Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Screening E3 Substrates Using a Live Phage Display Library

  • Zhengguang Guo ,

    Contributed equally to this work with: Zhengguang Guo, Xiaorong Wang

    Affiliations Department of Physiology and Pathophysiology, National Key Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences/School of Basic Medicine, Peking Union Medical College, Beijing, China, Department of Core Instrument Facility, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences/School of Basic Medicine, Peking Union Medical College, Beijing, China

    ⨯
  • Xiaorong Wang ,

    Contributed equally to this work with: Zhengguang Guo, Xiaorong Wang

    Affiliation Department of Physiology and Pathophysiology, National Key Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences/School of Basic Medicine, Peking Union Medical College, Beijing, China

    ⨯
  • Huihua Li,

    Affiliation Department of Pathology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences/School of Basic Medicine, Peking Union Medical College, Beijing, China

    ⨯
  • Youhe Gao

    gaoyouhe@pumc.edu.cn

    Affiliation Department of Physiology and Pathophysiology, National Key Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences/School of Basic Medicine, Peking Union Medical College, Beijing, China

    ⨯

Abstract

Ubiquitin ligases (E3s) determine specificity of ubiquitination by recognizing target substrates. However, most of their substrates are unknown. Most known substrates have been identified using distinct approaches in different laboratories. We developed a high-throughput strategy using a live phage display library as E3 substrates in in vitro screening. His-ubiquitinated phage, enriched with Ni-beads, could effectively infect E. coli for amplification. Sixteen natural potential substrates and many unnatural potential substrates of E3 MDM2 were identified through 4 independent screenings. Some substrates were identified in different independent experiments. Additionally, 10 of 12 selected candidates were ubiquitinated by MDM2 in vitro, and 3 novel substrates, DDX42, TP53RK and RPL36a were confirmed ex vivo. The whole strategy is rather simple and efficient. Non-degradation substrates can be discovered. This strategy can be extended to any E3s as long as the E3 does not ubiquitinate the empty phage.

Citation: Guo Z, Wang X, Li H, Gao Y (2013) Screening E3 Substrates Using a Live Phage Display Library. PLoS ONE 8(10): e76622. https://doi.org/10.1371/journal.pone.0076622

Editor: Irina U Agoulnik, Florida International University, United States of America

Received: May 11, 2013; Accepted: August 26, 2013; Published: October 4, 2013

Copyright: © 2013 Guo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by the National Basic Research Program of China (2012CB517606, 2013CB530805, 2011CB964901), the National High Technology Research and Development Program of China (2011AA020116), Program for Changjiang Scholars and Innovative Research Team in University-PCSIRT (IRT0909), and 111 Project (B08007). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Ubiquitination is a key mechanism in regulating many biological processes, such as proteasome degradation, endocytic trafficking, DNA repair, signal transduction and protein-protein interaction [1,2]. In these processes, ubiquitin ligases (E3s) determine the specificity of ubiquitination by recognizing target substrates [1,3-5]. However, most of their substrates that account for the biology are not known [1]. Identification of these substrates is a major challenge. Most substrates have been identified using distinct approaches in different laboratories. Better, faster and cheaper proteome-wide methods are needed to identify E3 substrates [1].

Recently, several powerful methods, including in vitro protein microarrays [2-6], ex vivo label-free [7] or stable isotope labeling by amino acids in cell culture (SILAC) [8,9] quantitative mass spectrometry and global protein stability (GPS) profiling [10,11] have been employed to identify E3 substrates. In vitro protein microarrays have been successfully used to explore substrates of the anaphase promoting complex (APC) [4], SMURF1 [3] and Nedd4 family E3s [2,5,6]. Thousands of candidate proteins were individually expressed, purified, and spotted on arrays. Then, the candidate proteins were incubated with a reaction mixture containing the E3 of interest and FITC-labeled ubiquitin under specific conditions. This is a high-throughput method and can be applied to low amounts of substrate [7] but is limited by the quantity and variety of candidate proteins covered by the arrays [2,5]. As ex vivo methods, the label-free and SILAC quantitative mass spectrometry strategies enable the identification of native substrates in physiologically relevant settings [7,8] and have no limitations on the quantity and variety of candidates. Another ex vivo method, global protein stability (GPS) profiling of a library of 8000 open reading frames (ORFs) coupled with either a flow cytometry enrichment strategy or a DNA microarray deconvolution strategy was employed to search for SCF substrates in mammalian cells [10,11]. Both the mass spectrometry-based and GPS profiling ex vivo methods could only identify the E3 substrates degraded by the proteasome; also, the possibility that candidate proteins decreased following the expression of functional E3 but not due to E3-mediated ubiquitination cannot be excluded. Simpler, more efficient systematic methods are needed to identify E3 substrates.

Phage display is a high-throughput method for the study of protein-protein, protein-peptide, and protein-DNA interactions that uses bacteriophages to connect proteins with the genetic information that encodes them [12]. Phage display has been used in screening enzyme inhibitors and substrates [13]. It was used to determine substrate specificity of proteases [14,15], kinases [16,17], and transglutaminases [18,19]. Here, we developed a screening method using a live phage display library as E3 substrates.

MDM2 was used as an example to explain and evaluate our strategy. The mdm2 gene was originally identified as one of three genes amplified in tumorigenic mouse cells derived from the NIH3T3 cell line [20,21]. MDM2 acts as an onco-protein that affects the cell cycle, apoptosis, and tumorigenesis through interactions with other proteins or ubiquitination of its substrates. MDM2 is located in many tissues, such as brain, placenta, uterus and lymph node, and is upregulated in many tumor issues [22-24]. It functions as an E3 that mediates ubiquitination of P53/TP53, leading to its proteasome-dependent degradation [25,26]. In addition to P53/TP53, many substrates of MDM2 have been identified individually in different laboratories; these substrates include CDKN1A [27-29], HIPK2 [30,31], RB1 [32-34], CDH1 [35], DLG4 [36], IGF1R [37], APEX1 [38], ADRBK1 [39], ARRB1 [40], ARRB2 [41], CREBBP [42], EID1 [43], IRS1 [40], JMY [44], KAT2B [45], KAT5 [46], MDM4 [47,48], NFATC2 [49], NOL3 [50], RPL26 [51] and itself [52].

Materials and Methods

1. Preparation of plasmids and mutagenesis

For the in vitro ubiquitination assay, candidate MDM2 substrates were amplified by PCR from phage clones and subcloned into EcoRI/HindIII sites or EcoRI/XhoI sites of the PET32b+ vector (69016-3, Novagen) (Madison, WI) and fused with His- and S-tags at the N-terminal of the substrates. For the ex vivo ubiquitination assay, the full-length human MDM2 (1-491aa) and MDM2Δring (1-435aa) were amplified by PCR from IMAGE: NM_002392.1 (GeneCopoeia Inc.) (Rockville, MD) and subcloned into the BamHI/XhoI sites of the pcDNA6 vector and fused with an N-terminal Flag tag. pEGFP-N1-RPL36AL-GFP, pEGFP-N1-TP5RK-GFP and pEGFP-N1-DDX42-GFP plasmids were all purchased from GeneChem (Shanghai, China), whose gene sequences were obtained from BC000741, BC009727 and BC015505 respectively, and subcloned into the pEGFP-N1 vector and fused with a C-terminal GFP tag. GFP-tagged TP53 plasmid was purchased from GeneCopoeia Inc. (Rockville, MD), whose gene sequence was obtained from IMAGE: NM_000546.2 and subcloned into the pReceiver-M98 vector and fused with a C-terminal eGFP tag. The HA-tagged ubiquitin plasmid and the His-Myc-tagged ubiquitin plasmid were kindly provided by Dr. Huihua Li (Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences).

