FBG1 is a promiscuous ubiquitin ligase that ...

Report

Cell Cycle 9:22, 4506-4517; November 15, 2010; © 2010 Landes Bioscience

FBG1 is a promiscuous ubiquitin ligase that sequesters APC2 and causes S-phase arrest Hsiang Wen,1 Namhun Kim,2 Ernesto J. Fuentes,3,4 Adam Mallinger,5 Pedro Gonzalez-Alegre6 and Kevin A. Glenn1,7,* 4

1 Department of Internal Medicine; Roy J. and Lucille A. Carver College of Medicine (CCOM); Iowa City, Iowa; 2Troy High School; Troy MI; 3Department of Biochemistry CCOM; Holden Comprehensive Cancer Center CCOM; 6Department of Neurology CCOM; 5Kansas City University of Medicine and Biosciences; Kansas City, Missouri; 7Veterans Affairs Medical Center; Iowa City, Iowa USA

Key words: FBG1, Cul1, Cul7, APC2, SCF, glycoprotein, degradation Abbreviations: Cul1, cullin 1; SCF, Skp1/Cul1/F-box protein, Roc1; FBG, F-box G domain

During cell proliferation, protein degradation is strictly regulated by the cell cycle and involves two complementary ubiquitin ligase complexes, the SCF (Skp, Cullin, F-box) and APC/C (Anaphase Promoting Complex/Cyclosome) ubiquitin ligases. SCF ligases are constitutively active and generally target only proteins after they have been selected for degradation, usually by phosphorylation. In contrast, APC/C complexes are themselves activated by phosphorylation and their substrates contain a targeting signal known as degron, a consensus amino acid sequence such as a D-Box. SCF complexes degrade proteins during the G1 phase. However, as DNA synthesis begins, the SCF complexes are degraded and APC/C complexes are activated. APC-2, a protein crucial to cell division, initiates anaphase by triggering the degradation of multiple proteins. This study explores an unexpected interaction between APC-2 and SCFFBG1. We found that FBG1 is a promiscuous ubiquitin ligase with many partners. Immunoprecipitation experiments demonstrate that FBG1 and APC2 interact directly. Mutagenesis-based experiments show that this interaction requires a D-Box found within the FBG1 F-box domain. Unexpectedly, we demonstrate that co-expression with FBG1 increases total APC2 levels. However, free APC2 is decreased, inhibiting cell proliferation. Finally, FACS analysis of cell populations expressing different forms of FBG1 demonstrate that this ubiquitin ligase induces S-phase arrest, illustrating the functional consequences of the interaction described. In summary, we have discovered a novel APC2 inhibitory activity of FBG1 independent from its function as ubiquitin ligase, providing the basis for future studies of FBG1 in aging and cancer.

Regulated protein degradation is critical for many cellular processes, such as aging, morphogenesis, signal transduction, transcription and cell cycle progression. The best studied mechanism of regulated protein degradation is ubiquitination, a process controlled by a complex interplay between a multitude of multimeric ubiquitin ligases. The key regulatory step of ubiquitination occurs at the level of attaching ubiquitin to the target, which occurs through a series of enzymatic steps. The first step is the activation of ubiquitin by the activating enzyme, E1. Activated ubiquitin is then transferred to a ubiquitin-conjugating enzyme, E2, that transfers ubiquitin to lysine residues of target proteins in a process facilitated by E3 ubiquitin ligases.1-3 E3 ubiquitin ligases specificity is determined both by partnering with particular E2s and by targeting specific substrates. The extent of specificity is reflected by the observation that the human genome likely encodes 527 E3s, compared to only 53 E2s.4,5 E3 ubiquitin ligases are broadly classified as either HECT-type or RING finger (Fig. 1).6 In the case of the latter, an E3 RINGfinger domain is essential for interacting with a specific E2. RING finger domains may be integral to the E3, as in Hrd1p/

Der3p or contained in a subunit of a multimeric complex, as in the anaphase promoting complex (APC) or the Skp1/Cul/F-box (SCF) complex, both of which control cell cycle progression.1,2,7 SCF ubiquitin ligases bind degradation targets via their F-box-protein subunit. F-box proteins are characterized by a bipartite structure that consists of a relatively invariant 42 amino acid F-box motif, involved in protein-protein interactions, and a highly variable C-terminal domain, which recognizes specific substrates. The human genome encodes 78 F-box proteins, broadly grouped according to homology within the C-terminal region. The F-box in the SCF complex recognizes Skp1, and one of several cullin proteins. Cullins act as scaffolding for the SCF complex and bind the ring-containing Roc protein, which bind an E2.8-11 Several E3 ubiquitin ligases, such as APC and SCF complexes, act in a highly coordinated manner to regulate cell cycle progression.12,13 Entry into S-phase occurs after the G1-S transition inhibitors, p27, p21 and p130 have been ubiquitinated by the SCFSkp2 complex.14 At the end of metaphase, the SCFβ-TrCP complex ubiquitinates the APC inhibitor Emi1, targeting it for

*Correspondence to: Kevin Glenn; Email: [email protected] Submitted: 07/25/10; Revised: 09/03/10; Accepted: 09/24/10 Previously published online: www.landesbioscience.com/journals/cc/article/13743 DOI: 10.4161/cc.9.22.13743 4506

Cell Cycle

Volume 9 Issue 22

report

Report

binds both APC2 and Cul7 with high affinity and Cul1 with low affinity. Furthermore, in contrast to all other ubiquitin ligases, FBG1 binds each of these cullins at unique sites within the F-box domain. We demonstrate that co-expression of FBG1 with Cul7 or APC2 stabilizes levels of both cullins. Finally, we show that FBG1 is unique among known F-box proteins in that it contains a non-canonical D-Box within its F-box domain, required for the growth arrest associated with FBG1 overexpression. Results Figure 1. Family of E3 ubiquitin ligases. Multimeric E3 ubiquitin ligases contain a separate RING containing protein and use a Cullin as a scaffold protein to bring the substrate binding protein the RING protein together. Cullins known to be used by the multimeric E3 ligases are shown on the right.

