Deubiquitinating c-Myc: USP36 steps up in the nucleolus

Abstract

Ubiquitination plays a key and complex role in the regulation of c-Myc stability, transactivation, and oncogenic activity. c-Myc is ubiquitinated by a number of ubiquitin ligases (E3s), such as SCF^Fbw7 and SCF^Skp2. Depending on the E3s, ubiquitination can either positively or negatively regulate c-Myc levels and activity. Meanwhile, c-Myc ubiquitination can be reversed by deubiquitination. An early study showed that USP28 deubiquitinates c-Myc via interacting with Fbw7α whereas a recent study reveals that USP37 deubiquitinates c-Myc independently of Fbw7 and c-Myc phosphorylation. Consequently, both USP28 and USP37 stabilize c-Myc and enhance its activity. We recently found the nucleolar USP36 as a novel c-Myc deubiquitinase that controls the end-point of c-Myc degradation pathway in the nucleolus. Here we briefly review the current understanding of ubiquitination and deubiquitination regulation of c-Myc and further discuss the USP36-c-Myc regulatory pathway.

Keywords:

• c-Myc
• deubiquitinating enzymes
• nucleolus
• USP36
• ubiquitination

Introduction

As a master transcription regulator, c-Myc (Myc, thereafter) controls the expression of most, if not all, actively transcribed genes in the human genome^1-4 and coordinates many cellular processes, including promoting cell proliferation and growth by enhancing ribosomal biogenesis/protein translation, nucleotide biosynthesis, metabolism, RNA processing and DNA replication, inducing apoptosis, senescence, differentiation, and angiogenesis, as well as regulating stem cell renewal.^5-7 Although cells cannot live without Myc, aberrant overexpression of Myc contributes to the development of malignancies in human and in mouse models.^6,8 Thus tightly regulated levels of Myc are essential for normal cell growth and proliferation. Among many mechanisms contributing to the deregulated overexpression of Myc, including transcription, gene amplification, and chromosome translocation, increasing studies have shown that Myc protein stabilization due to impaired Myc degradation pathway plays a key role in cancers.^9-13

The Myc protein contains a N-terminal unstructured transactivation domain (TAD) essential for Myc transactivation activity, which includes the conserved Myc box I (MBI) and MBII. The MBI contains 2 critical residues, Threonine 58 (T58) and Serine 62 (S62), whose dynamic phosphorylation controls Myc protein stability in response to cell growth signals.^5,6,9,10 The MBII is critical for recruiting key Myc transactivation co-activators such as TRAAP, GCN5, TIP48, TIP49, TIP60, CBP/p300, BAF53 as well as Skp2.^5,14,15 The central region of Myc contains the MBIII and MBIV followed by a PEST (rich in proline, glutamic acid, threonine, and proline residues) sequence and a nuclear localization signal (NLS), respectively. Located at the C-terminal region is a basic helix-loop-helix leucine zipper (bHLH-LZ) domain that dimerizes with the Max protein and binds to the E-box elements in target gene promoters.^5,6,14,15

Myc is a short-lived protein with a half-life less than 30 minutes in non-transformed cells.^10,16,17 The fast turnover of Myc protein is tightly controlled by the ubiquitin-proteasome system. Growth signal-controlled Myc turnover is largely through a phosphorylation-ubiquitination cascade. Phosphorylation of Myc at S62, triggered by the Ras-induced Raf-MEK-ERK kinase cascade and/or CDKs in response to growth signals, increases Myc stability. Phosphorylation of T58 occurs sequentially and is mediated by GSK3β, which is held in check by PI(3)K/Akt signaling, also in response to growth factor and receptor tyrosine kinase signaling.^10,12,18 When Ras signaling subsides, GSK3β becomes activated and starts to phosphorylate Myc at T58.^12,19-22 This phosphorylation facilitates the recruitment of a prolyl isomerase, Pin1, to catalyze the cis to trans isomerization at proline (P) 63 of Myc, leading to the subsequent conformational change that allows the PP2A phosphatase to dephosphorylate Myc at S62.²³ Finally, phosphorylated T58 serves as a dock to recruit the T58 phosphorylation-dependent ubiquitin ligase (E3) complex SCF^Fbw7 to mediate Myc ubiquitination and degradation through the proteasome system.^24–27

Likewise, there are many other ubiquitin E3s that mediate Myc ubiquitination, resulting in either activation or inhibition of Myc activity and either degradation or stabilization of Myc protein. Moreover, Myc ubiquitination can be reversed by deubiquitinating enzymes (DUBs). In this perspective, we will briefly review current progress toward understanding the ubiquitination-deubiquitination regulation of Myc stability and activity and discuss further our rec-ent finding of the novel nucleolar Myc deubiquitinating enzyme USP36 that positively regulates Myc levels and activity.

