Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Aug 15.
Published in final edited form as: Nat Med. 2012 Jul 22;18(8):10.1038/nm.2863. doi: 10.1038/nm.2863

MDM4 is a key therapeutic target in cutaneous melanoma

Agnieszka Gembarska 1,2, Flavie Luciani 1,2, Clare Fedele 3,4, Elisabeth A Russell 5, Michael Dewaele 1,2, Stéphanie Villar 6, Aleksandra Zwolinska 1,2, Sue Haupt 7,8, Job de Lange 9, Dana Yip 10, James Goydos 10, Jody J Haigh 11,12, Ygal Haupt 7,8,13, Lionel Larue 14, Aart Jochemsen 9, Hubing Shi 15,16, Gatien Moriceau 15,16, Roger S Lo 15,16, Ghanem Ghanem 17, Mark Shackleton 3,4,7, Federico Bernal 5, Jean-Christophe Marine 1,2
PMCID: PMC3744207  NIHMSID: NIHMS409993  PMID: 22820643

Abstract

The inactivation of the p53 tumor suppressor pathway, which often occurs through mutations in TP53 (encoding tumor protein 53) is a common step in human cancer. However, in melanoma—a highly chemotherapy-resistant disease—TP53 mutations are rare, raising the possibility that this cancer uses alternative ways to overcome p53-mediated tumor suppression. Here we show that Mdm4 p53 binding protein homolog (MDM4), a negative regulator of p53, is upregulated in a substantial proportion (∼65%) of stage I–IV human melanomas and that melanocyte-specific Mdm4 overexpression enhanced tumorigenesis in a mouse model of melanoma induced by the oncogene Nras. MDM4 promotes the survival of human metastatic melanoma by antagonizing p53 proapoptotic function. Notably, inhibition of the MDM4-p53 interaction restored p53 function in melanoma cells, resulting in increased sensitivity to cytotoxic chemotherapy and to inhibitors of the BRAF (V600E) oncogene. Our results identify MDM4 as a key determinant of impaired p53 function in human melanoma and designate MDM4 as a promising target for antimelanoma combination therapy.

Cutaneous melanoma is the leading cause of skin cancer–related deaths and is notorious for its resistance to therapy. Nevertheless, recent targeted therapy trials have been promising1,2. Notably, the Ras/Raf/MEK/ERK pathway has been identified as a major, druggable regulator of melanoma3. Activating NRAS mutations are commonly observed in human melanoma, usually affecting codon 61 (refs. 4,5). BRAF is also frequently mutated6, most commonly resulting in a glutamic acid for valine substitution at position 600 (V600E) (ref. 6).

BRAF (V600E), which results in constitutively overactive MAPK/ERK signaling and melanocyte hyperproliferation7, has been successfully exploited for targeted therapy. PLX4032 (vemurafenib), a selective RAF inhibitor, showed an unprecedented antitumor response rate in patients with BRAF (V600E) (ref. 8) and conferred an overall survival benefit in a pivotal phase 3 study9. Unfortunately, most patients rapidly acquire resistance to vemurafenib10, highlighting the urgent need for new treatment strategies of BRAF (V600E)-induced melanoma.

Restoration of the wild-type p53 tumor suppressor function has emerged as an attractive anticancer strategy11–13. However, usefulness of this approach in melanoma is unclear; although inactivating mutations or allelic loss of TP53 are common in human cancers14, the TP53 locus is intact in >95% of melanomas15. Nevertheless, increasing evidence supports a role for p53 in melanomagenesis, as loss of p53 cooperates with activated HRASV12G and BRAFV600E in mice16,17 and with oncogenic NRAS in zebrafish18, culminating in melanoma formation.

Cancers that retain wild-type p53 often find alternative ways to subvert p53 function, by deregulating upstream modulators and/or by inactivating downstream effectors19. For example, MDM2, which encodes an E3 ubiquitin ligase that controls p53 expression and function20, is amplified in 3–5% of human melanomas21.

Here we show that the p53 pathway is inactivated in the majority of cutaneous melanomas as a result of upregulated expression of MDM4, a key negative regulator of p53 (refs. 22,23). We show that targeting the MDM4-p53 interaction inhibits the growth of melanoma cells and markedly sensitizes them to conventional chemotherapeutics. Notably, MDM4 inhibition affects the growth of melanoma cells that have acquired resistance to BRAF inhibitors and synergizes with BRAF inhibitors to kill BRAF-mutant cells. These data identify the MDM4-p53 interaction as a new therapeutic target for melanoma treatment.

RESULTS

MDM4 is overexpressed in ∼65% of human melanomas

To investigate whether MDM4 overexpression contributes to p53 inactivation in human melanomas, we assessed MDM4 protein expression in normal skin and in melanomas by immunofluorescence. Although detectable in the cytoplasm of epidermal keratinocytes, MDM4 expression was very low to undetectable in normal melanocytes (Fig. 1a). Assessment of a panel of 54 primary cutaneous and metastatic melanomas showed variable MDM4 expression (from very low or undetectable to high), with MDM4 localizing to the nucleus in most cells (Fig. 1b). We observed that MDM4 expression was significantly increased in 68.5% of melanomas compared with normal melanocytes (Fig. 1c), including 56.7% of early stage (stage I or II) tumors and 83.3% of metastatic (stage III or IV) tumors (Fig. 1c). We found no correlation between age (calculated as age at resection of tumor) or pathologic subtype and MDM4 expression (Supplementary Table 1). There was, however, a statistically significant correlation between high MDM4 expression and late-stage (stage III or IV) metastatic disease (Supplementary Table 1).

Figure 1.

Figure 1

Open in a new tab

MDM4 is frequently overexpressed in human melanoma.(a) Immunofluorescence images of normal human skin co-stained for MDM4 (red) and the melanocyte marker Melan A (green). Nuclei are stained with DAPI (blue). High magnification images show that MDM4 was detected in the cytoplasm of keratinocytes (yellow arrows), but was very low to undetectable in normal melanocytes (white arrows). Scale bar, 25 μm. (b) Representative immunofluorescence staining of human melanomas showing low (top) and high (bottom) MDM4 expression. We confirmed melanoma status by Melan A (green) in the same section and S-100 in adjacent sections. Inset images in MDM4-high tumor show nuclear localization of MDM4 (yellow arrows) as determined by DAPI staining of the same section. Scale bars, 50 μm (10 μm in insets). (c) MDM4 protein expression, determined by immunofluorescence, in 54 human melanomas from clinical disease stages I–IV. Graphs represent the average MDM4 mean gray value (MGV) ± s.d. of at least 50 cells per sample. Asterisks show significantly increased MDM4 MGV compared to normal (N) melanocytes (*P < 0.01). (d) Immunoblotting analysis of expression levels of MDM2, MDM4 and p53 protein in human melanoma samples, in cancer cell lines and in congenital melanocytic nevi (CN). The samples are arranged into four clinically distinct subtypes: (primary cutaneous melanomas, regional dermal metastases, nodal metastases and distant metastases). MCF7, U2OS and SAOS-2 cells are reference controls and vinculin-specific immunoblotting is used to detect differences in sample loading.

We confirmed MDM4 overexpression in an additional cohort of 40 freshly isolated human melanomas by immunoblotting (Fig. 1d). Notably, in only 3 out of 40 of these samples were MDM4 mRNA levels comparable to or higher than those in the breast cancer cell line MCF-7, which is known to express high MDM4 mRNA levels (Supplementary Fig. 1)24. In contrast, MDM4 protein expression was comparable to or higher than that observed in MCF-7 cells in 65% of cases (Fig. 1d and Supplementary Table 2). Consistent with the immunofluorescence data, MDM4 protein expression was either undetectable or very low in normal melanocytes and in benign nevi (Fig. 1d and Supplementary Fig. 2a). Six out of ten primary cutaneous tumors had high MDM4 levels (Fig. 1d), supporting the possibility that MDM4 upregulation occurs early in melanomagenesis. In contrast, MDM2 protein expression levels ranged from undetectable to low in most cases (Fig. 1d). We only found MDM2 expression levels comparable to those in U2OS cells, an osteosarcoma cell line highly expressing MDM2, in one out of ten regional dermal metastases, one out of ten nodal metastases and four out of ten distant metastases (Fig. 1d and Supplementary Table 1). Overexpression of MDM2 and MDM4 co-occurred in only 2 out of 30 metastatic melanomas (stage IV) (Fig. 1d).

