Cyclin D1 is a key factor controlling cell cycle progression. It forms a complex with CDK4/6 and functions as a regulatory subunit in G1/S phase transition during cell cycle progression. Proteasome degradation is one of the critical post-translational regulatory mechanisms modulating steady-state protein levels of cyclin D1 during normal cell cycle progression (Qie and Diehl, 2020). Cullin-Ring complexes comprise the largest known family of ubiquitin ligases. Human cells express seven different cullins, CUL1, -2, -3, -4A, -4B, -5, and -7, and each of them nucleates a different ubiquitin ligase (Petroski and Deshaies, 2005). Recent studies demonstrate that F-box proteins Fbxw8, Fbx4, Fbxo31, and AMBRA1 interact with CUL1/7 and mediate cyclin D1 degradation in a phosphorylation-dependent manner (Chaikovsky et al., 2021; Kumar et al., 2005; Li et al., 2018; Maiani et al., 2021; Okabe et al., 2006; Santra et al., 2009; Simoneschi et al., 2021). However, it is not known if other E3 ligase(s) is also involved in maintaining steady-state protein levels of cyclin D1. In this study, we have screened an E3 ligase siRNA library and identified three additional cullin-associated E3 ligases, which mediate cyclin D1 ubiquitination and proteasome degradation. Our findings indicate multiple cullin-associated E3 ligases participate in the regulation of cyclin D1 stability in the cells.

Results and discussion

To determine if cullins are required for cyclin D1 degradation, CUL1–7 (CUL1, -2, -3, -4A, -4B, -5, and -7) expression plasmids were co-transfected with cyclin D1 into HEK293 cells. Forced expression of CUL1–7 significantly suppressed cyclin D1 protein levels (Figure 1—figure supplement 1A). In contrast, the levels of cyclin B1 (G2/M regulator) and cyclin A (S phase regulator) were not affected (Figure 1—figure supplement 1A), indicating that the CUL1–7 specifically regulate cyclin D1 protein levels. Consistent with these findings, RNAi-mediated knockdown of cullins significantly enhanced levels of endogenous cyclin D1 (Figure 1—figure supplement 1B). The knockdown efficiency of these siRNAs to each cullin mRNA was about 60–90% (Figure 1—figure supplement 2). In addition, the silencing specificity of each cullin siRNA has been verified through both western blotting and real-time PCR assays (Figure 1—figure supplements 3 and 4). Results of luciferase assay also showed that cullins inhibited cyclin D1 activity (Figure 1—figure supplement 5). In contrast, silencing of the cullins led to increased cyclin D1 activity (Figure 1—figure supplement 6).

We then determined whether cullin-induced cyclin D1 degradation is ubiquitin-dependent and performed ubiquitination assay. We found that forced expression of CUL1–7 resulted in polyubiquitination of wild-type (WT) cyclin D1 (Figure 1—figure supplement 7). It was noted that CUL4B and CUL7 exhibited the most significant effect on cyclin D1 ubiquitination compared with other cullins. In contrast, CUL1–7 had no effects on the ubiquitination of mutant cyclin D1 with phosphorylation defect (T286A) (Figure 1—figure supplement 7). It has been reported that Thr286 phosphorylation is required for cyclin D1 ubiquitination (Diehl et al., 1997). Our results suggest that cullin-induced cyclin D1 ubiquitination is phosphorylation-dependent. A proteasome degradation inhibitor, MG132, was used to further determine if the cullins-mediated cyclin D1 degradation is through a proteasome-dependent mechanism. MG132 significantly reversed cyclin D1 degradation induced by CUL1–7 (Figure 1—figure supplement 8). Compared with their effects on cyclin D1 protein degradation, CUL1–7 had no significant effects on cyclin D1 mRNA expression (Figure 1—figure supplement 9). These results demonstrated that these seven human cullins may play critical and redundant roles in cyclin D1 degradation through a phosphorylation-dependent mechanism.

Each cullin protein interacts with different multi-subunit ubiquitin ligases (Petroski and Deshaies, 2005). Hundreds of substrate receptor subunits associated with different cullin-Ring ligases have been characterized in mammalian cells (Petroski and Deshaies, 2005). To identify additional E3 ligases which are associated with cullin proteins and are involved in cyclin D1 degradation, we generated an siRNA library targeting known cullin-associated proteins. This library contains triplicated siRNAs targeting to specific genes encoding for 154 ligase subunits, including F-box, SOCS, BTB-containing proteins, and DDB proteins. We transfected these siRNAs with 3xE2F-Luc reporter construct into NIH3T3 cells. 3xE2F-Luc reporter specifically responds to cyclin D1 stimulation (Ohtani et al., 1995). After screening the siRNA library, we found that 3xE2F luciferase activity was enhanced by 24 E3 ligases (>1.5-fold increase) (Figure 1—figure supplement 10). These results suggest that multiple E3 ligases associated with different cullin proteins are involved in cyclin D1 degradation. We selected five E3 ligases, which showed highest stimulation activities on the 3xE2F-Luc reporter and found that all of them have the ability to reduce cyclin D1 protein levels in HEK293 cells (Figure 1—figure supplement 11). Among these E3 ligase subunits, Fbxw8 (specific for CUL1 and -7) has been reported to induce cyclin D1 degradation (Okabe et al., 2006). The roles of other E3 ligase subunits, such as Keap1 (associated with CUL3), DDB2 (associated with CUL4A and -4B), and WSB2 (associated with CUL2 and -5) in cyclin D1 degradation have not been reported.