Ubiquitination of Myc: Diverse roles in Regulating Myc Stability and Activity

As mentioned above, the SCF^Fbw7 E3 ligase mediates Myc ubiquitination and degradation upon growth signal deprivation. There are 3 Fbw7 isoforms located in distinct subcellular compartments: Fbw7β and Fbw7α are in the cytoplasm and the nucleoplasm, respectively, whereas Fbw7γ is in the nucleolus.²⁷ When overexpressed, all the 3 isoforms can mediate K48-linked Myc ubiquitination and degradation.²⁶ The T58 phosphorylation-dependent Myc degradation by the SCF^Fbw7 complex is critical for growth signal-controlled Myc turnover. T58 mutation results in stabilization and enhanced oncogenic activity of Myc, and T58 is one of the “hot spots” for mutation in a subset of Burkitt's lymphomas.^22,28-30 Mice harboring the Myc^T58A mutant develop cancer at a significantly higher penetrance and reduced latency than mice with the wild type c-Myc transgene.^31,32 Consistent with the role of Fbw7 in negatively regulating Myc, mutations and deletions of FBW7 are found in multiple human cancers.³³ Thus, T58 phosphorylation-dependent ubiquitination by SCF^Fbw7 clearly plays a key role in regulating Myc turnover and oncogenic activity. Yet, over the past decade, additional ubiquitin E3s have been identified as important Myc regulators (Fig. 1).

Figure 1. A diagram showing ubiquitin E3 ligases known to ubiquitinate Myc. Bars indicate the suppression of Myc activity, whereas arrows indicate the activation of Myc activity.

SCF^Skp2

The SCF^Skp2 ubiquitin E3 ubiquitinates Myc and targets Myc for proteasomal degradation.^34–36 Surprisingly, this degradation is associated with the increased activity of Myc and overexpression of Skp2 promotes Myc-induced cell cycle progression.^34,36 Skp2 binds to the MBII and bHLH-LZ domains of Myc^34,36 and acts as a co-activator of Myc. Although it remains unknown why Skp2 mediates both the proteasomal degradation and the transactivational activity of Myc, several pieces of evidence support a transcription-coupled proteasomal degradation mechanism. Myc recruits Skp2 and a number of proteasome subunits to Myc target gene promoters such as the cyclin D2 promoter,^34,35 suggesting that the 26S proteasome may degrade Myc at its target gene promoters once it finishes its role in activating transcription. Recently, it has been shown that the ubiquitination of Myc at the 6 lysines in the TAD is critical for the canonical transcriptional activity and transformation activity of Myc.³⁷ ARF inhibits the Myc-Skp2 interaction and Skp2-mediated Myc ubiquitination, leading to Myc stabilization and the switch from an oncogenic protein to an apoptotic inducer.³⁷ Supporting the transcription-coupled proteasomal degradation of Myc, we recently found that PIN1 promotes the turnover of Myc at target gene promoters while also increasing its transactivation activity.³⁸ Yet, other studies argued that Skp2 promotes Myc activity independently of the SCF complex, reporting that instead, it cooperates with Myc to recruit the Miz1 transcription factor and p300 coactivator to the target RhoA gene promoter, leading to the increase of RhoA expression and cell invasion and migration.³⁹ Similarly, overexpression of a Skp2 mutant that is unable to associate with the SCF complex also restores Myc activity that is down-regulated by the ERK-dependent reduction of Skp2.⁴⁰ Nevertheless, Skp2 promotes Myc's transactivation and transformation activity, while it can also ubiquitinate and degrades Myc.