Compared with primary melanocytes, we found that MDM4 was also elevated in 14 out of 16 patient-derived short-term cultures established from metastatic tumors, as well as in four out of four cell lines (A375, WM9, Mel-501, Lu1205) harboring wild-type p53 (Supplementary Fig. 2b,c and Supplementary Table 3). Consistent with the notion of a post-transcriptional mechanism being primarily responsible for MDM4 upregulation, MDM4 mRNA levels were higher than those in MCF-7 in only one of these cell lines (MM120; Supplementary Fig. 1b). As in freshly isolated human melanoma samples, MDM2 protein expression levels ranged from undetectable to low in the majority of short-term cultures (Supplementary Fig. 2b,c). We found that MDM2 levels were comparable to those in U2OS cells in four cell lines (MM011, MM034, MM061, MM117) and higher in only two cell lines (MM001, MM120; Supplementary Table 3). Notably, MDM2 was highly expressed in all (four out of four) well-established (Supplementary Fig. 2c) melanoma cell lines, suggesting that extended in vitro passage might induce MDM2 expression.

We determined TP53, BRAF and NRAS mutational status in the primary tumors and cell lines described above (Supplementary Tables 2 and 3). Consistent with previous reports15,25, TP53 mutations were rare (Supplementary Tables 2 and 3). Notably, we found that MDM4 was overexpressed at comparable frequencies in tumors and cell lines harboring either wild-type or mutated NRAS or BRAF (Supplementary Tables 2 and 3).

MDM4 promotes melanoma in mice

Constitutive expression of an activated form of human NRAS (NRASQ61K) in melanocytes leads to formation of aggressive melanomas in Tyr::NRASQ61/o transgenic mice (mice that carry one copy of the NRASQ61K transgene)26; however, melanomas develop only after a long latency and with low penetrance. To assess the role of p53 in these mice, we inactivated Trp53 in previously characterized null (–) and conditional (lox) knockout alleles27,28. Melanocyte-specific deletion of one or two alleles of Trp53 cooperated with NRASQ61K to induce spontaneous and aggressive melanoma formation (ref. 29 and Supplementary Fig. 3). Although indicative of p53’s role as a barrier to melanomagenesis in vivo, these experiments do not recapitulate the genetic makeup of human melanomas, most of which express wild-type p53 (ref. 15).

We reasoned that MDM4 could compromise p53 function in vivo to promote melanoma formation. We therefore induced Mdm4 overexpression in melanocytes of Tyr::NRASQ61/o and Tyr::Cre/o; Trp53lox/− mice by crossing these mice with the previously described Mdm4 conditional transgenic mice ROSA26-pCAGGs-(loxP-STOP-loxP)-Mdm4/o (mice carrying one copy of an Mdm4 conditional transgenic allele, referred to as Mdm4/o)30(Fig. 2a). We crossed these double-transgenic mice onto a Trp53-heterozygous background to facilitate spontaneous tumor development. As expected from the Trp53 heterozygosity of Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; Mdm4/o mice, some mice developed T cell lymphomas and sarcomas27 (data not shown). However, the vast majority of mice developed melanoma before developing other tumors (data not shown). We restricted further analyses to mice that were lymphoma- and sarcoma-free and that developed melanomas at least 3-mm thick. Quadruple-transgenic mice (Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; Mdm4/o) developed at least one melanoma significantly faster than controls (Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; o/o)(Fig. 2b). Melanomas were nodular, pigmented and positive for S-100 protein expression31, regardless of genotype (Fig. 2c). Tumors arising in Tyr::NRASQ61/o; Tyr::Cre/o; Mdm4/o mice retained high exogenous (myc-tagged) Mdm4 expression (Fig. 2d,e). Exogenous Mdm4 mRNA and protein expression levels were perfectly correlated.

Figure 2.

Figure 2

Open in a new tab

Mdm4 overexpression promotes melanoma in mice. (a) Schematic representation of the conditional ROSA-26-Mdm4 transgenic allele as described previously30. A single copy of the Mdm4 transgene is inserted into the ROSA26 locus by homologous recombination. Before Cre-mediated excision the pCAGG promoter drives the expression of the β-geo STOP cassette only. Upon Cre expression the floxed β-geo STOP cassette is excised and Myc-Mdm4-IRES-eGFP is transcribed. (b) Penetrance of malignant melanoma upon melanocyte-specific Mdm4 overexpression. Kaplan-Meier curves showing increased melanoma incidence in Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; Mdm4/o mice by comparison with the group of controls (Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; o/o and Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/+; o/o). (c) External views and H&E staining (middle) and immunohistochemistry (IHC) for S-100 (bottom) in the skin of mice with the indicated genotypes. Scale bars, 500 μm. (d) Reverse-transcription quantitative PCR (RT-qPCR) analysis of Mdm4 expression in tumor samples derived from mice with the indicated genotypes (samples A–C from Tyr::NRASQ61/o; Tyr::Cre/o; Trp53lox/−; samples 1–5 from Tyr::NRASQ61/o; Tyr::Cre/o; Trp53lox/+; samples a–d from Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; o/o; samples I–XI from Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; Mdm4/o). One primer pair allows specific detection of Myc-tagged exogenous Mdm4 (myc-Mdm4; gray bars) and the other pair detects both endogenous and exogenous Mdm4 expression (black bars). Elevated Mdm4 expression is observed at the transcriptional level in samples from Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; Mdm4/o mice. The data represent the means ± s.d. of three technical replicates. (e) Western blot analysis confirming high Myc-tagged exogenous Mdm4 expression in a Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; Mdm4/o (blue Mdm4/o) melanoma lesion and not in lesions from Tyr::NRASQ61/o; Tyr::Cre/o; Trp53+/−; o/o (blue o/o) or Tyr::NRASQ61/o; Tyr::Cre/o; Trp53lox/+ (Trp53lox/+). As previously described30, high expression levels of Mdm4 are seen in the brain of Sox2Cre/o; Mdm4/o mice (green Mdm4/o). Endogenous Mdm4 expression is also detected in the brain of Sox2Cre/o; o/o mice (green o/o). p53 and p21 protein levels were also examined in these samples. Vinculin serves as a loading control. (f) Penetrance of malignant melanoma upon melanocyte-specific overexpression of Mdm4 on a Trp53 wild-type background. Kaplan-Meier curves showing increased melanoma incidence in Tyr::NRASQ61/o; Tyr::Cre/o; Mdm4/o mice in comparison with the control group (Tyr::NRASQ61/o; Tyr::Cre/o; o/o) (P = 0.0156). (g) Percentage of the number of melanoma lesions per mouse in Tyr::NRASQ61/o; Tyr::Cre/o; Mdm4/o (Mdm4/o mice) and in the control groups (Tyr::NRASQ61/o; Tyr::Cre/o; o/o and Tyr::NRASQ61/o; o/o; Mdm4/o). These percentages differed significantly between the two cohorts (P < 0.05).

We next assessed whether Mdm4 overexpression also enhances melanoma development on a wild-type Trp53 background. Tyr::NRASQ61/o; Tyr::Cre/o; Mdm4/o had a nearly 100% incidence of melanoma with a much shorter average latency period than controls (Tyr::NRASQ61/o; Tyr::Cre/o; o/o) (Fig. 2f). On average, there were more lesions per mouse in Mdm4-overexpressing mice than in controls (Fig. 2g). Together, these data indicate that Mdm4 over-expression cooperates with the NRAS oncoprotein to promote aggressive melanoma formation in vivo.