Figure 2. Compared to the other FBG proteins, FBG1 has a higher affinity for Cul7 than Cul 1. Cos-7 cells co-transfected with vectors expressing various FLAG-tagged FBGs, empty vector (0) or un-transfected (Cos), plus vectors expressing HA-tagged Cul1 and Cul7 for 48 hours. Cell lysates were immunoprecipitated with anti-FLAG agarose beads, and separated by SDS-PAGE, and probed with anti-HA antibody. Input lanes represent 10% of cell lysate used for IP.

degradation.15,16 This, in turn, triggers activation of the APC/C complex, which remains active from mid-M phase until the end of G1 phase. The activated APC/C complex ubiquitinates B cyclins and other proteins, which allows the cell cycle to progress to the G1 phase. Finally, in a negative feedback step, the APC Cdh1 complex ubiquitinates Skp2 in the SCF complex.17,18 FBG1, like Skp2, is another member of the F box family of proteins.19-21 While Skp2 and many other ubiquitin ligases recognize phosphorylated substrates, the FBG family, comprised of FBG1-5, is unique in that it recognizes glycosylated proteins.22-27 In previous studies, we found that overexpressing FBG1 in CHO cells resulted in growth arrest.28 We and others also previously showed that FBG1 had a relatively low affinity for the SCF scaffold protein Cul1; in fact, rather than forming the canonical SCF complex, FBG1 primarily formed a dimer with Skp1.22,23,29-32 Finally we found that FBG1-mediated growth arrest required an intact F-box, suggesting that the growth arrest may be mediated through the FBG1-Skp1 interaction. Here, we show that in contrast to other members of the FBG family, FBG1 functionally interacts with several cullins. FBG1

www.landesbioscience.com

FBG1 interacts with different cullin proteins through different binding motifs. In our earlier work we noted that FBG1 had a low affinity for the SCF scaffold protein, Cul1.22,23,33 In fact, rather than forming the canonical SCF complex, it primarily formed a dimer with Skp1. Since Cul7 is the only other known cullin that binds F-box proteins (Fig. 1),34-36 and FBG1 bound Cul1 poorly, we tested whether FBG1 binds Cul7. As seen in Figure 2 co-expression of any FBG protein, especially FBG1 and FBG3, with Cul7 increases the steady-state level of Cul7. Unlike other FBG family members, FBG1 preferentially bound Cul7 with high affinity (Figs. 2 and 3). In contrast to FBG1, and like most F box proteins, FBGs 2–5 immunoprecipitated a significant fraction of Cul1 but not Cul7. We wanted to see if this increase in Cul7 leves was observed with other F box proteins or was it unique to FBG1. We coexpressed an irrelevant F-Box protein, Skp2 (S-phase kinaseassociated protein 2), and the Cul7 partner FBXW8 along with Cul1 and Cul7. Because these other F-box proteins were in different plasmid constructs we included a test protein, eGFP, to use as a transfection control. As seen in Figure 3 the ratio of eGFP to the loading control GAPDH, was similar for all the transfections indicating the transfection efficiency was similar. But again only FBG1 stabilized the levels of Cul7. Both Skp2 (‘S2’) and FBXW8 (‘W8’) failed to increase the steady state levels of Cul7 (Fig. 3). We were intrigued by the unexpected interaction between Cul7 and FBG1. We wanted to determine whether this interaction between Cul7 and FBG1 was physiologically relevant or merely an overexpression artifact. To do this we wanted to examine endogenous levels of Cul7 under conditions of high and low level FBG1 expression. The generation of a stable cell line expressing FBG1 was precluded by its effects on growth arrest. To circumvent this problem, we generated stable cell lines employing a system in which transgene expression is controlled by a tetracycline-responsive promoter, tet-FBG1 and, as a result, expressed when doxycycline is added to the media. We elected to use a neural-like cell line, PC6-3, because it expresses significant amounts of FBG1, indicating an adaptation to FBG1 expression.

Cell Cycle

4507

Induction of the tetracycline promoter with doxycycline results in a prompt increase in FBG1 production (Fig. 4). Paralleling the expression of FBG1, Cul7 which is virtually absent under normal conditions, dramatically increases after FBG1 induction, establishing the physiological relevance of the Cul7 and FBG1 interaction. To determine whether stabilization of Cul7 by FBG1 was affecting Cul7 activity, we measured the levels of a Cul7 substrate found in PC6-3 cells. As shown in Figure 4, p53 levels, a known Cul7 substrate,36-40 are unaffected by FBG1 induction. We also assayed Cul7 substrates, p53 and SV-40 T antigen levels36-40 in Cos-7 cells. FBG1 and Cul7 were transfected separately and together into Cos-7 cells (Fig. 5). Levels of FBG1 were not affected by co-transfection of Cul7. In contrast, as with our previous results, co-transfection of FBG1 resulted in a marked increase in Cul7 levels. In spite of the increase in Cul7 levels, the levels of the Cul 7 substrates, p53 and SV-40 T antigen, were not affected, again in agreement with the results we obtained in the PC6-3 cells (Fig. 4). Because FBG1 has a low affinity for Cul1, we examined the three amino acids in the F-box domain known to mediate this interaction (Table 1). Of the 53 F-box proteins for which some sequence data existed, only FBG1 has a proline in the third amino acid position. In other FBG proteins, this residue forms a hydrogen bond with tyrosine 139 of Cul1 (Fig. 6).10,41 A proline at this position would prevent this hydrogen bond, providing a reasonable explanation for the steep reduction in the interaction between FBG1 and Cul1 (Figs. 2 and 6). To test this possibility, we mutated either the third or the second and third amino acids in FBG1 to those found in other FBG proteins, either mutating the FBG1 F box to resemble that of FBG 2–3 (EN) or FBG 4–5 (PE). Figure 7 shows that both mutations did not modify the normal expression levels of wild-type FBG1 or its binding affinity for Cul7. However, they enhanced its interaction with Cul1. Since both mutants increased Cul1 binding without a corresponding reduction in Cul7 binding, these results suggest the FBG1 F box likely contains more than one cullin-binding motif. The D-box consensus sequence, RXX LXX XXD/N, has been shown to bind APC2, the cullin member of the APC complex.42-44 Sequence analysis shows that downstream of the F box, all FBGs contain a putative D-box domain: D-box 2 (Fig. 8). Additionally, only FBG1 contains a second D-box (D-box 1) within the F-box domain. To determine whether APC2 interacts with the D-box domain of FBG1, we co-transfected vectors expressing APC2 and FLAG-tagged versions of FBG1. Because the D-box in FBG1 was not a consensus D-box, we thought we might enhance activity and binding with APC2 by mutating the FBG1 D box to the consensus D Box found in cyclin B1 (‘BD’). Also included in the transfections was a wild-type FBG1 expression plasmid vector and a negative control in which the D-box motif was swapped for the same region found in FBG5 (‘27D’), which does not contain a D-box 1 (Table 2), and a control unrelated protein, FLAG-bacterial alkaline phosphatase (BAP). Cotransfection of any FBG with APC2 greatly enhanced the steady-state levels of APC2 compared to co-expression with BAP, similar to the results we obtained for Cul7. Anti-FLAG

4508

Figure 3. Among the F-box proteins tested only FBG1 stabilizes Cul7 levels. Cos-7 cells co-transfected with vectors expressing various F-box proteins including FBG1 (‘1’), FBG2 (‘2’), FBG5 (‘5’), FBXW8 (‘W8’), Skp2 (‘S2’), empty vector (0) or un-transfected (Cos), plus vectors expressing His-tagged Cul1 and Cul7 for 48 hours. Cell lysates and separated by SDS-PAGE, and probed with anti-HA antibody (for Cul1 and Cul7), or antibodies directed against GFP (transfection control) or GAPDH (loading control).

