitor. gametogenetin protein-1 (GGN-1), a germ cell–specific

1286 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Antizyme and Antizyme Inhibitor in Cancer

protein, although the functional consequences of this inter- Although high levels of ODC and polyamines are asso- action are not known (16). ciated with numerous types of neoplastic transformation, A putative fourth member of the antizyme family (anti- the role of antizyme and antizyme inhibitor in this process is zyme-4) was originally isolated from a human brain cDNA just emerging. It is clear from in vitro studies that increasing library, but it has not been well characterized. Yeast two- the ratio of antizyme inhibitor relative to antizyme within hybrid assays showed that antizyme-4 could also bind to the cell favors growth activation, whereas a decrease in the ODC and inhibit ODC enzymatic activity (17). The ability antizyme inhibitor/antizyme ratio favors growth repression. of antizyme-4 to promote ODC degradation or inhibit The pathways directly upstream of antizyme inhibitor polyamine uptake has not been examined yet. expression are largely unknown, and the antizyme inhibitor There are not only multiple isoforms of antizyme that promoter has not yet been studied in detail. Treatment with contribute to the complex regulation of the polyamine mitogens or growth factors leads to a very rapid increase in pathway but also multiple isoforms of antizyme inhibitor. antizyme inhibitor, prior even to the increase observed in The most predominant antizyme inhibitor is antizyme ODC, an immediate early gene (6). The antizyme inhibitor inhibitor-1 (AZIN-1), which is ubiquitously expressed at gene has been localized to chromosome 8q22.3 in human high levels and has been the most studied. Antizyme cells, a region previously recognized as a frequently ampli- inhibitor-2 (AZIN-2) was first identified in 2001 as an fied hotspot associated with decreased overall survival and ODC paralogue and termed ODCp or ODC-like (18). decreased time to distant metastases in primary breast Subsequent studies established that ODCp lacked enzymat- tumors (32). Antizyme inhibitor has been shown to be ic activity and seemed to function as a tissue-specific anti- highly expressed in gastric cancer (33), as well as in tumors zyme inhibitor in the brain and testes, where it is expressed at of the lung, liver, and ovary (33–35). In NIH3T3 cells, 6-fold or 23-fold greater levels than antizyme inhibitor-1, overexpression of antizyme inhibitor alone was sufficient to respectively (19, 20). Human antizyme inhibitor-2 retains transform cells and promote tumor growth in vivo (36). 45% identity and 66% similarity to antizyme inhibitor-1 at Using siRNA to suppress antizyme inhibitor levels in A549 the amino acid level (21) and has been shown to interact cells led to decreased cell proliferation and decreased poly- with all 3 characterized antizymes (19, 22, 23). Similar to amine levels; however, the ability of these cells to form antizyme inhibitor-1, overexpression of antizyme inhibitor- tumors in vivo was not determined (37). These results 2 has been shown to increase growth of NIH3T3 cells (23). suggest that decreasing the ratio of antizyme inhibitor to In the future, it will be interesting to determine whether this antizyme within the cell could have a significant inhibitory growth advantage is mediated primarily through the poly- effect on tumor growth. amine pathway and dependent on an intact antizyme- binding domain in antizyme inhibitor-2. To date, the Evidence from transgenic mice suggests that elevated majority of studies on the role of antizyme inhibitor in polyamine levels are associated with tumor tumors have been conducted on antizyme-1 and antizyme development inhibitor-1. For the remainder of this article, antizyme refers Numerous transgenic mouse models are available to to antizyme-1 and antizyme inhibitor refers to antizyme study the polyamine pathway, and they were summarized inhibitor-1. in a recent review (38). Several different mouse models with elevated ODC expression have been described. Polyamines and cancer Transgenic mice overexpressing a C-terminally truncated Because of its key role in promoting cell proliferation, ODC from the bovine keratin 5 (K5) and keratin 6 (K6) ODC is considered a potential oncogene. ODC is down- promoters have elevated ODC activity, as well as elevated stream of Myc and is one of the most rapidly induced genes levels of the polyamines putrescine and spermidine (39). upon growth stimulation (24). Elevated levels of ODC and These mice developed hair loss, formed dermal follicular polyamines have been associated with numerous types of cysts, and older mice frequently developed spontaneous neoplastic transformation, and ODC overexpression alone squamous neoplasms. These experiments were done, how- can induce cell transformation and in vivo tumor growth in ever, on founder and F1 mice with a mixed genetic back- NIH3T3 cells following subcutaneous implantation in ground, and subsequent breeding to either a B6 or FVB nude mice (25). ODC activity is induced by a wide range background did not result in spontaneous tumor forma- of chemical, environmental, and genetic cancer risk factors, tion (40). K6/ODC mice do develop tumors following including ultraviolet light, asbestos, and exposure to chem- treatment with carcinogens, exposure to UV radiation, or ical agents that induce skin carcinogenesis (26). ODC levels by breeding to transgenic mice that carry another onco- are also upregulated in intraepithelial neoplasias, the non- gene. K5/ODC and K6/ODC mice were much more invasive precursors of epithelial cancers. ODC expression is susceptible to tumor formation following a single treat- regulated by androgens in the prostate gland, and ODC ment with the carcinogen 7,12 dimethylbenz(a)anthracene levels are highly elevated in human prostate cancer (27) and (DMBA), and treatment with the tumor promoter phor- numerous other cancer types (28–30). Furthermore, a bol 12-myristate 13-acetate (PMA) was not required (40). meta-analysis of multiple microarray datasets established Subsequent treatment of existing tumors in the K6/ODC that the polyamine pathway is often highly altered in human mice with the irreversible ODC inhibitor alpha-difluoro- prostate cancer (31). methylornithine (DFMO) led to rapid tumor regression

