This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dang, J. Articles by Roussel, M. F. Search for Related Content PubMed PubMed Citation Articles by Dang, J. Articles by Roussel, M. F.

The RING Domain of Mdm2 Can Inhibit Cell Proliferation¹

Departments of Tumor Cell Biology [J. D., M-L. K., L. S., C. J. S., M. F. R.], Biochemistry [C. M. E.], and Howard Hughes Medical Institute [L. S., C. J. S.], St. Jude Children’s Research Hospital, Memphis, Tennessee 38105

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mdm2 is a p53-inducible phosphoprotein that negatively regulates p53 by binding to it and promoting its ubiquitin-mediated degradation. Alternatively spliced variants of Mdm2 have been isolated from human and mouse tumors, but their roles in tumorigenesis, if any, remain elusive. We cloned six alternatively spliced variants of Mdm2 from Eµ-Myc-induced mouse lymphomas, all of which lacked the NH₂-terminal p53-binding domain but conserved the remainder of the Mdm2 protein. Enforced expression of full-length Mdm2 in primary mouse embryo fibroblasts or bone marrow-derived, interleukin 7-dependent pre-B cells accelerated their proliferation, whereas unexpectedly, overexpression of truncated Mdm2 isoforms inhibited their growth. Truncated variants were active as inhibitors whether they localized predominantly to the nucleus or cytoplasm. Despite the absence of the p53-binding domain, growth inhibition remained strictly p53 dependent (but not p19^Arf dependent) and could be overcome by full-length Mdm2. The intact RING finger domain at the Mdm2 COOH terminus (amino acids 399–489) was necessary and sufficient for growth inhibition by truncated Mdm2 proteins and could physically interact with either the RING finger domain or central acidic region of full-length Mdm2. However, such interactions do not inhibit Mdm2 E3 ubiquitin ligase activity in vitro using p53 as a substrate. Expression of growth-inhibitory Mdm2 isoforms in tumors remains an enigma.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mdm2 was discovered as an amplified gene on murine double-minute chromosomes in a spontaneously transformed 3T3 cell line (1) . Subsequent analysis demonstrated that Mdm2 (Hdm2 in human) is overexpressed in 5–10% of human tumors (2 , 3) , and that its expression is not only able to immortalize primary rodent embryonic fibroblasts but also to transform them in cooperation with activated Ras (4 , 5) . Embryos lacking Mdm2 die in utero, but lethality is rescued in a p53-null genetic background, indicating that an essential function of Mdm2 is to negatively regulate p53 activity (6 , 7) . Oncogenic properties of Mdm2 are conferred, at least in part, by its ability to inactivate p53. Similar to inactivating p53 mutations, overexpression of Mdm2 also induces chromosome instability, inappropriate centrosome duplication, and changes in ploidy (8) .

Mdm2 binding to the p53 NH₂ terminus antagonizes p53 transcriptional activity (9, 10, 11) , inhibiting p53 acetylation and transactivation by interfering with p300/CBP (12 , 13) . Mdm2 also functions as an E3 ligase to ubiquitinate p53 (14, 15, 16) and to enforce its export from the nucleus to the cytoplasm, where it is degraded in proteasomes (17, 18, 19) . It is unlikely that Mdm2 E3 ubiquitin ligase activity alone is sufficient to trigger p53 proteolysis, because Mdm2 mono-ubiquitinates p53 at multiple sites but does not catalyze addition of polyubiquitin chains that are necessary for recognition by the proteasome (20) . One possibility is that mono-ubiquitination of p53 is required to expose a nuclear export signal, and that p53 polyubiquitination and degradation then proceed in the cytoplasm (21, 22, 23) . Because Mdm2 is a direct transcriptional target of p53, its expression acts in a negative feedback loop to terminate the p53 response (24) . However, Mdm2 is itself subject to positive regulation through Ras signaling (25) and to negative control by ATM-mediated phosphorylation (26) and through direct binding of the ARF tumor suppressor protein (27, 28, 29) . Apart from p53, ARF, and CBP/p300, Mdm2 has been found to directly associate with the retinoblastoma protein, E2F1, Numb, MTBP, p73, and ribosomal protein L5 (30 , 31) . Therefore, it is unlikely that p53 is the only physiological target of Mdm2. A homologue of Mdm2, Mdm4 (MdmX), although not a target of p53 transcriptional regulation, can also negatively regulate p53-mediated transcription (32, 33, 34, 35) . The recent demonstration that loss of Mdm4 in the mouse germ-line, like the disruption of Mdm2, results in embryonic lethality that is rescued on a p53-null background has led to the conclusion that Mdm2 and Mdm4 regulate p53 via different pathways (36) .

The integrity of the COOH-terminal RING finger domain of Mdm2 is necessary for both its E3 ubiquitin protein ligase activity (15 , 37, 38, 39) and RNA-binding activity (40) . Mutation of cysteine 464 to alanine disrupts the integrity of the RING finger and abolishes Hdm2-mediated p53 ubiquitination and nuclear export (14 , 21 , 22 , 37) . Mdm2 also mediates its own ubiquitination in a RING finger-dependent manner (Refs. 16 and 37 ), and Lysine 444 might be important for Mdm2 E3 ligase activity.⁵

A number of alternatively spliced variants of Mdm2 have been identified and isolated from both human and rodent tumor cells (41, 42, 43, 44, 45, 46, 47, 48, 49) . Expression of alternatively spliced Mdm2 transcripts correlates with high-grade malignancy in human ovarian tumors, bladder carcinomas, astrocytic tumors, and breast cancer (42 , 43 , 49) , irrespective of their p53 status. Of the characterized Mdm2 variants, most sustained deletions of the p53-binding domain. Some Mdm2 isoforms were reported to transform NIH-3T3 cells (42) . In Eµ-Myc transgenic mice, many of the B-cell lymphomas that arise sustain either p53 or Arf loss of function, with or without overexpression of Mdm2 (50) . Some lymphomas also expressed variant Mdm2 isoforms, which coexisted with full-length Mdm2, irrespective of whether the tumors sustained Arf deletions or p53 mutations. This prompted us to clone and characterize these variants and to examine their role in tumorigenesis.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Isolation of Variant Mdm2 cDNAs.
Full-length and alternatively spliced Mdm2 mRNAs were amplified by RT-PCR⁴ (ProSTAR First-Strand RT-PCR kit; Stratagene), using 1 µg of total RNA isolated from Eµ-Myc-induced lymphomas (Table 1 ; Ref. 50 ). Primers were chosen from the noncoding region of the published Mdm2 cDNA sequence [bp 169, sense (+), 5'-CCATCGATCACCGCGCTTCTCCTGCGGCC-3' and bp 1702, antisense (-) 5'-ATCGATATAAAATTCTATTTTTGTGAGCAGGTC-3'], including a ClaI site (underlined; Ref. 4 ). The amplified cDNA fragments were cloned into the pGEM-T Easy vector (Promega), and their nucleotide sequences were determined. Mammalian expression vectors were constructed by subcloning the cDNAs into pSRMSV-tkneo (generously provided by Drs. Charles Sawyers and Owen Witte UCLA, CA; Ref. 51 ) or MSCV-IRES-GFP retroviral vectors at an EcoRI site located immediately 3' to the long terminal repeat (52) .

