MDM4 (Murine Double Minute 4)
Mouse Mdm4 (murine double minute 4) was first identified in an attempt to isolate a novel human p53-binding protein by screening a mouse cDNA library (Shvarts et al. 1996). Subsequently, its human homolog, MDM4, was cloned through a screen that used mouse Mdm4 cDNA as probe (Shvarts et al. 1997). Human MDM4 protein is 90% similar to mouse Mdm4 (Shvarts et al. 1997), with the greatest homology in the N-terminal p53-binding domain, and is ubiquitously expressed in tissues and organs (Shvarts et al. 1996; Shvarts et al. 1997). Importantly, MDM4 is homologous to MDM2, a key E3 ubiquitin ligase that negatively regulates the tumor suppressor p53 (Shvarts et al. 1996) (see also the MDM2 section). Therefore, overexpression of MDM4 leads to the suppression of p53 activity (Shvarts et al. 1996). However, subsequent studies found that unlike MDM2, MDM4 does not have intrinsic E3 ubiquitin ligase activity (Stad et al. 2000). Moreover, unlike MDM2, MDM4 expression is not significantly induced by p53 activation (Shvarts et al. 1996).
There are several splicing isoforms of the MDM4 protein (Rallapalli et al. 1999). The most abundant and evolutionally conserved splice variant is MDM4-S, a 127-amino acid protein that omits exon 6 (exon 7 in mice) (Rallapalli et al. 1999). The deletion produces a frameshift and results in a truncated protein that contains only the p53-binding domain (Rallapalli et al. 1999). Unlike the full-length MDM4, MDM4-S shows nuclear localization and exhibits higher binding affinity for p53 (Rallapalli et al. 1999). When ectopically overexpressed, MDM4-S is able to repress p53 transcriptional activity more efficiently than full-length MDM4 can (Rallapalli et al. 1999, 2003), suggesting higher oncogenic potential of MDM4-S than that of its full-length counterpart. Indeed, high levels of MDM4-S have been reported in several cancer types, including soft-tissue sarcomas (Bartel et al. 2005). However, transgenic mice with the deletion of exon 6, thereby solely expressing Mdm4-S, showed elevated p53 activity, a phenotype analogous to Mdm4 knockout mice (Bardot et al. 2015). Moreover, forced induction of MDM4-S expression by antisense oligonucleotide-mediated exon 6 skipping suppressed the growth of tumor xenografts (Dewaele et al. 2016). Therefore, the physiological role of MDM4-S in cellular homoeostasis and tumorigenesis remains elusive.
The splice variant MDM4-A lacks the majority of the acidic domain, whereas the variant MDM4-G has an in-frame deletion of the p53-binding domain (de Graaf et al. 2003). The variant MDM4-211 only consists of the first 26 and last 138 amino acids of the full-length MDM4 protein (Giglio et al. 2005). Although MDM4-211 cannot bind p53 due to the loss of its p53-binding domain, it binds to and stabilizes MDM2 via the RING–RING interaction, suggesting an oncogenic function of this variant (Giglio et al. 2005). MDM4-XALT1 and XALT2 are products of alternative splicing induced by UV irradiation (Chandler et al. 2006). XALT1 retains only the p53-binding domain, whereas XALT2 lacks the p53-binding domain but contains the C-terminal RING-finger domain (Chandler et al. 2006). It remains to be fully elucidated how these splice variants play a role in regulating p53 activity.
MDM4 is largely localized in the cytoplasm. In response to genotoxic stimuli, MDM4 translocates to the nucleus where it binds to and inhibits p53. The RING-mediated interaction with MDM2 plays a pivotal role in the translocation of MDM4 to the nucleus as MDM2 is a nuclear protein (Migliorini et al. 2002a).
Mdm4 knockout mice are embryonically lethal due to fatal p53 activation during development. However, embryonic lethality can be rescued by concomitant deletion of p53 (Parant et al. 2001; Finch et al. 2002; Migliorini et al. 2002b), which genetically validates MDM4 as an essential inhibitor of p53. Conversely, overexpression of Mdm4 leads to the suppression of p53 activity and consequently, spontaneous tumor development (Xiong et al. 2010). These phenotypes are reminiscent of Mdm2-null and Mdm2-overexpression phenotypes (see the MDM2 section). Indeed, embryonic lethality in Mdm4 null mice can be rescued by Mdm2 overexpression (Steinman et al. 2005), suggesting that MDM4 and MDM2 have some functional overlap. Nevertheless, Mdm4 deficiency often leads to only modest p53 activation and less severe phenotypes compared to Mdm2 deficiency (Grier et al. 2006) – this difference may be ascribed to the p53-MDM2 feedback loop (see the MDM2 section). It should be noted that during early mouse embryogenesis, but not in adult mouse tissues, MDM2-MDM4 heterodimerization plays a critical role in the suppression of p53 activity since transgenic mice that express Mdm4 mutants defective in Mdm2-binding exhibit lethal p53 activation and embryonic lethality (Pant et al. 2011). Again, this phenotype can be rescued by p53-null alleles (Pant et al. 2011).