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1287

Olsen and Zetter

(41). Tumor initiation in K6/ODC mice was not limited to DMBA treatment, suggesting that these mice may be a A 1 2 3 4 5 6 B 1 2 3 4 5 6 87 useful model to identify potential human carcinogenic 1.0 1.0 compounds (42). Tumor development in K6/ODC mice was not limited to treatment with chemical carcinogens, 0.5 because crossing K6/ODC mice with mice overexpressing 0.5 v-Ha-ras led to development of spontaneous squamous cell skin tumors. Consistent with results from previous 0 models, these tumors also regressed upon ODC inhibition with DFMO (43, 44). These studies show that ODC 0 0.5 overexpression is sufficient to increase tumor development log2 copy number units log2 copy number units log2 copy in multiple models. Furthermore, the ability of DFMO to 1.0 reverse tumor growth suggests that interfering with poly- 0.5 amine homeostasis may prove to be an effective therapeutic strategy for tumor treatment. This theory is supported to Figure 3. Comparison of antizyme inhibitor and antizyme copy number in some extent by results from clinical trials in which DFMO normal versus cancer tissue. A, changes in antizyme inhibitor and, B, has been used in cancer therapy (27, 45). antizyme copy number were determined by meta-analysis of Oncomine The ability of antizyme to influence tumor growth in vivo data. For antizyme inhibitor, samples analyzed include the following: 1, has also been examined in some detail. Antizyme has been normal; 2, acute myeloid leukemia; 3, breast adenocarcinoma; 4, breast shown to have tumor-suppressive activity in numerous carcinoma; 5, ductal breast carcinoma; and 6, prostate cancer. For antizyme, the following tissue types were analyzed: 1, normal; 2, breast mouse models. In one model, antizyme expression was adenocarcinoma; 3, breast carcinoma; 4, dedifferentiated liposarcoma; 5, targeted to the skin using either the bovine K5 or K6 ductal breast carcinoma; 6, non–small cell lung carcinoma; 7, ovarian promoter (46). Transgenic K6/antizyme and K5/antizyme carcinoma; and 8, prostate carcinoma. mice developed normally but showed delayed tumor onset and decreased tumor numbers relative to wild-type controls after exposure to the chemical carcinogen DMBA and the Generation of conditional mouse models for genes in the tumor promoter 12-O-tetradecanoylphorbol-13-acetate polyamine pathway would contribute to our understanding (TPA; ref. 46). Expression of the K5 and K6 promoters of how the balance among antizyme inhibitor, antizyme, can sometimes occur in other epithelial cells, and K6/ and ODC influences tumor growth in vivo. antizyme and K5/antizyme mice were also shown to have a decreased incidence of forestomach tumors induced by Expression of antizyme inhibitor and antizyme in N-nitrosomethylbenzylamine (NMBA) (47). human cancer To date, only a few studies have addressed the role of Genomic studies have previously determined that anti- antizyme inhibitor in vivo, and many of these reports have zyme inhibitor expression is upregulated in human gastric focused on expression of antizyme inhibitor during normal cancer, ovarian cancer, breast cancer, and prostate cancer development (48). Attempts to make antizyme inhibitor (32–35). Current data from the Oncomine database also knockout mice established that homozygous antizyme in- suggest that antizyme inhibitor copy number is elevated in hibitor deletion results in neonatal lethality (49), and cancers of the bone marrow, breast, and prostate (Fig. 3A; antizyme inhibitor / embryos are currently being charac- ref. 51). Furthermore, studies collected in the Oncomine terized to better understand normal antizyme inhibitor database suggest that antizyme inhibitor expression is ele- þ function (50). Antizyme inhibitor / heterozygote mice vated in cancers of the breast (ductal invasive breast carci- are fertile and have a normal lifespan; they have no mor- noma), testis (seminoma), liver (hepatocellular carcinoma), phologic abnormalities, although they do display lower lung (adenocarcinoma and squamous cell carcinoma), skin ODC protein levels in the liver (49). One critical test of (melanoma), and prostate (carcinoma). Studies of antizyme whether antizyme inhibitor functions as an oncogene in vivo inhibitor expression in cancers of the brain (oligodendro- will be to determine whether transgenic mice with tissue- glioma) and bladder (superficial bladder cancer) were in- specific antizyme inhibitor overexpression have an increased conclusive, with some arrays showing moderately increased rate of tumor incidence, with or without prior carcinogen expression and others showing moderately decreased ex- treatment. For direct comparison to previous results from pression. Because antizyme inhibitor is known to promote ODC transgenic animals, such mice could be generated cell proliferation, increased antizyme inhibitor levels may using either the K5 or K6 promoters to explore the role of directly enhance tumorigenesis. antizyme inhibitor in skin carcinogenesis. It is likely that, Antizyme has been shown to function as a tumor suppres- similar to K6/ODC mice, K6/antizyme inhibitor mice sor in vivo, and it is possible that decreased antizyme levels will would display elevated levels of ODC and polyamines, also contribute to tumor development. A meta-analysis of and they may develop skin tumors following treatment current data in Oncomine suggests that antizyme copy with carcinogenic compounds. Furthermore, it will be number is slightly decreased in several types of breast tumors beneficial to determine whether treatment with DFMO compared with normal tissue (Fig. 3B). Antizyme copy would reverse tumor growth in K6/antizyme inhibitor mice. number may also be decreased in tumors of the lung, ovary,