2. Antibodies

All the antibodies used in our study were purchased from commercial sources: anti-S-tag (ab18616, Abcam) (Cambridge, UK), anti-T7-tag (ab9138, Abcam), anti-Flag-tag (DYKDDDDK) (M20008, Abmart) (Shanghai, China), anti-GFP-tag (M20004, Abmart), anti-HA-tag (M20003, Abmart) and anti-β-actin (A1978, Sigma) (St Louis, MO).

3. Protein expression and purification

The pET 32b+ constructs were expressed in the E. coli BL21 (DE3) strain. The expression of His-tagged proteins was induced by exposure to 0.2 mM IPTG at 30°C for 4-6 hours. The fusion proteins were enriched using MagExtractor(His-tag) (NPK-701, TOYOBO) (Osaka, Japan). The enriched proteins were visualized using SDS PAGE stained with Coomassie Blue and quantified using the Bradford method.

4. Ubiquitination screening using live phage as substrate

4.1. Positive and negative selections.

A T7Select Human Brain cDNA Library (70637-3, Novagen) was used for MDM2 substrate screening. Approximately 1.7×107 PFU phage was added to the in vitro ubiquitination system containing 110 ng ubiquitin activating enzyme (E1) (UBE1, human recombinant) (E-305, Boston Biochem) (Cambridge, MA), 300-500 ng ubiquitin conjugating enzyme (E2) (UbcH5b, human recombinant) (E2-622, Boston Biochem), 300-830 ng GST-MDM2 (E3-202, Boston Biochem) or GST, 2-3 µg His-ubiquitin (U530, Boston Biochem), 50 mM Tris-HCl (pH=7.4), 7.5 mM MgCl2, 3 mM ATP and 1 mM DTT in 20µL volumn. The reaction mixture was incubated at 30°C for 90 min and stopped by the addition of 80 µl Urea Buffer (containing 4 M urea, 10 mM Tris-HCl, 100 mM Na2HPO4, pH=8.0). Five microliters Ni-agarose beads were added to the reaction mixture and incubated for 1 h at room temperature. For negative selection, the supernatant was collected and amplified in E. coli. BLT5403 (69142, Novagen). For positive selection, the supernatant was removed, and the beads were washed with 100 µl Washing Buffer (containing 4 M urea, 10 mM Tris-HCl, 100 mM Na2HPO4, 1% Triton X-100, pH=8.0) for 5 times. The bound phage was eluted from the beads using 200 mM imidazole for 15 min. The eluted phages were amplified in E. coli. BLT5403. The phage titer of the amplified sub-library was determined and put into the in vitro ubiquitination system again for a second round of selection. After 3-5 rounds of selection, the phage titer of the eluent from the last positive selection was determined. Individual clones were selected from the titer plates, amplified by PCR using the T7Select UP and DOWN primers. PCR products were sequenced to obtain the potential E3-substrate coding sequences.

4.2. Trypsin treatment of the ubiquitinated phage.

Speculating that phage with a long poly-ubiquitin chain may have a lower E. coli infection efficiency. Poly-ubiquitinated phages were digested before infection in the hope that removing the long ubiquitin chain would equalize the infection efficiencies of both poly- and mono-ubiquitinated substrates. The T7 cDNA library was added to the in vitro ubiquitination system containing E1, E2, and E3 (MDM2). The reaction product was incubated with Ni-beads, and the beads were washed 5 times with Washing Buffer as described above. The bound phage was treated with 0.125% trypsin (cc035, MACGENE) (Beijing, China) at 37°C for 30 min. After that, the sample was centrifuged and the supernatant was collected. The remaining bound phages were eluted with 200 mM imidazole for 15 min. All the phages were collected from digestion and elution solutions and amplified in BLT5403.

5. In vitro ubiquitination assay

In vitro ubiquitination reactions were performed using 110 ng E1, 500 ng E2, 830 ng E3(GST-MDM2) or GST, 3 μg His-ubiquitin, 50 mM Tris-HCl (PH=7.4), 7.5 mM MgCl2, 3 mM ATP, 1 mM DTT and 100-200 ng substrate in 20µL volume. The reaction mixture was incubated at 30°C for 90 min and stopped by adding 5×SDS sample loading buffer (Genestar Biosolutions, Beijing, China). Proteins were separated by SDS PAGE and immunoblotted with anti-S-tag antibody.

6. Ex vivo ubiquitination and degradation assays

HEK293T cells, used for the ex vivo assays, were cultured under conventional conditions at 37.0°C with 5% carbon dioxide (CO2). HEK293T cells were transfected with Flag-tagged MDM2, HA-tagged ubiquitin and GFP-tagged substrates using I’ma Fect Transfection Reagent (IMA201101, IMAGEN). Harvested cells were lysed using RIPA lysis buffer (P1053, Applygen, containing 50 mM Tris-HCl, 150 mM NaCl, 1% NP-40, 0.1% SDS, pH=7.4) (Beijing, China) with 2 mM PMSF and sonication. For each immunoprecipitation, cell lysates were centrifuged at 12000rpm for 10min, and the supernatants were incubated with the appropriate antibodies at 4°C overnight. The protein G-agarose (P2009, Beyotime) (Shanghai, China) was added, allowed to bind the immunocomplex for 1-2 h, and then washed 3 times with cold PBS. Proteins were separated by SDS PAGE and immunoblotted with the appropriate antibodies. In experiments using a proteasome inhibitor, transfected cells were incubated with 10-20 µM MG132 (S1748, Beyotime) for 12 h before harvest.

7. Ingenuity pathways analysis (IPA)

IPA software (Ingenuity Systems, Redwood City, CA) was used to investigate possible interactions among all the identified substrates and MDM2. A protein network is a graphic representation of molecular relationship between molecules. Nodes represent molecules, and the lines represent direct and indirect biological relationship. All the lines are supported by at least one reference.

Results

1. E3-substrate screening using a live phage display library

The newly developed strategy involves the following steps, as illustrated in Figure 1.

thumbnail
Download:
Figure 1. Schematic of the proteomic strategy using a live phage display library as E3 substrates for screening.

In negative selection, the library that displayed human brain cDNA was incubated in the in vitro ubiquitination system containing E1, E2, and His-tagged ubiquitin without E3. The reaction product was incubated with Ni-beads. The unbound phage was collected and amplified in E. coli BLT5403. The amplified sublibrary was used for positive selection. In positive selection, the sublibrary was incubated in the in vitro ubiquitination system containing E1, E2, target E3 and His-tagged ubiquitin. The substrates that displayed on phages were mono- or poly-ubiquitinated by the E3 in the system. Then, the reaction products were incubated with Ni-beads. Phages with a His-ubiquitin tag were captured by Ni-beads. The bound phage was further eluted by imidazole and proliferated in BLT5403. The amplified sublibrary from positive selection was put into a second round of positive or negative selection. After several rounds of selection, individual clones were amplified by PCR and sequenced to obtain the potential E3-substrate coding sequences.

https://doi.org/10.1371/journal.pone.0076622.g001

First, to ensure the low background for successful screening in our strategy, E3 should not ubiquitinate the empty phage.

Second, Negative selection. Negative selection will eliminate the phage that bound the beads in the system without E3. The library that displayed human brain cDNA was incubated in the in vitro ubiquitination system containing E1, E2, and His-tagged ubiquitin without E3. The reaction product was incubated with Ni-beads. The unbound phage was collected and amplified in BLT5403. The amplified sub-library was used for positive selection.

Third, Positive selection. The sub-library was incubated in the in vitro ubiquitination system containing E1, E2, E3 and His-tagged ubiquitin. The substrates that displayed were mono- or poly-ubiquitinated by the E3 in the system. Then, the reaction products were incubated with Ni-beads. Phages with the His-ubiquitin tag were captured by Ni-beads. The bound phages were further eluted by imidazole and proliferated in BLT5403.

