A new layer of degradation mechanism for PR-Set7/Set8 during cell cycle

ABSTRACT

Set8 is critically involved in transcription regulation, cell cycle progression and genomic stability. Emerging evidence has revealed that E3 ubiquitin ligases such as CRL4^cdt2 and SCF^Skp2 regulate Set8 protein abundance. However, it is unclear whether other E3 ligase(s) could govern Set8 level for proper cell cycle progression in response to genotoxic stress such as UV irradiation. Recently, we report that the SCF^β-TRCP complex regulates Set8 protein stability by targeting it for ubiquitination and subsequent degradation. Notably, Set8 interacts with the SCF^β-TRCP E3 ligase complex. We further revealed a critical role of CKI in SCF^β-TRCP-mediated degradation of Set8. Mechanistically, CKI-mediated phosphorylation of Set8 at the S253 site promotes its destruction by SCF^β-TRCP. Importantly, SCF^β-TRCP-dependent Set8 destruction also contributes to the tight control of cell proliferation and cell cycle progression, in response to UV irradiation. Here, we summarize our new findings regarding the crucial role of β-TRCP in CKI-mediated Set8 degradation, which could provide new evidence to support that dysregulation of a tight regulatory network of Set8 could lead to aberrant cell cycle process.

Set8 (also known as PR-Set7 and SETD8) has been previously characterized as a histone methyltransferase that specifically monomethylates Lys (20) of histone H4 (H4K20).^1,2 Set8 is critically involved in multiple cellular processes including transcription regulation, cell cycle, DNA repair, genome integrity and tumor metastasis.³ The biological function of Set8 in cell cycle is largely exerted through suppressing DNA replication.^3,4 Further studies identified that Set8 could methylate non-histone proteins such as p53, Twist, Wnt-activated genes, PCNA (proliferating cell nuclear antigen), ERα (estrogen receptor) and AR (androgen receptor).⁵ These results indicate that Set8 may play a key role in the development and progression of cancers. In keeping with this notion, multiple studies have demonstrated that high levels of Set8 are associated with tumor metastasis and poor survival of breast cancer patients.^6,11 Similarly, Set8 overexpression led to shorter survival time, suggesting that Set8 might be a predictor of worse outcome for gastric cancer.⁶ Mechanistically, Set8 interacted with AR and promoted AR-mediated transcription activation, leading to enhanced prostate cancer cell proliferation.⁷ Notably, Set8 interacted with Twist and induced epithelial-mesenchymal transition (EMT), and subsequently led to enhanced cell invasion in breast cancer.⁸ Taken together, Set8 could play an oncogenic role to facilitate tumorigenesis.

Since Set8 is a key oncoprotein in a variety of human cancers, it is pivotal to determine how Set8 is regulated in cellular processes. A number of studies reported that Set8 is regulated at both the transcriptional level⁹ and post-translational level.¹⁰ Some enzymes have been validated to control Set8 modification, including kinases, SUMO (small ubiquitin-like modifier), and ubiquitin ligases. For example, the Ser29 residue of Set8 is phosphorylated by Cyclin B/Cdk1 (cyclin dependent kinase 1) during mitosis.¹¹ One E3 ligase CRL4^Cdt2(Cullin-RING ubiquitin ligase 4^Cdt2) controls the Set8 ubiquitination and degradation.^12-14 Moreover, the anaphase-promoting complex APC^Cdh1 was also confirmed to govern the ubiquitination and degradation of Set8 during the G1 phase.¹¹ Recently, microRNAs (miRNAs) have been validated to control the Set8 expression in human malignancies. It has been known that miRNAs, small non-coding RNAs, largely exert their biological functions through inhibiting the translation or targeting the mRNAs for degradation through direct binding to the 3′-UTR region of mRNAs.¹⁵ Several studies have revealed that Set8 is a direct target of miR-502.^16,17 In support of this concept, breast cancer patients with lower level of miR-502 have high level of Set8.¹⁷ Moreover, the low expression ratio of miR-502 to Set8 mRNA is correlated to relatively poor overall survival in breast cancer patients.¹⁷ Consistent with the findings, treatment with miR-502 inhibited cell proliferation and cell cycle and retarded cell migration, invasion and EMT partly through inhibition of Set8.¹⁷ Another elegant study discovered that miR-7 inhibited cell invasion and sensitized tumor cells to DNA damages in part through targeting Set8 in breast cancer cells.¹⁸ Interestingly, a polymorphism at the miR-502 binding site in the 3′ UTR of the Set8 gene is associated with the outcome of various types of human cancers including small-cell lung cancer, ovarian cancer, hepatocellular carcinoma, cervical cancer, and non-Hodgkin's lymphoma.^19-21 These experimental evidences together suggest a possible oncogenic role for Set8 in facilitating tumorigenesis.