1288 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Antizyme and Antizyme Inhibitor in Cancer

Table 1. SAGE analysis of antizyme inhibitor and antizyme expression in normal versus cancer tissue

Gene Tissue Normal Cancer Fold change

Antizyme inhibitor Prostate 4.67 2.89 16 7.55 3.43 Brain (ependymoma) 4.33 2.89 14.20 4.80 3.28 Breast 2.82 2.38 7.05 4.83 2.50 Liver 3.00 ND 7.33 4.51 2.44 Brain (meningioma) 4.33 2.89 7.50 4.95 1.73 Lung 4.00 ND 6.67 3.79 1.67 Pancreas 6.00 ND 7.75 3.10 1.29 Brain (glioblastoma) 4.33 2.89 5.58 3.63 1.29 Antizyme Bone Marrow 263.67 32.59 12.33 14.43 0.05 Muscle 107.00 5.66 42.00 ND 0.39 Lung 254.00 ND 140.67 24.42 0.55 Prostate 187.67 45.79 116.50 65.76 0.62 Liver 199.00 ND 133 15.87 0.67 Brain (oligodendroglioma) 103.00 56.95 78.00 41.01 0.76 Brain (ependymoma) 103.00 56.95 89.64 30.66 0.87

NOTE: SAGE Genie Digital Northern analysis from the CGAP (52) was used to assess antizyme inhibitor and antizyme mRNA expression levels in normal versus cancerous tissues from a variety of organs. The unique sequence TCTTTCACCC from the 30 untranslated region of antizyme inhibitor mRNA was used for analysis, and the sequence TTGTAATCGT was used for antizyme. The number of times the gene-specific sequence was detected out of every 200,000 tags was determined and then compared with the observed level in normal tissue. Values are given as average SD. Abbreviation: ND, not determined. and prostate, although there is significantly more variation in Functions of antizyme and antizyme inhibitor outside these datasets. of the polyamine pathway Expression of antizyme inhibitor and antizyme in various Intracellularly, the ratio of antizyme inhibitor to antizyme cancers can also be explored using SAGE Genie Digital is critically important for determining whether conditions Northern analysis provided through the National Cancer favor cell growth repression (high antizyme) or growth Institute's Cancer Genome Anatomy Project (CGAP; ref. activation (high antizyme inhibitor). Thus far, we have 52). As summarized in Table 1, antizyme inhibitor levels were primarily focused on how changes in the level of antizyme substantially elevated in cancers of the prostate, brain (epen- inhibitor or antizyme affect ODC and polyamine levels; dymoma), breast, and liver. Moderate increases were seen in however, both antizyme and antizyme inhibitor have been cancers of the lung and pancreas. Antizyme inhibitor levels shown to have additional functions outside of the poly- also may be increased in tissues from the stomach and kidney, amine pathway, which are discussed in detail below. This although data from normal tissues were not available for finding is particularly intriguing, because the therapeutic comparison. Although antizyme expression is regulated pri- potential of targeting antizyme and antizyme inhibitor may marily at the level of translation, decreased antizyme mRNA be influenced directly by these additional functions. transcription may result in decreased levels of antizyme protein and, consequently, increased tumorigenesis. As shown Antizyme-mediated degradation of cyclin D1 and other in Table 1, antizyme transcript levels seemed to be decreased proteins in tumors of the bone marrow, muscle, and lung, and The realization that antizyme has roles other than moderately decreased in tumors of the liver and prostate. regulating polyamine levels came with the observation that Although gene expression data imply changes in the antizyme can degrade proteins other than ODC. The first antizyme inhibitor/antizyme ratio in many human cancers, demonstration of this degradation was the finding that these data cannot be directly applied to protein levels, antizyme can degrade cyclin D1 in an ubiquitin-indepen- especially for antizyme, which is subject to posttranscrip- dent reaction, although cyclin D1 can also be degraded tional regulation. Confirmation of these results by immu- through the ubiquitin-mediated pathway. Antizyme was nohistochemical staining in tumor tissue microarrays would found not only to immunoprecipitate with cyclin D1 but not only validate the electronic gene expression analysis but also to mediate cyclin D1 degradation in an in vitro system also allow us to better understand whether misregulation of using purified proteasomes (53). Because cyclin D1 activity the antizyme/antizyme inhibitor pathway is a common is required for cell cycle passage, this effect provides a novel feature of numerous tumor types or is more specific to polyamine-independent mechanism for the growth re- certain cancers. pression observed in numerous cell lines upon antizyme