To further test the hypothesis that cyclin D1 degradation is mediated by multiple E3 ligases, we determined if different E3 ligases, Keap1, DDB2, and WSB2, are involved in cyclin D1 proteolysis. We first examined the interaction of these E3 ligases with cyclin D1 through co-immunoprecipitation (co-IP) assays. HEK293 cells were transiently transfected with HA-tagged Keap1 and treated with MG132 to prevent cyclin D1 degradation. Endogenous cyclin D1 was detected in the Keap1 immunoprecipitate, and this interaction was enhanced by the CUL3 (Figure 1A). Similarly, the interaction of DDB2 with endogenous cyclin D1 was enhanced by the CUL4A and -4B (Figure 1B). Consistent with these results, we also observed that CUL2 and -5 enhanced the interaction between WSB2 and endogenous cyclin D1 (Figure 1C). Rbx1 is associated with CUL1–7 and was used as a control in this study. The results of co-IP assay showed that Rbx1 interacted with endogenous cyclin D1 (Figure 1D).

Figure 1 with 12 supplements see all

Download asset Open asset

Figure 1—source data 1

Numerical data obtained during experiments represented in Figure 1.

https://cdn.elifesciences.org/articles/80327/elife-80327-fig1-data1-v2.zip

Download elife-80327-fig1-data1-v2.zip

Figure 1—source data 2

Original western blot files for Figure 1.

https://cdn.elifesciences.org/articles/80327/elife-80327-fig1-data2-v2.zip

Download elife-80327-fig1-data2-v2.zip

We then examined the effects of Keap1, DDB2, and WSB2 on cyclin D1 ubiquitination and degradation and used loss-of-function mutants of these E3 ligases to do experiments. In addition, we used Rbx1 as a control in these experiments. For Keap1, we used BTB domain deletion form which abolishes its binding with CUL3 (Furukawa and Xiong, 2005). For DDB2, we used the WD motif deletion form which blocks its binding to its substrates (Nag et al., 2001). For WSB2, we used the C-terminal deleted mutant which loses its binding to CUL2 and -5. And for Rbx1, we used the mutant form Rbx1^C53A/C56A which dramatically reduced the ligase activity (Ohta et al., 1999). We found that WT but not mutant Keap1 (F-box mutation) induced cyclin D1 ubiquitination (Figure 1E). In contrast, Keap1 had no effect on the ubiquitination of phosphorylation mutant cyclin D1 (T286A) (Figure 1E). Similarly, WT DDB2, WSB2, and Rbx1 induced cyclin D1 ubiquitination and mutant DDB2, WSB2, and Rbx1 had no effects on cyclin D1 ubiquitination (Figure 1F–H). In addition, Keap1, DDB2, WSB2, or Rbx1 had no effects on the ubiquitination of phosphorylation mutant cyclin D1 (T286A) (Figure 1E–H). These results indicate that Keap1, DDB2, and WSB2 interact with cyclin D1 and mediate cyclin D1 ubiquitination in a phosphorylation-dependent manner. We then determined the effects of these E3 ligases on cyclin D1 degradation by western blot analysis. Keap1, DDB2, WSB2, and Rbx1 induced WT but not T286A mutant cyclin D1 degradation (Figure 1I–L). In contrast, mutant Keap1, DDB2, WSB2, or Rbx1 had no effect on cyclin D1 degradation (Figure 1I–L). Consistent with these findings, silencing of Keap1, DDB2, WSB2, or Rbx1 resulted in the stabilization of cyclin D1 protein and significantly prolonged the half-life of cyclin D1 (Figure 1M and N). The silencing specificity of siRNA for each E3 ligase subunit has been verified through real-time PCR assays (Figure 1—figure supplement 12). To determine the effects of these E3 ligases on protein levels of endogenous cyclin D1, we transfected Keap1, DDB2, WSB2, and Rbx1 siRNA into human colon cancer cell line HCT-116 cells and found that knocking down of Keap1, DDB2, WSB2, or Rbx1 significantly enhanced endogenous cyclin D1 protein levels (Figure 2A, D, G, and J). Taken together, these findings indicate that cyclin D1 degradation is mediated by multiple E3 ligases which are associated with different cullin proteins.