HUWE1/ARF-BP1/HectH9

HUWE1 (also called ARF-BP1 and HectH9), a member of the HECT-domain containing ubiquitin E3s, is another Myc ubiquitin E3 that enhances Myc activity.⁴¹ HUWE1 binds to Myc at the TAD domain and mediates K63-linked Myc ubiquitination at the lysines near the NLS.⁴¹ This ubiquitination enhances the transcriptional activity of Myc by recruiting p300, without inducing Myc proteasomal degradation. Miz1 competes with Myc to interact with HUWE1 and inhibits HUWE1-mediated Myc ubiquitination. Recently it has been shown that small molecule inhibitors targeting HUWE1 inhibit the activity of Myc in colorectal cancer cells via stabilizing a Miz1-Myc repressive complex that suppresses Myc target genes.⁴²

SCF^FBXO28

Similar to HUWE1, SCF^FBXO28 also mediates non-proteolytic Myc ubiquitination.⁴³ During cell cycle progression to S phase, CDK1/2-mediated phosphorylation of FBXO28 at Ser 344 activates its ubiquitin E3 activity to polyubiquitinate Myc without inducing proteolytic Myc degradation. FBXO28 binds to Myc at the MBII and possible motifs upstream of the HLH motif. This Myc ubiquitination by SCF^FBXO28 stimulates Myc activity also by recruiting p300 to target gene promoters,⁴³ which requires the ubiquitination of several lysines within the 294–367 region of Myc known to be involved in ubiquitin-mediated Myc-p300 interaction.⁴¹ Consequently, FBXO28 promotes Myc-driven tumorigenesis and high expression of FBXO28 (as well as high Ser 344 phosphorylation) are associated with poor patient outcomes. Thus, SCF^FBXO28-mediated ubiquitination of Myc appears to play an important role in cell cycle regulation of Myc activity, transmitting CDK activity to Myc activity as cells progress to S phase. However, the detailed type of polyubiquitination and why this ubiquitination does not target Myc for degradation remains unknown. How this E3 interplays with other E3s in the cell cycle progression also warrants further investigation.

SCF^-TRCP

The fourth Myc E3 that does not mediate Myc degradation is the SCF^β-TRCP. Degradation of Myc by SCF^Fbw7 has been implicated in controlling Myc stability in G1 and early S phase of the cell cycle.¹² Popov et al. recently found that SCF^β-TRCP plays a role in Myc ubiquitination in the subsequent G2 cell cycle phases. This ubiquitination antagonizes SCF^Fbw7 to differentially ubiquitinate the N-terminus of Myc and stabilizes Myc.⁴⁴ While SCF^Fbw7 induces K48-linked ubiquitination, SCF^β-TRCP mediates K48/K63 heterotypic polyubiquitination of Myc and is required for Myc-dependent acceleration of cell cycle progression during S and G2 phases. This ubiquitination also requires the phosphodegron (a phospho-recognition sequence for β-TRCP binding at amino acids 278–283). Therefore, SCF^Fbw7 and SCF^β-TRCP E3s alternatively act on Myc during cell cycle progression. Consistently, PI3k/mTOR kinase inhibitors inhibit phosphorylation of β-TRCP, resulting in its proteasomal degradation in breast cancer cells, and this correlates with a reduction in Myc.⁴⁵

TRPC4AP/TRUSS

TRPC4AP (transient receptor potential cation channel, subfamily C, member 4-associated protein), also called TRUSS (tumor necrosis factor receptor-associated ubiquitous scaffolding and signaling protein), was identified as a novel Myc-interacting E3.⁴⁶ TRUSS interacts with both c-Myc and N-Myc at their C-terminal bHLH-LZ domain and promotes Myc ubiquitination. Further studies showed that TRUSS complex contains DDB1 (damage-specific DNA-binding protein 1), CUL4A (Cullin 4A), DDA1, and the small RING finger protein RBX1 (which is the receptor for a DDB1-CUL4 E3 ligase complex) and mediates the interaction of MYC with the DDB1-CUL4 E3 ligase for selective Myc degradation through the proteasome.⁴⁶ Consistently, TRUSS suppresses Myc transactivation and transformation activity and is downregulated in a number of cancers.⁴⁶ Interestingly, it was recently shown that TRUSS itself is ubiquitinated by Skp2 and degraded through the proteasome.⁴⁷ TRUSS is primarily expressed during the G1 phase of cell cycle and its level starts to decline with the expression of the S-phase specific E3 ligase Skp2. Overexpression of Skp2 correlates with the reduction of TRUSS in human cancers, providing support that Myc ubiquitin E3s coordinately regulate Myc and can inter-regulate each other during the cell cycle.