MDM4 promotes cell proliferation and survival of human melanoma

To test whether human metastatic melanoma cells depend on high MDM4 expression for their survival and proliferation we knocked down MDM4 in short-term cultured (MM011, MM031, MM117) and commonly used metastatic melanoma cell lines (A375, Lu1205). Our choice of cell lines was motivated by the presence of high protein expression levels of MDM4, a genetically intact TP53 pathway and different ratios of MDM4/MDM2 protein expression levels. We targeted different regions of MDM4 mRNA using two lentiviral constructs (data shown below and in Supplementary Fig. 4). MM031 expressed high MDM4 and low MDM2 protein expression levels, which best represents the vast majority of melanoma lesions; MM011 and MM117 highly express both MDM4 and MDM2, which is characteristic of only a few human distant-metastatic melanomas. Western blotting confirmed efficient MDM4 knockdown in all cell lines (Fig. 3a and Supplementary Fig. 5a). Cell viability and colony formation assays showed that MDM4 knockdown caused robust growth inhibition in all melanoma lines analyzed (Fig. 3b,c and Supplementary Fig. 5b,c).

Figure 3.

Figure 3

Open in a new tab

MDM4 promotes cell proliferation and survival of human melanoma. (a) Western blots of MDM4 and p53 protein expression levels in total protein lysates from metastatic melanoma cells infected with control knockdown (C), MDM4 knockdown (M), p53 knockdown (P), and both MDM4 knockdown and p53 knockdown (D) vectors, as well as in the MCF-7, U20S and SAOS-2 reference cell lines. The lines separate lanes that were not consecutive on the original western blot. (b) WST-1 assay of cell growth of the infected metastatic melanoma cells. Data represent the mean of three biological replicates. (c) Low-density colony formation assays 12 d after seeding, showing cell growth of the infected metastatic melanoma cells. The data are presented as the mean density of three different biological replicates ± s.d. (d) Fluorescence-activated cell sorting (FACS) analysis (dot plot) showing the percentage of BrdU-positive cells for one typical experiment in control knockdown (C), MDM4 knockdown (M), p53 knockdown (P) and MDM4/p53 double–knockdown (D) cultures. Data are from three independent infection experiments; C and M are significantly different (P values are indicated in the M dot plots). 7-AAD, 7-aminoactinomycin D. (e) Caspase3/7-Glo luminescence assays done in C, M, P and D cultures. The data are presented as the means of three different biological replicates ± s.d.; C and M and M and D are significantly different in all cell lines (P values are indicated on the graphs).

Cell cycle analysis showed a reproducible and significant decrease in BrdU incorporation upon MDM4 knockdown (Fig. 3d and Supplementary Fig. 5d). In addition to the reduced proliferation, increased cell death was evidenced by the increased sub-G1 fraction in the FACS profiles (data not shown). We found that cell death resulted, at least in part, from a reproducible and significant induction of caspase 3/7 activity (Fig. 3e and Supplementary Fig. 5).

Some of the parental melanoma cell lines (MM031, Lu1205, A375) efficiently formed tumors when injected into nude mice (8-week-old Rj:NMRI-nu (nu/nu) female mice). We show that MDM4 knockdown strongly reduces the transforming potential of these cells in vivo(Supplementary Figs. 5f and 6). These data indicate that MDM4 regulates the survival and proliferation of high-MDM4-expressing melanoma cells in vitro and in vivo.

MDM4 protects melanoma against p53-induced apoptosis

To test whether p53 inactivation underlies dependency of melanoma on high MDM4 expression, we knocked down both MDM4 and p53 in short-term cultured metastatic melanoma cell lines (MM011, MM031, MM117; Fig. 3a). Colony formation assays indicated that the MDM4 KD-mediated growth inhibition at least partly depends on p53 (Fig. 3c).

The two best-characterized p53-mediated antigrowth activities are induction of cell cycle arrest and apoptosis. Unexpectedly, p53 knockdown did not significantly rescue the proliferation of melanoma cells upon MDM4 knockdown (Fig. 3d). In contrast, p53 knockdown largely rescued MDM4 knockdown cells from apoptosis (Fig. 3e). Concomitant with apoptosis induction upon MDM4 knockdown, we observed increased recruitment of p53 to the promoters of its proapoptotic targets, BAX (encoding BCL2-associated X protein) and APAF1 (encoding apoptotic peptidase activating factor 1), and a marked increase in their transcription (Supplementary Fig. 7). The data suggest that high MDM4 levels inhibit apoptosis in human melanoma cells by antagonizing p53.

Inhibition of the MDM4-p53 interaction restores p53 activity

The reliance of melanoma cells on MDM4 for survival suggests that interfering with its interaction with p53 could restore apoptotic capability. SAH-p53-8 is a cell-penetrating, stabilized, α-helical peptide32 that binds with high affinity to MDM4 within its p53-binding pocket and is capable of disrupting p53-MDM4 complexes33. SAH-p53-8 has approximately 25-fold lower potency against MDM2 in vitro33. Unlike the MDM2-specific antagonist nutlin-3 (ref. 34), SAH-p53-8 can reactivate p53 and induce apoptosis in cancer cells that highly express MDM4 and wild-type p53 (ref. 33). To test whether direct inhibition of MDM4 is a viable therapeutic strategy for melanoma, we treated a series of melanoma cell lines, including the BRAF inhibitor–sensitive melanoma M238, M249 and M229 cell lines35, with increasing concentrations of SAH-p53-8. SAH-p53-8 potently inhibited growth of all melanoma lines carrying wild-type p53 (Fig. 4a and Supplementary Fig. 8). In contrast, melanoma cells with inactive p53 mutants did not respond to SAH-p53-8 treatment, indicating that SAH-p53-8-mediated growth inhibition depends on the presence of wild-type p53 (Fig. 4a). Moreover, SAH-p53-8 did not markedly affect normal melanocyte growth (Fig. 4a). SAH-p53-8, but not its biologically inactive point-mutant analog SAH-p53-8F19A, induces expression of p53-target genes, including BAX, APAF1 and BBC3 (also known as PUMA; encoding BLC2 binding component 3) in both MM031 and MM011 (Fig. 4). SAH-p53-8 induced apoptosis in all melanoma cell lines; this effect was significantly attenuated by knockdown or hypomorphic mutation of p53 (Fig. 4 and Supplementary Fig. 9).

Figure 4.

Figure 4

Open in a new tab

Inhibition of the MDM4-p53 interaction restores p53 activity in melanoma. (a) Cell viability of normal human melanocytes (N.M.) and melanoma cells harboring wild-type or mutant TP53 treated with increasing dose of SAH-p53-8 for 24 h. Data represent the mean ± s.d. of three biological replicates. (b–g) RT-qPCR mRNA expression analysis and Annexin V apoptosis assay. (b,d) mRNA expression of selected p53 target genes (CDKN1A, MDM2, BBC3, BAX, APAF1) in the MM031 (b) and MM011 (d) melanoma cell lines 24 h after SAH-p53-8 treatment (EC50 of SAH-p53-8 used per each cell line); SAH-p53-8F19A is used as a point mutant control and dimethyl sulfoxide (DMSO) as a vehicle control. The data are the means ± s.d. from three technical replicates. The values are normalized to the levels of mRNA expression in vehicle-treated cells (set to 1). (c,e–g) Annexin V apoptosis assay on MM031 and MM011 melanoma cells treated with vehicle, EC50 of SAH-p53-8 (16 μM for MM031 and 10 μM for MM011) or equivalent dose of nutlin-3 for 24 h. We incubated cells with Alexa Fluor 488–conjugated annexin V and propidium iodide (PI). (c,e) Mean percentage ± s.d. of apoptotic cells from three independent biological replicates; whereas we found that SAH-p53-8 (P = 0.01) but not nutlin-3 induced a significant apoptotic response in the MM031, MM011 responded to both agents (P = 0.008 and P = 0.02 for SAH-p53-8 and nutlin-3, respectively). (f,g) FACS data from a representative experiment are shown. Values represent the percentage of cells that are positive for either Annexin-5 or PI. (h,i) Viability of MM031 (h, left) and MM011 (i, left) melanoma cell lines treated with 0.5–20 μM SAH-p53-8, nutlin-3, or an equimolar combination for 24 h. Data are means ± s.d. for experiments done in at least triplicate. Dose-effect synergy analyses of MM031 (h, right) and MM011 (i, right) melanoma cells treated with 0.5–20 μM SAH-p53-8, nutlin-3, or an equimolar combination. The EC50 values for each treatment are indicated. (j) SAH-p53-8 overcomes MDM4-mediated p53 suppression and blocks tumor growth in vivo. Tumor volume of cohorts of MM031 xenograft mice treated with vehicle (5% (vol/vol) DMSO in D5W) or 10 mg kg−1 SAH-p53-8 by i.v. injection daily for 5 consecutive days. Data are means ± s.d. of seven different biological replicates (P < 0.01 on day 12). Bottom, external views of representative tumor-bearing mice on day 12. Scale bar, 1.5 cm.