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1289

Olsen and Zetter

overexpression. Since this initial report, antizyme has been Furthermore, the protein levels of both DNA methyltrans- found to regulate degradation of numerous other proteins ferase 3B and the histone H3K9me-specific methyltrans- involved in cell cycle regulation, including Smad 1 (54), ferase G9a were decreased upon antizyme expression, Aurora A (55), and Mps1 (56, 57), a protein that regulates providing additional evidence for a direct link between centrosome duplication. The effect of antizyme and anti- changes in antizyme levels and methylation status. Whether zyme inhibitor on centrosomes is discussed in detail in the observed effect of antizyme on gene methylation could the following section. It is likely that additional targets of be exploited for the therapeutic reactivation of silenced antizyme-mediated degradation remain to be identified. genes in tumors remains to be determined, but it has It is also possible that antizyme binds to ODC with higher intriguing implications for future cancer therapy. affinity than it does to cyclin D1 or other cell cycle mediators. Precise measurement of the binding constants Nonpolyamine functions of antizyme inhibitor between antizyme and its interacting proteins, for example, Antizyme inhibitor has also been shown to have functions by surface plasmon resonance, would greatly enhance our beyond antizyme binding and regulating antizyme func- understanding of the relative importance of these interac- tions. When the antizyme-binding region of antizyme in- tions within the cell. Furthermore, under conditions in hibitor was deleted, overexpression of the mutant protein which antizyme levels are elevated, degradation of proteins (AZIDAZ) was still able to enhance cell proliferation, al- other than ODC may exert a greater cellular effect. though not to the same extent as overexpression of wild-type antizyme inhibitor (61). This finding suggests that antizyme Antizyme and antizyme inhibitor function at inhibitor has antizyme-independent functions, perhaps me- centrosomes diated by preventing degradation of cyclin D1 or other cell In addition to cytoplasmic and nuclear sites, antizyme and cycle regulators. Antizyme inhibitor also partially localizes to antizyme inhibitor localize, in part, to centrosomes (58). the centrosome in a variety of mammalian cells as described These results suggest that centrosomal antizyme and anti- above, and overexpression of antizyme inhibitor can induce zyme inhibitor may have an important role in regulating centriole amplification. Whether antizyme inhibitor directly normal centriole duplication. This role was confirmed by promotes centrosome duplication by interfering with anti- studies in which antizyme overexpression caused a decrease zyme-mediated functions or through some other mechanism in the number of cells with excess centrosomes in a variety of has yet to be fully explored. cell types, whereas antizyme inhibitor overexpression stimulated excess centrosome duplication (58). Centrosome Can altering the antizyme inhibitor–to–antizyme amplification is a common feature of numerous solid tumors balance promote tumor growth? and seems to be an early event in tumorigenesis. These results From the experiments described above, it is clear that suggest that centrosome duplication may be mediated in alteration of the antizyme inhibitor/antizyme ratio within part by antizyme. This effect provides a novel mechanism the cell could have a direct effect not only on rates of cell whereby changes in the antizyme inhibitor/antizyme axis proliferation but also on centrosome duplication and gene affect additional compounds of tumor behavior. Thus far, methylation. Alterations in the antizyme inhibitor/antizyme antizyme has been reported to be involved in degradation of ratio can also affect the uptake of extracellular polyamines. Aurora A and Mps1, 2 kinases that promote centrosome Because antizyme inhibitor inhibits antizyme, increased duplication. It is likely that antizyme and antizyme inhibitor antizyme inhibitor expression would be expected to result are involved in directly regulating ubiquitin-independent in increased ODC activity and increased polyamine uptake. degradation of multiple proteins that regulate centrosome These effects are expected to be somewhat transitory, duplication and that this takes place directly at the centro- however, because increased polyamine uptake will induce some. It is possible that altered polyamine levels affect expression of antizyme, which will inhibit polyamine trans- centrosome duplication, but this has not yet been shown. port. The ratio of antizyme/antizyme inhibitor within the cell will determine the precise effect on polyamine transport Antizyme and DNA methylation following an increase in the antizyme inhibitor level (62). In addition to direct effects of antizyme on cell cycle Numerous experiments thus far have shown that constitu- progression, antizyme has also been suggested to act as a tive overexpression of antizyme inhibitor results in increased tumor suppressor by altering DNA methylation. Ectopic polyamine uptake and increased accumulation of polyamine expression of antizyme in the hamster oral keratinocyte cell analogs (36, 63). This increase may mimic the situations line HCPC-1 led to a reduction in ODC activity and in some human cancers. Measurement of polyamine trans- demethylation of 5-methyl cytosines at CCGG sites (59). port and accumulation of polyamine analogs following This intriguing result suggests that antizyme overexpression inducible expression of antizyme or antizyme inhibitor could lead to reactivation of cellular genes that had been would clarify the short- and long-term effects of these previously silenced by hypermethylation during cancer proteins on polyamine transport. development. Later mechanistic studies showed that anti- The results described in this review argue that changes zyme overexpression was directly associated with hypo- in cell proliferation following an increase in the antizyme methylation of CpG islands in genomic DNA and inhibitor level would lead to cell transformation and in- histone H3 lysine 9 dimethylation (H3K9me2; ref. 60). creased tumor growth, via decreased antizyme-mediated