TRIM32

Schwamborn et al⁴⁸ has shown that TRIM32 is an Ring-finger ubiquitin ligase for Myc that controls Myc protein stability. The RING finger mutant TRIM32^C24A failed to ubiquitinate and degrade Myc. TRIM32-mediated Myc degradation is essential for TRIM32 to induce neuronal differentiation. However, whether and how TRIM32 plays a role in tumorigenesis by regulating Myc remains unclear.

Pirh2

Pirh2 is originally shown to be a p53 ubiquitin ligase. To analyze the in vivo function of Pirh2, Hakem et al⁴⁹ generated Pirh2 deletion mice. Consistent with the increased levels of p53, increased p53-dependent apoptosis was observed in the mice in response to DNA damage. They also found that Myc levels were increased in these mice. Subsequent study revealed that Pirh2 associated with Myc through 2 contacts: the C-terminal Ring-finger containing domains of Pirh2 interacts with the Myc MBII whereas the N-terminal domain of Pirh2 interacts with the bHLH-LZ domain of Myc.⁴⁹ Downregulation of Pirh2 is associated with poor outcome of a number of human cancers including breast, ovarian, and lung cancers.⁴⁹ Genetic deletion of Pirh2 in mice also results in plasma cell hyperplasia associated with the increased Myc, γ-globulinemia, kidney failure, premature death, tumorigenesis, and shortened lifespan.⁴⁹ Approximately 25% of Pirh2^-/- and 17% of Pirh2^+/- sick mice also developed solid tumors including sarcomas, liver, testes, mammary and lung tumors, indicating that Pirh2 is a tumor suppressor in vivo.

CHIP

CHIP (carboxyl terminus of Hsc70-interacting protein) is a chaperone-associated U-box-containing E3 ligase that links a chaperone to the 26S proteasome by ubiquitinating chaperone substrates that then direct the chaperone toward the proteasome.⁵⁰ Overexpression of CHIP could accelerate the turnover rate of Myc protein, whereas knockdown of CHIP by RNAi stabilizes Myc.⁵¹ CHIP interacts with and ubiquitinates Myc and suppresses Myc activity. Interestingly, the association between CHIP and Myc is dependent on the chaperone, particularly Hsp70. Further study needs to clarify the interaction between Myc and CHIP and whether CHIP regulates Myc levels under physiological conditions.⁵¹ Overall, the activity of CHIP to inhibit Myc is consistent with its putative tumor suppressor function.

SCF^FBXO32

Recently, it has been shown that SCF^FBXO32 promotes Myc ubiquitination through K48-linked ubiquitin chains and targets it for proteasome degradation, independently of its phosphorylation at either T58 or S62.⁵² FBXO32 interacts with the MBII, MBIV and PEST motifs. The ubiquitination site is mapped at K326 on Myc. Unlike FBXO28, FBXO32 suppresses Myc transactivation activity and inhibits cell growth. Moreover, FBXO32 itself is a Myc target gene, forming a negative regulatory loop.

Deubiquitination of Myc: When and Where?

Adding to the complexity of the ubiquitination regulation of Myc, deubiquitination regulation has emerged as another important mechanism for the control of Myc. Similar to many other reversible posttranslational modifications, ubiquitination of Myc can be reversed by deubiquitination catalyzed by DUBs. The human genome encodes approximately 95 DUBs that are classified into 5 families.^53,54 Ubiquitin-specific proteases (USPs), ubiquitin C-terminal hydrolases (UCHs), ovarian tumor associated proteases (OTUs), Machado-Joseph disease (or Josephin domain) proteins (MJDs), and JAB1/MPN/MOV34 proteins (JAMMs). The USP family member USP28 is the first DUB discovered that deubiquitinates and stabilizes Myc.⁵⁵ Knockdown of USP28 reduces the levels and activity of Myc. Interestingly, USP28 binds to Myc through Fbw7α in the nucleoplasm, but not Fbw7γ in the nucleolus, and reverses Fbw7α-, but not Fbw7γ-, mediated Myc ubiquitination and degradation,⁵⁵ suggesting that it deubiquitinates and stabilizes Myc in the nucleus, but not in the nucleolus. Consistently, USP28 overexpression is found in colon and breast cancers and USP28-mediated stabilization of Myc is essential for the cancer cell proliferation observed in the study.⁵⁵ In response to DNA damage, USP28 dissociates from Fbw7α, allowing Myc to be degraded by Fbw7α, providing a mechanism for Myc degradation upon DNA damage.⁵⁶