Figure 2 with 4 supplements see all

Download asset Open asset

Figure 2—source data 1

Numerical data obtained during experiments represented in Figure 2.

https://cdn.elifesciences.org/articles/80327/elife-80327-fig2-data1-v2.xlsx

Download elife-80327-fig2-data1-v2.xlsx

Figure 2—source data 2

Original western blot files for Figure 2.

https://cdn.elifesciences.org/articles/80327/elife-80327-fig2-data2-v2.zip

Download elife-80327-fig2-data2-v2.zip

We then determined whether these E3 ligases affect the function of cyclin D1 during normal cell cycle progression. Over-expression of Keap1, DDB2, WSB2, or Rbx1 markedly inhibited the growth of HCT-116 cells. In contrast, these E3 ligases had no effect on cell proliferation in the cells stably transfected with T286A mutant cyclin D1 (Figure 2B, E, H, and K). Then, we performed fluorescence-activated cell sorting (FACS) analysis and treated cells with nocodazole to synchronize the cell division cycle. We found that in HCT-116 cells stably transfected with the WT cyclin D1, forced expression of Keap1, DDB2, WSB2, or Rbx1 prevented the nocodazole-mediated G2/M block and subsequently resulted in accumulation of cells in G1 phase (Figure 2—figure supplement 1). In contrast, Keap1, DDB2, WSB2, or Rbx1 failed to induce efficient cell cycle arrest at the G1 phase in the mutant cyclin D1 (T286A) transfected cells (Figure 2B, E, H, and K). The results indicate that these cullin-associated ubiquitin ligases promoted cyclin D1 degradation and subsequently decreased cell cycle progression rate in a phosphorylation-dependent manner. Moreover, forced expression of Keap1, DDB2, WSB2, or Rbx1 also blocked DNA synthesis in WT cyclin D1 transfected cells but not mutant cyclin D1 (T286A) transfected cells (Figure 2C, F, I, and L). Consistent with these findings, we also observed that decreased phospho-Rb protein levels in the cells in which Keap1, DDB2, WSB2, or Rbx1 were co-transfected with WT cyclin D1 but not the mutant cyclin D1 (T286) (Figure 2—figure supplement 2). It was reported that Lysine 269 is essential for cyclin D1 ubiquitination (Barbash et al., 2009). To determine if mutation of Lysine 269 will affect cyclin D1 degradation induced by newly identified E3 ligases in the current study, WT or mutant cyclin D1 (K269R) expression plasmids were co-transfected with Keap1 or DDB2 expression plasmid into HEK293 cells. AMBRA1 expression plasmid, a well known E3 ligase targeting cyclin D1, was used as a control in this experiment. 48 hr after transfection, changes in cyclin D1 protein levels were detected by the western blot analysis. We found that expression of Keap1 or DDB2 in 293 cells reduced WT but not K269R mutant cyclin D1 protein levels (Figure 2—figure supplement 3). To determine if these newly identified E3 ligases directly interact cyclin D1, we performed in vitro ubiquitination assay and used AMBRA1 as a positive control. We found that cyclin D1 ubiquitination ladders were observed when Keap1 or DDB2 were added, suggesting that Keap1 and DDB2 could directly interact with cyclin D1. In contrast, WSB2 only has weak effect to directly interact with cyclin D1 (Figure 2—figure supplement 4).

Our findings indicate that cyclin D1 degradation is mediated by multiple E3 ligases, which are associated with different cullin proteins, including CUL2, -3, -4A, -4B, and -5. Several F-box proteins, Fbxw8, Fbxo4, Fbxo31, and AMBRA1, have previously been reported to mediate cyclin D1 degradation, and these three E3 ligases interact with cullin 1, 4, or 7 (Chaikovsky et al., 2021; Kumar et al., 2005; Li et al., 2018; Maiani et al., 2021; Okabe et al., 2006; Santra et al., 2009; Simoneschi et al., 2021). In our study, we found that Keap1 (works together with CUL3), DDB2 (works together with CUL4A and -4B), and WSB2 (works together with CUL2 and -5), which functions as the Ring subunit binding protein interacting with the C-terminal domains of the cullins (Petroski and Deshaies, 2005), participated in the regulation of cyclin D1 proteolysis and affected the cell cycle progression (Figure 2M). The cullins interact with substrate receptor subunits to form ubiquitin ligase complex and mediate cyclin D1 degradation through a phosphorylation-dependent mechanism. This redundant cyclin D1 regulatory mechanism may function to ensure the normal G1-S phase transition and cell cycle progression in the cell.

Compared with transcriptional regulatory mechanism, post-translational regulation controls protein stability and plays a major role in the response to exogenous signals. It has been demonstrated that chemical reagents or γ-irradiation, which induce DNA damage, result in decreased cyclin D1 protein stability (Choo et al., 2009). DDB2 functions as a decisive factor for cell fate after DNA damage. DDB2-deficient cells are resistant to apoptosis in response to a variety of DNA damaging agents and also undergo cell cycle arrest (Stoyanova et al., 2009). Our results raise the possibility that DDB2 may regulate cell cycle pace after DNA damage through regulation of cyclin D1 protein stability.