It is known that ubiquitination leads to degradation of proteins, which controls a plethora of cellular functions such as cell proliferation, cell cycle and survival.^22-25 This process is catalyzed by a group of enzymes including ubiquitin activating enzyme (E1), an ubiquitin-conjugating enzyme (E2), and an ubiquitin ligase (E3). It is noteworthy that the E3 can specifically determine the substrates for ubiquitination and more than 600 E3 ligases have been reported in human genome.^26,27 Among these E3 enzymes, the CRLs are the largest E3 ligase family in eukaryotes.²⁸ It has been known that Cullins act as a scaffold protein, and its C-terminus is associated with the RING finger protein RBX1/ROC1 or RBX2/ROC2. The N-terminal region of Cullins binds with Cullin-specific adaptor protein to target diverse substrates.^29,30 Furthermore, CRLs activity is regulated by Nedd8 post-translational modification known as neddylation. Interestingly, this process can be reversed by COP9 signalosome (CSN)-mediated deneddylation, leading to inactivation of CRLs.^30-32

Among eight Cullins (CUL1-7 and PARC), Cullin4 (CUL4) has been well studied. Cullin4 consists of 2 members, CUL4A and CUL4B, which share extensively sequence homology and functional redundancy.^28,30 CUL4 is expressed aberrantly in a wide range of human tumors and involves in tumor-related changes including cell cycle, DNA damage repair, histone methylation and oncoproteins turnover.^28,30 In particular, the aberrant expression of CUL4A has been identified in breast cancer, squamous cell carcinoma, pleural mesothelioma and non-small cell lung cancer.^33-37 Intriguingly, CUL4A induced EMT and promoted cancer metastasis in part via regulation of ZEB1 (zinc finger E-box-binding homebox 1) expression.³⁸ Strikingly, overexpression of CUL4 is associated with poor outcome in node-negative breast cancer.³⁹ In support of this notion, depletion of CUL4A inhibited cell proliferation, colony formation and breast cancer development.⁴⁰ Notably, Liu et al. generated Cul4a conditional knockout mice and found that skin-specific Cul4a ablation enhanced resistance to UV-induced skin carcinogenesis due to potential augmentation of cellular DNA repair proficiency.⁴¹ Moreover, transgenic mice with cre-inducible overexpression of the Cul4a gene exhibited hyperplasia.⁴² Therefore, CUL4A could play an oncogenic role in the tumorigenesis.

Emerging evidence has suggested that the abnormal cell cycle is a common feature of cancer cells. Remarkably, CRL4 has been validated to control cell cycle through regulation of multiple proteins including cell cycle proteins (Cyclins), cyclin-dependent kinases (CDKs), and cyclin-dependent kinase inhibitors (CDKIs), which are critical for cell cycle process.³⁰ For example, Cyclin E is a key regulator of the entry from G1 phase to S phase and its aberrant expression is observed in several types of human cancers.^43,44 Studies have discovered that CRL4 regulated the degradation of Cyclin E.^45,46 Overexpression of CRL4 decreased Cyclin E protein level, while down-regulation of CRL4 increased Cyclin E expression.^45,46 Clearly, p21 is an inhibitor of CDK and prevents the occurrence of the S phase as well as regulates the tumor development and progression.⁴⁷ Consistently, silencing CRL4 by its siRNA increased the level of p21, indicating that CRL4 may regulate p21 expression.^48,49 In addition, p27 and p16, 2 key CDKIs, are also regulated by CRL4.^46,50,⁵¹ These reports indicate that CRL4 exerts its oncogenic function partly through dysregulation of cell cycle.