1290 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Antizyme and Antizyme Inhibitor in Cancer

degradation of ODC and other cell cycle regulatory pro- induces G1 cell cycle arrest (72), and DFMO treatment teins. Although centrosome amplification is a common impairs development of neuroblastoma in Em-Myc transgenic feature of numerous solid tumors, many cell lines have mice (73). On the basis of these results, an ongoing phase I the ability to cluster extra centrosomes and still undergo a clinical trial is being conducted to determine the efficacy of pseudo-bipolar cell division (64). Nevertheless, excess cen- DFMO-etoposide in neuroblastoma (NCT01059071). trosomes can contribute to increased genomic instability We believe that targeting antizyme inhibitor for cancer and higher rates of aneuploidy and may lead to other therapy may actually prove to be a better strategy than chromosomal changes that enhance cell transformation targeting ODC, because such a treatment would affect cell (65). We are only beginning to understand the effect of proliferation not only through the polyamine pathway but antizyme overexpression on gene methylation, and further also through additional cell cycle mediators. As previously studies will be needed to determine the relative importance described, one of the limitations associated with using of those effects during tumor growth in vivo. DFMO to inhibit ODC is that DFMO does not prevent cells from taking up polyamines from the microenviron- Could we target antizyme inhibitor for cancer therapy? ment. It is possible that inhibiting antizyme inhibitor would On the basis of its critical role in regulating cell growth have this same limitation; therefore, a combination of anti- and transformation, there has been long-term interest in zyme inhibitor inhibition and an agent to block polyamine targeting ODC as a tumor treatment, and several ODC uptake may have more therapeutic effects than antizyme inhibitors are now in clinical trials (45, 66–68). Despite inhibitor inhibition alone. Similar studies have already dramatic effects in cell culture systems, early monotherapy shown that combining DFMO treatment with polyamine trials with DFMO were largely unsuccessful, perhaps be- transport inhibitors can deplete intracellular polyamine cause of the ability of cells to enhance polyamine uptake to pools in cultured cells and can effectively inhibit tumor bypass the effect of DFMO treatment. It is possible that growth in the K6/ODC mouse model of squamous cell future combinations of DFMO with either polyamine carcinoma (74). The pathway involved in polyamine uptake transport inhibitors or cytotoxic drugs would provide more in mammalian cells has not yet been fully characterized, and effective therapeutic treatment for many cancer types. identifying the proteins involved could have a significant Considerable recent progress has been made in targeting impact on the design of current and future therapeutics (75). the polyamine pathway in neuroblastoma (69). One frequent event in neuroblastoma development is Myc-N amplification Disclosure of Potential Conflicts of Interest (70), and ODC is known to be a downstream target of Myc. Tumor microarray analysis of 88 neuroblastoma tumors No potential conflicts of interest were disclosed. showed that elevated ODC mRNA levels were associated with highly undifferentiated stage IV tumors and were pre- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance dictive of decreased overall survival probability and poor with 18 U.S.C. Section 1734 solely to indicate this fact. disease prognosis (71). In contrast, elevated levels of antizyme-2 were associated with several indicators of good prog- Received April 21, 2011; revised July 18, 2011; accepted August 1, 2011; nosis (71). Treatment of neuroblastoma cells with DFMO published OnlineFirst August 17, 2011.

References 1. Morgan DDL. Polyamine protocols. Totowa, New Jersey: Humana 9. Albeck S, Dym O, Unger T, Snapir Z, Bercovich Z, Kahana C. Press; 1998. Crystallographic and biochemical studies revealing the structural 2. Kahana C, Asher G, Shaul Y. Mechanisms of protein degradation: an basis for antizyme inhibitor function. Protein Sci 2008;17:793–802. odyssey with ODC. Cell Cycle 2005;4:1461–4. 10. Coffino P. Antizyme, a mediator of ubiquitin-independent proteasomal 3. Coffino P. Regulation of cellular polyamines by antizyme. Nat Rev Mol degradation. Biochimie 2001;83:319–23. Cell Biol 2001;2:188–94. 11. Zhu C, Lang DW, Coffino P. Antizyme2 is a negative regulator of 4. Murakami Y, Matsufuji S, Nishiyama M, Hayashi S. Properties and ornithine decarboxylase and polyamine transport. J Biol Chem fluctuations in vivo of rat liver antizyme inhibitor. Biochem J 1999;274:26425–30. 1989;259:839–45. 12. Chen H, MacDonald A, Coffino P. Structural elements of antizymes 1 5. Murakami Y, Ichiba T, Matsufuji S, Hayashi S. Cloning of antizyme and 2 are required for proteasomal degradation of ornithine decar- inhibitor, a highly homologous protein to ornithine decarboxylase. J boxylase. J Biol Chem 2002;277:45957–61. Biol Chem 1996;271:3340–2. 13. Snapir Z, Keren-Paz A, Bercovich Z, Kahana C. Antizyme 3 inhibits 6. Nilsson J, Grahn B, Heby O. Antizyme inhibitor is rapidly induced polyamine uptake and ornithine decarboxylase (ODC) activity, but in growth-stimulated mouse fibroblasts and releases ornithine does not stimulate ODC degradation. Biochem J 2009;419:99–103, decarboxylase from antizyme suppression. Biochem J 2000;346: 1 p following. 699–704. 14. Ivanov IP, Rohrwasser A, Terreros DA, Gesteland RF, Atkins JF. 7. Tobias KE, Kahana C. Intersubunit location of the active site of Discovery of a spermatogenesis stage-specific ornithine decar- mammalian ornithine decarboxylase as determined by hybridization boxylase antizyme: antizyme 3. Proc Natl Acad Sci U S A 2000;97: of site-directed mutants. Biochemistry 1993;32:5842–7. 4808–13. 8. Fujita K, Murakami Y, Hayashi S. A macromolecular inhibitor 15. Hakovirta H, Keiski A, Toppari J, HalmekytoM,AlhonenL,J€ anne€ J, of the antizyme to ornithine decarboxylase. Biochem J 1982;204: et al. Polyamines and regulation of spermatogenesis: selective 647–52. stimulation of late spermatogonia in transgenic mice overexpres-