MDM4 is regulated by multiple kinases (Fig. 1). Phosphorylation at Ser96 by Cyclin-dependent kinase 1 (CDK1) is required for nuclear export of MDM4 (Elias et al. 2005). Interestingly, phosphorylation at Ser96 also increases nuclear export of MDM2 (Elias et al. 2005), indicating that while MDM2 regulates the nuclear localization of MDM4, MDM4 conversely controls the cytoplasmic localization of MDM2. Phosphorylation of MDM4 at Ser289 by Casein Kinase 1α (CK1α) increases MDM4 affinity for p53, thus resulting in the suppression of p53 activity (Chen et al. 2005b). Genotoxic stress causes c-ABL-mediated phosphorylation of Tyr55 and Tyr99 on MDM4, which prohibits the association between p53 and MDM4 (Zuckerman et al. 2009). DNA damage also promotes phosphorylation of MDM4 at Ser342, Ser367, and Ser403 by checkpoint kinases 1 and 2 (CHK1/CHK2) and ataxia-telangiectasia mutated (ATM) kinase (Chen et al. 2005a; Pereg et al. 2006). Phosphorylation of these Ser residues facilitates ubiquitination and degradation of MDM4 protein. Independent of DNA damage, the serine/threonine kinase AKT also phosphorylates MDM4 at Ser367 to create a binding site for the adaptor protein 14-3-3 (Lopez-Pajares et al. 2008). 14-3-3 stabilizes MDM4 and MDM2, resulting in enhanced inhibition of p53 activity (Lopez-Pajares et al. 2008). Furthermore, MDM4 can be SUMOylated on K254 and K379 (Pan and Chen 2005); however, the functional significance of MDM4 SUMOylation remains unclear.
MDM4 also has p53-independent functions. Independently of p53 and MDM2, MDM4 interacts with the Mre11-Rad50-Nbs1 (MRN) complex and inhibits double-strand DNA repair (Carrillo et al. 2015), suggesting a p53-independent oncogenic function of MDM4. Conversely, loss of MDM4 leads to genome instability independently of p53 (Matijasevic et al. 2008), indicating that MDM4 can also act as a tumor suppressor. Much remains to be learned about p53-independent functions of MDM4.
MDM4 in Cancer
The tumor suppressor protein p53 is genetically inactivated in ∼50% of human tumors. However, in tumor cells with wild-type p53, p53 activity is often suppressed in various ways. As a critical inhibitor of p53, MDM4 is often amplified or overexpressed in these tumors, which include but is not limited to malignant gliomas (Riemenschneider et al. 1999), metastatic melanomas (Gembarska et al. 2012), and human retinoblastomas (Laurie et al. 2006). It should be noted that MDM4 levels are also elevated in a subset of cancers that have lost wild-type p53 alleles (Carrillo et al. 2015). This may be ascribed to p53-independent oncogenic functions of MDM4. Elevated levels of short-spliced forms of MDM4 are also found in various cancers.
Several MDM4 inhibitors that interrupt MDM4-p53 interaction have been developed in an effort to reactivate wild type p53 in cancer cells (see Wade and Wahl 2009, for review). The disruption of MDM4-p53 interaction by such inhibitors frees p53 protein from MDM4-mediated suppression, which leads to the activation of p53 and increased sensitivity to DNA-damaging agents in cancer cells. Importantly, MDM2 inhibitors, including Nutlin-3 and its family compounds, that interfere with MDM2-p53 binding do not display significant binding affinity for MDM4, thus rendering them ineffective for killing cancer cells that overexpress MDM4 (Hu et al. 2006; Graves et al. 2012). Although several MDM2 inhibitors are being examined in clinical trials, inhibitors that target MDM4 have not been tested clinically. Given that MDM2 and MDM4 cooperate to inhibit p53 and that MDM4 has p53-/MDM2-independent functions, a new compound or strategy that effectively inhibits MDM4 is anticipated.