A FLAG-tagged UbcH5 construct was generated by PCR using human UbcH5 cDNA and the following primers: FLAG-H5–1+, 5'-GACCATG-GACTACAAGGACGACGATGACAAGGCGCTGAAGAGGATTCAGAA-AGAA-3' (NcoI site underlined); H5(-), 5'-GGATCCTTACATTGAATATTTCTGAGTCCATTC-3' (BamHI site underlined). The PCR product was subcloned into pGEM-T Easy, excised with EcoRI, and subcloned into the pVL-1393 baculovirus expression vector (PharMingen). Baculovirus constructs containing full-length Mdm2 and mutants were generated by excision with EcoRI of the Mdm2 inserts from pGEM-T Easy or the MSCV-IRES-GFP vector and subcloning them into the EcoRI site of pVL1393.

Cell Culture and Viral Vector Production.
WT, p53-null, Arf-null or Cip1-null MEFs were isolated from 13.5 day midgestation mouse embryos as described previously (53) and cultured at early passages in DMEM containing 10% FBS, 2 mM glutamine, 0.1 mM nonessential amino acids, 55 mM 2-mercaptoethanol, and 10 µg/ml gentamicin, in 8% CO₂ humidified incubators. Primary pre-B cells were derived from mouse bone marrow harvested from femurs and tibias of WT animals, expanded, and maintained on feeders of NIH3T3 cells expressing human IL-7 (T220-29) or in liquid culture with recombinant human IL-7 in RPMI 1640 containing 5% FBS, 2 mM glutamine, 55 mM ß-mercaptoethanol, penicillin/streptomycin, in 10% CO₂ humidified incubators, as described previously (50 , 54) . NIH3T3 cells stably expressing a zinc-inducible p19^Arf protein (pMTCB6-HA-Arf) were generated by transfection of a mammalian expression vector (pMTCB6) in which the HA-tagged Arf cDNA was cloned downstream of the sheep metallothionein promoter (MT1). Cells were maintained in DMEM containing 10% FBS, 2 mM glutamine, penicillin/streptomycin, and 400 µg/ml of G418. Production of high titer ecotropic viruses in 293T cells and infections of MEFs (53) and pre-B cells (50 , 54) were carried out as described. Spodoptera frugiperda (Sf9) cells were grown at 24°C in Grace’s medium (Life Technologies, Inc.) supplemented with 5% FBS and infected with baculoviruses as described previously (54) .

Protein Extraction, Immunoprecipitation, and Immunoblotting.
MEFs were trypsinized, and after two washes with PBS, were lysed in ice-cold Tween 20 lysis buffer [50 mM HEPES (pH 7.5), 200 mM NaCl, 1 mM EDTA, 0.1% Tween 20, 1 mM phenylmethylsulfonyl fluoride, 0.4 unit of aprotinin/ml, 1 mM NaF, 10 mM ß-glycerophosphate, and 0.1 mM sodium orthovanadate] and left on ice for 1 h. After sonication at 4°C (7 s x 2), cellular debris was removed by centrifugation in a microcentrifuge at 14,000 rpm for 15 min at 4°C. Proteins (200 µg/lane) were electrophoretically separated on denaturing polyacrylamide gels containing SDS and transferred onto membranes (Osmonics, Westborough, MA). Membranes were immunoblotted with affinity-purified rabbit polyclonal antibodies to mouse p19^Arf (55) , a monoclonal antibody directed to mouse Mdm2 (2A10) generously provided by Dr. Arnold Levine (Rockefeller University, New York, NY); a monoclonal antibody to the HA epitope (generously provided by Dr. Albert Reynolds, Vanderbilt University, Nashville, TN); a monoclonal antibody (9E10) recognizing the NH₂-terminal epitope of c-Myc (56) ; or with commercial antibodies directed to mouse p53 (Ab-7; Oncogene Research Products), mouse p21^Cip1 (F-5; Santa Cruz Biotechnology, Santa Cruz, CA), Mdm2 (SMP14; Santa Cruz Biotechnology), or FLAG (anti-FLAG, M2; Sigma Chemical Co.). Sequential immunoprecipitation and immunoblotting were performed as described (57) .

Immunofluorescence.
Procedures were described in detail previously (18) . Cells (3 x 10⁴) plated on coverslips were fixed and permeabilized in cold acetone:methanol (1:1, v/v) for 15 min at -20°C. Coverslips were air dried, blocked with 10% FBS in PBS, and stained with the 2A10 antibody to Mdm2, followed by a goat antimouse antibody conjugated with fluorescein (Amersham). Nuclei were visualized by 4',6-diamidino-2-phenylindole staining.

In Vitro Ubiquitination Assay.
The ubiquitination assay was based on protocols published previously (14 , 16 , 58) . p53 was ubiquitinated in 50 µl of total volume reactions containing Sf9 lysates containing E1 (10 µg of lysate protein), FLAG-UbcH5 (10 µg of lysate protein), Mdm2 (10 µg of lysate protein), and p53 (10 µg of lysate protein) in 100 mM Tris-HCl (pH 7.5), 5 mM MgCl₂, 0.6 mM DTT plus 15 µM ubiquitin (Sigma Chemical Co.), 2 mM ATP, 10 mM ß-glycerol phosphate, 5 µg/ml aprotinin, and 5 µg/ml ubiquitin-aldehyde (Boston Biochemical). Reactions were allowed to proceed at 25°C for 2 h. Products were resolved on 7.5% denaturing polyacrylamide gels, transferred to membranes, and immunoblotted with anti-p53 antibodies (Ab-7) visualized by ECL (Amersham). A His₆-tagged, NH₂-terminal p19^Arf peptide (N37) expressed in Escherichia coli (57) was purified on a nickel (Ni2⁺) column according to the manufacturer’s instructions (Qiagen) and added (10 µg) as an Mdm2 inhibitor. Mdm2 mutants V4, FLAG-M3, or FLAG-M4 from Sf9 lysates were also added as indicated.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Characterization of Alternatively Spliced Mdm2 Variants.
RNA was extracted from Eµ-Myc-induced B-cell tumors, the p53 and p19^Arf status of which had been determined previously (Ref. 50 ; Table 1 ). cDNAs encoding various Mdm2 isoforms were amplified from total RNA by RT-PCR and subcloned, and their nucleotide sequences were determined. This yielded six alternatively spliced Mdm2 variants (designated V1 to V6) in addition to full-length Mdm2 cDNA (Fig. 1A) . Single or multiple Mdm2 isoforms were expressed in different B-cell tumors, regardless of whether full-length Mdm2 was overexpressed or not and without any obvious correlation with WT or mutant p53 status (Table 1) . Alignment of cDNA sequences of all variants with that of Mdm2 predicted that five of the six Mdm2 isoforms (V1 to V5) would not encode portions of the p53-binding domain (mapped within the first 120 NH₂-terminal amino acids) because of deletions of exons 3, 5, or 4–8. The one exception was the V6 isoform that lacked sequences encoded by exon 8 alone (residues 134–165). Variants V2 to V5 retained the entire COOH-terminal portion of Mdm2 from amino acids 166 to 489, including nuclear localization (NLS) and nuclear export (NES) signals located between amino acids 178 to 195, the p19^Arf-binding domain (amino acids 210 to 304), zinc finger (amino acids 303 to 320), and RING finger domain (amino acids 436–476; Fig. 1A ). Isoform V1 contained an additional in-frame deletion of amino acids 231–240.