β-TRCP (beta-transducin repeats containing proteins) belongs to the SCF (Skp1-Cullin-1-F-box protein) type of E3 ligase complexes. The SCF ligase complex included Skp1, Cullin-1, Rbx1/Roc1, and one of the 69 F-box proteins. Notably, β-TRCP recognizes the consensus sequence D-pS-G-X-X-pS (X represents any amino acid) degron of substrates.⁵² Moreover, phosphorylation of both serine residues by specific kinases is required for β-TRCP-mediated degradation.⁵² Dysregulation of β-TRCP is involved in regulation of several cellular processed such as cell cycle, apoptosis, invasion, and tumorigenesis through recognizing specific substrates including EMI-1 (early mitotic inhibitor-1),^53,54 Wee1,⁵⁵ and Cdc25A (cell division cycle 25 homolog A).^56,57 For example, the degradation of EMI-1 by β-TRCP induced mitotic catastrophe.^53,54 β-TRCP targeted Wee1, a Cdc2 inhibitory kinase, leading to abnormal mitosis.⁵⁵ Biologically, β-TRCP regulated Cdc25A to promote the cell cycle progression.^56,57 Moreover, β-TRCP recognized and degraded BTG (B-cell translocation gene) and thus controlled the cell cycle and cell proliferation.⁵⁸ The centriolar protein Plk4 (polo-like kinase 4) is a regulator of centriole biogenesis, and β-TRCP can bind and degrade Plk4, which is phosphorylated by itself.^59,60 Another elegant report shows that Plk1 can phosphorylate Bora and promote its degradation by β-TRCP, leading to subsequent regulation of mitotic progression.⁶¹ Similarly, β-TRCP was reported to regulate centrosome separation partly via targeting CEP68 for degradation.⁶² Interestingly, after exposure to UV irradiation, Securin is degraded by β-TRCP and results in cell cycle arrest.⁶³ Moreover, REST (repressor element-1-silencing transcription factor), which participates in cell cycle, is also degraded by β-TRCP.⁶⁴ In line with this observation, β-TRCP-mediated degradation of Claspin is important for the efficient and timely termination of the DNA replication checkpoint.⁶⁵ Intriguingly, inhibition of Claspin proteolysis led to subsequent activation of Chk1 and attenuated the recovery from the DNA replication stress response, thereby delaying entry into mitosis.⁶⁵ These findings strongly suggest that β-TRCP plays a critical role in regulation of cell cycle progression.

A number of evidence indicates that β-TRCP could be an oncoprotein in certain types of human cancers. In support of this notion, overexpression of β-TRCP has been observed in a variety of human cancers such as colorectal cancer,⁶⁶ hepatoblastoma,⁶⁷ pancreatic cancer,⁶⁸ and melanoma.⁶⁹ Consistent with an oncogenic role in human cancers, β-TRCP enhanced cell growth and tumor growth in mice.^70,71 Surprisingly, studies have argued that β-TRCP could also play tumor suppressive role in a tissue-specific manner. In keeping with this note, mutations of β-TRCP were observed in several human cancers including gastric cancer,^72,73 prostate cancer⁷⁴ and breast cancer.⁷⁵ Since β-TRCP targets both oncoproteins and tumor suppressors for proteolysis, β-TRCP could play an oncogenic or tumor suppressive role in the tissue-specific or cellular context-dependent manner. Without a doubt, conditional engineered animal models are necessary to further parse the exact function of β-TRCP in tumorigenesis.

As we described before, Set8 is a cell-cycle-regulated enzyme and promotes chromatin compaction.¹⁰ CRL4 E3 ligase ubiquitinated and degraded Set8 in S phase of the cell cycle, which required the interaction between PCNA and Set8.^13,14 The inhibition of the CRL4-Cdt2-PCNA-Set8 degradation axis delayed the progression from G2 phase and cell proliferation.^13,14 Moreover, the dysregulation of the CRL4-Cdt2-PCNA-Set8 axis led to DNA damage and induction of p53 and p53-transactivated proapoptotic genes.¹³ Additionally, depletion of CRL4 (or Cdt2) increased Set8 stability and the level of aberrant H4K20me1. A checkpoint-mediated G2 arrest happened due to the dysregulation of Set8, and destruction of Set8 preserved genome stability through preventing aberrant chromatin compaction during DNA synthesis.¹⁴ Furthermore, the ultraviolet irradiation accelerated the degradation of Set8 in the CRL4 (Cdt2) ubiquitination way.⁷⁶ Interestingly, there is a cross-regulation between Cullin 4 and Cullin 1 E3 ubiquitin ligases. For example, Cdt2 is autoubiquitylated by the CRL4A E3 ubiquitin ligase and it is also additionally polyubiquitylated and degraded by CRL1-FBXO11.⁷⁷ CRL1 (FBXO11)-mediated degradation of Cdt2 promoted the stabilization of Set8 and p21.⁷⁷ The degradation of Cdt2 also stimulated the migration of epithelial cells and subsequently the stabilization of Set8.⁷⁷ One study showed that Cdt2 is a prognostic marker and a therapeutic target in melanoma. With the treatment of pevonedistat, Cdt2 is suppressed and melanoma cell growth is promoted due to the stabilization of Set8.⁷⁸ Taken together, these results demonstrate that CRL4 (Cdt2)-dependent cell-cycle regulation of Set8 for the maintenance of cell viability and chromatin compaction.