MDM4 was identified as a gene product homologous to MDM2 and is ubiquitously expressed in various tissues and organs. The primary function of MDM4 is to inhibit the tumor suppressor protein p53, although p53-independent roles of MDM4 have also been documented. MDM4 overexpression or amplification is often associated with cancers. Several splicing variants of MDM4 have also been identified, some of which may modulate p53 suppression mediated by the full-length MDM4 and potentially play a role in tumorigenesis. Several MDM4 inhibitors have been developed. Although some of the inhibitors showed significant efficacy in tumor suppression in cell culture and animal models, they have yet to achieve complete clinical success.
The work in the Kurokawa laboratory is supported by an NCI Career Development Award R00 CA140948 (to M.K.), an American Cancer Society Institutional Research Grant IRG-82-003-30 (to M.K.), a Norris Cotton Cancer Center Prouty grant (to M.K.), and a Norris Cotton Cancer Center Holland Award (to M.K.).
- Bardot B, Bouarich-Bourimi R, Leemput J, Lejour V, Hamon A, Plancke L, Jochemsen AG, Simeonova I, Fang M, Toledo F. Mice engineered for an obligatory Mdm4 exon skipping express higher levels of the Mdm4-S isoform but exhibit increased p53 activity. Oncogene. 2015;34:2943–8. doi: 10.1038/onc.2014.230.PubMedCrossRefGoogle Scholar
- Bartel F, Schulz J, Böhnke A, Blümke K, Kappler M, Bache M, Schmidt H, Würl P, Taubert H, Hauptmann S. Significance of HDMX-S (or MDM4) mRNA splice variant overexpression and HDMX gene amplification on primary soft tissue sarcoma prognosis. Int J Cancer. 2005;117:469–75. doi: 10.1002/ijc.21206.PubMedCrossRefGoogle Scholar
- Elias B, Laine A, Ronai Z. Phosphorylation of MdmX by CDK2/Cdc2(p34) is required for nuclear export of Mdm2. Oncogene. 2005;24:2574–9. doi: 10.1038/sj.onc.1208488.
- Giglio S, Mancini F, Gentiletti F, Sparaco G, Felicioni L, Barassi F, Martella C, Prodosmo A, Iacovelli S, Buttitta F, Farsetti A, Soddu S, Marchetti A, Sacchi A, Pontecorvi A, Moretti F. Identification of an aberrantly spliced form of HDMX in human tumors: a new mechanism for HDM2 stabilization. Cancer Res. 2005;65:9687–94. doi: 10.1158/0008-5472.CAN-05-0450.PubMedCrossRefGoogle Scholar
- Graves B, Thompson T, Xia M, Janson C, Lukacs C, Deo D, Di Lello P, Fry D, Garvie C, Huang KS, Gao L, Tovar C, Lovey A, Wanner J, Vassilev LT. Activation of the p53 pathway by small-molecule-induced MDM2 and MDMX dimerization. Proc Natl Acad Sci USA. 2012;109:11788–93. doi: 10.1073/pnas.1203789109.PubMedPubMedCentralCrossRefGoogle Scholar
- Pant V, Xiong S, Iwakuma T, Quintás-Cardama A, Lozano G. Heterodimerization of Mdm2 and Mdm4 is critical for regulating p53 activity during embryogenesis but dispensable for p53 and Mdm2 stability. Proc Natl Acad Sci USA. 2011;108:11995–2000. doi: 10.1073/pnas.1102241108.PubMedPubMedCentralCrossRefGoogle Scholar
- Pereg Y, Lam S, Teunisse A, Biton S, Meulmeester E, Mittelman L, Buscemi G, Okamoto K, Taya Y, Shiloh Y, Jochemsen AG. Differential roles of ATM- and Chk2-mediated phosphorylations of Hdmx in response to DNA damage. Mol Cell Biol. 2006;26:6819–31. doi: 10.1128/MCB.00562-06.PubMedPubMedCentralCrossRefGoogle Scholar
- Riemenschneider MJ, Büschges R, Wolter M, Reifenberger J, Boström J, Kraus JA, Schlegel U, Reifenberger G. Amplification and overexpression of the MDM4 (MDMX) gene from 1q32 in a subset of malignant gliomas without TP53 mutation or MDM2 amplification. Cancer Res. 1999;59(24):6091–6.PubMedGoogle Scholar