Consistent with the fact that high MDM4 expression levels are associated with a reduced apoptotic response to MDM2 antagonists36,37, nutlin-3 induced relatively low apoptotic activity in MM011 cells despite high expression levels of MDM2 in these cells, and it induced practically no activity in MM031 cells (Fig. 4e–g).

We next treated melanoma cells with SAH-p53-8, nutlin-3 or a 1:1 stoichiometric combination of both compounds (Fig. 4h,i). As expected, nutlin-3 had only a marginal cytotoxic effect on MM031 cells. In contrast, we observed marked growth inhibition in MM031 cells in response to SAH-p53-8 alone (Fig. 4h), and we saw only a very modest increase in cytotoxicity upon combination with nutlin-3 (combination index = 0.90; combination index is calculated as described in ref. 33). MM011 cells, however, showed sensitivity to the single agents (Fig. 4i) and the two compounds synergize strongly, leading to enhanced cytotoxicity (combination index = 0.32). The mutant SAH-p53-8F19A peptide did not induce any measurable cytotoxic effects in any of the cell lines (Fig. 4h,i). Together our data indicate that all human melanoma cell lines analyzed are either highly sensitive to MDM4 inhibition either alone or in combination with MDM2 inhibition.

To test the therapeutic potential of inhibiting the p53-MDM4 interaction in vivo, we evaluated the activity of SAH-p53-8 in a MM031 mouse xenograft model. Although tumor growth rate was unaffected in vehicle-treated (vehicle was 5% (vol/vol) DMSO in D5W) mice, intravenous treatment with SAH-p53-8 significantly suppressed tumor growth (Fig. 4j). As previously shown, histological examination of SAH-p53-8-treated mice indicated no obvious toxicity33. Collectively, these data show the therapeutic potential of inhibiting the MDM4-p53 interaction in melanoma.

SAH-p53-8 sensitizes melanoma to chemotherapy and BRAF (V600E) inhibition

DNA-damaging agents such as Cisplatin or melphalan have yielded low response rates in the treatment of melanomas38. As these agents rely partly on activating the p53 pathway, we hypothesized that co-treatment with p53-MDM4 inhibitors may enhance their potency. We therefore investigated the effects of Cisplatin and melphalan alone or in combination with either nutlin-3 or SAH-p53-8 on MM011 and MM031 cell line growth. Although treatment with these agents alone yielded variable growth inhibition, cytotoxic effects were greatly potentiated by co-treatment with SAH-p53-8 (Fig. 5a–d and Supplementary Fig. 10). Transcriptional induction of p53 targets was more robust in cells treated with both Cisplatin and SAH-p53-8 than in cells treated with single agents (Fig. 5e,f). As expected, MM031 cells expressing low levels of MDM2 responded more favorably to the combination of alkylating agents with SAH-p53-8 than to co-treatment with nutlin-3 (Fig. 5a,c). In contrast, high expression levels of MDM2 rendered MM011 cells at least partly sensitive to nutlin-3 (Fig. 5b,d); identical results were obtained with melphalan (Supplementary Fig. 10). Collectively, these data suggest that targeting the MDM4-p53 interaction sensitizes melanoma cells to DNA-damaging agents.

Figure 5.

Figure 5

Open in a new tab

Therapeutic potential of targeting the MDM4-p53 pathway. (a–d) Targeting the MDM4-p53 interaction sensitizes metastatic melanoma cells to chemotherapy. We treated MM031 (a,c) and MM011 (b,d) melanoma cells with 6.25–50 μM Cisplatin with or without an EC50 dose of SAH-p53-8 or with or without a corresponding dose of nutlin-3. (a,b) Cell viability was measured at 24 h; data are means ± s.d. of at least three biological replicates. Cells treated with SAH-p53-8 show a statistically significant decrease in cell viability (P < 0.001) compared with cells treated with Cisplatin alone. (c,d) Dose effect synergy analyses of MM031 (c) and MM011 (d) melanoma cells treated with 6,25-50 μM Cisplatin with or without 0,5-20 μM SAH-p53-8 or nutlin-3. (e,f) RT-qPCR mRNA expression analysis of selected p53 target genes in the MM031 (e) and MM011 (f) melanoma cell lines 24 h after treatment. The data represent the means ± s.d. from three technical replicates. The values are normalized to the level of mRNA expression in vehicle-treated cells (set to 1). CDKN1A, MDM2 and BBC3 mRNA expression levels differed significantly in cells treated with both Cisplatin and SAH-p53-8 compared with cells treated with the single agents (P < 0.05) in both cell lines.

Overcoming widespread resistance to BRAF inhibitors39 is likely to require targeting of multiple signaling pathways. We investigated whether targeting the MDM4-p53 interaction could affect the growth of BRAF inhibitor–sensitive parental melanoma cell lines (M238 and M249) as well as BRAF-resistant sublines (M238R1 and M249R4) that were artificially derived by chronic exposure to a BRAF inhibitor35. All cell lines express high MDM4 protein expression levels and harbor wild-type p53. Treatment of both parental and BRAF-resistant cell lines with increasing doses of SAH-p53-8 led to decreased viability, indicating that cells that have acquired drug resistance to BRAF inhibitor remain largely sensitive to MDM4-p53 targeting (Fig. 6a). We saw a comparable increase in expression of p53 and its target p21 in the parental and resistant lines upon SAH-p53-8 exposure (Fig. 6b). Notably, SAH-p53-8 exposure led to a marked decrease in MDM4 protein expression levels (Fig. 6b).

Figure 6.

Figure 6

Open in a new tab

Targeting the MDM4-p53 pathway sensitizes melanoma cells to a BRAFV600E-inhibitor. (a) We exposed Isogenic BRAF inhibitor–sensitive parental and BRAF inhibitor–resistant cell lines to increasing doses of SAH-p53-8 (7.5–30 μM). We measured cell viability 24 h after SAH-p53-8 treatment. Data represent the means ± s.d. of three biological replicates. (b) MDM4, p53 and p21 protein levels assessed by western blotting analysis in nontreated (NT) and SAH-p53-8 (S)-treated human melanoma cells M238, M238R1, M249 and M249R4, which were described previously35. We used MCF7, U2OS and SAOS-2 cells as reference controls; vinculin is a loading control. P; parental, R; BRAF inhibitor–resistant subline. Long and Short indicate long and short exposure, respectively. (c) Dose-effect synergy analyses of melanoma cells treated with 5 μM PLX4032 and an EC50 dose of SAH-p53-8 (18 and 8.5 μM). Data are means ± s.d. of at least three biological replicates. Response to combination treatment differed significantly from the response to single agents (P < 0.01) in both cell lines.

To investigate whether combined BRAF (V600E) and MDM4 inhibition synergizes to induce cytotoxicity, we treated BRAF inhibitor–sensitive lines with a BRAF inhibitor, PLX4032, and with SAH-p53-8 individually or in combination (Fig. 6c). PLX4032 alone was sufficient to decrease the viability of these cells (Fig. 6c). Co-treatment with PLX4032 and SAH-p53-8 significantly enhanced the effect when compared with each individual compound (Fig. 6c).