Interestingly, a recent study has increased the complexity of this USP28-Fbw7α-Myc axis. Schulein-Volk et al⁵⁷ demonstrated that USP28 has a dose-dependent and dual regulation of Fbw7 function. An elegant USP28 knockout mouse study revealed that USP28 primarily controls the stability of Fbw7 by antagonizing its auto-ubiquitination and degradation. Thus, while heterozygous deletion of USP28 can still maintain stable Fbw7 to ubiquitinate and degrade its substrates, complete deletion of USP28 triggers Fbw7 auto-destruction, resulting in the stabilization of the Fbw7 substrates such as Myc and cyclin E.⁵⁷ Thus both overexpression of USP28 and its complete deletion stabilize Myc and promote cell transformation.⁵⁷ Further complexity was revealed in an intestine study, where it was shown that deletion of USP28 rescued the cytotoxic effects of Fbw7-deficiency in primary fibroblasts and meliorated the hyperproliferation and the impaired goblet and Paneth cell differentiation in mice with intestine-specific deletion of Fbw7 (Fbw7^ΔIEC).⁵⁸ This correlates with correcting the aberrant accumulation of Fbw7 substrates such as Myc, c-Jun and NICD1. These results suggest that USP28 can function independent of a direct interaction with Fbw7. Using peptide binding assays, it was shown that USP28 binds to the same Myc phosphodegron recognized by Fbw7 in MBI; yet USP28 binds to the non-phosphorylated form whereas Fbw7 binds to the phosphorylated form of this Myc degron.⁵⁸ These studies demonstrate a complex regulation of Myc by the USP28-Fbw7 interplay, suggesting an important role in the dynamic and finely tuned regulation of Myc activity necessary for normal cellular function.

Recently, USP37 was found to be another DUB that deubiquitinates and stabilizes Myc.⁵⁹ Interestingly, USP37 regulation of Myc does not require either Fbw7 or the phosphorylation of Myc at T58. Instead, it directly interacts with Myc at the MBIII region. Functionally, USP37 promotes cell proliferation and metabolism by stabilizing Myc and it is also overexpressed in tested human lung cancers.⁵⁹ Whether the activity of USP37 coordinates with that of USP28 requires further study.

Both USP28 and USP37 regulate Myc stability in the nucleoplasm. Studies indicate that Myc is mainly degraded in the nucleolus and proteasome inhibition leads to accumulation of Myc in the nucleolus.^27,60,61 Importantly, Myc-mediated cell growth and tumorigenesis is well tied with its role in enhancing ribosome biogenesis in the nucleolus.^62,63 Thus, the nucleolus certainly plays a key role in regulating Myc levels and oncogenic activity and it is important to understand whether and how Myc can be deubiquitinated in the nucleolus. Toward this goal, we recently discovered the nucleolar USP36 as a novel Myc DUB and a positive Myc regulator in the nucleolus.⁶⁴

USP36 is a Novel Nucleolar DUB for Myc

USP36 was initially identified as a DUB overexpressed in ovarian cancers.⁶⁵ Later, it was found that USP36 is mainly located in the nucleolus and regulates the stability of nucleolar proteins nucleophosmin and fibrillarian, both of which are involved in rRNA processing and ribosomal biogenesis.^66,67 Knockdown of USP36 impaired rRNA synthesis, processing and ribosomal biogenesis in the nucleolus.⁶⁷ Further supporting this role, a yeast homolog ubp10 controls the stability of the largest subunit of RNA polymerase I and is essential for RNA Pol I activity and yeast cell growth; and Ubp10 deletion-induced yeast growth retardation can be rescued by human USP36.⁶⁸ Like many nucleolar proteins, USP36 likely shuttles between the nucleolus and the nucleoplasm and a small fraction of USP36 is located in the nucleoplasm,⁶⁴ indicating that USP36 may have additional functions in the nucleoplasm. Further, it has been shown that the Drosophila USP36 homolog (dUSP36) regulates autophagy⁶⁹ and the NFκB-associated inflammation signaling by targeting the IMD proteins,⁷⁰ suggesting that USP36 executes multiple functions in cells.