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1291

Olsen and Zetter

sing the human ornithine decarboxylase gene. Mol Endocrinol 37. Choi KS, Suh YH, Kim WH, Lee TH, Jung MH. Stable siRNA-mediated 1993;7:1430–6. silencing of antizyme inhibitor: regulation of ornithine decarboxylase 16. Zhang J, Wang Y, Zhou Y, Cao Z, Huang P, Lu B. Yeast two-hybrid activity. Biochem Biophys Res Commun 2005;328:206–12. screens imply that GGNBP1, GGNBP2 and OAZ3 are potential inter- 38. Feith DJ. Carcinogenesis studies in mice with genetically engineered action partners of testicular germ cell-specific protein GGN1. FEBS alterations in polyamine metabolism. Methods Mol Biol 2011;720: Lett 2005;579:559–66. 129–41. 17. Mangold U, Leberer E. Regulation of all members of the antizyme 39. Megosh L, Gilmour SK, Rosson D, Soler AP, Blessing M, Sawicki JA, family by antizyme inhibitor. Biochem J 2005;385:21–8. et al. Increased frequency of spontaneous skin tumors in transgenic 18. Pitkanen€ LT, Heiskala M, Andersson LC. Expression of a novel human mice which overexpress ornithine decarboxylase. Cancer Res 1995; ornithine decarboxylase-like protein in the central nervous system and 55:4205–9. testes. Biochem Biophys Res Commun 2001;287:1051–7. 40. O’Brien TG, Megosh LC, Gilliard G, Soler AP. Ornithine decarboxylase 19. Lopez-Contreras AJ, Lopez-Garcia C, Jimenez-Cervantes C, Cre- overexpression is a sufficient condition for tumor promotion in mouse mades A, Peñafiel R. Mouse ornithine decarboxylase-like gene skin. Cancer Res 1997;57:2630–7. encodes an antizyme inhibitor devoid of ornithine and arginine dec- 41. Peralta Soler A, Gilliard G, Megosh L, George K, O’Brien TG. Poly- arboxylating activity. J Biol Chem 2006;281:30896–906. amines regulate expression of the neoplastic phenotype in mouse 20. Lopez-Contreras AJ, Ramos-Molina B, Cremades A, Peñafiel R. skin. Cancer Res 1998;58:1654–9. Antizyme inhibitor 2 (AZIN2/ODCp) stimulates polyamine uptake in 42. Chen Y, Megosh LC, Gilmour SK, Sawicki JA, O’Brien TG. K6/ODC mammalian cells. J Biol Chem 2008;283:20761–9. transgenic mice as a sensitive model for carcinogen identification. 21. Lopez-Contreras AJ, Ramos-Molina B, Cremades A, Peñafiel R. Toxicol Lett 2000;116:27–35. Antizyme inhibitor 2: molecular, cellular and physiological aspects. 43. Smith MK, Trempus CS, Gilmour SK. Co-operation between follicular Amino Acids 2010;38:603–11. ornithine decarboxylase and v-Ha-ras induces spontaneous papillo- 22. Kanerva K, Makitie€ LT, Pelander A, Heiskala M, Andersson LC. Human mas and malignant conversion in transgenic skin. Carcinogenesis ornithine decarboxylase paralogue (ODCp) is an antizyme inhibitor but 1998;19:1409–15. not an arginine decarboxylase. Biochem J 2008;409:187–92. 44. Lan L, Trempus C, Gilmour SK. Inhibition of ornithine decarboxy- 23. Snapir Z, Keren-Paz A, Bercovich Z, Kahana C. ODCp, a brain- and lase (ODC) decreases tumor vascularization and reverses spontaneous testis-specific ornithine decarboxylase paralogue, functions as an tumors in ODC/Ras transgenic mice. Cancer Res 2000;60:5696–703. antizyme inhibitor, although less efficiently than AzI1. Biochem J 45. BaileyHH,KimK,VermaAK,SielaffK,LarsonPO,SnowS,etal. 2008;410:613–9. A randomized, double-blind, placebo-controlled phase 3 skin can- 24. Pegg AE. Regulation of ornithine decarboxylase. J Biol Chem 2006; cer prevention study of alpha-difluoromethylornithine in subjects 281:14529–32. with previous history of skin cancer. Cancer Prev Res (Phila) 2010; 25. Auvinen M, Paasinen A, Andersson LC, Holtt€ a€ E. Ornithine decarbox- 3:35–47. ylase activity is critical for cell transformation. Nature 1992;360:355–8. 46. Feith DJ, Shantz LM, Pegg AE. Targeted antizyme expression in the 26. O’Brien TG, Simsiman RC, Boutwell RK. Induction of the polyamine- skin of transgenic mice reduces tumor promoter induction of ornithine biosynthetic enzymes in mouse epidermis by tumor-promoting decarboxylase and decreases sensitivity to chemical carcinogenesis. agents. Cancer Res 1975;35:1662–70. Cancer Res 2001;61:6073–81. 