Figure 4. Induction of FBG1 increases Cul7 levels but does not increase degradation of p53. PC6-3 cells stably transfected with FBG1 under control of the tetracycline promoter was induced with doxycycline (‘+’) to produce FBG1 or left un-induced (‘-’). Cells were plated at time 0; doxycycline or vehicle was added at day 1, and cells were harvested on day 2. 100 μg of cell lysate were separated on SDS-PAGE gels and probed with indicated antibodies. Tubulin served as loading control.

antibodies co-immunoprecipitated significantly more APC2 from lysates containing either wild-type FBG1 or the BD mutant (Fig. 9), suggesting that the D-box1, and not D-box 2, mediates the strongest interaction with APC2. BAP immunoprecipitated no APC2, as expected. The immunoprecipitation of minor

Cell Cycle

Volume 9 Issue 22

Table 1. Cul1 binding amino acids bolded within the consensus F-box sequence Consensus

LXXLPXE

Fbxo1 (cyclin F)

ILSLPED

Fbxl1 (SKP2)

WDSLPDE

Fbw8

DIQLPYE

Fbxo45

GGRLPSR

FBG5

LSQLPPE

FBG4

LDALPPE

FBG3

INELPEN

FBG2

INELPEN

FBG1

LDELPEP

FBG1 (PE)

LDELPPE

FBG1 (EN)

LDELPEN

Only FBG1 has a P in the 3rd amino acid position, a position necessary to form a hydrogen bond with T139 in Cul1. Figure 5. FBG1 stabilizes Cul7 levels without affecting endogenous Cul7 substrates. Cos-7 cells co-transfected with vectors expressing FBG1 or Cul 7 as indicated. After 48 hours cells were harvested, and cell lysates separated by SDS-PAGE, and probed with anti-HA antibody (for Cul7), or antibodies directed against p53, SV-40 T antigen (‘T-Ag’), FLAG (for FBG1) or tubulin (loading control).

amount of APC2 with the 27D mutant probably represent binding to D-box-2. FBG1 expression influences cell cycle progression through interactions with APC2. In an earlier study, we showed that FBG1 overexpression caused a growth arrest in CHO cells.28 To determine whether this effect was a consequence of APC2 binding via its D-box, we overexpressed wild-type FBG1 and the BD and 27D mutants (Fig. 10) in Cos7 cells, assaying for growth arrest by cell counts 24–72 hours later. While overexpressing both wild-type and the BD mutant caused profound growth arrest, the 27D mutant had no effect (Fig. 10). Moreover, mutating the D-box 1 to match the consensus sequence had no further effect on growth arrest, suggesting that the non-consensus D-box found in FBG1 is sufficient for APC2 binding. Mutating D-box 1 to an unrelated sequence completely blocked growth arrest, suggesting that the APC2 interaction and growth arrest are both mediated by this D-box. Furthermore, cells transfected with vectors expressing either wild-type FBG1 or BD contained a larger fraction of cells with giant nuclei, compared to those transfected with vectors expressing the 27D mutant or the BAP control (data not shown), suggesting cell cycle arrest in S phase. We were surprised to find that co-expression of APC2 plus FBG1 also stabilizes APC2 (Fig. 9), suggesting that APC2 and FBG1 form an inactive but stable complex which does not result in the ubiquitin-mediated degradation of either protein. If this is true, the steady-state level of free, unbound APC2 should decrease. To test this hypothesis, we monitored the growth of cells transfected with FBG1 and increasing amounts of APC2, including an empty vector to control for transfection. As shown in Figure 11, transfection with equal amounts of all three vectors results in growth inhibition (‘222’), while doubling the amount


of APC2 mitigates this effect in a dose-dependent manner (‘240’) until the growth inhibition is eliminated using a 3:1 ratio of APC2 to FBG1 (‘260’). This provides strong evidence that the FBG1 D-box regulated growth inhibition is directly mediated by APC2 levels. To establish the biological significance of this effect, we compared the growth arrest induced by FBG1 to that observed with FBXL1 (‘SKP2’), another F-box protein known to control cell cycle.18,45-47 Transfecting vectors expressing FBG1 and SKP2 affected proliferation similarly when compared to the negative control (Fig. 12). Thus, under our conditions, the effect FBG1 on cell cycle control is as significant as SKP2. We also examined the growth characteristics of a PC6-3 line stably transfected with tet-FBG1. Cells were plated at day 0, and transgene expression was induced with doxycycline or no-doxycycline control media at day 1. As shown in Figure 13, one day after induction (day 2), the doxycycline-treated cells exhibited profound growth arrest compared to the vehicle-treated cells. At day 3, this affect was even more pronounced. To determine if the growth arrest caused by FBG1 was due to cell cycle inhibition, we performed fluorescence-activated cell sorting (FACS) of Cos-7 cells transiently transfected with wildtype FBG1 or with the 27D mutant. At day 3 post infection, a representative 2-D plot of cells expressing FBG1 (Fig. 14A) or 27D (Fig. 14B) demonstrates that overexpression of wildtype FBG1 but not the 27D mutant arrests cells in S-phase and decreases the population of cells in G0 /G1 and G2-M (Fig. 14). To determine whether FBG1 was interfering with the APC/C complex we preformed a co-transfection of APC2 with wildtype FBG1 or with the 27D mutant or the negative control BAP. As shown in input lanes in Figure 15 cotransfection of either FBG1 or 27D greatly enhanced the APC2 levels compared to BAP, similar to our previous result (Fig. 9) and similar to the results we obtained with Cul7. The inputs also revealed that there was a slight decrease in Cdh1 levels in both the FBG1 and 27D lanes compared to BAP. Cdc20 levels were also examined, but no reliable change was observed (data not shown). When proteins

Cell Cycle

4509

co-precipitating with APC2 were examined, we found that as expected no BAP was pulled down and that both FBG1 and 27D immunoprecipitated with APC2. Surprisingly, though the input lanes indicated equal amounts of transfected FBG1 and 27D, the APC2 immunoprecipitation pulled down more 27D than FBG1. While there were differences in Cdh1 in the inputs, we consistently saw equal amounts of Cdh1 immunoprecipitated with APC2. Consistent with the absence in a change in the APC2Chd1 complex, we also found no changes in levels of substrate proteins, cyclin B1 or Cdc6 (data not shown). Discussion

Figure 6. The third amino acid in the PXE sequence is critical to stabilize the Cul1- F box interaction. The structure of the Cul1-Rbx1-Skp1-FboxSkp2 SCF ubiquitin ligase complex (PDB code 1LDK) is shown for reference (ref) (A). The interface between the Cul1 and FboxSkp2 is expanded in (B) and residues critical for protein-protein interactions are highlighted. Residues 3,113–3,115 (PDE) within helix 1 of the FboxSkp2 domain were mutated in silico to PEP, the sequence found in FBG1 F-box (C). Mutation of E3115 to P predicts that the side-chain hydrogen bond between E3115 and Y139 will be disrupted.