View larger version (38K): [in this window] [in a new window]   Fig. 1. Structure, expression, and binding characteristics of Mdm2 variants isolated from Eµ-Myc-induced B-cell mouse lymphomas. A, schematic structure of Mdm2 variants isolated from Eµ-Myc-induced B-cell lymphomas. p53 binding (amino acids 19–102), p19Arf-binding (amino acids 210–304), and RING finger (RD; amino acids 436–476) domains are designated by unshaded boxes. NLSs, NESs, and sequences required for nucleolar localization (NrLS) are shown as black boxes. Arrows, predicted translation ATG start codons. Deleted sequences within Mdm2 variants are shown as lines. Cellular localization of full-length Mdm2 and each variant was determined by indirect immunofluorescence with a monoclonal antibody against Mdm2 (2A10). ND, undetermined. B, primary MEFs at early passage were infected with a control retrovirus vector (Lane 1), or with vectors expressing full-length Mdm2 (Lane 2), or Mdm2 variants (V1 to V6, Lanes 3–8). Cells were lysed 72 h after infection, and proteins were detected by direct immunoblotting (IB) with a monoclonal antibody to Mdm2 (2A10). Right, protein markers (in thousands). C and D, Sf9 cells were coinfected with baculoviruses encoding p53 (C) or p19Arf (D) together with full-length Mdm2 (Lanes 1–3), Mdm2 variant V2 (Lanes 4–6), or Mdm2 variant V4 (Lanes 7–9). Cell lysates were precipitated with normal rabbit serum (NRS; C, Lanes 1, 4, and 7), a monoclonal antibody against a Myc epitope (9E10; D, Lanes 1, 4, and 7), a monoclonal antibody to Mdm2 (2A10; Lanes 2, 5, and 8), or antibodies against p53 (Ab-1) or p19Arf (Lanes 3, 6, and 9). Proteins electrophoretically resolved on denaturing gels were blotted with monoclonal antibody to Mdm2, 2A10. IB, immunoblot.

Mdm2 isoforms V3 and V5 encoded M_r 55,000 proteins that were predicted to initiate at codon 198 (Fig. 1A) . Therefore, these variants lacked the NLS and NES and, in agreement, immunofluorescence analysis (data not shown) revealed that both proteins were predominantly expressed in the cytoplasm (summarized in Fig. 1A ). Deletion of exon 5 in V1 changes its reading frame, also forcing internal initiation at methionine 198 and probably accounting for the low level of protein detected (Fig. 1B , Lane 3). In addition, the downstream in-frame deletion resulted in production of a cytoplasmic polypeptide smaller than the V3 and V5 isoforms (Fig. 1B , Lane 3). V6 lacked exon 8, again resulting in disruption of its reading frame, but unlike V1, its expression was undetectable. V2 lacked residues specified by exon 3 and encoded a M_r 75,000 protein, the translation of which was likely initiated from methionine 50 (Fig. 1B , Lane 4). This isoform is identical to that previously identified as p76^Mdm2 by others (59) . The V4 protein (Fig. 1B , Lane 6) initiates at methionine 1, but it contains a large in-frame deletion from amino acids 11 to 155, as well as deletion of serine 207. As predicted, both the V2 and V4 isoforms were predominantly localized to the nucleus of infected MEFs (Fig. 1A) . Of interest, although endogenous Mdm2 expression was not detected in MEFs infected with the naked vector (Fig. 1B , Lane 1), we observed a modest but consistent increase in its expression in cells expressing the truncated Mdm2 variants (Fig. 1B , Lanes 3–7, and see below).

Nucleotide sequence analysis predicted that all Mdm2 variants would be unable to bind p53 but would still interact with p19^Arf. To confirm this, insect Sf9 cells were coinfected with baculoviruses encoding either WT Mdm2 or variants V2 or V4, together with baculoviruses encoding WT p53 (Fig. 1C) or p19^Arf (Fig. 1D) . Lysates of infected cells were then either precipitated with control antibodies (NRS or 9E10), monoclonal antibodies to Mdm2 (2A10) or p53 (Ab-1), or antibodies to the mouse p19^Arf COOH terminus, and proteins were separated on denaturing gels were immunoblotted with antibody to Mdm2. As expected, the full-length Mdm2 protein coprecipitated with either p53 or p19^Arf (Fig. 1, C and D , Lanes 1–3). In contrast, Mdm2 variants V2 and V4 were unable to form complexes with p53 but retained the ability to bind p19^Arf (Fig. 1, C and D , Lanes 4–9).

Overexpression of Truncated Mdm2 Isoforms Inhibits Cell Proliferation.
Because the Mdm2 variants were expressed in B-cell tumors, we suspected that they might compete with WT Mdm2 for p19^Arf binding, facilitating the ability of endogenous Mdm2 to antagonize p53, and thereby accelerating cell proliferation. To test this, we used retroviral vectors to introduce either full-length Mdm2 or variant isoforms into an engineered NIH-3T3 cell line in which p19^Arf expression can be conditionally up-regulated by addition of zinc to the culture medium. Two days after infection, p19^Arf expression was induced by addition of 100 µM zinc, and cell proliferation was monitored using long-term colony assays. Contrary to our expectations, WT Mdm2 bypassed Arf-induced growth arrest, whereas the variants did not (data not shown), inconsistent with the idea that truncated Mdm2 variants antagonize p19^Arf function in this manner.

Others reported that Mdm2 variants isolated from human tumor cells were able to transform NIH-3T3 cells (42) . Although the NIH-3T3 cell line used in our studies lacks the Ink4a/Arf locus (55) and is readily transformed by oncogenic Ras, enforced expression of variants V2, V4, or V5 in these cells did not transform them but instead decreased their growth rate in a manner similar to that seen with primary cell strains (data not shown, but see below).

We next tested the effects of the enforced expression of each Mdm2 variant on the proliferation of both primary MEFs and mouse bone marrow-derived pre-B cells. First, early-passage mouse primary MEFs (Fig. 2A) were infected with retroviruses encoding full-length Mdm2, variants V2, V4, or V5, or the empty control vector. The MSCV-IRES-GFP vector carries the gene encoding GFP in cis, which was used as an internal control to monitor infection efficiency. Three to 4 days after infection, GFP-positive cells were seeded at 2 x 10⁴ per dish, and proliferation was monitored by counting cells (triplicate plates/day) for 8 days thereafter. MEFs infected with full-length Mdm2 proliferated considerably more rapidly than cells infected with the empty vector (Fig. 2A) , became smaller in size (data not shown), and arrested at confluence by day 8. These results contrast directly with those reported by others who concluded that Hdm2 overexpression inhibited the proliferation of NIH-3T3 cells (60) . In contrast, MEFs infected with variants V2, V4, and V5 grew at a much reduced rate and eventually stopped proliferating before becoming confluent (Fig. 2A) . Both nuclear (V2 and V4) and cytoplasmic (V5) Mdm2 variants had inhibitory effects on cell proliferation. Growth-arrested MEFs remained viable for at least 4 weeks in culture and were morphologically flat and enlarged (data not shown). Although the rate of growth inhibition by each variant differed, V4 showed the strongest effect and was chosen for subsequent experiments.

View larger version (21K): [in this window] [in a new window]   Fig. 2. Overexpression of Mdm2 variants inhibits cell growth. Primary early-passage, WT MEFs (passage 4; A), WT bone marrow-derived pre-B cells (B) or primary MEFs from mice lacking p53 (C, p53-/-), Cip1 (D, p21-/-), or Arf (E, Arf-/-) were infected with retroviruses expressing Mdm2 full-length (wt, ) and variants (V2, x; V4, ; V5, ) or an empty vector MSCV-IRES-GFP (Vector, ) and seeded at 2 x 104 in 60-mm diameter culture dishes 72–96 h after infection. Growth rates were determined by counting cell numbers from triplicate cultures every day for 8 days (MEFs) or every other day for 12 days (pre-B cells). F, primary WT MEFs were infected with retroviruses expressing full-length Mdm2 (WT, Lane 2), Mdm2 variant (V4, Lane 3), or empty vector (Vect, Lane 1) and lysed 72 h after infection. Proteins were separated on denaturing gels and blotted with antibodies against Mdm2, p53, p21Cip1, and p19Arf as indicated.