Fourth, combination of multiple rounds of positive and negative selection. Multiple rounds of positive selection would further enrich phages that were ubiquitinated by MDM2. The amplified sub-library from positive selection was put into the second round of positive or negative selection. After several rounds of selection, individual clones were PCR amplified and sequenced to obtain the potential E3-substrate coding sequences.

2. Screening MDM2 substrates

To test this substrate screening strategy, nonspecific ubiquitination of empty phage (T7Select 10-3, which were kindly provided by Lina Zhang (Beijing Tuberculosis and Thoracic Tumor Research Institute)) by the ubiquitin ligase MDM2 was evaluated. As shown in Figure 2A, empty phage, when as the substrate in the in vitro ubiquitination system containing GST-MDM2 as E3, had low nonspecific absorption to the Ni-beads (1.2×103 PFU), as did empty phages not subjected to the reaction (1.2×103 PFU), compared to the input (2×107 PFU). Western blotting with anti-T7 tag was performed to further validate that the T7-tagged coat protein 10B of the empty phage was not ubiquitinated by MDM2 (Figure 2B). MDM2 did not ubiquitinate empty phage.

thumbnail
Download:
Figure 2. Ubiquitin ligase MDM2 does not ubiquitinate the empty phage.

A. Empty phage, after the reaction as substrate in the in vitro ubiquitination system containing GST-MDM2 as E3, had low nonspecific absorption to the Ni-beads (1.2×103 PFU), as did empty phages not subjected to the reaction (1.2×103 PFU), compared to the input (2×107 PFU). B. MDM2 did not ubiquitinate the coat protein of T7 phage in Western blot analysis.

https://doi.org/10.1371/journal.pone.0076622.g002

Overall, 107 empty phages and T7 cDNA phage library were added into separate in vitro ubiquitination systems; the number of eluted T7 cDNA library phage (26×102 PFU) was a few times higher than that of eluted empty phage (6×102 PFU). Therefore the estimated false positive rate of one round of positive selection without negative selection is 23% (6/26). Negative selection and multiple rounds of positive selection will further reduce false positive rate. Substituting different enzymes for E3 might produce different false positive rates.

Four independent experiments were used to screen MDM2 substrates. The sequences of positive and negative selection in each experiment are described in Table 1 row 2. The in vitro ubiquitination system is described in Table 1 rows 3-6. In Experiments 3 and 4, higher concentrations of E1, E2, E3 and His-ubiquitin but fewer rounds of selection were used. In Experiment 4, each positive selection round was followed by trypsin treatment prior to infection, as described above (Table 1 row 7).

Experiment1  Experiment2  Experiment3  Experiment4  Different cloneDirect associationOther family memberNetworkin vitro validationex vivo validation
Selection sequenceNPPPNPPNPNPNPNPPPNPPP
Reaction systemE1110ng110ng110ng110ng
E2300ng300ng500ng500ng
E3300ng300ng830ng830ng
His-Ubiquitin2ug2ug4ug4ug
Trypsin treatment---+
Total clones40408353
Natural proteins2263628
NOLC1154520++++NA
DDX42122-++++
MAP22-+++NA
NUCKS12++++NA
RRP11+NANA
RBBP611-++++NA
PDS51+NANA
RPL36A1++++
UBTF2++NANA
TP53RK1+++
RPL1543++++NA
NPIPL31+NANA
HMGN11--NA
MAP1LC3A1+NANA
PRDM21++-NA
C12orf352-++NA
Non-natural proteins18344725

Table 1. All substrates identified in 4 independent screenings.

Each N or P represents one round of negative or positive selection, respectively; +: positive result; -: negative result; NA: do not tested.
CSV
Download CSV

After four independent screenings, 216 clones were obtained. Of these clones, 92 encode proteins with correct ORFs of known proteins, for a total of 16 proteins, as shown in Table 1 column 1. All the raw data are shown in Tables S1-S4 and Text S1. NOLC1 was identified in all four experiments. DDX42 was identified in Experiments 1 and 3. RBBP6 was identified in Experiments 1 and 2. RPL15 was identified in Experiments 3 and 4. Four proteins (NOLC1, RPL15, NUCKS1, UBTF) were identified in different clones encoding the same protein (Table 1 column 6). In total, 8 proteins were identified at least twice. These proteins are therefore identified with high-confidence as putative MDM2 substrates. These proteins were validated with in vitro and ex vivo experiments, discussed in detail later in the results.

We analyzed the 16 potential MDM2 substrates using IPA software. As shown in Figure 3, 15 proteins were involved in the MDM2-P53 Network (Table 1 column 9). Of these 15 proteins, 4 have previously been determined to interact or have a functional association with MDM2 by independent laboratories (Table 1 column 7). Endogenous PACT (RBBP6) [53] can interact with MDM2 and enhance MDM2-mediated ubiquitination and degradation of P53, and increase the P53-MDM2 affinity. Endogenous NOLC1 [54] interacts with the promoter of MDM2 together with P53 to activate the MDM2 promoter in NPC cells. A large-scale protein interaction study using an mRNA display showed NUSCK1 and MAP2 interact with MDM2 [55]. Another 11 proteins, which were indirectly associated with MDM2, also existed in the MDM2-P53 network. For example, PRDM2 interacts with P53 and regulates the proliferation of monocytic leukemia cells via activation of P53 [56]. PRDM2 also interacts with Rb1, another MDM2 substrate [57,58]. P53 posttranscriptionally down-regulates autophagy protein LC3 (MAP1LC3), supporting cancer cell survival under prolonged starvation [59]. Another large scale protein interaction study by mass spectrometry discovered that PDS5B interacts with MYC, a significant MDM2 regulator [60]. TP53RK binds to TP53, and phosphorylates P53 [TP53] protein to phosphorylated (S15) P53 [TP53] protein [61]. Mutant RB1 (C706F) interacts with and decreases DNA binding activity of human UBF1 (UBTF) protein [62-64]. P53 prevents the interaction between SL1 and UBTF, and represses RNA Pol I transcription activity [65]. Chromatin-associated factor RRP1B interacts with RPL5, a MDM2 binding protein [66]. An important MDM2 regulator, PTEN increases expression of mouse RPL44 [RPL36a] mRNA in mouse mammary gland tissue [67]. The family members of 4 proteins are known physiological substrates or interactors of MDM2 (Table 1 column 8). RPL36a and RPL15 are members of the RPL family. As members of this family, L5 [68], L11 [69] and L23 [70,71] interact with MDM2 to inhibit MDM2-mediated P53 ubiquitination. Another member of this family, L26 [51], binds to MDM2, which leads to its polyubiquitinylation and proteasomal degradation. DDX42 is a member of the DEAD box (DDX) family, one member of which, DDX17 [72], combines with p300 and P/CAF to stimulate MDM2 promoters. PRDM2 is a member of the PRDM (PRDI-BF1 and RIZ domain containing) family, one member of which, PRDM5 [73], down-regulates MDM2 gene expression to inhibit tumor cell clonogenicity and cell proliferation.

thumbnail
Download:
Figure 3. Protein network analysis of MDM2 substrates by Ingenuity Pathway Analysis software (IPA).