In this work, we demonstrate that FBG1, through the presence of non-canonical motifs and interactions, influences the cell cycle, thus potentially playing an important role in biological processes such as cancer and senescence. First, we demonstrate that the presence of specific amino acid residues in its F box domain expands the cullin-binding repertoire of FBG1 when compared to other F box proteins. Second, FBG1 binds and stabilizes APC2, the cullin member of the APC complex, sequestering it and decreasing its unbound levels. Third, expression of wild type FBG1, but not of mutant forms in which APC2 binding is abolished, causes growth arrest. Finally, the growth arrest caused by FBG1 is due to cell cycle inhibition. Thus, our studies demonstrate that FBG1 is a unique F box protein that has evolved into a putatively important player in mammalian cell cycle control. FBG1 appears to be unique among the over 500 different E3 ubiquitin ligases in that, while most E3s recognize a small number of phosphorylated substrates, FBG1 binds all misfolded glycoprotein thus far tested. Thus, the potential list of FBG1 substrates numbers in the thousands. Here, we also show that FBG1 is promiscuous in its choice of adapter proteins; in addition, FBG1 is found in both the nucleus and cytoplasm, further broadening the potential targets. Those E3 ligases that use adaptor proteins generally bind only one. For F-box ubiquitin ligases, X-ray crystallography has revealed that three amino acids in the F-box domain, PXE, are important for binding the adaptor protein Cul1 to form the SCF (Skp1/Cul/F box) complex. The proline lies at the beginning of the first alpha helix in the F-box. This is significant because prolines are poor helix-forming amino acids since they restrict the backbone dihedral angle, introduce steric hindrance, and do not form alpha-helix backbone side chains; thus, prolines tend be found at the ends of helices. P113 in SkpKP2 (an F-box protein) forms a hydrophobic interaction with T54 in the second α-helix of Cul1 (H2). The third amino acid, glutamic acid (E), interacts

4510

Figure 7. Mutating the FBG1 F box to contain an E or N in the 3rd amino acid position enhances Cul1 binding. Cos-7 cells were cotransfected with FLAG tagged FBG1 or mutants and HA tagged Cul1 and Cul7 for 48 hours. After cell lysis the proteins were immunoprecipitated with anti-FLAG agarose beads, and separated by SDS-PAGE and probed with anti-HA antibody.

with two amino acids in helix 5 of Cul1 forming a hydrogen bond with tyrosine 139, (Y139) and a salt bridge with histidine 143 (H143). In FBG1, this critical glutamic acid is replaced by another proline, potentially interfering with cullin binding on several levels and providing the structural basis for its unique interactions. First, there is no longer the ability to form a hydrogen bond with Y139 (a loss of ½ kcal/mol) and second, the salt bridge with H143 is lost (subtracting another 1–3 kcal/mol). These energy losses probably greatly reduce the stability of the Cul1-FBG1 interaction. Furthermore, the proline substitution probably interferes with the alpha helix formation and greatly reduces the chance of a Cul1-FBG1 interaction. Any Cul1 binding likely comes from a tertiary FBG1-Skp1-Cul1 interaction. Therefore, mutating this proline in FBG1 to a glutamic acid or

Cell Cycle

Volume 9 Issue 22

Figure 8. CLUSTAL W (1.8) multiple sequence alignment of the FBG family members reveals two D-box domains in FBG1. The regions for the D boxes are highlighted in yellow. Three amino acids critical for Cul1 binding are also shown in green.

an asparagine, both of which can form hydrogen bonds, restores Cul1 binding. The FBG1 gene and the proline in the 3rd amino acid position is conserved in human, dog, cow, mouse, rat and chicken suggesting that this reduce binding affinity has been conserved in evolution. There are few examples of F-box proteins that bind proteins other than Cul1. In the first instance, FBW8 binds very little Cul1 and instead binds Cul7 with much higher affinity. The FBW8 F-box motif, which is known to bind Cul1, contains a glutamic acid at the third position, but the Cul7 binding site is undefined.35,36 In a second example, FBXO45 was shown to lack an amino acid critical for cullin binding: the third residue in the F-box PXE motif is replaced with an R. This abrogates cullin binding by the F-box; and FBXO45 instead binds another RING


type ligase, called PAM (Protein Associated with Myc) through an SPRY domain.41 Skp2 is also known to bind APC2 in addition to Cul1; binding Cul1 via an F-box region and APC2 via a different motif. Here, we show an unexpected promiscuity for FBG1, as it binds at least three different adapters: Cul1 (with low affinity), Cul7 and APC2. In addition, we demonstrate the first known occurrence of a D-box within an F-box domain. Furthermore, we found that the affinity of FBG1 for Cul1 can be increased without affecting Cul7 binding, suggesting the presence of distinct binding domains. Hence, the FBG1 F-box domain is tripartite, containing three potential Cull sub-domains, allowing this unique E3 ligase to bind multiple adapter proteins and underscoring its potential unique biological relevance among the F box ubiquitin ligases.