In addition, cyclin D1 is abnormally up-regulated in many different types of human cancers as a proto-oncogene, including breast, lung, oesophagus, bladder, and lymphoid cancers (Gautschi et al., 2007). Genetic lesions, such as gene amplification or mutations, can only account for a portion of the cases for the abnormal up-regulation of cyclin D1 in these cancers. Growing evidences have shown that defective regulation at the post-translational level may be an important reason resulting in the increased cyclin D1 stability. The half-life of cyclin D1 in normal cells is relatively short (~20 min) due to ubiquitin-proteasome degradation regulated by a number of E3 ligases. In the current study we found that silencing of these ubiquitin ligases resulted in increased stability and accumulation of cyclin D1 in human cancer cells leading to the stimulation of cell cycle progression. In fact, a number of these cullins-associated E3 ligases, such as Fbxo31, Keap1, DDB2, and the cullins, have been demonstrated to function as tumor suppressors or components of tumor suppressor complexes (Fay et al., 2003). These tumor suppressors may exert their functions through controlling the steady-state protein levels of cyclin D1. Loss-of-function mutations of these E3 ligases lead to spontaneous tumor development (Ohta et al., 2008).

All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 1, Figure 2 and the figure supplements.

1. 1. Barbash O
   2. Egan E
   3. Pontano LL
   4. Kosak J
   5. Diehl JA

   (2009) Lysine 269 is essential for cyclin D1 ubiquitylation by the SCF(Fbx4/alphaB-crystallin) ligase and subsequent proteasome-dependent degradation

   Oncogene 28:4317–4325.

   https://doi.org/10.1038/onc.2009.287

   • PubMed
   • Google Scholar
2. 1. Chaikovsky AC
   2. Li C
   3. Jeng EE
   4. Loebell S
   5. Lee MC
   6. Murray CW
   7. Cheng R
   8. Demeter J
   9. Swaney DL
   10. Chen SH
   11. Newton BW
   12. Johnson JR
   13. Drainas AP
   14. Shue YT
   15. Seoane JA
   16. Srinivasan P
   17. He A
   18. Yoshida A
   19. Hipkins SQ
   20. McCrea E
   21. Poltorack CD
   22. Krogan NJ
   23. Diehl JA
   24. Kong C
   25. Jackson PK
   26. Curtis C
   27. Petrov DA
   28. Bassik MC
   29. Winslow MM
   30. Sage J

   (2021) The AMBRA1 E3 ligase adaptor regulates the stability of cyclin D

   Nature 592:794–798.

   https://doi.org/10.1038/s41586-021-03474-7

   • PubMed
   • Google Scholar
3. 1. Choo DW
   2. Baek HJ
   3. Motoyama N
   4. Cho KH
   5. Kim HS
   6. Kim SS

   (2009) ATM is required for rapid degradation of cyclin D1 in response to gamma-irradiation

   Biochemical and Biophysical Research Communications 378:847–850.

   https://doi.org/10.1016/j.bbrc.2008.11.132

   • PubMed
   • Google Scholar
4. 1. Diehl JA
   2. Zindy F
   3. Sherr CJ

   (1997) Inhibition of cyclin D1 phosphorylation on threonine-286 prevents its rapid degradation via the ubiquitin-proteasome pathway

   Genes & Development 11:957–972.

   https://doi.org/10.1101/gad.11.8.957

   • PubMed
   • Google Scholar
5. 1. Fay MJ
   2. Longo KA
   3. Karathanasis GA
   4. Shope DM
   5. Mandernach CJ
   6. Leong JR
   7. Hicks A
   8. Pherson K
   9. Husain A

   (2003) Analysis of CUL-5 expression in breast epithelial cells, breast cancer cell lines, normal tissues and tumor tissues

   Molecular Cancer 2:40.

   https://doi.org/10.1186/1476-4598-2-40

   • PubMed
   • Google Scholar
6. 1. Furukawa M
   2. Xiong Y

   (2005) BTB protein Keap1 targets antioxidant transcription factor Nrf2 for ubiquitination by the Cullin 3-Roc1 ligase

   Molecular and Cellular Biology 25:162–171.

   https://doi.org/10.1128/MCB.25.1.162-171.2005

   • PubMed
   • Google Scholar
7. 1. Gautschi O
   2. Ratschiller D
   3. Gugger M
   4. Betticher DC
   5. Heighway J

   (2007) Cyclin D1 in non-small cell lung cancer: a key driver of malignant transformation

   Lung Cancer 55:1–14.