DISCUSSION

Unlike in most other tumor types, TP53 mutations are rare in melanomas, making the relevance of the p53 pathway in melanomagenesis controversial. Given our observation of MDM4 overexpression in about 65% of melanomas, we contend that MDM4 overexpression is an important oncogenic event that alters p53 function in melanoma in a large proportion of patients.

Because MDM4 overexpression is primarily detected at the protein level, its detection inevitably escaped previous investigations that focused on transcriptomic analyses. This raises a possibility that MDM4 overexpression might be an even more frequent oncogenic event than previously anticipated in other tumor types as well23.

At physiological levels MDM4 serves as a constitutive buffer against untoward p53 activity23,40,41; its oncogenic activity results from its ability to attenuate p53 function23,24. Consistent with this, we found that most melanoma cells depend on high MDM4 expression to survive by keeping the proapoptotic functions of p53 in check. Furthermore, we found that high MDM4 expression levels are also required, in a p53-independent manner, by melanoma cells for proliferation in culture. This raises the possibility that MDM4 possesses additional oncogenic functions, independent of p53 suppression. Although p53-independent function(s) may contribute to MDM4-mediated oncogenesis, we found that targeting the MDM4-p53 interaction was sufficient to suppress the growth of metastatic melanoma cells with wild-type p53.

Restoration of p53 function has been extensively pursued as a therapeutic approach in cancers in which, as in most melanomas, p53 function is compromised although the TP53 locus remains intact42–44. Several efforts have focused on blocking MDM2 as a strategy to reactivate p53 (refs.34,45–48), despite several caveats49. One major limitation of MDM2-specific therapy is that tumor cells in which MDM4 expression is high but MDM2 is low respond poorly to MDM2 inhibition13,33. We found that most human melanoma cell lines express high MDM4 and low MDM2 levels and indeed respond very poorly to treatment with the MDM2-specific antagonist nutlin-3. In contrast, these cells are extremely sensitive to SAH-p53-8, thus indicating that pharmacological disruption of the MDM4-p53 interaction is required to induce p53 signaling in these cells. Note, however, that SAH-p53-8 can also suppress MDM2-p53 binding to some extent. It therefore cannot be ruled out that SAH-p53-8 effectiveness against melanoma cells with a high MDM4/MDM2 ratio depends, at least partly, on MDM2 targeting. Melanomas that express high levels of both MDM4 and MDM2 are rare, although we found that cell lines with this profile were sensitive to either SAH-p53-8 or nutlin-3 but even more sensitive to combination treatment. Therefore, whereas the majority of melanoma patients may respond poorly to single MDM2 inhibitors, they could benefit greatly from treatment with single MDM4 or dual MDM4/MDM2 antagonists.

Recent data argue that MDM4 inhibition is a more feasible and less hazardous strategy than MDM2 inhibition for restoring p53 function. For instance, genetic ablation of Mdm2 in normal adult tissues, including those that are quiescent and fully differentiated, leads to pathologies because of the induction of p53-dependent cell death20. Hence, toxicity after systemic exposure to potent MDM2 inhibitors is a serious concern. In contrast, ablation of Mdm4 in most adult somatic tissues leads to either modest or no phenotypic consequences40,41.

Most patients with metastatic melanoma do not respond to conventional chemotherapy regimens50. Our results show that inhibition of the MDM4-p53 interaction greatly potentiates the cytotoxic effects of two widely used chemotherapeutic agents. MDM4 inhibitors could therefore be used to enhance the effectiveness of conventional therapeutic agents in the treatment of melanoma.

Similarly, despite the initial success with BRAF inhibitors, chronic treatment is invariably associated with the development of drug resistance39. Management of BRAF inhibitor resistance is likely to be achieved through combination therapy that targets multiple pathways. We have shown that MDM4 is overexpressed in melanomas harboring mutations in BRAF and NRAS alike, that MDM4-p53 inhibition equally affects the growth of NRAS- and BRAF-mutant melanoma cells and, notably, that MDM4-p53 inhibition is equally effective at inhibiting growth of BRAF-mutant cells before or after they acquire resistance to BRAF inhibitors. We also show that combined treatment with MDM4-p53 inhibitors and BRAF inhibitors synergizes to kill melanoma cells that are sensitive to BRAF inhibition.

Our findings suggest that MDM4-p53 targeting offers a promising approach to improve the clinical benefits of BRAF inhibition. Crucially, our in vitro results predict that MDM4-p53 targeted therapy may be effective in cases when BRAF inhibitor resistance has already been acquired and in melanomas with NRAS mutations, for which no specific inhibitors exist. Prioritization should therefore be given to the preclinical development of strategies that inhibit MDM4-mediated suppression of p53 function in human melanoma.

ONLINE METHODS

Cell culture

MM001, MM011, MM031 and MM117 cells were cultured in F10 medium with 5% FBS (FBS, HyClone) and 5% calf bovine serum (HyClone). A375 and Lu1205 cells were cultured in, respectively, RPMI and DMEM medium with 10% FBS (Sigma). M238, M248, M229 and their BRAF inhibitor resistant sublines were cultured in DMEM plus 5% FBS (HyClone).

Viral transduction

To knock down MDM4 we used two different lentiviral shRNA expression vectors (M:5′-GTGCAGAGGAAAGTTCCAC and M(2): 5′-ACAGTCCTTCAGCTATTT). One of these vectors was described previously51 and the other was obtained from the Mission shRNA library (Sigma). The p53 knockdown and shRNA control vectors have also been described previously51.

Cell growth and viability

Cell growth was measured using the WST-1 (Roche) or CellTiter-Glo (Promega) assays. Cells were seeded in triplicate in 96-well plates, and treated with chemotherapeutics: Cisplatin (Sigma) or melphalan (Sigma), SAH-p53-8 and SAH-p53-8F19A (ref. 33), BRAF inhibitor (PLX4032, Selleck Chemicals) or nutlin-3 (Johnson & Johnson) for 24 h. Luminescence was measured in a microplate reader (Victor; PerkinElmer).

Colony-formation assays

Cells were plated at a density between 2 × 103 and 32 × 103 cells per six-well plate and cultured for 12 days. The cells were washed with PBS 1×, fixed and stained 5 min with a 1% crystal violet in 35% methanol solution.

Flow cytometry and apoptosis assays

Cells were incubated for 2 h with BrdU (final concentration 10 μM), washed in PBS and fixed in ice-cold 70% ethanol. Cells were labeled for BrdU incorporation with an APC BrdU flow kit (BD Pharmingen no. 557892), according to the manufacturer’s protocol and resus-pended in 300 ml of PBS containing 20 ml of 7-AAD. For the annexin V apoptosis assay, melanoma cells were plated at a density of 1.5 × 106 per well in a six-well plate and were treated with SAH-p53-8 or nutlin-3 for 24 h in Opti-MEM. Cells were harvested, washed with PBS, and incubated with Alexa Fluor 488–conjugated annexin V and propidium iodide. Flow cytometry was done in the FACSCanto (BD Biosciences). Apoptosis was also measured using the Caspase-Glo 3/7 assay (Promega).