Cell Cycle

4511

Counterintuitively, we found that co-expression of FBG1 with either Cul7 or APC2 stabilized the adaptor protein, rather that triggering its degradation. The stabilization of the adaptor protein was probably via forming an inactive complex between the adaptor and FBG1. As far as we could determine this inactive complex had no effect on other adaptors partners or substrate proteins, or ability to form an active complex with other proteins. Hence, this inactive complex was likely a method on intracellular storage. The stabilization of both an F box protein and adaptor has also been reported for the Fbxo45 PAM complex.41 We and other have reported a similar phenomena with a stable FBG1-Skp1 dimer identified in the mouse cochlea,30-33,48 where sequestration by FBG1 likely reduces the level of free Skp1.33 In the FBG1 knockout mouse, total Skp1 levels are reduced in this tissue, likely to maintain constant levels of free Skp1. By analogy, FBG1 may sequester either Cul7 or APC2 into an inactive complex, allowing the adaptor protein to be present to rapidly bind other available E3 ligase subunits when required. This would allow the cell to rapidly adapt to changing levels of E3 ligases, a response that may be crucial for cell cycle progression. By forming an inactive complex with FBG1, the activity of Cul7 and APC2 on target proteins may be finely controlled at the post-translational level. Consistent with this, our examination of substrates for both Cul7 and APC2 revealed no large difference in substrate protein levels. The reduction in Cdh1 levels observed after cotransfection of APC2 with FBG1 suggests that an APC2Cdh1 complex increases Cdh1 levels. Transfection of FBG1 binds APC2, reduces free APC2 levels and reducing the formation of APC2-Cdh1 complexes, and hence lowers Chd1 levels. Proteins with strong D boxes are rapidly polyubiquitinated in a highly processive manner, and these proteins would be fairly insensitive to levels of APC2.49 Conversely, proteins with a weak D box are ubiquitinated in a more distributive manner and are very sensitive to APC2 expression. Hence, by tightly regulating the availability of “free” APC2, FBG1 would modulate the degradation of weak D-box containing proteins, helping set the order of ubiquitination of APC2 substrates. As a result, proteins with strong D-boxes would be degraded early and quickly because they are processively ubiquitinated, but the degradation of those containing weak D-boxes would be delayed, as they must undergo multiple rounds of distributive ubiquitination. This early and late timing of protein degradation would be reinforced by sequestering APC2 into an inactive complex with FBG1, as it will require time to remove all the APC2 from the inactive complex, re-enforcing the late degradation of proteins with weak D-boxes. The altered ratio of APC2 to Cdh1 that we observed may also alter the timing of protein degradation. Additionally FBG1 sequestering of APC2 may result in the transfer of APC2 from different cellular pools which determine the order of destruction of various APC2 substrates.47 Thus, FBG1 would add another level of complexity to the regulation of APC2 mediated protein degradation. An alternative view is that APC2 binds and stabilizes free FBG1 for later use. The ER and nuclear envelope dissolve by the end of prophase, releasing many ER and nuclear envelope proteins to the cytoplasm. A significant fraction of these proteins are probably misfolded and need to be degraded. FBG1 is known

4512

Table 2. D-box domains in APC/C substrates and FBG1 with FBG1 mutants Consensus

RXXLXXXXD/N

Securin

KATRKALGTVNRA

CyclinA2

PRTRAALAVLKSG

Aurora-A

PSQRPMLREVLEH

Skp2

IVRRPKLNRENFP

Geminin

SVPRRTLKMIQPS

PLK1

PSNRKPLTVLNKG

UbcH10

YDVRTILLSIQSL

p21

RRGRDELGGGRRP

Nek2A

LKERKFLSLASNP

TACC3

LGERSGLKPPLRK

Cyclin B1

LRPRTALGDIGNK

FBG1 D-box 2(122-131)

KRRRNLLRNPCGE

FBG1 D-box 1(72-80)

QACRLVCLRWKEL

FBG1 (BD)

QACRLVLLRIKNL

FBG1 (27D)

QACRLVCRRWRAL

Red color indicates position of critical residue and bold indicates critical amino acid identity. FBG! has a non-consensus D-box as the second critical residue, L, is one amino acid C-terminal to the consensus location.

Figure 9. Only FBG1 wild type and BD mutant co-immunoprecipitates with APC2. Cos-7 cells were co transfected with FLAG tagged FBG1 or mutant, or BAP as a negative control. After 48 hours, the proteins were immunoprecipitated with anti-FLAG agarose beads, and separated by SDS-PAGE. The blot was probed with an antibody to APC2, FLAG or tubulin as indicated. Input lanes represent 10% of cell lysate used for IP.

to bind misfolded glycoproteins. At this point of the cell cycle, APC2 forms an E3 ligase complex with CDC20 and other proteins to degrade mitotic substrates. Hence, FBG1 levels would increase, as they are no longer sequestered by APC2, allowing FBG1 to clear the old and newly formed cells from the influx of misfolded glycoproteins. We previously hypothesized that FBG1-mediated growth arrest was via the p21 pathway or due to Skp1 sequestration. However, the robust interaction found between APC2 and FBG1 suggested that the growth arrest is mediated directly by APC2. Direct interactions with APC/C complexes without the need for adaptor proteins such have Cdc20 have previously been reported.50,51 Mutation of D-box 1 in FBG1 to a nonfunctional D box, 27D, had some residual binding to APC2, probably due

Cell Cycle

Volume 9 Issue 22

Figure 10. FBG1 and FBG1 (BD) cause growth arrest in Cos-7 cell lines. Cos-7 cells were transfected with FBG1, mutant plasmids or control plasmid (BAP). Cells were harvested and counted beginning one day post transfection.

Figure 11. FBG1-mediated growth arrest is mitigated by APC2 overexpression. Cos7 cells transfected with the indicated constructs where numbers indicated micrograms of DNA used, first number is FBG1, second number is APC2, and last number represents vector DNA. Cells were counted beginning one day after transfection.

the second downstream D-box 2. However, this mutant 27D, failed to show any growth arrest. This is quite similar to the result of mutating the Skp2 D box, where loss of growth inhibition occurred despite some residual binding to Cdh 1.18 The loss of growth arrest we detected upon APC2 overexpression also strongly supports a direct interaction with APC2.


There are multiple pieces of experimental evidence that, together, suggest an important role for FBG1 on cell cycle control and cancer. First, high levels of FBG1 are constitutively expressed in the brain, which is post mitotic, and the pancreas and testes which contain a mixture of mitotic and post mitotic areas.52 FBG1 is also expressed in other organs but at tenfold

Cell Cycle

4513

Figure 12. FBG1 and Skp2 produce same level of growth inhibition. Cos-7 cells were transfected with FBG1 or Skp2. Cells were harvested and counted beginning one day post transfection.