   https://doi.org/10.1016/j.lungcan.2006.09.024

   • PubMed
   • Google Scholar
8. 1. Kumar R
   2. Neilsen PM
   3. Crawford J
   4. McKirdy R
   5. Lee J
   6. Powell JA
   7. Saif Z
   8. Martin JM
   9. Lombaerts M
   10. Cornelisse CJ
   11. Cleton-Jansen AM
   12. Callen DF

   (2005) FBXO31 is the chromosome 16q24.3 senescence gene, a candidate breast tumor suppressor, and a component of an SCF complex

   Cancer Research 65:11304–11313.

   https://doi.org/10.1158/0008-5472.CAN-05-0936

   • PubMed
   • Google Scholar
9. 1. Li Y
   2. Jin K
   3. Bunker E
   4. Zhang X
   5. Luo X
   6. Liu X
   7. Hao B

   (2018) Structural basis of the phosphorylation-independent recognition of cyclin D1 by the SCF^FBXO31 ubiquitin ligase

   PNAS 115:319–324.

   https://doi.org/10.1073/pnas.1708677115

   • PubMed
   • Google Scholar
10. 1. Maiani E
    2. Milletti G
    3. Nazio F
    4. Holdgaard SG
    5. Bartkova J
    6. Rizza S
    7. Cianfanelli V
    8. Lorente M
    9. Simoneschi D
    10. Di Marco M
    11. D’Acunzo P
    12. Di Leo L
    13. Rasmussen R
    14. Montagna C
    15. Raciti M
    16. De Stefanis C
    17. Gabicagogeascoa E
    18. Rona G
    19. Salvador N
    20. Pupo E
    21. Merchut-Maya JM
    22. Daniel CJ
    23. Carinci M
    24. Cesarini V
    25. O’sullivan A
    26. Jeong Y-T
    27. Bordi M
    28. Russo F
    29. Campello S
    30. Gallo A
    31. Filomeni G
    32. Lanzetti L
    33. Sears RC
    34. Hamerlik P
    35. Bartolazzi A
    36. Hynds RE
    37. Pearce DR
    38. Swanton C
    39. Pagano M
    40. Velasco G
    41. Papaleo E
    42. De Zio D
    43. Maya-Mendoza A
    44. Locatelli F
    45. Bartek J
    46. Cecconi F

    (2021) AMBRA1 regulates cyclin D to guard S-phase entry and genomic integrity

    Nature 592:799–803.

    https://doi.org/10.1038/s41586-021-03422-5

    • PubMed
    • Google Scholar
11. 1. Nag A
    2. Datta A
    3. Yoo K
    4. Bhattacharyya D
    5. Chakrabortty A
    6. Wang X
    7. Slagle BL
    8. Costa RH
    9. Raychaudhuri P

    (2001) DDB2 induces nuclear accumulation of the hepatitis B virus X protein independently of binding to DDB1

    Journal of Virology 75:10383–10392.

    https://doi.org/10.1128/JVI.75.21.10383-10392.2001

    • PubMed
    • Google Scholar
12. 1. Ohta T
    2. Michel JJ
    3. Schottelius AJ
    4. Xiong Y

    (1999) ROC1, a homolog of APC11, represents a family of cullin partners with an associated ubiquitin ligase activity

    Molecular Cell 3:535–541.

    https://doi.org/10.1016/s1097-2765(00)80482-7

    • PubMed
    • Google Scholar
13. 1. Ohta T
    2. Iijima K
    3. Miyamoto M
    4. Nakahara I
    5. Tanaka H
    6. Ohtsuji M
    7. Suzuki T
    8. Kobayashi A
    9. Yokota J
    10. Sakiyama T
    11. Shibata T
    12. Yamamoto M
    13. Hirohashi S

    (2008) Loss of Keap1 function activates Nrf2 and provides advantages for lung cancer cell growth

    Cancer Research 68:1303–1309.

    https://doi.org/10.1158/0008-5472.CAN-07-5003

    • PubMed
    • Google Scholar
14. 1. Ohtani K
    2. DeGregori J
    3. Nevins JR

    (1995) Regulation of the cyclin E gene by transcription factor E2F1

    PNAS 92:12146–12150.

    https://doi.org/10.1073/pnas.92.26.12146

    • PubMed
    • Google Scholar
15. 1. Okabe H
    2. Lee SH
    3. Phuchareon J
    4. Albertson DG
    5. McCormick F
    6. Tetsu O

    (2006) A critical role for FBXW8 and MAPK in cyclin D1 degradation and cancer cell proliferation

    PLOS ONE 1:e128.

    https://doi.org/10.1371/journal.pone.0000128

    • PubMed
    • Google Scholar
16. 1. Petroski MD
    2. Deshaies RJ

    (2005) Function and regulation of cullin-RING ubiquitin ligases

    Nature Reviews. Molecular Cell Biology 6:9–20.

    https://doi.org/10.1038/nrm1547

    • PubMed
    • Google Scholar
17. 1. Qie S
    2. Diehl JA

    (2020) Cyclin D degradation by E3 ligases in cancer progression and treatment

    Seminars in Cancer Biology 67:159–170.