Western blotting analysis

Cells were lysed in Giordano buffer (50 mM Tris-HCl (pH 7.4); 250 mM NaCl; 0,1% Triton X-100; 5 mM EDTA) containing phosphatase and protease inhibitors (Sigma). 40 mg of each sample was fractionated by SDS-PAGE (Invitrogen; NuPAGE Novex 4–12% Bis-Tris gel) and transferred to a PVDF membrane (iBlot dry blotting system). Membranes were blocked in Tris-buffered saline, 0.2% Tween-20 (TBST) containing 5% or 10% nonfat dry milk, and were subsequently incubated with the appropriate primary antibody. Membranes were subsequently incubated with either horseradish peroxidase–conjugated horse mouse-specific or goat rabbit-specific secondary antibody (Cell Signaling). Proteins were detected by enhanced chemiluminescence (ECL) western blotting detection reagents (Amersham Biosciences) or SuperSignal West Femto maximum sensitivity substrate (Thermo Scientific). The following primary antibodies were used: rabbit MDM4-specific (IHC-00108, Bethyl Laboratories, 1:5000), mouse Mdm2-specific (mixture of a 1:10 dilution of the homemade mouse monoclonal antibodies 2A10 and 4B2 or SMP14, Santa Cruz Biotechnology, sc-965, 1:1000), mouse p53-specific (DO-1, Santa Cruz Biotechnology, sc-126, 1:500), mouse p21-specific (F-5, Santa Cruz Biotechnology, sc-6246, 1:500), mouse β-tubulin–specific (Sigma-Aldrich, T2200, 1:10000), mouse vinculin-specific (Sigma-Aldrich, V9131, 1:10000).

Quantitative real-time PCR

RNA was isolated using the RNeasy minikit (Qiagen) according to the manufacturer’s protocol. The RNA was quantified using a NanoDrop 1000 (Thermo Scientific). 2 mg of total RNA was reverse-transcribed using the High-Capacity cDNA reverse transcription kit (Applied Biosystems). Quantitative reverse-transcriptase PCR (RT-qPCR) assays were done using Fast SYBR Green 2× master mix according to the manufacturer’s instructions (Applied Biosystems). For normalization, the geometric mean of at least three reference genes was used. For mouse samples, TaqMan probes were designed by Applied Biosystems. For human samples the primers used were as follows: APAF1 (Fwd, 5′-CCTGTTGTCTCTTCTTCCAGTGT-3′, Rev, 5-AAAACAACTGGCCTCTGTGG-3′), BAX (Fwd, 5′-ATGTTTTCTGA CGGCAACTTC-3′, Rev, 5-ATCAGTTCCGGCACCTTG-3′), MDM4 (Fwd, 5′-AGGTGCGCAAGGTGAAATGT-3′, Rev, 5-CCATATGCTGCTCCTG CTGAT-3′), MDM2 (Fwd, 5′-AGGAGATTTGTTTGGCGTGC-3′, Rev, 5-TGAGTCCGATGATTCCTGCTG-3′), BBC3 (Fwd, 5′-GACCTCAAC GCACAGTA-3′, Rev, 5-CTAATTGGGCTCCATCT-3′), CDKN1A (Fwd, 5′-AGCAGAGGAAGACCATGTGGA-3′, Rev, 5-AATCTGTCATGCTGGT CTGCC-3′). The following reference genes were used: GAPDH (Fwd, 5′-TG CCATGTAGACCCCTTGAAG-3′, Rev, 5-ATGGTACATGACAAGGTGC GG-3′), HMBS (Fwd, 5′-GGCAATGCGGCTGCAA-3′, Rev, 5-GGGTACC CACGCGAATCAC-3′), RPL13a (Fwd, 5′-CCTGGAGGAGAAGAGGAAA GAGA-3′, Rev, 5-TTGAGGACCTCTGTGTATTTGTCAA-3′), TBP (Fwd, 5′-CGGCTGTTTAACTTCGCTTC-3′, Rev, 5′-CACACGCCAAGAAACAG TGA-3′). Gene expression levels and errors in the gene expression levels were calculated using qBasePLUS 1.0 analysis software52.

Determining BRAF, NRAS and p53 status

200 ng of cDNA was used as a template for a PCR reaction using the following primers: BRAF (Fwd, 5′-AGCACCTACACCTCAGCAGTTACA-3′, Rev, 5′-ACAGGTATCC TCGTCCCACCATAA-3′), NRAS (Fwd, 5′-ACAAACTGGTGGTGGTTG GA-3′, Rev, 5′-TGGCCATCCCATACAACCCT-3′). The purified PCR products were sequenced using following nested primers: Primer BRAF (Nested) 5′-AGGGCATGGATTACTTACACGCCA-3′, Primer NRAS (Nested) 5′-ACTCGCTTAATCTGCTCCCTGT-3′. Genomic DNA was screened for TP53 mutations at IARC. Exons 2–10 of TP53 were analyzed by PCR and direct sequencing as described in the protocol available at (http://www-p53.iarc.fr/Download/TP53_DirectSequencing_IARC.pdf). Mutations were screened on both DNA strands and were confirmed in an independent PCR product. The effect of mutations on protein function was determined according to annotations provided in the IARC TP53 database (http://www-p53.iarc.fr)53.

Histology and IHC

Tissues were fixed overnight in 4% paraformaldehyde, dehydrated, paraffin embedded, sectioned (6 mm) and stained with H&E. For IHC, slides were bleached for 5 h in 10% H2O2 solution and stained with antibodies against S-100 (rabbit, Z0311, 1:300; Dako). Detection was done with the secondary goat rabbit-specific antibody (E0432, 1:500; Dako) combined with the incubation in an ABC reagent (Vector). Sections were counterstained with hematoxylin.

Immunofluorescence

Ethics approval was obtained from the Peter MacCallum Cancer Centre Ethics Committee (10/02) for all human tissues and clinicopatho-logical data used in this study. Informed consent was obtained for all human tissues. Formalin-fixed, paraffin-embedded tissue sections used for immune-fluorescence analyses were supplied by TissuePath (normal) and the Melbourne Melanoma Project (melanomas). Informed consent was obtained for all human tissues. Tissue sections of 4 mm were dewaxed and antigens were retrieved at 125 °C for 3 min in epitope retrieval solution, pH 9 (Novocastra). For MDM4/Melan A coimmunofluorescence in melanomas, sections were stained on a Dako autostainer. Sections were co-incubated with antibodies against MDM4 (rabbit, IHC-00108, Bethyl Laboratories, 0.5 μg ml−1) and Melan A (mouse, M2-9E3, Santa Cruz Biotechnologies, 2 μg ml−1) for 2 h at room temperature before detection with rabbit-specific and mouse-specific IgG2b Alexa Fluor secondary antibodies (Molecular Probes; 1:500) for 1 h. Nuclei were counterstained with DAPI (Roche; 20 μg ml−1). MDM4 staining in normal skin was imaged on a Leica SP5 confocal microscope. For image acquisition and analysis of MDM4 expression in melanomas, representative fields were imaged on an Olympus BX-61 microscope at the same exposure and camera settings (including normal skin as control). At least 50 melanoma cells (or normal melanocytes) per sample were individually assessed for mean gray value (MGV) (measure of fluorescence) using ImageJ analysis software by drawing a single line through each cell intercepting the brightest point. MDM4 overexpression was defined as any melanoma sample showing a statistically significantly increase (P < 0.01) in MDM4 MGV compared to normal melanocytes, as determined by an unpaired Student’s t-test (two-tailed).

Mice

All animal experiments were carried out in accordance with the guidelines of the University of Leuven Animal Care and Use ethical Committee. All experiments were carried out using pure C57BL/6 males housed in the KU Leuven conventional mouse facility. The subsequent primer sequences were used for genotyping: Trp53-null: (Fwd, 5′-ACACACCTGTAGCTCCAGCAC-3′, Rev, (in exon 5) 5′-AGCGTCTCACGACCTCCGTC-3′, Rev, (in the probe) 5′-GTGTTCCGGCTGTCAGCGCA-3′), Trp53-floxed (Fwd, 5′-AAGGG GTATGAGGGACAAGG-3′, Rev, 5′-GAAGACAGAAAAGGGGAGGG-3′), Tyr::NRASQ61K (Fwd, 5′-GGCGAAGGCTTCCTCTGTGT-3′, Rev, 5′-GGCC AGTTCGTGGGCTTGT-3′), Tyr::Cre (Fwd, 5′-GTCACTCCAGGGGTTGC TGG-3′, Rev, 5′-CCGCCGCATAACCAGTGA-3′. Genotyping of Mdm4/o and Sox2-Cre mice was done as previously described30,54.