Figure 13. PC6-3 cells stably transfected with FBG1 show growth arrest upon FBG1 induction. A PC6-3 clone stably transfected with a full length FBG1 plasmid under control of the tetracycline promoter was induced with doxycycline (squares) or untreated (diamonds). Cells were plated at time 0; doxycycline or vehicle was added at day 1.

lower levels. Second, in contrast to other FBG family members, FBG1 is not expressed during embryonic development and its expression increases in the brain during the first ten months of life. This is consistent with FBG1 causing an S-phase cell cycle arrest to maintain the developed brain in a post-mitotic state. Third, FBG1 is expressed in embryonic stem cells and embryonic

4514

cancer cells.53,54 Examination of GEO expression profiles55 accession GSE3189 revealed that FBG1 is downregulated in melanomas. Fourth, this ubiquitin ligase degrades the adhesion receptor SHPS-1, which may be involved in tumor metastasis in melanomas and other cancers.56-58 Finally, we show here that FBG1 interacts with APC2 and Cul7, both of which play important roles in

Cell Cycle

Volume 9 Issue 22

in 6-well dishes. Beginning 24 hours post transfection, cells were harvested, incubated with 1% Trypan blue and counted in a hemocytometer or using the automated Countess (Invitrogen) cell counter (both methods gave equivalent results). For immunoprecipitation experiments, Cos-7 cells were transfected as above and harvested at 72 hours post-transfection in RIPA buffer with protease inhibitors (Complete, Roche). PC6-3 inducible clonal cell lines cells were generated as previously described in reference 59. 6-His ΔFbox FBG1 was made using a 5'primer encoding a KpnI site and 6 His sequence that annealed to nucleotide 243 in the rat FBG1 sequence of Figure 14. FBG1 causes cell cycle arrest at S phase with reduced populations in G0/G1 and HA-pcDNA3 Fbx2,60 and a 3' SP6 primer. After G2-M. Cos-7 cells transfected with FBG1(WT) (A) or FBG1(27D) (B). 3 days post transfection subcloning, conformation of correct sequence in the cells were harvested, stained with DAPI, and sorted by FACS. Numbers below the a shuttle vector, the PCR product removed with figures represent the fraction of cells in each phase of the cell cycle. a Hind III/EcoRV digestion, gel purified and was ligated into the Hind III and EcorV sites in pcDNA5/TO. After transfection, PC6-3 stable lines were selected in 150 μg/mL Hygromycin for 3 weeks. Selected clones were cultured at 37°C, 5% CO2 in RPMI1640 with 5% FCS and 10% Horse Serum (HyClone, Logan UT), 1% P/S, 2 μg/mL Blasticidin and 100 μg/ml Hygromycin (Roche) on collagencoated plates (BD Biosciences), as previously described. Cells were induced with doxycycline 1.5 μg/ml (Sigma) for 24 hours and then counted as described for Cos-7 cells. Figure 15. FBG1 increase the ratio of APC2 to Chd1, but does not affect APC2-CDh1 complex formaPlasmid constructs. Using tion. Cos-7 cells co-transfected with vectors expressing APC2 and wild type FBG1 (‘FBG1’), mutant FBG HA-pcDNA3 Fbx2,60 plasmid as the plasmids or control plasmid (BAP) for 48 hours. 200 υg of cell lysates were used for immunoprecipatatemplate DNA, 6His ΔFbox FBG1 tion using anti-APC2 antibody. Cell lysates and co-immunoprecipitating proteins were and separated by SDS-PAGE, and probed with indicated antibodies. was created using a unique forward primer and a SP6 reverse primer. cellular transformation. For all of these reasons, we speculate that Forward: 5'-AGG CGG TAC CAC CAT GGG TCG GCA FBG1 may play a significant role in cell cycle control and perhaps TCA TCA TCA TCA TCA TGG TCC TGG TTG TCA GCA cancer development. GGA GGG GCT GGT GCC T-3'. Using pCMV-FLAG-FBG1 plasmid as the template DNA, Materials and Methods FBG1 (BD) was created using primers Forward: 5'-GCC TGC CGC CTG GTG CTC CTG CGC Cell culture, transfection and cell counts. All reagents were from ATC AAG AAC CTG GTG GAC GGC GCC-3', Invitrogen (Carlsbad, CA) unless otherwise specified. Cos-7 cells Reverse: 5'-GGC GCC GTC CAC CAG GTT CTT GAT were cultured at 37°C, 5% CO2 in DMEM with 10% Fetal Calf GCG CAG GAG CAC CAG GCG GCA GGC-3', and FBG1 Serum (FCS, HyClone, Logan UT) and 1% penicillin (10,000 (27D) using primers Forward: 5'-TGC CGC CTG GTG TGC CGG CGC TGG units/ml)/streptomycin (10,000 μg/ml) (P/S) as previously described. Cells were grown on 10 cm2 plates to 80% confluent, AGG GCG CTG GTG GAC GGC GCC-3', Reverse: 5'-GGC GCC GYC CAC CAG CGC CCY CCA and transfected with 6 μg DNA using Lipofectamine-Plus, as previously described. For cell counting experiments Cos-7 cells GCG CCG GCA CAC CAG GCG GCA-3' were transfected with 6 μg pCMV-FLAG-BAP (Sigma,) or 6 μg with the QuikChange protocol (Stratagene). PCR was performed pCMV-FLAG-FBG1 constructs and 0.1 μg of eGFP as a trans- for 18 cycles with a denaturing temperature of 95°C, an annealfection control (Clonetech). At 24 hours post transfection each ing temperature of 63°C, and an extension temperature of 72°C. 10 cm2 plate of cells was trypsinized and re-plated into 9 wells The sequences of all cloned inserts were verified.


Cell Cycle

4515

Immunoprecipitations. Two days after transfection the media from a 10 cm plate was aspirated, and the cells washed in Hanks Balanced Salt Solution. Then 1.0 mL of RIPA with protease inhibitor (Complete, Roche) was added and the cells incubated on ice for 30 minutes, the cells were then scrapped off. Cells and buffer were centrifuged at 16,000x g for 30 minutes. Automated immmunoprecipiations (IPs) were performed with the MAGic Sample Processor (Invitrogen) using 500 ul of cell lysate and 4 uL rabbit polyclonal antibody to APC2 (#550362 BD Pharmingen) using protein A Dynabeads, with a sample and antibody binding time of one hour at room temperature, followed by a 40 μL elution. Manual IPs were performed using 300 μL aliquot of cell lysate incubated with 40 μL of antibody conjugated beads (EZview Red ANTI-FLAG M2 Affinity Gel, Sigma), with endend rotation for 1 hour at 4°C. The beads were then centrifuged at 8,200 x g for 10 minutes washed three times, and eluted with 40 μL of 3x FLAG peptide 150 ng/μL (Sigma). Westerns. Cell lysates and immunoprecipitations were separated by 10% SDS-PAGE, transferred to PVDF membranes in 9.5 min at 25 volts using the iBlot gel transfer device (Invitrogen). Blots were washed and probed in the Bench Pro 41000 card processor using (Invitrogen) rabbit polyclonal antibodies recognizing APC2 (Abcam; diluted 1:200), Cdh1 (Abcam, diluted 1:100), p53 (Santa Cruz, Sc6243, diluted 1:200), mouse monoclonal antibodies recognizing HA (Zymed; diluted 1:300), mouse References 1. 2.