The genes shaded reds are substrates selected in our strategy. Solid lines represent direct interactions, dotted lines represent indirect interactions. Arrows from one node to another indicate that this node acts upon the other. Lines without arrows represent binding. Node shapes are: vertical diamond means enzyme; dotted rectangle means ion channel; inverted triangle means kinase; horizontal diamond means peptidase; triangle means phosphatase; horizontal oval means transcription regulator; double-circle means complex/group; trapezium means microRNA; semicircle means mature microRNA; circle means other;.

https://doi.org/10.1371/journal.pone.0076622.g003

The remaining 124 clones do not encode natural proteins with correct ORFs. Some clones match the genome sequence but not the cDNA sequence, some match the non-coding region of the cDNA, some match reverse cDNA sequences, some have frameshifts, and others match neither the genome nor the cDNA sequences. These results indicated that the library also displayed other peptides. Twelve polypeptides were identified at least twice in the 124 clones, as shown in Table S5. One was identified in different independent experiments, and at least six were identified in different clones that encode the same peptides. These peptides are high-confidence unnatural substrates of MDM2. Interestingly, many clones contain poly-lysine, which likely occurred for several reasons. MDM2 has Asp/Glu-rich (acidic residue-rich) region at 243-301 aa and might interact with basic residue-rich poly-lysine peptides and cause its ubiquitination. The enrichment in the cDNA library of the AAA codon derived from the poly-A region of mRNA, which encodes lysine, may be the cause of poly-lysine peptides in the library.

3. In vitro validation of MDM2 substrates

One potential caveat of this strategy is that the N-terminal phage coat protein fusion might affect protein folding and ubiquitination. To validate that the identified substrates are ubiquitinated by MDM2, 11 natural novel potential substrates and 1 unnatural potential substrates of MDM2 were selected from the phage display screening and inserted into the bacterial expression plasmid PET32b+. The empty vector was used as a negative control. Each recombinant protein was fused with a tandem His and S tag at the N-terminus for purification and detection. Each recombinant protein was added to the in vitro ubiquitination reaction containing E1, E2 (UbcH5B), E3 (GST-MDM2 or GST) and ubiquitin. Ubiquitination of the potential substrates was detected by anti-S tag antibody.

As shown in Figure 4 and summarized in Table 1 column 10, 10 of the 12 potential substrates identified in the phage display screening were ubiquitinated by the ubiquitin ligase. Among them, RPL15a was mono-ubiquitinated by MDM2 in the in vitro ubiquitination assay. MAP2 was bi-ubiquitinated by MDM2 in the in vitro ubiquitination assay. NOLC1, NUSCK1 and RPL36a were mainly oligo-ubiquitinated by MDM2. TP53RK, DDX42, C12orf35, RBBP6, and the unnatural potential substrate were poly-ubiquitinated by GST-MDM2 compared to GST in solution. However, the ubiquitination of the PRDM2 and HMGN1 was not detected. The protein expressed by the empty vector of PET32b+, used as a negative control, was not ubiquitinated by MDM2.

thumbnail
Download:
Figure 4. In vitro validation of candidate MDM2 substrates selected by the phage display strategy.

Candidate substrates TP53RK (A), DDX42 (B), RBBP6 (C), C12orf35 (D), the unnatural substrate 1# (E), RPL15 (F), MAP2 (G), NUSCK1 (H), NOLC1 (I), RPL36a (J), PRDM2 (K), HMGN1 (L) and the negative control (M) were purified and subjected to traditional liquid ubiquitination reactions. They were then separated by SDS PAGE and immunoblotted with an anti-S tag antibody. All candidate substrates were ubiquitinated by MDM2, except PRDM2 (K) and HMGN1 (L). *: ubiquitination bands. SE: short exposure. LE: long exposure.

https://doi.org/10.1371/journal.pone.0076622.g004

4. Ex vivo confirmation of MDM2 substrates

Three novel potential substrates of MDM2, including DDX42, RPL36a, and TP53RK, were further tested in ex vivo ubiquitination assays. In the ex vivo ubiquitination assay, Flag-tagged MDM2 and MDM2Δring were transfected into HEK293T cells with GFP-tagged putative substrates and HA-tagged ubiquitin. MDM2Δring is a truncated form of MDM2 without E3 activity. Putative full-length substrates were immunoprecipitated from cell lysates with anti-GFP antibody and immunoblotted with anti-HA antibody to detect their ubiquitination. As shown in Figure 5 and summarized in Table 1 column 11, in the presence of Flag-MDM2, DDX42 (Figure 5A) and TP53RK (Figure 5B) were significantly ubiquitinated by MDM2 compared to Flag-MDM2 non-overexpressed cells, as indicated by the high-molecular-weight ladders and smear, while very faint ubiquitination bands were observed when MDM2Δring was over-expressed. RPL36a (Figure 5C) was ubiquitinated when MDM2 was co-expressed, while weaker ubiquitination bands were also observed in the MDM2Δring lane. In fact, RPL36a poly-ubiquitination increased by more than 2 fold in MDM2 over-expressed cells, while only 1 fold increase was observed in MDM2Δring over expressed cells (the raw quantification data was shown in Table S6). Empty vector of GV142 expressed only GFP protein, which was used as a negative control, was not ubiquitinated by MDM2, indicating that the GFP tag was not ubiquitinated (Figure 5D); P53, a well-known substrate of MDM2, which was used as a positive control, was significantly ubiquitinated by MDM2 (Figure 5E).

thumbnail
Download:
Figure 5. Confirmation of MDM2 substrates in mammalian cells by an ex vivo ubiquitination assay.

HEK293T cells were transfected with the GFP-tagged substrates alone or co-transfected with Flag-tagged MDM2 (either WT or a catalytically inactive MDM2Δring mutant). The GFP-tagged substrate was immunoprecipitated from cell lysates with the anti-GFP antibody and immunoblotted with the anti-HA antibody to detect its ubiquitination. DDX42 (A) and TP53RK (B) were ubiquitinated when MDM2 was co-expressed. TP53RK and DDX42 were ubiquitinated when MDM2 was co-expressed, whereas only very faint ubiquitination bands were observed in the MDM2Δring lane. RPL36a (C) was ubiquitinated when MDM2 was co-expressed, while weaker ubiquitination bands were observed in the MDM2Δring lane. To calculate the relative extent of RPL36a poly-ubiquitination, the poly-ubiquitination signal of RPL36a (poly-Ub-RPL36a) was quantified using ImageJ software, divided by the corresponding GFP-RPL36a signal and normalized to 1.0. HEK293T (D) cells were transfected with empty vector of GV142 expressed only GFP protein (negative control) alone or with co-transfected Flag-tagged MDM2 (either WT or a catalytically inactive MDM2Δring mutant). The GFP was immunoprecipitated from cell lysates with the anti-GFP antibody and immunoblotted with the anti-HA antibody to detect its ubiquitination. GFP itself was not ubiquitinated by MDM2. HA-ubiquitin (E) and GFP-P53 (positive control) were co-transfected into HEK293T cells. The GFP-P53 was immunoprecipitated from cell lysates with the anti-GFP antibody and immunoblotted with the anti-HA antibody to detect its ubiquitination. P53 was ubiquitinated when MDM2 was co-expressed. GFP tagged TP53RK (F) or RPL36a (G) were transfected into HEK293T cells alone or co-transfected with Flag-tagged MDM2. The protein content of the lysates of the transfected cells were separated and immunoblotted using the anti-GFP antibody. The expression of MDM2 induced the significant degradation of transfected TP53RK and RPL36a. When the transfected cells were treated with 10 or 20 μM MG132, the degradation of TP53RK and RPL36a were blocked. *: ubiquitination bands.

https://doi.org/10.1371/journal.pone.0076622.g005

The ubiquitination of cellular proteins often leads to their proteasome-dependent degradation. We evaluated whether MDM2-induced ubiquitination causes the degradation of DDX42, RPL36a and TP53RK. GFP-tagged DDX42, RPL36a and TP53RK were co-transfected into HEK293T cells with Flag-tagged MDM2. Proteins in the lysates of transfected cells were separated and immunoblotted with anti-GFP antibody. As shown in Figure 5, the over-expression of MDM2 induced the significant degradation of GFP-tagged TP53RK (Figure 5F) and RPL36a (Figure 5G) (the raw quantification data was shown in Table S7), while degradation was blocked when treated with 10 or 20 µM MG132 for 12 h. Therefore, MDM2 promoted the degradation of TP53RK and RPL36a in a proteasome-dependent manner. However, the degradation of DDX42 mediated by MDM2 was not detectable (Figure S1). This finding suggested that the ubiquitination of DDX42 does not lead to its degradation. The ubiquitination of DDX42 might serve as a nonproteolytic signal in cellular processes.