27. Gerner EW, Meyskens FL Jr. Polyamines and cancer: old molecules, 47. Fong LY, Feith DJ, Pegg AE. Antizyme overexpression in transgenic new understanding. Nat Rev Cancer 2004;4:781–92. mice reduces cell proliferation, increases apoptosis, and reduces 28. Hogarty MD, Norris MD, Davis K, Liu X, Evageliou NF, Hayes CS, et al. N-nitrosomethylbenzylamine-induced forestomach carcinogenesis. ODC1 is a critical determinant of MYCN oncogenesis and a thera- Cancer Res 2003;63:3945–54. peutic target in neuroblastoma. Cancer Res 2008;68:9735–45. 48. Murakami Y, Suzuki J, Samejima K, Oka T. Developmental alterations 29. Yoshida M, Hayashi H, Taira M, Isono K. Elevated expression of the in expression and subcellular localization of antizyme and antizyme ornithine decarboxylase gene in human esophageal cancer. Cancer inhibitor and their functional importance in the murine mammary Res 1992;52:6671–5. gland. Amino Acids 2010;38:591–601. 30. Hayes CS, DeFeo K, Lan L, Paul B, Sell C, Gilmour SK. Elevated levels 49. Tang H, Ariki K, Ohkido M, Murakami Y, Matsufuji S, Li Z, et al. Role of of ornithine decarboxylase cooperate with Raf/ERK activation to ornithine decarboxylase antizyme inhibitor in vivo. Genes Cells convert normal keratinocytes into invasive malignant cells. Oncogene 2009;14:79–87. 2006;25:1543–53. 50. Wan T, Hu Y, Zhang W, Huang A, Yamamura K, Tang H. Changes in 31. Rhodes DR, Barrette TR, Rubin MA, Ghosh D, Chinnaiyan AM. Meta- liver gene expression of Azin1 knock-out mice. Z Naturforsch C analysis of microarrays: interstudy validation of gene expression 2010;65:519–27. profiles reveals pathway dysregulation in prostate cancer. Cancer 51. Oncomine [database on the Internet]. Ann Arbor (MI): Res 2002;62:4427–33. Compendia Bioscience. 2007–2010 [cited 2010 Nov. 11]. Avail- 32. Chin SF, Teschendorff AE, Marioni JC, Wang Y, Barbosa-Morais NL, able from: https://www.oncomine.org. Thorne NP, et al. High-resolution aCGH and expression profiling 52. SAGE Genie [website on the Internet]. Cancer Genome Anatomy identifies a novel genomic subtype of ER negative breast cancer. Project (CGAP), National Cancer Institute. [cited 2011 Mar. 24]. Genome Biol 2007;8:R215. Available from: http://cgap.nci.nih.gov/SAGE. 33. Jung MH, Kim SC, Jeon GA, Kim SH, Kim Y, Choi KS, et al. Iden- 53. Newman RM, Mobascher A, Mangold U, Koike C, Diah S, Schmidt M, tification of differentially expressed genes in normal and tumor human et al. Antizyme targets cyclin D1 for degradation. A novel mechanism gastric tissue. Genomics 2000;69:281–6. for cell growth repression. J Biol Chem 2004;279:41504–11. 34. Schaner ME, Ross DT, Ciaravino G, Sorlie T, Troyanskaya O, Diehn M, 54. Gruendler C, Lin Y, Farley J, Wang T. Proteasomal degradation of et al. Gene expression patterns in ovarian carcinomas. Mol Biol Cell Smad1 induced by bone morphogenetic proteins. J Biol Chem 2003;14:4376–86. 2001;276:46533–43. 35. van Duin M, van Marion R, Vissers K, Watson JE, van Weerden WM, 55. Lim SK, Gopalan G. Antizyme1 mediates AURKAIP1-dependent deg- Schroder€ FH, et al. High-resolution array comparative genomic hy- radation of Aurora-A. Oncogene 2007;26:6593–603. bridization of chromosome arm 8q: evaluation of genetic progression 56. Kasbek C, Yang CH, Fisk HA. Antizyme restrains centrosome ampli- markers for prostate cancer. Genes Chromosomes Cancer 2005;44: fication by regulating the accumulation of Mps1 at centrosomes. Mol 438–49. Biol Cell 2010;21:3878–89. 36. Keren-Paz A, Bercovich Z, Porat Z, Erez O, Brener O, Kahana C. 57. Kasbek C, Yang CH, Yusof AM, Chapman HM, Winey M, Fisk HA. Overexpression of antizyme-inhibitor in NIH3T3 fibroblasts provides Preventing the degradation of mps1 at centrosomes is sufficient to growth advantage through neutralization of antizyme functions. On- cause centrosome reduplication in human cells. Mol Biol Cell cogene 2006;25:5163–72. 2007;18:4457–69.