3.

4.

5. 6.

7.

8.

9.

10.

11.

12.

Ciechanover A. The ubiquitin-proteasome pathway: on protein death and cell life. EMBO J 1998; 17:7151-60. Ciechanover A, Schwartz AL. The ubiquitin-proteasome pathway: the complexity and myriad functions of proteins death. Proc Natl Acad Sci USA 1998; 95:2727-30. Scheffner M, Nuber U, Huibregtse JM. Protein ubiquitination involving an E1-E2-E3 enzyme ubiquitin thioester cascade. Nature 1995; 373:81-3. Glickman MH, Ciechanover A. The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiol Rev 2002; 82:373-428. Semple CA. The comparative proteomics of ubiquitination in mouse. Genome Res 2003; 13:1389-94. Bays NW, Gardner RG, Seelig LP, Joazeiro CA, Hampton RY. Hrd1p/Der3p is a membrane-anchored ubiquitin ligase required for ER-associated degradation. Nat Cell Biol 2001; 3:24-9. Jackson PK, Eldridge AG, Freed E, Furstenthal L, Hsu JY, Kaiser BK, et al. The lore of the RINGs: substrate recognition and catalysis by ubiquitin ligases. Trends in cell biology 2000; 10:429-39. Craig KL, Tyers M. The F-box: a new motif for ubiquitin dependent proteolysis in cell cycle regulation and signal transduction. Progress in Biophysics & Molecular Biology 1999; 72:299-328. Patton EE, Willems AR, Tyers M. Combinatorial control in ubiquitin-dependent proteolysis: don’t Skp the F-box hypothesis. Trends in Genetics 1998; 14:236-43. Zheng N, Schulman BA, Song L, Miller JJ, Jeffrey PD, Wang P, et al. Structure of the Cul1-Rbx1-Skp1-F boxSkp2 SCF ubiquitin ligase complex. Nature 2002; 416:703-9. Bai C, Sen P, Hofmann K, Ma L, Goebl M, Harper JW, et al. SKP1 connects cell cycle regulators to the ubiquitin proteolysis machinery through a novel motif, the F-box. Cell 1996; 86:263-74. Vodermaier HC. APC/C and SCF: controlling each other and the cell cycle. Curr Biol 2004; 14:787-96.

4516

monoclonal antibodies recognizing FLAG (Sigma; diluted 1:1,000), mouse monoclonal antibodies against SV-40 T antigen (Santa Cruz,Sc-148, diluted 1:200) or mouse monoclonal antibodies recognizing tubulin (Sigma; 1:1,000). Blots were probed with peroxidase conjugated goat anti-rabbit or anti-mouse secondary antibodies diluted 1:5,000. Bands were visualized with SuperSignal West Femto or Pico (Pierce, Rockford Il) enhanced chemiluminescent assay. Molecular modeling. Figure 3 was produced using the molecular graphics program PyMOL (DeLano Scientific, LLC). The coordinates from the Cul1-Rbx1-Skp1-FboxSkp2 SCF ubiquitin ligase structure (PDB code 1LDK) 61 were used as a template for modeling. Amino acid residues 3,113–3,115 (PDE) within helix 1 of the FboxSkp2 domain were mutated in silico to the corresponding amino acids PEP found in FBG1 F-box. Flow cytometry. While still attached on the plates, cells were washed twice with 10 mM EDTA in PBS and then incubated at 37°C for 5 min. Cells were gently scraped off and re-suspended in PBS, 10 mM EDTA and centrifuged at 500 g for 5 minutes. Cell pellets were re-suspended in PBS, 10 mM EDTA, 10 μg/ ml DAPI with 0.6% NP40 (Nonidet P40). A single-cell suspension was made by filtering through a 70 μM cell strainer (BD Falcon MA) and sorted on a Becton Dickinson LSR II flow cytometer with a 355 nm laser (http://www.healthcare.uiowa. edu/CoreFacilities/FlowCytometry).

13. Vodermaier HC, Peters JM. APC activators caught by their tails? Cell Cycle 2004; 3:265-6. 14. Reed SI. Ratchets and clocks: the cell cycle, ubiquitylation and protein turnover. Nat Rev Mol Cell Biol 2003; 4:855-64. 15. Guardavaccaro D, Kudo Y, Boulaire J, Barchi M, Busino L, Donzelli M, et al. Control of Meiotic and Mitotic Progression by the F Box Protein [beta]-Trcp1 In Vivo. Developmental Cell 2003; 4:799-812. 16. Margottin-Goguet F, Hsu JY, Loktev A, Hsieh HM, Reimann JD, Jackson PK. Prophase destruction of Emi1 by the SCF(betaTrCP/Slimb) ubiquitin ligase activates the anaphase promoting complex to allow progression beyond prometaphase. Dev Cell 2003; 4:813-26. 17. Bashir T, Dorrello NV, Amador V, Guardavaccaro D, Pagano M. Control of the SCF(Skp2-Cks1) ubiquitin ligase by the APC/C(Cdh1) ubiquitin ligase. Nature 2004; 428:190-3. 18. Wei W, Ayad NG, Wan Y, Zhang GJ, Kirschner MW, Kaelin WG Jr. Degradation of the SCF component Skp2 in cell cycle phase G1 by the anaphase-promoting complex. Nature 2004; 428:194-8. 19. Cenciarelli C, Chiaur DS, Guardavaccaro D, Parks W, Vidal M, Pagano M. Identification of a family of human F-box proteins. Current Biology 1999; 9:1177-9. 20. Ilyin GP, Serandour AL, Pigeon C, Rialland M, Glaise D, Guguen-Guillouzo C. A new subfamily of structurally related human F-box proteins. Gene 2002; 296:11-20. 21. Winston JT, Koepp DM, Zhu C, Elledge SJ, Harper JW. A family of mammalian F-box proteins. Current Biology 1999; 9:1180-2. 22. Glenn KA, Nelson RF, Wen HM, Mallinger AJ, Paulson HL. Diversity in tissue expression, substrate binding and SCF complex formation for a lectin family of ubiquitin ligases. J Biol Chem 2008; 283:12717-29. 23. Nelson RF, Glenn KA, Miller VM, Wen H, Paulson HL. A Novel Route for F-box Protein-mediated Ubiquitination Links CHIP to Glycoprotein Quality Control. J Biol Chem 2006; 281:20242-51.