Discussion

This strategy is notably different from other approaches in the following respects. First, the whole strategy is simple and efficient. The expression and purification of numerous candidate proteins were unnecessary; the process of phage display screening was very simple and not time-consuming. The whole selection process will be finished less than two weeks, and the cost for sequencing DNA of phage clones is much lower than protein identification. Second, the low false-positive rate (2 of 12) of the screen is favorable compared with those of most high through-put microarray-based approaches [2,11]. Third, not only poly-ubiquitinated and degradation substrates but also the mono-/oligo-ubiquitinated and non-degradation substrates of the target E3s can be discovered. Assembly of a chain of at least four ubiquitins linked together via their Lys48 residue leads to the 26S proteasome-dependent degradation of targeted cellular proteins; in contrast, mono-ubiquitination or poly-ubiquitination with chains linked together via Lys63 serve as nonproteolytic signals in intracellular trafficking, DNA repair, and signal transduction pathways [1]. Cell-surface transmembrane molecules are often mono-ubiquitinated for endocytosis [74]. Histones are usually mono-ubiquitinated and associated with signaling or structural marking [75]. Mono-ubiquitination is also involved in virus budding [74]. K63 linkage of ubiquitin is known to be involved in DNA damage recognition of DNA double-strand breaks. These types of protein modification are important but cannot be easily detected by ex vivo high-throughput methods, such as label-free [7] or SILAC [8,9] quantitative mass spectrometry and GPS profiling [10,11] because of the non-degradation fate of the substrates. Additionally, in our strategy, more mono-ubiquitination or oligo-ubiquitination substrates of MDM2 were found. This phenomenon might because the mono- or oligo-ubiquitination modified phages might have higher amplification efficiency in host cells and these clones would have more chance to be identified. We also found that poly-ubiquitination of DDX42 by MDM2 does not cause its degradation.

Previous studies have identified 23 proteins as bona fide substrates of MDM2. However, most of these substrates were not identified using our strategy. This discrepancy likely occurred for several reasons. Some of these substrates (Cadherin-1，Dlg4，IGF1R) are membrane proteins, which are difficult to display on the surface of phages. Some of them are very low abundance proteins, which were not included in the library we used. Some of the substrates might require post-translational modifications, which would be lacking in the in vitro ubiquitination system, to be recognized by MDM2. Lastly, it has been noticed in previous studies that only a fraction of MDM2 substrates are ubiquitinated in any given in vitro system, so several independent selections per target E3 may be necessary to improve the coverage of substrates.

The vector of T7 cDNA library was T7 select 10-3, which could theoretically display up to 1200aa protein. The range of molecular weights of identified natural substrates was from 57aa to 342aa. Many of them were truncated proteins but not full-length proteins. This strategy was suitable to screen substrates of small proteins or truncated large proteins, which could be further validated in the full-length form.

Four different conditions were used to screen the MDM2 substrates: low enzyme concentration and more rounds of positive selection (Experiment 1), more rounds of negative selection (Experiment 2), high enzyme concentration and fewer rounds of selection (Experiment 3), and trypsin treatment during selection (Experiment 4). Some substrates were identified in different independent experiments, and some substrates were identified in only one experiment. Many substrates were identified in both Experiments 1 and 3. Therefore, both Experiment 1 and Experiment 3 were effective, suggesting that the enzyme concentration and rounds of selection were variable in our strategy. Only a few substrates were identified in Experiment 2, which might due to the loss of substrates in excessive rounds of negative selection. We speculated that phage with a long poly-ubiquitin chain may have a lower E. coli infection efficiency. Poly-ubiquitinated phages were digested by 0.125% trypsin treatment at 37°C for 30 min in the hope that removing the long ubiquitin chain would equalize the infection efficiencies of both poly- and mono-ubiquitinated substrates in Experiment 4. Theoretically, more poly-ubiquitinated substrates of MDM2 should be identified in Experiment 4. In fact, only one novel poly-ubiquitinated substrate (C12orf35) and many mono- and oligo-ubiquitinated substrates (RPL15 and NOLC1) were identified, indicating that the trypsin treatment did not improve the identification of poly-ubiquitinated substrates. The ubiquitination of HMGN1 and PRDM2 were not detected in the in vitro ubiquitination system, indicating that trypsin treatment might introduce false-positive results. Overall, we suggested that more flexible screening conditions could be used. More importantly, several independent selections should be performed for target E3s to expand the coverage of substrates. The results generated from independent experiments are more reliable. High-quality phage libraries will also help us to identify more natural substrates of E3s in our strategy.

Many unnatural substrates were identified in the study. These unnatural substrates could be used to find the consensus of substrates, which might help us to characterize the substrates recognition mechanism. Additionally, non-natural MDM2 substrates might have the potential role to be the competitive inhibitor of MDM2.

This strategy can be extended to substrate discovery for any E3s, as long as the target E3 does not ubiquitinate the empty phage and the E3-substrate recognition does not depend on cellular post-translational modifications which are unfit for the in vitro ubiquitination system. There are a series of ubiquitin-like proteins (UBLs), such as SUMOs and NEDD8, involved in post-translational protein modification. The mechanisms of SUMOylation and NEDDylation are similar to that of ubiquitin and are also mediated by corresponding enzymes (E1, E2, E3) [76]. SUMOylation and NEDDylation of target proteins have been successfully characterized by in vitro systems. Therefore, our strategy can also be potentially expanded to identify the targets of SUMOylation and NEDDylation. Live phage display libraries could be used as substrates for any protein modification enzyme screening, as long as the enzyme does not modify the empty phage, the modification to the phage does not affect its infection of host cells, and the modified phage can be purified.

Supporting Information

Figure S1.

MDM2 did not induce the significant degradation of transfected DDX42. GFP tagged DDX42 was transfected into HEK293T cells alone or co-transfected with Flag-tagged MDM2. The protein content of the lysates of the transfected cells were separated and immunoblotted using the anti-GFP antibody. The expression of MDM2 did not induce the significant degradation of transfected DDX42.

https://doi.org/10.1371/journal.pone.0076622.s001

(DOC)

Table S1.

Encoding sequence and encoding protein of clones selected in Experiment 1.

https://doi.org/10.1371/journal.pone.0076622.s002

(DOC)

Table S2.

Encoding sequence and encoding protein of clones selected in Experiment 2.

https://doi.org/10.1371/journal.pone.0076622.s003

(DOC)

Table S3.

Encoding sequence and encoding protein of clones selected in Experiment 3.

https://doi.org/10.1371/journal.pone.0076622.s004

(DOC)

Table S4.

Encoding sequence and encoding protein of clones selected in Experiment 4.

https://doi.org/10.1371/journal.pone.0076622.s005

(DOC)

Table S5.

Unnatural peptides identified at least twice in four independent screenings. +: positive result; -: nagetive result; NA: do not tested.

https://doi.org/10.1371/journal.pone.0076622.s006

(DOC)

Table S6.

Quantification of exogenous ubiquitinated RPL36a using ImageJ in MDM2 overexpressed HEK293T cells.

https://doi.org/10.1371/journal.pone.0076622.s007

(DOC)

Table S7.

Quantification of exogenous RPL36a using ImageJ in MDM2 overexpressed HEK293T cells.

https://doi.org/10.1371/journal.pone.0076622.s008

(DOC)

Text S1.