Xenograft experiments

Eight-week-old Rj:NMRI-nu (nu/nu) female mice were injected subcutaneously with human cell lines in sterile PBS. MM031 xenografts were established by injecting 107 cells, whereas Lu1205 and A375 xenografts were established by injecting 106 cells. For the knockdown experiments, tumor growth was monitored with a caliper twice a week, and the volume was calculated using the following formula V = a × b2 × 0.5, where a is the largest and b the smallest diameter of the tumor. The SAH-p53-8 experiment was done on MM031 cell line–derived tumors with an average volume of 200 mm3. Cohorts (n = 7) were treated with vehicle (5% DMSO in D5W) or SAH-p53-8 (10 mg kg−1), once daily for 5 consecutive d by i.v. injection, and tumor volume was monitored daily for a period of 12 d.

Statistical analyses

Numerical values are reported as means ± s.d. For comparisons between two groups, P values were calculated using either paired or an unpaired two-tailed Student’s t tests. For the Kaplan-Meier curves, P values were determined with a log-rank Mantel-Cox test. P values of 0.05 or less were considered statistically significant.

Supplementary Material

Supporting Data

Acknowledgments

We thank O.Van Goethem for excellent technical assistance. We thank P. Agostinis, P. Wolter and M. Skipper for helpful discussions and comments on the manuscript. We thank M. Cario-Andre and A. Taïed for materials from human nevi. A. Zwolinska is a recipient of a ‘Het Fonds Wetenschappelijk Onderzoek-Vlaanderen (FWO)’ scholarship. C. Fedele was supported by a Clare Oliver Memorial Fellowship from the Victorian Cancer Agency. M. Shackleton was supported by fellowships from Pfizer Australia and the Victorian Endowment for Science, Knowledge and Innovation (VESKI). Y. and S. Haupt were supported by the Australian National Health and Medical Research Council (nos. 509197, 1026990 and 628426) and VESKI. R.S. Lo was supported by Stand Up to Cancer, the Joint Center for Translational Medicine, the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research, and the Seaver Institute. This work was supported by the Intramural Research Program of the US National Institutes of Health and the National Cancer Institute, the Association for International Cancer Research (AICR), the Melbourne Melanoma Project, the Victorian Cancer Agency and the ‘Belgian Foundation against Cancer’.

Footnotes

AUTHOR CONTRIBUTIONS A.G. did experimental work, developed the hypothesis, analyzed the data and coordinated the project. F.L. did experimental work and analyzed the data. C.F. conducted immunofluorescence analyses in normal human skin and melanomas and analyzed the data. E.A.R. conducted cellular assays. M.D. conducted experimental work. S.V. determined the p53 status of melanoma cell lines and primary tumors. A.Z. did the chromatin immunoprecipitation experiments and analyzed the data. S.H. contributed to the development of MDM4 immunohistochemistry. J.d.L. generated MDM4 knockdown lentiviral vectors. D.Y. and J.G. obtained primary human melanoma samples. J.J.H. contributed to the mouse work and experimental design. H.S. and G.M. generated and characterized BRAF inhibitor–resistant cell lines. F.B. produced and supplied SAH-p53-8 and SAH-p53-8F19A. Y.H., L.L., A.J., R.S.L., G.G., M.S. and F.B. discussed the hypothesis and contributed to data interpretation and experimental design. J.-C.M. conceived the hypothesis, led the project, interpreted the data and wrote the manuscript.

COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests.

Note: Supplementary information is available in the online version of the paper.