Importantly, Set8 has been reported to be degraded after UV exposure, but the exact mechanism remains largely unclear. Our study first investigated that Set8 interacts with the SCF^β-TRCP E3 ligase complex.⁷⁹ The co-immunoprecipitation experiments are performed to find that exogenously expressed Set8 was associated with endogenous β-TRCP1, and β-TRCP1 mutant (R474A) reduced the interaction between the β-TRCP1 and Set8.⁵² Moreover, our results indicated that Set8 interacted with SCF components including Skp1 and Rbx1.⁷⁹ Importantly, depletion of endogenous β-TRCP extended the half-life of endogenous Set8 protein and increased the level of Set8 protein. In support of the notion that both Cullin1 and Cullin 4 are involved in controlling Set8 stability, the protein level of Set8 was upregulated after depletion of either Cullin1 or Cullin4. It is known that substrates are phosphorylated by kinase and then degraded in the SCF type of E3 ligases-dependent way.^52,80 We found that CKI (casein kinase I) phosphorylated Set8 at Ser253 and subsequently triggered its destruction by SCF^β-TRCP, suggesting that CKI could be a key upstream kinase to govern Set8 degradation.⁷⁹ Consistently, the inactivation of CKI also disrupted the interaction between Set8 and β-TRCP1.

Previous studies showed that Set8 levels are significantly reduced after DNA damage.^13,14 In line with this concept, we also observed that the levels of endogenous Set8 are significantly reduced in cells after UV treatment.⁷⁹ Intriguingly, UV-triggered reduction of Set8 was partially reversed by depletion of endogenous β-TRCP1. Biologically, overexpression of Set8 inhibited cell proliferation.¹⁴ In line with this notion, we also found that ectopic expression of Set8^WT inhibited cell proliferation, while the Set8^S253A mutants caused a more marked effect in cell proliferation inhibition.⁷⁹ Accumulating evidence indicated that acute expression of an oncogene such as Ha-Ras,⁸¹ Akt,^82,83 BRAF^84,85 or HIF (hypoxia-inducible factor 1),⁸⁶ could lead to cell growth arrest but not accelerate cell proliferation. This phenomenon is called OIS (onocogene-induced senescence). In concert with these observations, deletion of tumor suppressor PTEN (phosphatase and tensin homolog 10)^87,88 or VHL (Von Hippel-lindau)⁸⁹ also led to the onset of senescence, which could be due to the aberrant expression of pAkt and HIF oncoproteins. Ectopic expression of BRAF directly caused senescence in melanocytes.⁸⁴ Logically, the disruption of CRL4^cdt2 or the SCF^β-TRCP pathway led to Set8 stabilization, and subsequently triggered cell growth arrest partly via OIS mechanism.

In summary, there are at least 2 major ways for Set8 degradation during the cell cycle transitions. Importantly, different E3 ligases could regulate different cell cycle regulators such as p21, Cdt2, and Set8. Set8 could be governed by Skp2 and β-TRCP in G1 phase, whereas Set8 destruction is controlled by CRL4^cdt2 in S phase (Fig. 1). Although these findings dissect the regulation of Set8 by 3 E3 ligases, how these ligases fine-tune the control of the timely destruction of Set8 to ensure proper cell cycle process is not elucidated. Moreover, tissue specific knockout mice or transgenic mice are required to further determine contribution of Set8 in tumorigenesis. As Set8 is tightly associated with tumorigenesis, targeting Set8 could be a potential strategy for the treatment of human cancers. To this end, one study has discovered that Set8 could be regulated by estradiol and bisphenol A in ovarian cancer cells.⁹⁰ Moreover, a natural compound curcumin has been reported to exert its anti-tumor activity through regulation of Set8 in pancreatic cancer.⁹¹ Further investigations are needed to develop and discover new safer inhibitors of Set8 for human cancer patients with overexpressing Set8.

Figure 1.

A schematic model illustrating how β-TRCP, Skp2 and Cdt2 govern the protein stability of Set8 during the cell cycle procession. In G1 phase, Set8 could be controlled by SCF^Skp2 and SCF^β-TRCP. However, in S phase, Set8 destruction is largely governed by CRL4^Cdt2.

Disclosure of potential conflicts of interest

The authors declare no conflict of interest.

Funding

This work is supported in part by the NIH grants to W.W. (GM094777 and CA177910). This work was supported by grant from National Natural Science Foundation of China (NSFC number 81572936) and the priority academic program development of Jiangsu higher education institutions.