Raw DNA sequencing results.

https://doi.org/10.1371/journal.pone.0076622.s009

(TXT)

Acknowledgments

We thank Dr. Qimin Zhan for kindly providing Myc-tagged TP53 plasmid and Dr. Lina Zhang for kindly providing empty phage (T7Select 10-3).

Author Contributions

Conceived and designed the experiments: ZG YG. Performed the experiments: XW ZG. Analyzed the data: ZG. Contributed reagents/materials/analysis tools: HL. Wrote the manuscript: ZG YG.

References

  1. 1. Deshaies RJ, Joazeiro CA (2009) RING domain E3 ubiquitin ligases. Annu Rev Biochem 78: 399-434. doi:https://doi.org/10.1146/annurev.biochem.78.101807.093809. PubMed: 19489725.
  2. 2. Lu JY, Lin YY, Qian J, Tao SC, Zhu J et al. (2008) Functional dissection of a HECT ubiquitin E3 ligase. Mol Cell Proteomics 7: 35-45. PubMed: 17951556.
  3. 3. Andrews PS, Schneider S, Yang E, Michaels M, Chen H et al. (2010) Identification of substrates of SMURF1 ubiquitin ligase activity utilizing protein microarrays. Assay Drug Dev Technol 8: 471-487. doi:https://doi.org/10.1089/adt.2009.0264. PubMed: 20804422.
  4. 4. Merbl Y, Kirschner MW (2009) Large-scale detection of ubiquitination substrates using cell extracts and protein microarrays. Proc Natl Acad Sci U S A 106: 2543-2548. doi:https://doi.org/10.1073/pnas.0812892106. PubMed: 19181856.
  5. 5. Gupta R, Kus B, Fladd C, Wasmuth J, Tonikian R et al. (2007) Ubiquitination screen using protein microarrays for comprehensive identification of Rsp5 substrates in yeast. Mol Syst Biol 3: 116. PubMed: 17551511.
  6. 6. Persaud A, Alberts P, Amsen EM, Xiong X, Wasmuth J et al. (2009) Comparison of substrate specificity of the ubiquitin ligases Nedd4 and Nedd4-2 using proteome arrays. Mol Syst Biol 5: 333. PubMed: 19953087.
  7. 7. Burande CF, Heuzé ML, Lamsoul I, Monsarrat B, Uttenweiler-Joseph S et al. (2009) A label-free quantitative proteomics strategy to identify E3 ubiquitin ligase substrates targeted to proteasome degradation. Mol Cell Proteomics 8: 1719-1727. doi:https://doi.org/10.1074/mcp.M800410-MCP200. PubMed: 19376791.
  8. 8. Hör S, Ziv T, Admon A, Lehner PJ (2009) Stable isotope labeling by amino acids in cell culture and differential plasma membrane proteome quantitation identify new substrates for the MARCH9 transmembrane E3 ligase. Mol Cell Proteomics 8: 1959-1971. doi:https://doi.org/10.1074/mcp.M900174-MCP200. PubMed: 19457934.
  9. 9. Bartee E, McCormack A, Früh K (2006) Quantitative membrane proteomics reveals new cellular targets of viral immune modulators. PLOS Pathog 2: e107. doi:https://doi.org/10.1371/journal.ppat.0020107. PubMed: 17238276.
  10. 10. Yen HC, Xu Q, Chou DM, Zhao Z, Elledge SJ (2008) Global protein stability profiling in mammalian cells. Science 322: 918-923. doi:https://doi.org/10.1126/science.1160489. PubMed: 18988847.
  11. 11. Yen HC, Elledge SJ (2008) Identification of SCF ubiquitin ligase substrates by global protein stability profiling. Science 322: 923-929. doi:https://doi.org/10.1126/science.1160462. PubMed: 18988848.
  12. 12. Smith GP (1985) Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface. Science 228: 1315-1317. doi:https://doi.org/10.1126/science.4001944. PubMed: 4001944.
  13. 13. Benhar I (2001) Biotechnological applications of phage and cell display. Biotechnol Adv 19: 1-33. doi:https://doi.org/10.1016/S0734-9750(00)00054-9. PubMed: 14538090.
  14. 14. Matthews DJ, Wells JA (1993) Substrate phage: selection of protease substrates by monovalent phage display. Science 260: 1113-1117. doi:https://doi.org/10.1126/science.8493554. PubMed: 8493554.
  15. 15. Ding L, Coombs GS, Strandberg L, Navre M, Corey DR et al. (1995) Origins of the specificity of tissue-type plasminogen activator. Proc Natl Acad Sci U S A 92: 7627-7631. doi:https://doi.org/10.1073/pnas.92.17.7627. PubMed: 7644467.
  16. 16. Schmitz R, Baumann G, Gram H (1996) Catalytic specificity of phosphotyrosine kinases Blk, Lyn, c-Src and Syk as assessed by phage display. J Mol Biol 260: 664-677. doi:https://doi.org/10.1006/jmbi.1996.0429. PubMed: 8709147.
  17. 17. Dente L, Vetriani C, Zucconi A, Pelicci G, Lanfrancone L et al. (1997) Modified phage peptide libraries as a tool to study specificity of phosphorylation and recognition of tyrosine containing peptides. J Mol Biol 269: 694-703. doi:https://doi.org/10.1006/jmbi.1997.1073. PubMed: 9223634.
  18. 18. Sugimura Y, Hosono M, Wada F, Yoshimura T, Maki M et al. (2006) Screening for the preferred substrate sequence of transglutaminase using a phage-displayed peptide library: identification of peptide substrates for TGASE 2 and Factor XIIIA. J Biol Chem 281: 17699-17706. doi:https://doi.org/10.1074/jbc.M513538200. PubMed: 16636049.
  19. 19. Hitomi K, Kitamura M, Sugimura Y (2009) Preferred substrate sequences for transglutaminase 2: screening using a phage-displayed peptide library. Amino Acids 36: 619-624. doi:https://doi.org/10.1007/s00726-008-0126-6. PubMed: 18651094.
  20. 20. Cahilly-Snyder L, Yang-Feng T, Francke U, George DL (1987) Molecular analysis and chromosomal mapping of amplified genes isolated from a transformed mouse 3T3 cell line. Somat Cell Mol Genet 13: 235-244. doi:https://doi.org/10.1007/BF01535205. PubMed: 3474784.
  21. 21. Fakharzadeh SS, Trusko SP, George DL (1991) Tumorigenic potential associated with enhanced expression of a gene that is amplified in a mouse tumor cell line. EMBO J 10: 1565-1569. PubMed: 2026149.
  22. 22. Momand J, Jung D, Wilczynski S, Niland J (1998) The MDM2 gene amplification database. Nucleic Acids Res 26: 3453-3459. doi:https://doi.org/10.1093/nar/26.15.3453. PubMed: 9671804.
  23. 23. Cordon-Cardo C, Latres E, Drobnjak M, Oliva MR, Pollack D et al. (1994) Molecular abnormalities of mdm2 and p53 genes in adult soft tissue sarcomas. Cancer Res 54: 794-799. PubMed: 8306343.
  24. 24. Landers JE, Haines DS, Strauss JF 3rd, George DL (1994) Enhanced translation: a novel mechanism of mdm2 oncogene overexpression identified in human tumor cells. Oncogene 9: 2745-2750. PubMed: 8058341.
  25. 25. Haupt Y, Maya R, Kazaz A, Oren M (1997) Mdm2 promotes the rapid degradation of p53. Nature 387: 296-299. doi:https://doi.org/10.1038/387296a0. PubMed: 9153395.