References

  • 1.Chin L, Garraway LA, Fisher DE. Malignant melanoma: genetics and therapeutics in the genomic era. Genes Dev. 2006;20:2149–2182. doi: 10.1101/gad.1437206. [DOI] [PubMed] [Google Scholar]
  • 2.Ibrahim N, Haluska FG. Molecular pathogenesis of cutaneous melanocytic neoplasms. Annu Rev Pathol. 2009;4:551–579. doi: 10.1146/annurev.pathol.3.121806.151541. [DOI] [PubMed] [Google Scholar]
  • 3.Cohen C, et al. Mitogen-actived protein kinase activation is an early event in melanoma progression. Clin Cancer Res. 2002;8:3728–3733. [PubMed] [Google Scholar]
  • 4.Demunter A, Stas M, Degreef H, De Wolf-Peeters C, van den Oord JJ. Analysis of N- and K-ras mutations in the distinctive tumor progression phases of melanoma. J Invest Dermatol. 2001;117:1483–1489. doi: 10.1046/j.0022-202x.2001.01601.x. [DOI] [PubMed] [Google Scholar]
  • 5.Papp T, et al. Mutational analysis of the N-ras, p53, p16INK4a, CDK4, and MC1R genes in human congenital melanocytic naevi. J Med Genet. 1999;36:610–614. [PMC free article] [PubMed] [Google Scholar]
  • 6.Davies H, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417:949–954. doi: 10.1038/nature00766. [DOI] [PubMed] [Google Scholar]
  • 7.Gray-Schopfer VC, da Rocha Dias S, Marais R. The role of B-RAF in melanoma. Cancer Metastasis Rev. 2005;24:165–183. doi: 10.1007/s10555-005-5865-1. [DOI] [PubMed] [Google Scholar]
  • 8.Flaherty KT, et al. Inhibition of mutated, activated BRAF in metastatic melanoma. N Engl J Med. 2010;363:809–819. doi: 10.1056/NEJMoa1002011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Chapman PB, et al. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med. 2011;364:2507–2516. doi: 10.1056/NEJMoa1103782. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Nazarian R, et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature. 2010;468:973–977. doi: 10.1038/nature09626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Lane DP, Cheok CF, Lain S. p53-based cancer therapy. Cold Spring Harb Perspect Biol. 2010;2:a001222. doi: 10.1101/cshperspect.a001222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Brown CJ, Lain S, Verma CS, Fersht AR, Lane DP. Awakening guardian angels: drugging the p53 pathway. Nat Rev Cancer. 2009;9:862–873. doi: 10.1038/nrc2763. [DOI] [PubMed] [Google Scholar]
  • 13.Wade M, Wahl GM. Targeting Mdm2 and Mdmx in cancer therapy: better living through medicinal chemistry? Mol Cancer Res. 2009;7:1–11. doi: 10.1158/1541-7786.MCR-08-0423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Vousden KH, Prives C. Blinded by the light: the growing complexity of p53. Cell. 2009;137:413–431. doi: 10.1016/j.cell.2009.04.037. [DOI] [PubMed] [Google Scholar]
  • 15.Chin L. The genetics of malignant melanoma: lessons from mouse and man. Nat Rev Cancer. 2003;3:559–570. doi: 10.1038/nrc1145. [DOI] [PubMed] [Google Scholar]
  • 16.Bardeesy N, et al. Dual inactivation of RB and p53 pathways in RAS-induced melanomas. Mol Cell Biol. 2001;21:2144–2153. doi: 10.1128/MCB.21.6.2144-2153.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Goel VK, et al. Melanocytic nevus-like hyperplasia and melanoma in transgenic BRAFV600E mice. Oncogene. 2009;28:2289–2298. doi: 10.1038/onc.2009.95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Dovey M, White RM, Zon LI. Oncogenic NRAS cooperates with p53 loss to generate melanoma in zebrafish. Zebrafish. 2009;6:397–404. doi: 10.1089/zeb.2009.0606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature. 2000;408:307–310. doi: 10.1038/35042675. [DOI] [PubMed] [Google Scholar]
  • 20.Marine JC, Lozano G. Mdm2-mediated ubiquitylation: p53 and beyond. Cell Death Differ. 2010;17:93–102. doi: 10.1038/cdd.2009.68. [DOI] [PubMed] [Google Scholar]
  • 21.Muthusamy V, et al. Amplification of CDK4 and MDM2 in malignant melanoma. Genes Chromosom. Cancer. 2006;45:447–454. doi: 10.1002/gcc.20310. [DOI] [PubMed] [Google Scholar]
  • 22.Marine JC, Jochemsen AG. Mdmx as an essential regulator of p53 activity. Biochem Biophys Res Commun. 2005;331:750–760. doi: 10.1016/j.bbrc.2005.03.151. [DOI] [PubMed] [Google Scholar]
  • 23.Toledo F, Wahl GM. Regulating the p53 pathway: in vitro hypotheses, in vivo veritas. Nat Rev Cancer. 2006;6:909–923. doi: 10.1038/nrc2012. [DOI] [PubMed] [Google Scholar]
  • 24.Danovi D, et al. Amplification of Mdmx (or Mdm4) directly contributes to tumor formation by inhibiting p53 tumor suppressor activity. Mol Cell Biol. 2004;24:5835–5843. doi: 10.1128/MCB.24.13.5835-5843.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Houben R, et al. High-level expression of wild-type p53 in melanoma cells is frequently associated with inactivity in p53 reporter gene assays. PLoS ONE. 2011;6:e22096. doi: 10.1371/journal.pone.0022096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ackermann J, et al. Metastasizing melanoma formation caused by expression of activated N-RasQ61K on an INK4a-deficient background. Cancer Res. 2005;65:4005–4011. doi: 10.1158/0008-5472.CAN-04-2970. [DOI] [PubMed] [Google Scholar]
  • 27.Jacks T, et al. Tumor spectrum analysis in p53-mutant mice. Curr Biol. 1994;4:1–7. doi: 10.1016/s0960-9822(00)00002-6. [DOI] [PubMed] [Google Scholar]
  • 28.Jonkers J, et al. Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer. Nat Genet. 2001;29:418–425. doi: 10.1038/ng747. [DOI] [PubMed] [Google Scholar]
  • 29.Ferguson B, et al. Differential roles of the pRb and Arf/p53 pathways in murine naevus and melanoma genesis. Pigment Cell Melanoma Res. 2010;23:771–780. doi: 10.1111/j.1755-148X.2010.00752.x. [DOI] [PubMed] [Google Scholar]
  • 30.De Clercq S, et al. Widespread overexpression of epitope-tagged Mdm4 does not accelerate tumor formation in vivo. Mol Cell Biol. 2010;30:5394–5405. doi: 10.1128/MCB.00330-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Ohsie SJ, Sarantopoulos GP, Cochran AJ, Binder SW. Immunohistochemical characteristics of melanoma. J Cutan Pathol. 2008;35:433–444. doi: 10.1111/j.1600-0560.2007.00891.x. [DOI] [PubMed] [Google Scholar]
  • 32.Bernal F, Tyler AF, Korsmeyer SJ, Walensky LD, Verdine GL. Reactivation of the p53 tumor suppressor pathway by a stapled p53 peptide. J Am Chem Soc. 2007;129:2456–2457. doi: 10.1021/ja0693587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Bernal F, et al. A stapled p53 helix overcomes HDMX-mediated suppression of p53. Cancer Cell. 2010;18:411–422. doi: 10.1016/j.ccr.2010.10.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Vassilev LT, et al. In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science. 2004;303:844–848. doi: 10.1126/science.1092472. [DOI] [PubMed] [Google Scholar]
  • 35.Nazarian R, et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature. 2010;468:973–977. doi: 10.1038/nature09626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Patton JT, et al. Levels of HdmX expression dictate the sensitivity of normal and transformed cells to Nutlin-3. Cancer Res. 2006;66:3169–3176. doi: 10.1158/0008-5472.CAN-05-3832. [DOI] [PubMed] [Google Scholar]
  • 37.Wade M, Wong ET, Tang M, Stommel JM, Wahl GM. Hdmx modulates the outcome of p53 activation in human tumor cells. J Biol Chem. 2006;281:33036–33044. doi: 10.1074/jbc.M605405200. [DOI] [PubMed] [Google Scholar]
  • 38.Chapman PB, et al. Phase III multicenter randomized trial of the Dartmouth regimen versus dacarbazine in patients with metastatic melanoma. J Clin Oncol. 1999;17:2745–2751. doi: 10.1200/JCO.1999.17.9.2745. [DOI] [PubMed] [Google Scholar]
  • 39.Fedorenko IV, Paraiso KH, Smalley KS. Acquired and intrinsic BRAF inhibitor resistance in BRAF V600E mutant melanoma. Biochem Pharmacol. 2011;82:201–209. doi: 10.1016/j.bcp.2011.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Garcia D, et al. Validation of MdmX as a therapeutic target for reactivating p53 in tumors. Genes Dev. 2011;25:1746–1757. doi: 10.1101/gad.16722111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Marine JC, Dyer MA, Jochemsen AG. MDMX: from bench to bedside. J Cell Sci. 2007;120:371–378. doi: 10.1242/jcs.03362. [DOI] [PubMed] [Google Scholar]
  • 42.Martins CP, Brown-Swigart L, Evan GI. Modeling the therapeutic efficacy of p53 restoration in tumors. Cell. 2006;127:1323–1334. doi: 10.1016/j.cell.2006.12.007. [DOI] [PubMed] [Google Scholar]
  • 43.Ventura A, et al. Restoration of p53 function leads to tumour regression in vivo. Nature. 2007;445:661–665. doi: 10.1038/nature05541. [DOI] [PubMed] [Google Scholar]
  • 44.Xue W, et al. Senescence and tumour clearance is triggered by p53 restoration in murine liver carcinomas. Nature. 2007;445:656–660. doi: 10.1038/nature05529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Canner JA, et al. MI-63: a novel small-molecule inhibitor targets MDM2 and induces apoptosis in embryonal and alveolar rhabdomyosarcoma cells with wild-type p53. Br J Cancer. 2009;101:774–781. doi: 10.1038/sj.bjc.6605199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Hedström E, Issaeva N, Enge M, Selivanova G. Tumor-specific induction of apoptosis by a p53-reactivating compound. Exp Cell Res. 2009;315:451–461. doi: 10.1016/j.yexcr.2008.11.009. [DOI] [PubMed] [Google Scholar]
  • 47.Issaeva N, et al. Small molecule RITA binds to p53, blocks p53-HDM-2 interaction and activates p53 function in tumors. Nat Med. 2004;10:1321–1328. doi: 10.1038/nm1146. [DOI] [PubMed] [Google Scholar]
  • 48.Shangary S, et al. Reactivation of p53 by a specific MDM2 antagonist (MI-43) leads to p21-mediated cell cycle arrest and selective cell death in colon cancer. Mol Cancer Ther. 2008;7:1533–1542. doi: 10.1158/1535-7163.MCT-08-0140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Marine JC. MDM2 and MDMX in cancer and development. Curr Top Dev Biol. 2011;94:45–75. doi: 10.1016/B978-0-12-380916-2.00003-6. [DOI] [PubMed] [Google Scholar]
  • 50.Flaherty KT. Chemotherapy and targeted therapy combinations in advanced melanoma. Clin Cancer Res. 2006;12:2366s–2370s. doi: 10.1158/1078-0432.CCR-05-2505. [DOI] [PubMed] [Google Scholar]
  • 51.Lam S, et al. Role of Mdm4 in drug sensitivity of breast cancer cells. Oncogene. 2010;29:2415–2426. doi: 10.1038/onc.2009.522. [DOI] [PubMed] [Google Scholar]
  • 52.Vandesompele J, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 2002;3 doi: 10.1186/gb-2002-3-7-research0034. RESEARCH0034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Petitjean A, et al. Impact of mutant p53 functional properties on TP53 mutation patterns and tumor phenotype: lessons from recent developments in the IARC TP53 database. Hum Mutat. 2007;28:622–629. doi: 10.1002/humu.20495. [DOI] [PubMed] [Google Scholar]
  • 54.Migliorini D, et al. Mdm4 (Mdmx) regulates p53-induced growth arrest and neuronal cell death during early embryonic mouse development. Mol Cell Biol. 2002;22:5527–5538. doi: 10.1128/MCB.22.15.5527-5538.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Data
NIHMS409993-supplement-Supporting_Data.pdf (8.4MB, pdf)

RESOURCES