Mdm2 Variants Inhibit Cell Proliferation in a p53-dependent Manner.
Expression of Mdm2 variants in a subset of Eµ-Myc-induced tumors that retained WT p53, Mdm2, and Arf prompted us to determine whether their ability to inhibit growth might depend on p53, p19^Arf, or p21^Cip1, the latter a direct transcriptional target of p53. We expressed either full-length Mdm2 or V4 in early-passage MEFs derived from mice lacking p53, both p53 and Mdm2, Cip1, or Arf and tested their effects on cell proliferation. The rates of proliferation of p53-null MEFs (Fig. 2C) or of those lacking both p53 and Mdm2 (data not shown) were unaffected by introduction of either full-length Mdm2 or V4. Similar results were obtained with the V2 and V5 isoforms in these cells (data not shown). Therefore, although truncated Mdm2 variants cannot interact with p53 directly, their ability to inhibit cell proliferation was still p53 dependent. The effect of V4 was partially compromised in Cip1-null and Arf-null MEFs (compare effects of V4 in Fig. 2, D and E , versus A), indicating that although the absence of p21^Cip1 or p19^Arf facilitates faster cell proliferation, neither is strictly required for V4-mediated inhibition.

The strict dependency on p53 and partial contributions of p21^Cip1 and p19^Arf for growth inhibition suggested that the variant Mdm2 isoforms might affect the expression of p53, p19^Arf, Mdm2, and p21^Cip1 in infected cells. Indeed, p76^Mdm2 (V2) was found previously to antagonize the function of WT Mdm2, increasing the levels and activity of p53 (59) . In agreement, we had observed modest increases in endogenous Mdm2 levels in cells coexpressing the truncated Mdm2 variants (Fig. 1B) . Three days after infection with the control vector or those encoding Mdm2 or V4, early-passage, WT MEFs were lysed, and levels of p53, p21^Cip1, and p19^Arf were determined by immunoblotting (Fig. 2F) . The very low levels of p53 expressed in primary MEFs were not detectably changed in cells infected with WT Mdm2 (Fig. 2F , Lane 2) but were slightly elevated in cells expressing V4 (Lane 3) compared with those infected with the control vector (Lane 1). More obviously, p21^Cip1 expression was significantly induced by V4, whereas Wt Mdm2, but not V4, increased p19^Arf levels. The latter results are consistent with previous findings that active p53 feeds back to repress Arf expression (61 , 62) . These data are consistent with the idea that Mdm2 variants trigger a p53 response that slows cell growth, and that growth retardation depends in part on p21^Cip1.

The RING Domain Is Necessary and Sufficient for Growth Inhibition by Mdm2 Variants.
To define the domain(s) responsible for growth arrest by the Mdm2 variants, we prepared additional Mdm2 mutants (Fig. 3A) . NH₂-terminal truncation mutants lacked 197 (M1), 298 (M2), or 398 (M3) residues, leaving an intact RING domain from amino acids 399 to 489 (M3). The truncation mutant (M4) differed from M3 by codeletion of COOH-terminal residues 464–489. Each construct was tagged at its NH₂ terminus with a FLAG or HA epitope preceded by an initiator methionine codon, cloned into the MSCV-IRES-GFP retroviral vector, and introduced into early-passage MEFs (Fig. 3B) . Expression of each mutant construct was confirmed by immunoblotting with antibodies to Mdm2 or the epitope tags, revealing variable levels of protein overexpression (Fig. 3C) . Similar to the natural Mdm2 variants recovered from B-cell tumors, mutants M1, M2, and M3 each inhibited cell growth when expressed in WT MEFs (Fig. 3B) . However, M4, containing a disrupted RING finger domain, was completely devoid of inhibitory activity.

View larger version (16K): [in this window] [in a new window]   Fig. 3. The RING finger domain of Mdm2 inhibits cell growth. A, schematic structure of Mdm2 deletion mutants (M1–M4) and M3 RING domain point mutants (M5 and M6). p53-binding, p19Arf-binding, and RING finger (RD) domains are designated by unshaded boxes. The NLS, NES, and sequences required for nucleolar localization (NrLS) are shown as black boxes. Epitopes recognized by Mdm2 antibodies, 2A10 and SMP14, are indicated as black bars above the top schematic. Arrows, predicted ATG translation start codons. RING finger domain mutations generated by PCR are indicated for M5 and M6. B, primary WT MEFs infected with retroviruses expressing full-length Mdm2 (WT, ), mutants (M1, ; M2, ; M3, ; M4, ; M5, x; M6, ), or an empty vector () were grown and counted as described in Fig. 2 . Bars, SE. C, protein expression was confirmed in lysates made from WT MEFs after infection with retroviruses encoding Mdm2 mutants (M1–M6, Lanes 1–6). Immunoblotting (IB) was performed with antibody 2A10 to Mdm2 for Mdm2 mutants M1 and M2 or with an antibody to the FLAG tag for FLAG-M3 to FLAG-M6.

Interaction of RING-containing, Truncated Mdm2 Isoforms with Full-length Mdm2.
Although growth inhibition by Mdm2 variants is strictly p53 dependent, all such isoforms lacked a functional p53-binding domain, implying that Mdm2 variants interact with regulators of p53 rather than p53 itself. Given that p19^Arf was not required for growth inhibition, endogenous Mdm2 was the most obvious candidate. Indeed, MEFs coinfected with the smallest RING-containing construct M3 together with full-length Mdm2 grew at almost the same rate as MEFs infected with full-length Mdm2 alone (Fig. 4A) , indicating that M3 and Mdm2 functionally compete with one another in this regard. The integrity of the Mdm2 RING domain was essential for full-length Mdm2 to override the inhibitory effects of M3, because Mdm2 mutants lacking the complete COOH terminus (464–489) or containing an alanine for cysteine mutation at codon 462 were unable to reverse M3-induced growth inhibition (Fig. 4A) .