  26. 26. Kubbutat MH, Jones SN, Vousden KH (1997) Regulation of p53 stability by Mdm2. Nature 387: 299-303. doi:https://doi.org/10.1038/387299a0. PubMed: 9153396.
  27. 27. Jin Y, Lee H, Zeng SX, Dai MS, Lu H (2003) MDM2 promotes p21waf1/cip1 proteasomal turnover independently of ubiquitylation. EMBO J 22: 6365-6377. doi:https://doi.org/10.1093/emboj/cdg600. PubMed: 14633995.
  28. 28. Zhang Z, Wang H, Li M, Agrawal S, Chen X et al. (2004) MDM2 is a negative regulator of p21WAF1/CIP1, independent of p53. J Biol Chem 279: 16000-16006. doi:https://doi.org/10.1074/jbc.M312264200. PubMed: 14761977.
  29. 29. Xu H, Zhang Z, Li M, Zhang R (2010) MDM2 promotes proteasomal degradation of p21Waf1 via a conformation change. J Biol Chem 285: 18407-18414. doi:https://doi.org/10.1074/jbc.M109.059568. PubMed: 20308078.
  30. 30. Di Stefano V, Mattiussi M, Sacchi A, D’Orazi G (2005) HIPK2 inhibits both MDM2 gene and protein by, respectively, p53-dependent and independent regulations. FEBS Lett 579: 5473-5480. doi:https://doi.org/10.1016/j.febslet.2005.09.008. PubMed: 16212962.
  31. 31. Rinaldo C, Prodosmo A, Mancini F, Iacovelli S, Sacchi A et al. (2007) MDM2-regulated degradation of HIPK2 prevents p53Ser46 phosphorylation and DNA damage-induced apoptosis. Mol Cell 25: 739-750. doi:https://doi.org/10.1016/j.molcel.2007.02.008. PubMed: 17349959.
  32. 32. Uchida C, Miwa S, Kitagawa K, Hattori T, Isobe T et al. (2005) Enhanced Mdm2 activity inhibits pRB function via ubiquitin-dependent degradation. EMBO J 24: 160-169. doi:https://doi.org/10.1038/sj.emboj.7600486. PubMed: 15577944.
  33. 33. Sdek P, Ying H, Chang DL, Qiu W, Zheng H et al. (2005) MDM2 promotes proteasome-dependent ubiquitin-independent degradation of retinoblastoma protein. Mol Cell 20: 699-708. doi:https://doi.org/10.1016/j.molcel.2005.10.017. PubMed: 16337594.
  34. 34. Miwa S, Uchida C, Kitagawa K, Hattori T, Oda T et al. (2006) Mdm2-mediated pRB downregulation is involved in carcinogenesis in a p53-independent manner. Biochem Biophys Res Commun 340: 54-61. doi:https://doi.org/10.1016/j.bbrc.2005.11.148. PubMed: 16343421.
  35. 35. Yang JY, Zong CS, Xia W, Wei Y, Ali-Seyed M et al. (2006) MDM2 promotes cell motility and invasiveness by regulating E-cadherin degradation. Mol Cell Biol 26: 7269-7282. doi:https://doi.org/10.1128/MCB.00172-06. PubMed: 16980628.
  36. 36. Colledge M, Snyder EM, Crozier RA, Soderling JA, Jin Y et al. (2003) Ubiquitination regulates PSD-95 degradation and AMPA receptor surface expression. Neuron 40: 595-607. doi:https://doi.org/10.1016/S0896-6273(03)00687-1. PubMed: 14642282.
  37. 37. Girnita L, Girnita A, Larsson O (2003) Mdm2-dependent ubiquitination and degradation of the insulin-like growth factor 1 receptor. Proc Natl Acad Sci U S A 100: 8247-8252. doi:https://doi.org/10.1073/pnas.1431613100. PubMed: 12821780.
  38. 38. Busso CS, Iwakuma T, Izumi T (2009) Ubiquitination of mammalian AP endonuclease (APE1) regulated by the p53-MDM2 signaling pathway. Oncogene 28: 1616-1625. doi:https://doi.org/10.1038/onc.2009.5. PubMed: 19219073.
  39. 39. Salcedo A, Mayor F Jr., Penela P (2006) Mdm2 is involved in the ubiquitination and degradation of G-protein-coupled receptor kinase 2. EMBO J 25: 4752-4762. doi:https://doi.org/10.1038/sj.emboj.7601351. PubMed: 17006543.
  40. 40. Usui I, Imamura T, Huang J, Satoh H, Shenoy SK et al. (2004) beta-arrestin-1 competitively inhibits insulin-induced ubiquitination and degradation of insulin receptor substrate 1. Mol Cell Biol 24: 8929-8937. doi:https://doi.org/10.1128/MCB.24.20.8929-8937.2004. PubMed: 15456867.
  41. 41. Wang P, Gao H, Ni Y, Wang B, Wu Y et al. (2003) Beta-arrestin 2 functions as a G-protein-coupled receptor-activated regulator of oncoprotein Mdm2. J Biol Chem 278: 6363-6370. doi:https://doi.org/10.1074/jbc.M210350200. PubMed: 12488444.
  42. 42. Sánchez-Molina S, Oliva JL, García-Vargas S, Valls E, Rojas JM et al. (2006) The histone acetyltransferases CBP/p300 are degraded in NIH 3T3 cells by activation of Ras signalling pathway. Biochem J 398: 215-224. doi:https://doi.org/10.1042/BJ20060052. PubMed: 16704373.
  43. 43. Miyake S, Sellers WR, Safran M, Li X, Zhao W et al. (2000) Cells degrade a novel inhibitor of differentiation with E1A-like properties upon exiting the cell cycle. Mol Cell Biol 20: 8889-8902. doi:https://doi.org/10.1128/MCB.20.23.8889-8902.2000. PubMed: 11073989.
  44. 44. Coutts AS, Boulahbel H, Graham A, La Thangue NB (2007) Mdm2 targets the p53 transcription cofactor JMY for degradation. EMBO Rep 8: 84-90. doi:https://doi.org/10.1038/sj.embor.7400855. PubMed: 17170761.
  45. 45. Jin Y, Zeng SX, Lee H, Lu H (2004) MDM2 mediates p300/CREB-binding protein-associated factor ubiquitination and degradation. J Biol Chem 279: 20035-20043. doi:https://doi.org/10.1074/jbc.M309916200. PubMed: 14769800.
  46. 46. Legube G, Linares LK, Lemercier C, Scheffner M, Khochbin S et al. (2002) Tip60 is targeted to proteasome-mediated degradation by Mdm2 and accumulates after UV irradiation. EMBO J 21: 1704-1712. doi:https://doi.org/10.1093/emboj/21.7.1704. PubMed: 11927554.
  47. 47. Pan Y, Chen J (2003) MDM2 promotes ubiquitination and degradation of MDMX. Mol Cell Biol 23: 5113-5121. doi:https://doi.org/10.1128/MCB.23.15.5113-5121.2003. PubMed: 12860999.
  48. 48. Kawai H, Wiederschain D, Kitao H, Stuart J, Tsai KK et al. (2003) DNA damage-induced MDMX degradation is mediated by MDM2. J Biol Chem 278: 45946-45953. doi:https://doi.org/10.1074/jbc.M308295200. PubMed: 12963717.
  49. 49. Wyszomierski SL, Yu D (2005) A knotty turnabout?: Akt1 as a metastasis suppressor. Cancer Cell 8: 437-439. doi:https://doi.org/10.1016/j.ccr.2005.11.006. PubMed: 16338656.
  50. 50. Foo RS, Chan LK, Kitsis RN, Bennett MR (2007) Ubiquitination and degradation of the anti-apoptotic protein ARC by MDM2. J Biol Chem 282: 5529-5535. PubMed: 17142834.
  51. 51. Ofir-Rosenfeld Y, Boggs K, Michael D, Kastan MB, Oren M (2008) Mdm2 regulates p53 mRNA translation through inhibitory interactions with ribosomal protein L26. Mol Cell 32: 180-189. doi:https://doi.org/10.1016/j.molcel.2008.08.031. PubMed: 18951086.