We found that USP36 directly deubiquitinates and stabilizes Myc, and this can occur in the nucleolus. USP36 directly binds to Myc, and it interacts specifically with Fbw7γ in the nucleolus, but not Fbw7α in the nucleoplasm, and it forms a tertiary complex with Myc and Fbw7γ.⁶⁴ Consistently, ablation of USP36 significantly reduced the levels of Myc and inhibited cell proliferation. Its overexpression was also seen in many human cancers such as breast and lung cancers, implying the oncogenic nature of USP36.⁶⁴ As Myc is a master regulator of ribosomal biogenesis, our results further support the role of USP36 in ribosomal biogenesis and demonstrate USP36 as a critical Myc regulator.⁶⁴ Yet, many questions still remain.

Although USP36 regulates Myc in the nucleolus, it increases the steady state levels of Myc in both the nucleoplasm and the nucleolus,⁶⁴ suggesting that the deubiquitinated nucleolar Myc can shuttle back to the nucleoplasm. This is consistent with the ability of USP36 to enhance Myc transactivation activity of Pol II-driven transcription. These data do not completely exclude the possibility that USP36 might also directly deubiquitinate Myc in the nucleoplasm, as a small fraction of USP36 is localized in the nucleoplasm and it can directly interact with Myc without the need of Fbw7. Also our ubiquitination assays were conducted in cells treated with proteasome inhibitor, which results in the accumulation of the ubiquitinated Myc in the nucleolus.^27,60,61 The nucleolar-localization defective mutant of USP36 could solve this issue. Most likely, at normal homeostatic conditions, USP36 mainly controls Myc ubiquitination and turnover in the nucleolus. It will be interesting to determine whether USP36 localization and its regulation of Myc changes under stress conditions and whether post-translational modifications can regulate the cellular localization and function of USP36.

It is also important to further elucidate the interplay between USP36 and Fbw7. We show that USP36 interacts specifically with Fbw7γ and thus the detailed interaction needs to be analyzed. Why it only interacts with Fbw7γ, but not Fbw7α, will be interesting to determine. Do USP36 and Fbw7γ mutually regulate each other, given that an ubiquitin E3 often complexes with a corresponding DUB to dynamically regulate a substrate? We indeed observed that siRNA-mediated knockdown of Fbw7γ or all Fbw7 isoforms increases the levels of USP36 (Fig. 2A) without affecting its nucleolar localization (Fig. 2B), suggesting that Fbw7 may ubiquitinate and destabilize USP36. Conversely, it is interesting to test whether USP36 could suppress Fbw7γ autoubiquitination and stabilize it, similar to the role of USP28 on Fbw7.^57,58 It is also important to differentiate the role of USP36 as a DUB to directly deubiquitinate Myc or as an Fbw7γ inhibitor to suppress Fbw7γ-mediated Myc ubiquitination, or a combined effect of both.

Figure 2. Fbw7γ regulates the levels of USP36, but not its nucleolar localization. Hela cells transfected with scrambled, Fbw7γ specific siRNA or pan-siRNA targeting all Fbw7 isoforms were subjected to immunoblot to detect the indicated proteins (A) and IF staining with anti-USP36 (green) or B23 (red) antibodies (B).