View larger version (22K): [in this window] [in a new window]   Fig. 4. Mdm2 rescues RING finger domain-mediated growth inhibition. A, primary WT MEFs were infected with retroviruses expressing full-length Mdm2 (WT, ) or FLAG-tagged M3 (), or were coinfected with both (x), with M3 plus Mdm2 (464–489; ), or with M3 plus Mdm2 (C462A; ). Growth rates were determined by counting cell numbers from triplicate cultures every day for 8 days. B, NIH3T3 cells containing a zinc-inducible Arf protein under the control of the metallothionein promoter (pMTCB6HA-ARF) were infected with a retrovirus encoding FLAG-tagged M3 (Lane 2). Uninfected cells were used as controls (WT, Lane 1). Zinc (100 µM) was added to the cultures for an additional 24 h to induce p19Arf, after which cell lysates were prepared. Proteins precipitated with anti-FLAG (M2, FLAG) were separated on denaturing gels and blotted with antibody to Mdm2 (2A10) and anti-FLAG. IP, immunoprecipitation; IB, immunoblot. C, insect Sf9 cells were coinfected with baculoviruses expressing FLAG-tagged M3 and HA-tagged full-length Mdm2 (HA-WT), FLAG-M3 and HA-tagged Mdm2 with a truncation of the RING domain (HA-464–489), or HA-tagged full-length Mdm2 and FLAG-tagged M4 (FLAG-M4). Cell lysates were precipitated with anti-Myc (9E10) used as a negative control (Lanes 1, 4, and 7), antibodies to Mdm2 (SMP14, Lanes 2 and 5; 2A10, Lane 9), or with anti-FLAG (M2, Lanes 3, 6, and 8). Electrophoretically separated proteins were immunoblotted with anti-Mdm2 (2A10) and anti-FLAG (M2). IP, immunoprecipitation; IB, immunoblot. D, insect Sf9 cells were coinfected with baculoviruses expressing HA-tagged M3 (HA-M3) and FLAG-tagged M3 (FLAG-M3, Lanes 1–3) or infected with individual baculoviruses (Lanes 4–7) and lysed 48 h after infection. Cell lysates were precipitated with anti-Myc (9E10, Lane 1), anti-HA (HA, Lanes 2, 4, and 6), and anti-FLAG (Lanes 3, 5, and 7), and immunoblotted with anti-FLAG (Lanes 1–5) or anti-HA (Lanes 6 and 7). IP, immunoprecipitation; IB, immunoblot. E, insect Sf9 cells were coinfected with baculoviruses expressing HA-tagged Mdm2 mutants, HA 1–50, 469–489 (Lanes 1–3), or HA 305–489 (Lanes 4–6), together with FLAG-M3. Cells were lysed 48 h after infection. Lysates were precipitated with anti-Myc (9E10, Lanes 1 and 4), anti-HA (Lanes 2 and 5), and anti-FLAG (Lanes 3 and 6), and separated proteins were immunoblotted with anti-FLAG and anti-HA. IB, immunoblot.

To identify the second interaction site outside of the Mdm2 RING domain, two additional Mdm2 deletion mutants were coexpressed with FLAG-tagged M3 in Sf9 cells (Fig. 4E) . One of these contained deletions in both the p53-binding and RING domains (1–50, 464–489), and the other lacked residues 305–489, leaving the p53-binding and acidic domains intact. Coprecipitation experiments revealed that both mutants interacted with FLAG-M3 (Fig. 4E) , suggesting that, in addition to RING-RING interactions, FLAG-M3 can also bind to the acidic domain of Mdm2.

To confirm that similar interactions could occur in mammalian cells, we introduced M3 alone into the NIH-3T3 cell line that conditionally expressed zinc-inducible p19^Arf. Induction of p19^Arf strongly increased p53 levels and led to Mdm2 expression. Precipitation with the anti-FLAG antibody demonstrated that endogenous Mdm2 protein coprecipitated with M3 under these conditions (Fig. 4B , Lane 2). Similar results were obtained using HA-tagged M3 (data not shown).

Mdm2 Variants Do Not Affect Ubiquitination of p53 Mediated by Mdm2.
If inhibition of cell proliferation by the overexpressed Mdm2 RING domain depends upon direct binding to endogenous Mdm2, a potential consequence might be disruption of Mdm2 E3 ubiquitin protein ligase activity, leading, in turn, to p53-dependent growth retardation. We therefore tested whether Mdm2 variants could inhibit Mdm2-directed p53 ubiquitination in reconstituted in vitro enzyme reactions containing E1 and E2 (UbcH5) enzymes plus recombinant full-length Mdm2. Multiple ubiquitinated forms of p53 were resolved on denaturing gels and detected with an antibody to p53 (Fig. 5A , Lane 1). This assay does not distinguish between forms of p53 mono-ubiquitinated on multiple lysine acceptor sites from forms containing tandem ubiquitin chains at single sites (polyubiquitination). However, only mono-ubiquitinated forms have been documented in such assays (20) . Expression of Mdm2, V4, M3 and M4 proteins in Sf9 cells was confirmed by immunoblotting (Fig. 5B) . Addition of full-length Mdm2 to the reaction (10 µg of Sf9 lysate as shown in Fig. 5B ) was essential for ubiquitination of p53 (Fig. 5A , Lane 1), which was completely inhibited by addition of 1 or 10 µg of a peptide representing the NH₂-terminal 37 amino acids (N37) of p19^Arf (Fig. 5A , Lanes 2 and 3). Mutant Mdm2 proteins failed to induce ubiquitination of p53 (Fig. 5A , Lanes 6, 9, and 11) and did not inhibit p53 ubiquitination mediated by full-length Mdm2 (Fig. 5A , Lanes 4, 5, 7, 8, and 10). It should be noted that lysates containing equal quantities of total protein (10 µg) expressed far more V4 than full-length Mdm2 (Fig. 5B) . Moreover, the mutant Mdm2 proteins were unable to reverse inhibition of Mdm2-dependent ubiquitination by p19^Arf N37 (Fig. 5, C and D) . Therefore, induction of p53 in cells overexpressing the growth-inhibitory Mdm2 variants is not likely to be attributable to their ability to directly inhibit p53 ubiquitination by full-length Mdm2.

View larger version (48K): [in this window] [in a new window]   Fig. 5. In vitro ubiquitination of p53 is unaffected by truncated Mdm2 isoforms. A, in vitro ubiquitination of p53 was performed with recombinant proteins expressed in Sf9 cells, including full-length Mdm2, the V4 variant, the RING domain (M3), or a COOH-terminally truncated RING domain mutant (M4). N37, representing the first 37 NH2-terminal amino acids of p19Arf, was prepared in bacteria using a chemically synthesized minigene template (58) . Ladders of ubiquitinated forms of p53 were visualized with antibody to p53. Sf9 lysates containing Mdm2 (10 µg of total protein) or lysates expressing M3, M4, or V4 (1 or 10 µg of total protein as indicated at the top of the panel) were used. B, expression of proteins in Sf9 cells infected with full-length Mdm2 and variant V4 was confirmed by immunoblotting (IB) with antibody 2A10 to Mdm2, whereas protein expression from Sf9 cells infected with FLAG-tagged M3 and FLAG-tagged M4 was confirmed by immunoblotting with anti-FLAG (M2). Uninfected cells were used as controls. C, in vitro p53 ubiquitination assay with full-length Mdm2 alone (Lanes 3–5) or together with variant Mdm2 V4 (Lanes 8–10) with increasing concentration of N37-p19Arf (Lanes 4, 5 and 9, 10). D, p53 ubiquitination assay with full-length Mdm2 together with the RING domain (M3, Lanes 3–5) or a RING domain mutant (M4, Lanes 7–9) with increasing quantity of N37-p19Arf (Lanes 4, 5 and 7–9).

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Because the ability of NH₂-terminally truncated Mdm2 isoforms to inhibit cell proliferation required p53 but could not be mediated through a direct interaction with p53 itself, we reasoned that their interaction with full-length Mdm2 might be required. In agreement with this idea, Mdm2 variants were observed to associate directly with full-length Mdm2. Growth inhibition was completely abrogated by a single point mutation at cysteine 462 that prevents Mdm2 from ubiquitinating both p53 and itself (16 , 37) as well as by a partial COOH-terminal deletion (amino acids 464–489), both of which disrupt the structure of the RING finger domain. Therefore, the integrity of the RING domain was required for a functional interaction. Moreover, the isolated RING domain, when expressed alone, was found to be growth inhibitory. In contrast, elimination of lysine 444 was without effect, implying that such modification is not required for growth inhibition. Mutation of the cryptic nucleolar localization sequence (amino acids 464–471) within the RING finger domain of Mdm2, which is required for p19^Arf-mediated sequestration in the nucleolus, did not affect growth inhibition, suggesting that nucleolar translocation of Mdm2 must not be necessary either (data not shown; Refs. 18 , 57 , and 63 ).