    https://doi.org/10.1016/j.semcancer.2020.01.012

    • PubMed
    • Google Scholar
18. 1. Santra MK
    2. Wajapeyee N
    3. Green MR

    (2009) F-box protein FBXO31 mediates cyclin D1 degradation to induce G1 arrest after DNA damage

    Nature 459:722–725.

    https://doi.org/10.1038/nature08011

    • PubMed
    • Google Scholar
19. 1. Simoneschi D
    2. Rona G
    3. Zhou N
    4. Jeong YT
    5. Jiang S
    6. Milletti G
    7. Arbini AA
    8. O’Sullivan A
    9. Wang AA
    10. Nithikasem S
    11. Keegan S
    12. Siu Y
    13. Cianfanelli V
    14. Maiani E
    15. Nazio F
    16. Cecconi F
    17. Boccalatte F
    18. Fenyö D
    19. Jones DR
    20. Busino L
    21. Pagano M

    (2021) CRL4AMBRA1 is a master regulator of D-type cyclins

    Nature 592:789–793.

    https://doi.org/10.1038/s41586-021-03445-y

    • Google Scholar
20. 1. Stoyanova T
    2. Roy N
    3. Kopanja D
    4. Bagchi S
    5. Raychaudhuri P

    (2009) DDB2 decides cell fate following DNA damage

    PNAS 106:10690–10695.

    https://doi.org/10.1073/pnas.0812254106

    • PubMed
    • Google Scholar
21. 1. Zhang M
    2. Xie R
    3. Hou W
    4. Wang B
    5. Shen R
    6. Wang X
    7. Wang Q
    8. Zhu T
    9. Jonason JH
    10. Chen D

    (2009) PTHrP prevents chondrocyte premature hypertrophy by inducing cyclin-D1-dependent Runx2 and Runx3 phosphorylation, ubiquitylation and proteasomal degradation

    Journal of Cell Science 122:1382–1389.

    https://doi.org/10.1242/jcs.040709

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Multiple Cullin-Associated E3 Ligases Regulate Cyclin D1 Protein Stability" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Mone Zaidi as the Senior Editor. The following individual involved in the review of your submission has agreed to reveal their identity: Chao Xie (Reviewer #1).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

This interesting set of data provides new understanding of the role of cullin-associated E3 ligases in regulating cyclin D1 protein stability. The data within the manuscript largely support most conclusions. However, there are major concerns that need to be addressed.

1) The exact mechanism of how cyclin D1 is ubiquitinated and degraded is incomplete. It is important for the authors to show direct interactions and in vitro ubiquitylation assays. It is also important to put the findings in a coherent model that discusses the existing literature showing that CRL4-AMBRA1 is the major E3 ligase under most tested conditions.

2) It is necessary for the authors to include appropriate controls for the co-IP and CHX chase experiments.

Reviewer #2 (Recommendations for the authors):

The lack of identified lysine residue mutant within cyclin D1 that is deficient for ubiquitination is considered a significant weakness for the manuscript.

The in vitro ubiquitination assays of cyclin D1 by different combinations of cullin-E3 ligases will further strengthen the major claim of the manuscript.

Reviewer #3 (Recommendations for the authors):

1) One major oversight of the study has been not to include the studies describing CRL4-AMBRA1 as a major E3 ligase degrading cyclin D (1-3) (Simoneschi et al., Nature 2021; Maiani et al., Nature 2021; Chaikovsky et al., Nature 2021). These studies provide convincing evidence that CUL4 and AMBRA1, but not other cullins or substrate adaptors regulate endogenous cyclin D levels in several cell lines including RPE-1, U2OS, and HCT-116 cells, and show functional relevance during human development and cancer. Some of the results presented In these studies contradict the results presented in this short report (e.g. Simoneschi et al. show in Figure 1 that knockdown of only CUL4A/B and AMBRA1, but no other cullin or substrate adaptor, including DDB2, increase cyclin D1 levels in HCT-116 cells). Thus, the authors need to mention these previous findings in their introduction. This would obviously change the scope of their study and require them to address the discrepancies between these previous and their findings and also provide experimental evidence for when and how CRLs other than CRL4-AMBRA1 is important. This would ultimately result in a completely new study.

2) Overstatements:

– The authors claim that no other CRLs than CRL1/7 with substrate adaptors FBXW8, FBXO4, and FBXO31 has been implicated in Cyclin D degradation, which is not the case (see above).

– In the abstract they also claim that KEAP1, DDB2, and WSB2 directly interact with cyclin D1, something they did not test. They only provide evidence that these proteins, when ectopically expressed, can be in a complex with cyclin D1 in cells. These co-IP experiments also lack important controls (see below point 4). Direct interaction can only be claimed by in vitro reconstitution with recombinantly purified proteins.