Another important question is whether the USP36- Fbw7γ axis interplays with the USP28- Fbw7α axis to regulate Myc. Both USP28 and USP36 deubiquitinate Myc and they interact with Fbw7α and Fbw7γ, respectively. Interestingly, USP36 can inhibit Myc degradation mediated by either Fbw7α or Fbw7γ, suggesting that Fbw7α ubiquitination of Myc in the nucleoplasm collaboratively contributes to Fbw7γ-mediated Myc ubiquitination in the nucleolus, either by adding priming ubiquitins or by facilitating translocation of Myc to the nucleolus, which would be antagonized by USP28.⁷¹ Indeed, we observed that overexpression of USP36 and USP28 cooperatively further increase Myc levels compared to individual overexpression of USP36 or USP28 (Fig. 3). Thus, it is important in future studies to examine whether the USP28-Fbw7α and USP36-Fbw7γ axis regulate Myc ubiquitination sequentially at the respective cell compartments and how they may interplay to dynamically control Myc ubiquitination and degradation (Fig. 4).

Figure 3. USP36 and USP28 collaboratively enhance Myc levels. 293 cells transfected with V5-USP36, Flag-USP28, or both plasmids were assayed by immunoblot using anti-Myc, anti-V5 and anti-Flag antibodies as indicated.

Figure 4. A diagram showing Fbw7 ubiquitinating and coordinating DUBs known to stabilize Myc. USP28 and USP37 act in the nucleoplasm whereas USP36 deubiquitinates Myc in the nucleolus.

Finally, we show that USP36 itself is a Myc target gene, suggesting that USP36 forms a positive feedback regulatory loop with Myc and plays an important role in rapid Myc stabilization upon growth stimulation. Indeed, we show that knockdown of USP36 drastically inhibited Myc induction following serum stimulation. Intriguingly, following its peak induction at 2 hours after serum stimulation, Myc levels decrease, while USP36 remains at relatively high levels for up to 12 hours,⁶⁴ due to the nature of its longer half-life (data not shown). This indicates that USP36 may lose its activity to target Myc before its level declines, although the underlying mechanism is not known. It will be interesting to examine whether posttranslational modifications of USP36 may be involved. Also, USP36 levels are likely regulated through cell cycle progression and/or USP36 regulation of Myc may be regulated by the cell cycle. Nevertheless, USP36 induction by Myc and in turn stabilization of Myc by USP36 suggest an elegant feed-forward regulatory loop that ensures a rapid control of Myc levels and activity following growth stimulation or in response to other extrinsic or intrinsic cues.

Conclusions

Emerging evidence indicates that deubiquitination regulation of Myc is equally important for the control of Myc protein stability and activity as ubiquitination regulation. Whether other DUBs besides the above mentioned USPs could down-regulate Myc activity with or without affecting Myc stability similar to some of the ubiquitin E3s is an interesting question to be explored. Does USP36 co-regulate Myc with ubiquitin E3s other than Fbw7 during alternate cell cycle phases or under different physiological conditions? Does it co-regulate Myc with DUBs other than USP28? It has been shown that SCF^β-TRCP antagonizes SCF^Fbw7 to ubiquitinate Myc at the N-terminus. Does USP36 deubiquitinate Myc that is ubiquitinated by SCF^β-TRCP or SCF^Skp2 in the nucleus? Finally, it is important to study the USP36 regulation of Myc using in vivo models to further understand its role in regulating Myc, nucleolar dynamics and ribosomal biogenesis. Myc is a master regulator of ribosome biogenesis; Myc-overexpressing cells require or are adapted to the elevated ribosomal biogenesis/translation activity for uncontrolled cell proliferation and growth (oncogenic addiction). As USP36 is mainly localized in the nucleolus^64,66 and plays a role in ribosome biogenesis,^66,68 knockdown or inhibition of USP36 may severely attenuate ribosomal biogenesis (by directly inhibiting Myc and other nucleolar targets^66,68 involved in ribosomal biogenesis), leading to a synthetic lethality in Myc-addicted cells. It will be interesting to investigate whether USP36 regulates other nucleolar proteins involved in rRNA biogenesis and whether it has a nucleolar scaffold protein function independent of its DUB activity. Nevertheless, our data showing that USP36 stabilizes Myc and its levels correlate with the levels of Myc in tumors suggest that targeting USP36 could be a means to develop novel cancer therapeutics.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by NIH/NCI grants R01 CA186241 to M-S. D. and R. S. and R01 CA160474 to M-S. D., and R01 CA 100855 to R. S., as well as a grant from the Medical Research Foundation (MRF) of Oregon to X.-X. S.