Deletion mapping studies indicated that the isolated RING domain of Mdm2 was capable of binding the full-length Mdm2 protein. Binding was mediated by direct RING-RING interactions as well as by an association of the isolated RING domain with the acidic domain of Mdm2. Other investigators reported that full-length Hdm2 was growth inhibitory when expressed in NIH-3T3 cells (60) , whereas we obtained the opposite result, i.e., overexpression of Mdm2 accelerated proliferation. In their studies, the inhibitory domain of Hdm2 was mapped to residues 155 to 324, which includes the central Mdm2 acidic domain but not the COOH-terminal RING. Again, our results do not agree, and the basis for these discrepancies remains unresolved.

Because RING fingers appear to have no intrinsic E3 protein ligase activity of their own, the simplest interpretation is that an interaction between the full-length and variant Mdm2 proteins might impair the E3 ligase activity of Mdm2. However, this now seems unlikely, because neither the expression of variant Mdm2 V4 nor expression of the RING domain alone inhibited Mdm2-mediated p53 ubiquitination in an in vitro assay. Another possibility is that overexpression of Mdm2 RING domains in cells can sequester the E2 Ubc enzymes that are required for Mdm2 E3 ligase activity. We do not favor this for several reasons: (a) E2 enzymes interact with many E3s and are not generally thought to be rate-limiting in vivo; and (b) the crystal structure of the Cbl E3 ligase indicates that E2 binding is coordinated both by residues within the RING domain and through additional contacts elsewhere in the protein (64) . In addition, overexpression of UbcH5 or UbcH7 together with Mdm2 variants failed to rescue their growth inhibitory effects (data not shown). Therefore, the mechanism by which truncated Mdm2 isoforms inhibit cell growth remains undefined.

Despite the ability of truncated Mdm2 isoforms to associate directly with Mdm2, an important caveat is a lack of formal proof that Mdm2 is, in fact, their critical target. There is no simple way to test whether Mdm2 is required, because cells from Mdm2-null mice cannot be propagated unless they also lack p53 (6 , 7) . It may well prove that Mdm2-related Mdm4 (MdmX) or other proteins that interact with Mdm2 are responsible for the observed effects. In agreement with data of others (65) , we confirmed that the RING finger domain of Mdm2 could interact with Mdm4 and vice versa (data not shown). The exact role of Mdm4 in inhibiting p53 function remains unclear, but unlike Mdm2, Mdm4 is not a p53-inducible gene, does not seem to be expressed at particularly high levels during p53 stress responses, and although it inhibits p53-dependent transcription, Mdm4 is not thought to catalyze p53 degradation (34 , 39) . Disruption of Mdm4 in the mouse germ-line leads to embryonic lethality accompanied by growth arrest, but not apoptosis, and these effects are rescued on a p53-null background (36) . Intriguingly, disruption of the Mdm4 gene resulted in production of a truncated product that likely encodes the RING domain. This formally leaves open the possibility that embryonic lethality results from a gain of function (overexpression of the Mdm4 RING) versus Mdm4 loss.

Whatever the exact mechanisms, enforced overexpression of the truncated Mdm2 variants led to p53 activation, which could be reversed by simultaneous overexpression of full-length Mdm2. Similarly, a recent study showed that Mdm2 rescues cell growth arrest mediated by another Mdm2 binding protein, MTBP (66) . MEFs transduced by Mdm2 variants alone arrested irreversibly after several days, remained viable for as long as 4 weeks in culture, and assumed an enlarged and flat morphology reminiscent of senescent fibroblasts. Interestingly, although p21^Cip1 induction in response to Mdm2 variants was relatively robust, the induced levels of endogenous Mdm2 were significantly lower than those usually observed in cells undergoing a p53-dependent stress response. It may be that the ability of Mdm2 variants to interact directly with full-length Mdm2 somehow affects the ability of p53 to activate the Mdm2 feedback loop that normally cancels the p53 response.

Several studies showed that alternatively spliced and mutant Mdm2 variants are found in many types of human tumors, including invasive breast cancer (47) , late-stage and high-grade ovarian and bladder carcinomas (42) , and liposarcomas (48) . Although the role of these Mdm2 variants in the onset or late stages of tumors remains elusive, their occurrence and persistence in both human and mouse tumors suggest that they somehow contribute to tumorigenicity. Our results reveal that when overexpressed in primary MEFs or pre-B cells, these Mdm2 variants paradoxically inhibit rather than promote cell growth. We have considered several possibilities to rationalize these results: (a) it may prove that, similar to activated Ras (67) , the enforced expression of these Mdm2 variants triggers growth arrest in primary cells, but in collaboration with Myc, promotes proliferation; and (b) alternatively, expression of truncated Mdm2 variants might allow certain cells to escape Eµ-Myc-induced apoptosis during early stages of lymphomagenesis, after which subsequent genetic changes then allow the expansion of this resistant population. However, we found that enforced expression of Mdm2 variants in early-passage MEFs did not inhibit Myc-ER-induced apoptosis in response to tamoxifen (data not shown). Conceivably, the variants might even be induced as part of a surveillance mechanism to prevent cell proliferation in response to oncogenic signals, whereas subsequent selection for Mdm2 overexpression, Arf loss, or p53 mutations would bypass their effects, leaving them as inert vestigial markers during later stages in tumor development. Whatever the explanation, RING finger domains can act as potent growth inhibitors.

	ACKNOWLEDGMENTS

 
We thank Dr. John Cleveland for many helpful discussions and critical review of the manuscript, Jason Weber for initial help designing the oligonucleotides for PCR amplification of the Mdm2 variants from B-cell tumors, Frederique Zindy for preparing MEFs, and David Randle for preparing mouse bone marrow-derived pre-B cells. We are indebted to Rose Mathew and Camulous Hornsby for excellent technical assistance.

	FOOTNOTES

 
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 with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ This work was supported in part by NIH Grants CA-71907 (to M. F. R.), Cancer Center Core Grant CA-21765, and by the American Lebanese Syrian Associated Charities of St. Jude Children’s Research Hospital. C. J. S. is an investigator of the Howard Hughes Medical Institute.

² Present address: Eppley Cancer Institute, University of Nebraska Medical Center, 986805 Nebraska Medical Center, Omaha, NE 68198-6805.

³ To whom requests for reprints should be addressed, at Department of Tumor Cell Biology, DTRT 5006C, Mail Stop 350, St. Jude Children’s Research Hospital, 332 North Lauderdale, Memphis, TN 38105. Phone: (901) 495-3481/3597; Fax: (901) 495-2381; E-mail: martine.roussel{at}stjude.org

⁴ The abbreviations used are: RT-PCR, reverse transcription-PCR; HA, hemagglutinin; WT, wild type; MEF, mouse embryo fibroblast; FBS, fetal bovine serum; IL, interleukin; NLS, nuclear localization sequence; NES, nuclear export sequence; IRES, internal ribosomal entry site; GFP, green fluorescent protein; Ubc, ubiquitin conjugating.

⁵ H. Yasuda, personal communication.