  52. 52. Buschmann T, Fuchs SY, Lee CG, Pan ZQ, Ronai Z (2000) SUMO-1 modification of Mdm2 prevents its self-ubiquitination and increases Mdm2 ability to ubiquitinate p53. Cell 101: 753-762. doi:https://doi.org/10.1016/S0092-8674(00)80887-9. PubMed: 10892746.
  53. 53. Li L, Deng B, Xing G, Teng Y, Tian C et al. (2007) PACT is a negative regulator of p53 and essential for cell growth and embryonic development. Proc Natl Acad Sci U S A 104: 7951-7956. doi:https://doi.org/10.1073/pnas.0701916104. PubMed: 17470788.
  54. 54. Hwang YC, Lu TY, Huang DY, Kuo YS, Kao CF et al. (2009) NOLC1, an enhancer of nasopharyngeal carcinoma progression, is essential for TP53 to regulate MDM2 expression. Am J Pathol 175: 342-354. doi:https://doi.org/10.2353/ajpath.2009.080931. PubMed: 19541936.
  55. 55. Miyamoto-Sato E, Fujimori S, Ishizaka M, Hirai N, Masuoka K et al. (2010) A comprehensive resource of interacting protein regions for refining human transcription factor networks. PLOS ONE 5: e9289. doi:https://doi.org/10.1371/journal.pone.0009289. PubMed: 20195357.
  56. 56. Shadat NM, Koide N, Khuda II, Dagvadorj J, Tumurkhuu G et al. (2010) Retinoblastoma protein-interacting zinc finger 1 (RIZ1) regulates the proliferation of monocytic leukemia cells via activation of p53. Cancer Invest 28: 806-812. doi:https://doi.org/10.3109/07357907.2010.494323. PubMed: 20594067.
  57. 57. Buyse IM, Shao G, Huang S (1995) The retinoblastoma protein binds to RIZ, a zinc-finger protein that shares an epitope with the adenovirus E1A protein. Proc Natl Acad Sci U S A 92: 4467-4471. doi:https://doi.org/10.1073/pnas.92.10.4467. PubMed: 7538672.
  58. 58. Buyse IM, Huang S (1997) In vitro analysis of the E1A-homologous sequences of RIZ. J Virol 71: 6200-6203. PubMed: 9223517.
  59. 59. Scherz-Shouval R, Weidberg H, Gonen C, Wilder S, Elazar Z et al. (2010) p53-dependent regulation of autophagy protein LC3 supports cancer cell survival under prolonged starvation. Proc Natl Acad Sci U S A 107: 18511-18516. doi:https://doi.org/10.1073/pnas.1006124107. PubMed: 20937856.
  60. 60. Ewing RM, Chu P, Elisma F, Li H, Taylor P et al. (2007) Large-scale mapping of human protein-protein interactions by mass spectrometry. Mol Syst Biol 3: 89. PubMed: 17353931.
  61. 61. Abe Y, Matsumoto S, Wei S, Nezu K, Miyoshi A et al. (2001) Cloning and characterization of a p53-related protein kinase expressed in interleukin-2-activated cytotoxic T-cells, epithelial tumor cell lines, and the testes. J Biol Chem 276: 44003-44011. doi:https://doi.org/10.1074/jbc.M105669200. PubMed: 11546806.
  62. 62. Hannan KM, Hannan RD, Smith SD, Jefferson LS, Lun M et al. (2000) Rb and p130 regulate RNA polymerase I transcription: Rb disrupts the interaction between UBF and SL-1. Oncogene 19: 4988-4999. doi:https://doi.org/10.1038/sj.onc.1203875. PubMed: 11042686.
  63. 63. Cavanaugh AH, Hempel WM, Taylor LJ, Rogalsky V, Todorov G et al. (1995) Activity of RNA polymerase I transcription factor UBF blocked by Rb gene product. Nature 374: 177-180. doi:https://doi.org/10.1038/374177a0. PubMed: 7877691.
  64. 64. Voit R, Schäfer K, Grummt I (1997) Mechanism of repression of RNA polymerase I transcription by the retinoblastoma protein. Mol Cell Biol 17: 4230-4237. PubMed: 9234680.
  65. 65. Zhai W, Comai L (2000) Repression of RNA polymerase I transcription by the tumor suppressor p53. Mol Cell Biol 20: 5930-5938. doi:https://doi.org/10.1128/MCB.20.16.5930-5938.2000. PubMed: 10913176.
  66. 66. Crawford NP, Yang H, Mattaini KR, Hunter KW (2009) The metastasis efficiency modifier ribosomal RNA processing 1 homolog B (RRP1B) is a chromatin-associated factor. J Biol Chem 284: 28660-28673. doi:https://doi.org/10.1074/jbc.M109.023457. PubMed: 19710015.
  67. 67. Dupont J, Renou JP, Shani M, Hennighausen L, LeRoith D (2002) PTEN overexpression suppresses proliferation and differentiation and enhances apoptosis of the mouse mammary epithelium. J Clin Invest 110: 815-825. doi:https://doi.org/10.1172/JCI200213829. PubMed: 12235113.
  68. 68. Dai MS, Lu H (2004) Inhibition of MDM2-mediated p53 ubiquitination and degradation by ribosomal protein L5. J Biol Chem 279: 44475-44482. doi:https://doi.org/10.1074/jbc.M403722200. PubMed: 15308643.
  69. 69. Lohrum MA, Ludwig RL, Kubbutat MH, Hanlon M, Vousden KH (2003) Regulation of HDM2 activity by the ribosomal protein L11. Cancer Cell 3: 577-587. doi:https://doi.org/10.1016/S1535-6108(03)00134-X. PubMed: 12842086.
  70. 70. Dai MS, Zeng SX, Jin Y, Sun XX, David L et al. (2004) Ribosomal protein L23 activates p53 by inhibiting MDM2 function in response to ribosomal perturbation but not to translation inhibition. Mol Cell Biol 24: 7654-7668. doi:https://doi.org/10.1128/MCB.24.17.7654-7668.2004. PubMed: 15314173.
  71. 71. Jin A, Itahana K, O’Keefe K, Zhang Y (2004) Inhibition of HDM2 and activation of p53 by ribosomal protein L23. Mol Cell Biol 24: 7669-7680. doi:https://doi.org/10.1128/MCB.24.17.7669-7680.2004. PubMed: 15314174.
  72. 72. Shin S, Janknecht R (2007) Concerted activation of the Mdm2 promoter by p72 RNA helicase and the coactivators p300 and P/CAF. J Cell Biochem 101: 1252-1265. doi:https://doi.org/10.1002/jcb.21250. PubMed: 17226766.
  73. 73. Shu XS, Geng H, Li L, Ying J, Ma C et al. (2011) The epigenetic modifier PRDM5 functions as a tumor suppressor through modulating WNT/beta-catenin signaling and is frequently silenced in multiple tumors. PLOS ONE 6: e27346. doi:https://doi.org/10.1371/journal.pone.0027346. PubMed: 22087297.
  74. 74. Hicke L (2001) Protein regulation by monoubiquitin. Nat Rev Mol Cell Biol 2: 195-201. doi:https://doi.org/10.1038/35056583. PubMed: 11265249.
  75. 75. Hofmann K (2009) Ubiquitin-binding domains and their role in the DNA damage response. DNA Repair (Amst) 8: 544-556. doi:https://doi.org/10.1016/j.dnarep.2009.01.003. PubMed: 19213613.
  76. 76. Hochstrasser M (2009) Origin and function of ubiquitin-like proteins. Nature 458: 422-429. doi:https://doi.org/10.1038/nature07958. PubMed: 19325621.