1292 Mol Cancer Res; 9(10) October 2011 Molecular Cancer Research

Antizyme and Antizyme Inhibitor in Cancer

58. Mangold U, Hayakawa H, Coughlin M, Munger€ K, Zetter BR. Anti- the prevention of sporadic colorectal adenomas: a randomized zyme, a mediator of ubiquitin-independent proteasomal degradation placebo-controlled, double-blind trial. Cancer Prev Res (Phila) and its inhibitor localize to centrosomes and modulate centriole 2008;1:32–8. amplification. Oncogene 2008;27:604–13. 68. Shantz LM, Levin VA. Regulation of ornithine decarboxylase during 59. Tsuji T, Usui S, Aida T, Tachikawa T, Hu GF, Sasaki A, et al. Induction oncogenic transformation: mechanisms and therapeutic potential. of epithelial differentiation and DNA demethylation in hamster malig- Amino Acids 2007;33:213–23. nant oral keratinocyte by ornithine decarboxylase antizyme. Onco- 69. Evageliou NF, Hogarty MD. Disrupting polyamine homeostasis as a gene 2001;20:24–33. therapeutic strategy for neuroblastoma. Clin Cancer Res 2009;15: 60. Yamamoto D, Shima K, Matsuo K, Nishioka T, Chen CY, Hu GF, et al. 5956–61. Ornithine decarboxylase antizyme induces hypomethylation of ge- 70. Look AT, Hayes FA, Shuster JJ, Douglass EC, Castleberry RP, nome DNA and histone H3 lysine 9 dimethylation (H3K9me2) in human Bowman LC, et al. Clinical relevance of tumor cell ploidy and N- oral cancer cell line. PLoS ONE 2010;5:e12554. myc gene amplification in childhood neuroblastoma: a Pediatric 61. Kim SW, Mangold U, Waghorne C, Mobascher A, Shantz L, Banyard Oncology Group study. J Clin Oncol 1991;9:581–91. J, et al. Regulation of cell proliferation by the antizyme inhibitor: 71. Geerts D, Koster J, Albert D, Koomoa DL, Feith DJ, Pegg AE, evidence for an antizyme-independent mechanism. J Cell Sci et al. The polyamine metabolism genes ornithine decarboxylase 2006;119:2583–91. and antizyme 2 predict aggressive behavior in neuroblastomas 62. Mitchell JL, Thane TK, Sequeira JM, Thokala R. Unusual aspects of with and without MYCN amplification. Int J Cancer 2010;126: the polyamine transport system affect the design of strategies for use 2012–24. of polyamine analogues in chemotherapy. Biochem Soc Trans 72. Wallick CJ, Gamper I, Thorne M, Feith DJ, Takasaki KY, Wilson 2007;35:318–21. SM, et al. Key role for p27Kip1, retinoblastoma protein Rb, and 63. Mitchell JL, Thane TK, Sequeira JM, Marton LJ, Thokala R. Antizyme MYCN in polyamine inhibitor-induced G1 cell cycle arrest in and antizyme inhibitor activities influence cellular responses to poly- MYCN-amplified human neuroblastoma cells. Oncogene 2005;24: amine analogs. Amino Acids 2007;33:291–7. 5606–18. 64. Kwon M, Godinho SA, Chandhok NS, Ganem NJ, Azioune A, Thery M, 73. Rounbehler RJ, Li W, Hall MA, Yang C, Fallahi M, Cleveland JL. et al. Mechanisms to suppress multipolar divisions in cancer cells with Targeting ornithine decarboxylase impairs development of MYCN- extra centrosomes. Genes Dev 2008;22:2189–203. amplified neuroblastoma. Cancer Res 2009;69:547–53. 65. Ganem NJ, Godinho SA, Pellman D. A mechanism linking extra 74. Burns MR, Graminski GF, Weeks RS, Chen Y, O’Brien TG. Lipophilic centrosomes to chromosomal instability. Nature 2009;460:278–82. lysine-spermine conjugates are potent polyamine transport inhibitors 66. Bachrach U. Polyamines and cancer: minireview article. Amino Acids for use in combination with a polyamine biosynthesis inhibitor. J Med 2004;26:307–9. Chem 2009;52:1983–93. 67. Meyskens FL Jr, McLaren CE, Pelot D, Fujikawa-Brooks S, Car- 75. Palmer AJ, Wallace HM. The polyamine transport system as a target penter PM, Hawk E, et al. Difluoromethylornithine plus sulindac for for anticancer drug development. Amino Acids 2010;38:415–22.

www.aacrjournals.org Mol Cancer Res; 9(10) October 2011 1293

Evidence of a Role for Antizyme and Antizyme Inhibitor as Regulators of Human Cancer

Rachelle R. Olsen and Bruce R. Zetter

Mol Cancer Res 2011;9:1285-1293. Published OnlineFirst August 17, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-11-0178

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2011/08/18/1541-7786.MCR-11-0178.DC1

Cited articles This article cites 72 articles, 35 of which you can access for free at: http://mcr.aacrjournals.org/content/9/10/1285.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/9/10/1285.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/9/10/1285. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.