Received 9/ 5/01. Accepted 12/14/01.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Cahilly-Snyder L., Yang-Feng T., Francke U., George D. L. Molecular analysis and chromosomal mapping of amplified genes isolated from a transformed mouse 3T3 cell line. Somatic Cell Mol. Gen., 13: 235-244, 1987.
2. Momand J., Jung D., Wilczynski S., Niland J. The MDM2 gene amplification database. Nucleic Acids Res., 26: 3453-3459, 1998.[Abstract/Free Full Text]
3. Juven-Gershon T., Oren M. Mdm2: the ups and downs. Mol. Med., 5: 71-83, 1999.[Medline]
4. Fakharzadeh S. S., Trusko S. P., George D. L. Tumorigenic potential associated with enhanced expression of a gene that is amplified in a mouse tumor cell line. EMBO J., 10: 1565-1569, 1991.[Medline]
5. Finlay C. A. The mdm-2 oncogene can overcome wild-type p53 suppression of transformed cell growth. Mol. Cell. Biol., 13: 301-306, 1993.[Abstract/Free Full Text]
6. Montes de Oca Luna R., Wagner D. S., Lozano G. Rescue of early embryonic lethality in mdm2-deficient mice by deletion of p53. Nature (Lond.), 378: 203-206, 1995.[Medline]
7. Jones S. N., Roe A. E., Donehower L. A., Bradley A. Rescue of embryonic lethality in Mdm-2-deficient mice by absence of p53. Nature (Lond.), 378: 206-208, 1995.[Medline]
8. Carroll P. E., Okuda M., Horn H. F., Biddinger P., Stambrook P. J., Gleich L. L., Li Y-Q., Tarapore P., Fukasawa K. Centrosome hyperamplification in human cancer: chromosome instability induced by p53 mutation and/or Mdm2 overexpression. Oncogene, 18: 1935-1944, 1999.[Medline]
9. Momand J., Zambetti G. P., Olson D. C., George D., Levine A. J. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell, 69: 1237-1245, 1992.[Medline]
10. Oliner J. D., Pietenpol J. A., Thiagalingam S., Gyuris J., Kinzler K. W., Vogelstein B. Oncoprotein MDM2 conceals the activation domain of tumour suppressor p53. Nature (Lond.), 362: 857-860, 1993.[Medline]
11. Chen J., Marechal V., Levine A. J. Mapping of the p53 and mdm-2 interaction domains. Mol. Cell. Biol., 13: 4107-4114, 1993.[Abstract/Free Full Text]
12. Kobet E., Zeng X., Zhu Y., Keller D., Lu H. MDM2 inhibits p300-mediated p53 acetylation and activation by forming a ternary complex with the two proteins. Proc. Natl. Acad. Sci. USA, 97: 12547-12552, 2000.[Abstract/Free Full Text]
13. Ito A., Lai C-H., Zhao X., Saito S., Hamilton M. H., Appella E., Yao T-P. p300/CBP-mediated p53 acetylation is commonly induced by p53-activating agents and inhibited by MDM2. EMBO J., 20: 1331-1340, 2001.[Medline]
14. Honda R., Tanaka H., Yasuda H. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett., 420: 25-27, 1997.[Medline]
15. Honda R., Yasuda H. Association of p19^ARF with Mdm2 inhibits ubiquitin ligase activity of MDM2 for tumor suppressor p53. EMBO J., 18: 22-27, 1999.[Medline]
16. Honda R., Yasuda H. Activity of MDM2, a ubiquitin ligase, toward p53 or itself is dependent on the RING finger domain of the ligase. Oncogene, 19: 1473-1476, 2000.[Medline]
17. Tao W., Levine A. J. Nucleocytoplasmic shuttling of oncoprotein Hdm2 is required for Hdm2-mediated degradation of p53. Proc. Natl. Acad. Sci. USA, 96: 3077-3080, 1999.[Abstract/Free Full Text]
18. Weber J. D., Taylor L. J., Roussel M. F., Sherr C. J., Bar-Sagi D. Nucleolar Arf sequesters Mdm2 and activates p53. Nat. Cell Biol., 1: 20-26, 1999.[Medline]
19. Lu W., Pochampally R., Chen L., Traidej M., Wang Y., Chen J. Nuclear exclusion of p53 in a subset of tumors requires MDM2 function. Oncogene, 19: 232-240, 2000.[Medline]
20. Lai Z., Ferry K. V., Diamond M. A., Wee K. E., Kim Y. B., Ma J., Yang T., Benfield P. A., Copeland R. A., Auger K. R. Human Mdm2 mediates multiple monoubiquitination of p53 by a mechanism requiring enzyme isomerization. J. Biol. Chem., 276: 31357-31367, 2001.[Abstract/Free Full Text]
21. Boyd S. D., Tsai K. Y., Jacks T. An intact HDM2 RING-finger domain is required for nuclear exclusion of p53. Nat. Cell Biol., 2: 563-568, 2000.[Medline]
22. Geyer R. K., Yu Z. K., Maki C. G. The MDM2 RING-finger domain is required to promote p53 nuclear export. Nat. Cell Biol., 2: 569-573, 2000.[Medline]
23. Stommel J. M., Marchenko N. D., Jimenez G. S., Moll U. M., Hope T. J., Wahl G. M. A leucine-rich nuclear export signal in the p53 tetramerization domain: regulation of subcellular localization and p53 activity by NES masking. EMBO J., 18: 1660-1672, 1999.[Medline]
24. Wu X., Bayle J. H., Olson D., Levine A. J. The p53-mdm-2 autoregulatory feedback loop. Genes Dev., 7: 1126-1132, 1993.[Abstract/Free Full Text]
25. Ries S., Biederer C., Woods D., Shifman O., Shirasawa S., Sasazuki T., McMahon M., Oren M., McCormick F. Opposing effects of Ras on p53: transcriptional activation of mdm2 and induction of p19^ARF. Cell, 103: 321-330, 2000.[Medline]
26. Maya R., Balass M., Kim S-T., Shkedy D., Leal J-F. M., Shifman O., Moas M., Buschmann T., Ronai Z., Shiloh Y., Kastan M. B., Katzir E., Oren M. ATM-dependent phosphorylation of Mdm2 on serine 395: role in p53 activation by DNA damage. Genes Dev., 15: 1067-1077, 2001.[Abstract/Free Full Text]
27. Sherr C. J. Tumor surveillance via the ARF-p53 pathway. Genes Dev., 12: 2984-2991, 1998.[Free Full Text]
28. Sherr C. J., Weber J. D. The ARF/p53 pathway. Curr. Opin. Genet. Dev., 10: 94-99, 2000.[Medline]
29. Zhang Y., Xiong Y. Control of p53 ubiquitination and nuclear export by MDM2 and ARF. Cell Growth Differ., 12: 175-186, 2001.[Abstract/Free Full Text]
30. Momand J., Wu H-H., Dasgupta G. MDM2—master regulator of the p53 tumor suppressor protein. Gene (Amst.), 242: 15-29, 2000.[Medline]
31. Daujat S., Neel H., Piette J. MDM2: life without p53. Trends Genet., 17: 459-464, 2001.[Medline]
32. Shvarts A., Steegenga W. T., Riteco N., van Laar T., Dekker P., Bazuine M., van Ham R. C. A., van der Houven van Oordt W., Hateboer G., van der Eb A. J., Jochemsen A. G. MDMX: a novel p53-binding protein with some functional properties of MDM2. EMBO J., 15: 5349-5357, 1996.[Medline]
33. Böttger V., Böttger A., Garcia-Echeverria C., Ramos Y. FM., van der Eb A. J., Jochemsen A. G., Lane D. P. Comparative study of the p53-mdm2 and p53-MDMX interfaces. Oncogene, 18: 189-199, 1999.[Medline]