– Throughout the manuscript effects with RBX1 are presented as novel findings; however, RBX1 is an integral part of all CRLs and thus should be presented as a positive control

3) Another important issue that arises from this study is how different CRLs would recognize cyclin D structurally. The authors draw a model (Figure 2M), in which all four different types of CRLs recognize cyclin D by the same mechanism using different substrate adaptors. This would be rather surprising, given that different substrate adaptors are generally thought to use unique manners to bind specific sets of substrates. Given that the authors do no provide evidence for direct interaction and ubiquitylation, It is equally conceivable that the effects they are observing are occurring through indirect mechanisms and unknown substrates (e.g. AMBRA1).

4) In addition to these major conceptual weaknesses, there are also missing controls for several experiments. To name a few: the anti-substrate adaptor co-IP experiments (Figure 1A-C) are lacking loading controls that show that the same amount of substrate adaptor was enriched in the IP. The ubiquitylation assays (Figure 1E-H) are missing controls that show that the same amount of ubiquitin was immunoprecipitated in each condition. The CHX assays (Figure 1M) need to be quantified and half-lives should be determined.

Essential revisions:

This interesting set of data provides new understanding of the role of cullin-associated E3 ligases in regulating cyclin D1 protein stability. The data within the manuscript largely support most conclusions. However, there are major concerns that need to be addressed.

1) The exact mechanism of how cyclin D1 is ubiquitinated and degraded is incomplete. It is important for the authors to show direct interactions and in vitro ubiquitylation assays. It is also important to put the findings in a coherent model that discusses the existing literature showing that CRL4-AMBRA1 is the major E3 ligase under most tested conditions.

In the revised manuscript we have performed in vitro ubiquitination assays and used AMBRA1 as a positive control in this assay. We found that Keap1, DDB2, and WSB2 can induce cyclin D1 ubiquitination. Especially, Keap1-induced cyclin D1 ubiquitination ladder is similar to AMBRA1-induced cyclin D1 ubiquitination ladder. In contrast, no clear ubiquitination ladder was observed in Rbx1 group (Figure 2—figure supplement 4). In addition, we have also added those key publications about the role of CRL4-AMBRA1 in cyclin D1 degradation in the revised manuscript as the reviewer suggested.

2) It is necessary for the authors to include appropriate controls for the co-IP and CHX chase experiments.

We have added ubiquitin input controls for co-IP (Figure 1E-H) and quantified the half-lives for CHX assays (Figure 1N).

Reviewer #2 (Recommendations for the authors):

The lack of identified lysine residue mutant within cyclin D1 that is deficient for ubiquitination is considered a significant weakness for the manuscript.

It was reported that Lysine 269 is essential for cyclin D1 ubiquitination (Barbash et al., 2009). WT or mutant cyclin D1 (K269R) expression plasmids were co-transfected with Keap1, DDB2, or AMBRA1 expression plasmid into HEK293 cells. 48 hours after transfection, changes in cyclin D1 protein levels were detected by the western blot analysis. We found that expression of WT cyclin D1, but not K269R mutant cyclin D1, was decreased in HEK293 cells transfected with Keap1, DDB2, or AMBRA1 plasmid, suggesting that Lysine 269 is essential for cyclin D1 ubiquitination.

The in vitro ubiquitination assays of cyclin D1 by different combinations of cullin-E3 ligases will further strengthen the major claim of the manuscript.

We have performed in vitro ubiquitination assay as the reviewer suggested. The results demonstrated that Keap1, DDB2, or WSB2 can induce cyclin D1 ubiquitination. Especially, Keap1 induced cyclin D1 ubiquitination and formed ubiquitination ladder similar to AMBRA1-induced cyclin D1 ubiquitination ladder. In contrast, no clear ubiquitination ladder was observed in Rbx1 group (Figure S16).

Reviewer #3 (Recommendations for the authors):

1) One major oversight of the study has been not to include the studies describing CRL4-AMBRA1 as a major E3 ligase degrading cyclin D (1-3) (Simoneschi et al., Nature 2021; Maiani et al., Nature 2021; Chaikovsky et al., Nature 2021). These studies provide convincing evidence that CUL4 and AMBRA1, but not other cullins or substrate adaptors regulate endogenous cyclin D levels in several cell lines including RPE-1, U2OS, and HCT-116 cells, and show functional relevance during human development and cancer. Some of the results presented In these studies contradict the results presented in this short report (e.g. Simoneschi et al. show in Figure 1 that knockdown of only CUL4A/B and AMBRA1, but no other cullin or substrate adaptor, including DDB2, increase cyclin D1 levels in HCT-116 cells). Thus, the authors need to mention these previous findings in their introduction. This would obviously change the scope of their study and require them to address the discrepancies between these previous and their findings and also provide experimental evidence for when and how CRLs other than CRL4-AMBRA1 is important. This would ultimately result in a completely new study.