34. Stad R., Ramos Y. F. M., Little N., Grivell S., Attema J., van der Eb A. J., Jochemsen A. G. Hdmx stabilizes Mdm2 and p53. J. Biol. Chem., 275: 28039-28044, 2000.[Abstract/Free Full Text]
35. Stad R., Little N. A., Xirodimas D. P., Frenk R., van der Eb A. J., Lane D. P., Saville M. K., Jochemsen A. G. Mdmx stabilizes p53 and Mdm2 via two distinct mechanisms. EMBO Rep., 21: 1029-1034, 2001.
36. Parant J., Chavez-Reyes A., Little N. A., Yan W., Reinke V., Jochemsen A. G., Lozano G. Rescue of embryonic lethality in Mdm4-null mice by loss of Trp53 suggests a nonoverlapping pathway with MDM2 to regulate p53. Nat. Genet., 29: 92-95, 2001.[Medline]
37. Fang S., Jensen J. P., Ludwig R. L., Vousden K. H., Weissman A. M. Mdm2 is a RING finger-dependent ubiquitin protein ligase for itself and p53. J. Biol. Chem., 275: 8945-8951, 2000.[Abstract/Free Full Text]
38. Kubbutat M. H. G., Ludwig R. L., Levine A. J., Vousden K. H. Analysis of the degradation function of Mdm2. Cell Growth Differ., 10: 87-92, 1999.[Abstract/Free Full Text]
39. Jackson M. W., Berberich S. J. MdmX protects p53 from Mdm2-mediated degradation. Mol. Cell. Biol., 20: 1001-1007, 2000.[Abstract/Free Full Text]
40. Elenbaas B., Dobbelstein M., Roth J., Shenk T., Levine A. J. The MDM2 oncoprotein binds specifically to RNA through its RING finger domain. Mol. Med., 2: 439-451, 1996.[Medline]
41. Olson D. C., Marechal V., Momand J., Chen J., Romocki C., Levine A. J. Identification and characterization of multiple mdm-2 protein and mdm-2-p53 protein complexes. Oncogene, 8: 2353-2360, 1993.[Medline]
42. Sigalas I., Calvert A. H., Anderson J. J., Neal D. E., Lunec J. Alternatively spliced mdm2 transcripts with loss of p53 binding domain sequences: transforming ability and frequent detection in human cancer. Nat. Med., 2: 912-917, 1996.[Medline]
43. Matsumoto R., Tada M., Nozaki M., Zhang C-L., Sawamura Y., Abe H. Short alternative splice transcripts of the mdm2 oncogene correlate to malignancy in human astrocytic neoplasms. Cancer Res., 58: 609-613, 1998.[Abstract/Free Full Text]
44. Kraus A., Neff F., Behn M., Schuermann M., Muenkel K., Schlegel J. Expression of alternatively spliced mdm2 transcripts correlates with stabilized wild-type p53 protein in human glioblastoma cells. Int. J. Cancer, 80: 930-934, 1999.[Medline]
45. Evans S. C., Viswanathan M., Grier J. D., Narayana M., El-Naggar A. K., Lozano G. An alternatively spliced HDM2 product increases p53 activity by inhibiting HDM2. Oncogene, 20: 4041-4049, 2001.[Medline]
46. Bartel F., Meye A., Würl P., Kappler M., Bache M., Lautenschläger C., Grünbaum U., Schmidt H., Taubert H. Amplification of the MDM2 gene, but not expression of splice variants of MDM2 MRNA, is associated with prognosis in soft tissue sarcoma. Int. J. Cancer, 95: 168-175, 2001.[Medline]
47. Lukas J., Gao D-Q., Keshmeshian M., Wen W-H., Tsao-Wei D., Rosenberg S., Press M. F. Alternative and aberrant messenger RNA splicing of the mdm2 oncogene in invasive breast cancer. Cancer Res., 61: 3212-3219, 2001.[Abstract/Free Full Text]
48. Tamborini E., Della Torre G., Lavarino C., Azzelli A., Carpinelli P., Pierotti M. A., Pilotti S. Analysis of the molecular species generated by MDM2 gene amplification in liposarcomas. Int. J. Cancer, 92: 790-796, 2001.[Medline]
49. Hori M., Shimazaki J., Inagawa S., Itabashi M., Hori M. Alternatively spliced MDM2 transcripts in human breast cancer in relation to tumor necrosis and lymph node involvement. Path. Int., 50: 786-792, 2000.
50. Eischen C. M., Weber J. D., Roussel M. F., Sherr C. J., Cleveland J. L. Disruption of the ARF-Mdm2–p53 tumor suppressor pathway in Myc-induced lymphomagenesis. Genes Dev., 13: 2658-2669, 1999.[Abstract/Free Full Text]
51. Roussel M. F., Theodoras A. M., Pagano M., Sherr C. J. Rescue of defective mitogenic signaling by D-type cyclins. Proc. Natl. Acad. Sci. USA, 92: 6837-6841, 1995.[Abstract/Free Full Text]
52. Persons D. A., Allay J. A., Allay E. R., Smeyne R. J., Ashmun R. A., Sorrentino B. P., Nienhuis A. W. Retroviral-mediated transfer of the green fluorescent protein gene into murine hematopoietic cells facilitates scoring and selection of transduced progenitors in vitro and identification of genetically modified cells in vivo. Blood, 90: 1777-1786, 1997.[Abstract/Free Full Text]
53. Zindy F., Eischen C. M., Randle D., Kamijo T., Cleveland J. L., Sherr C. J., Roussel M. F. Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev., 12: 2424-2433, 1998.[Abstract/Free Full Text]
54. Randle D. H., Zindy F., Sherr C. J., Roussel M. F. Differential effects of p19^Arf and p16^Ink4a loss on senescence of murine bone marrow-derived pre-B cells and macrophages. Proc. Natl. Acad. Sci. USA, 98: 9654-9659, 2001.[Abstract/Free Full Text]
55. Quelle D. E., Zindy F., Ashmun R. A., Sherr C. J. Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell, 83: 993-1000, 1995.[Medline]
56. Evan G. I., Lewis G. K., Ramsay G., Bishop J. M. Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product. Mol. Cell. Biol., 5: 3610-3616, 1985.[Abstract/Free Full Text]
57. Weber J. D., Kuo M-L., Bothner B., DiGiammarino E. L., Kriwacki R. W., Roussel M. F., Sherr C. J. Cooperative signals governing ARF-Mdm2 interaction and nucleolar localization of the complex. Mol. Cell. Biol., 20: 2517-2528, 2000.[Abstract/Free Full Text]
58. Skowyra D., Craig K. L., Tyers M., Elledge S. J., Harper J. W. F-box proteins are receptors that recruit phosphorylated substrates to the SCF ubiquitin-ligase complex. Cell, 91: 209-219, 1997.[Medline]
59. Perry M. E., Mendrysa S. M., Saucedo L. J., Tannous P., Holubar M. p76^MDM2 inhibits the ability of p90^MDM2 to destabilize p53. J. Biol. Chem., 275: 5733-5738, 2000.[Abstract/Free Full Text]
60. Brown D. R., Thomas C. A., Deb S. P. The human oncoprotein MDM2 arrests the cell cycle: elimination of its cell-cycle-inhibitory function induces tumorigenesis. EMBO J., 17: 2513-2525, 1998.[Medline]
61. Kamijo T., Weber J. D., Zambetti G., Zindy F., Roussel M. F., Sherr C. J. Functional and physical interactions of the ARF tumor suppressor with p53 and Mdm2. Proc. Natl. Acad. Sci. USA, 95: 8292-8297, 1998.[Abstract/Free Full Text]
62. Stott F. J., Bates S., James M. C., McConnell B. B., Starborg M., Brookes S., Palmero I., Ryan K., Hara E., Vousden K. H., Peters G. The alterna