We agree that those findings about CRL4-AMBRA1 in cyclin D1 degradation provide most critical information in the regulation of steady-state protein levels of cyclin D1 in the cell. In addition, we have used CRL4-AMBRA1 as a positive control in our studies during revision of our manuscript.

2) Overstatements:

– The authors claim that no other CRLs than CRL1/7 with substrate adaptors FBXW8, FBXO4, and FBXO31 has been implicated in Cyclin D degradation, which is not the case (see above).

We have added the information about critical role of CRL4-AMBRA1 in regulation of cyclin D1 protein stability in the revised manuscript.

– In the abstract they also claim that KEAP1, DDB2, and WSB2 directly interact with cyclin D1, something they did not test. They only provide evidence that these proteins, when ectopically expressed, can be in a complex with cyclin D1 in cells. These co-IP experiments also lack important controls (see below point 4). Direct interaction can only be claimed by in vitro reconstitution with recombinantly purified proteins.

We have performed experiments to determine if Keap1, DDB2, or WSB2 could directly interact with cyclin D1 and induce cyclin D1 ubiquitination and we found that Keap1, DDB2, and WSB2 can induce cyclin D1 ubiquitination. Especially, Keap1 induced cyclin D1 ubiquitination and formed ubiquitination ladder similar to AMBRA1-induced cyclin D1 ubiquitination ladder. In contrast, no clear ubiquitination ladder was observed in Rbx1 group (Figure 2—figure supplement 4). We also added input control in the co-IP assays as the reviewer suggested.

– Throughout the manuscript effects with RBX1 are presented as novel findings; however, RBX1 is an integral part of all CRLs and thus should be presented as a positive control

We have presented Rbx1 as a control in the revised manuscript as the reviewer suggested.

3) Another important issue that arises from this study is how different CRLs would recognize cyclin D structurally. The authors draw a model (Figure 2M), in which all four different types of CRLs recognize cyclin D by the same mechanism using different substrate adaptors. This would be rather surprising, given that different substrate adaptors are generally thought to use unique manners to bind specific sets of substrates. Given that the authors do no provide evidence for direct interaction and ubiquitylation, It is equally conceivable that the effects they are observing are occurring through indirect mechanisms and unknown substrates (e.g. AMBRA1).

In the revised manuscript, we have performed in vitro ubiquitination assay as the reviewer suggested. Together, with co-IP assay and in vitro ubiquitination assay, our findings suggest that some E3 ligase that we have identified could directly interact with cyclin D1 and some others probably affect cyclin D1 protein stability by indirect mechanisms.

4) In addition to these major conceptual weaknesses, there are also missing controls for several experiments. To name a few: the anti-substrate adaptor co-IP experiments (Figure 1A-C) are lacking loading controls that show that the same amount of substrate adaptor was enriched in the IP. The ubiquitylation assays (Figure 1E-H) are missing controls that show that the same amount of ubiquitin was immunoprecipitated in each condition. The CHX assays (Figure 1M) need to be quantified and half-lives should be determined.

We have added proper controls for co-IP assays and ubiquitination assays and quantified the half-lives for CHX assays in the revised manuscript as the reviewer suggested.

Reference

Barbash, O., Egan, E., Pontano, L.L., Kosak, J., and Diehl, J.A. (2009). Lysine 269 is essential for cyclin D1 ubiquitylation by the SCF(Fbx4/alphaB-crystallin) ligase and subsequent proteasome-dependent degradation. Oncogene 28, 4317-4325.

Author details

1. Ke Lu

   Research Center for Computer-aided Drug Discovery, Chinese Academy of Sciences, Shenzhen, China

   Contribution

   Data curation, Formal analysis, Methodology

   Contributed equally with

   Ming Zhang

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1641-4748

2. Ming Zhang

   Department of Oncology, Johns Hopkins University, Baltimore, United States

   Contribution

   Data curation, Formal analysis, Methodology

   Contributed equally with

   Ke Lu

   Competing interests

   No competing interests declared

3. Guizheng Wei

   Research Center for Computer-aided Drug Discovery, Chinese Academy of Sciences, Shenzhen, China

   Contribution

   Investigation

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0009-0006-0479-5076

4. Guozhi Xiao

   Department of Biochemistry, Southern University of Science and Technology, Shenzhen, China

   Contribution

   Formal analysis, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4269-2450

5. Liping Tong

   Research Center for Computer-aided Drug Discovery, Chinese Academy of Sciences, Shenzhen, China

   Contribution

   Formal analysis, Project administration, Writing - review and editing

   Competing interests

   No competing interests declared

6. Di Chen

   Research Center for Computer-aided Drug Discovery, Chinese Academy of Sciences, Shenzhen, China

   Contribution

   Conceptualization, Funding acquisition, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   di.chen@siat.ac.cn

   Competing interests

   Reviewing editor, eLife

   "This ORCID iD identifies the author of this article:" 0000-0002-4258-3457

• 397

  Page views

• 107

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.