Cell Cycle

24. Yoshida Y. A novel role for N-glycans in the ERAD system. J Biochem (Tokyo) 2003; 134:183-90. 25. Yoshida Y, Adachi E, Fukiya K, Iwai K, Tanaka K. Glycoprotein-specific ubiquitin ligases recognize N-glycans in unfolded substrates. EMBO Rep 2005; 6:239-44. 26. Yoshida Y, Chiba T, Tokunaga F, Kawasaki H, Iwai K, Suzuki T, et al. E3 ubiquitin ligase that recognizes sugar chains. Nature 2002; 418:438-42. 27. Yoshida Y, Tokunaga F, Chiba T, Iwai K, Tanaka K, Tai T. Fbs2 is a new member of the E3 ubiquitin ligase family that recognizes sugar chains. J Biol Chem 2003; 278:43877-84. 28. Erhardt JA, Hynicka W, DiBenedetto A, Shen N, Stone N, Paulson H, et al. A novel F box protein, NFB42, is highly enriched in neurons and induces growth arrest. J Biol Chem 1998; 273:35222-7. 29. Nelson RF, Glenn KA, Zhang Y, Wen H, Knutson T, Gouvion CM, et al. Selective Cochlear Degeneration in Mice Lacking the F-Box Protein, Fbx2, a GlycoproteinSpecific Ubiquitin Ligase Subunit. J Neurosci 2007; 5163-71. 30. Tan A, Tanner JJ, Henzl MT. Energetics of OCP1OCP2 complex formation. Biophys Chem 2008; 134:64-71. 31. Thalmann R, Henzl MT, Killick R, Ignatova EG, Thalmann I. Toward an understanding of cochlear homeostasis: the impact of location and the role of OCP1 and OCP2. Acta Otolaryngol 2003; 123:203-8. 32. Henzl MT, O’Neal J, Killick R, Thalmann I, Thalmann R. OCP1, an F-box protein, co-localizes with OCP2/ SKP1 in the cochlear epithelial gap junction region. Hear Res 2001; 157:100-11. 33. Nelson RF, Glenn KA, Zhang Y, Wen H, Knutson T, Gouvion CM, et al. Selective Cochlear Degeneration in Mice Lacking the F-Box Protein, Fbx2, a Glycoprotein-Specific Ubiquitin Ligase Subunit 10.1523/JNEUROSCI.0206-07.2007. J Neurosci 2007; 27:5163-71.

Volume 9 Issue 22

34. Yamoah K, Oashi T, Sarikas A, Gazdoiu S, Osman R, Pan ZQ. Autoinhibitory regulation of SCF-mediated ubiquitination by human cullin 1’s C-terminal tail. Proc Natl Acad Sci USA 2008; 105:12230-5. 35. Xu X, Sarikas A, Dias-Santagata DC, Dolios G, Lafontant PJ, Tsai SC, et al. The CUL7 E3 ubiquitin ligase targets insulin receptor substrate 1 for ubiquitindependent degradation. Mol Cell 2008; 30:403-14. 36. Sarikas A, Xu X, Field LJ, Pan ZQ. The cullin7 E3 ubiquitin ligase: a novel player in growth control. Cell Cycle 2008; 7:3154-61. 37. Jung P, Verdoodt B, Bailey A, Yates JR, III, Menssen A, Hermeking H. Induction of Cullin 7 by DNA damage attenuates p53 function 10.1073/pnas.0609467104. PNAS 2007:0609467104. 38. Kasper JS, Kuwabara H, Arai T, Ali SH, DeCaprio JA. Simian virus 40 large T antigen’s association with the CUL7 SCF complex contributes to cellular transformation. J Virol 2005; 79:11685-92. 39. Kasper JS, Arai T, DeCaprio JA. A novel p53-binding domain in CUL7. Biochem Biophys Res Commun 2006; 348:132-8. 40. Dowell JD, Tsai SC, Dias-Santagata DC, Nakajima H, Wang Z, Zhu W, et al. Expression of a mutant p193/ CUL7 molecule confers resistance to MG132- and etoposide-induced apoptosis independent of p53 or Parc binding. Biochim Biophys Acta 2007; 1773:358-66. 41. Saiga T, Fukuda T, Matsumoto M, Tada H, Okano HJ, Okano H, et al. Fbxo45 Forms a Novel Ubiquitin Ligase Complex and Is Required for Neuronal Development. Mol Cell Biol 2009; 29:3529-43. 42. Glotzer M, Murray AW, Kirschner MW. Cyclin is degraded by the ubiquitin pathway. Nature 1991; 349:132-8. 43. Burton JL, Tsakraklides V, Solomon MJ. Assembly of an APC-Cdh1-substrate complex is stimulated by engagement of a destruction box. Mol Cell 2005; 18:533-42.


44. Kraft C, Vodermaier HC, Maurer-Stroh S, Eisenhaber F, Peters JM. The WD40 propeller domain of Cdh1 functions as a destruction box receptor for APC/C substrates. Mol Cell 2005; 18:543-53. 45. Kamura T, Hara T, Kotoshiba S, Yada M, Ishida N, Imaki H, et al. Degradation of p57(Kip2) mediated by SCFSkp2-dependent ubiquitylation. Proceedings of the National Academy of Sciences of the United States of America 2003; 100:10231-6. 46. Nakayama K, Nagahama H, Minamishima YA, Matsumoto M, Nakamichi I, Kitagawa K, et al. Targeted disruption of Skp2 results in accumulation of cyclin E and p27Kip1, polyploidy and centrosome overduplication. EMBO J 2000; 19:2069-81. 47. van Leuken R, Clijsters L, Wolthuis R. To cell cycle, swing the APC/C. Biochimica et Biophysica Acta (BBA)—Reviews on Cancer 2008; 1786:49-59. 48. Henzl MT, Thalmann I, Thalmann R. OCP2 exists as a dimer in the organ of Corti. Hear Res 1998; 126:37-46. 49. Rape M, Reddy SK, Kirschner MW. The processivity of multiubiquitination by the APC determines the order of substrate degradation. Cell 2006; 124:89-103. 50. Eytan E, Moshe Y, Braunstein I, Hershko A. Roles of the anaphase-promoting complex/cyclosome and of its activator Cdc20 in functional substrate binding. Proc Natl Acad Sci USA 2006; 103:2081-6. 51. Yamano H, Gannon J, Mahbubani H, Hunt T. Cell cycle-regulated recognition of the destruction box of cyclin B by the APC/C in Xenopus egg extracts. Mol Cell 2004; 13:137-47. 52. Peng Sw, Zhu Ly, Chen M, Zhang M, Li Dz, Fu Yc, et al. Heterogeneity in Mitotic Activity and Telomere Length Implies an Important Role of Young Islets in the Maintenance of Islet Mass in the Adult Pancreas. Endocrinology 2009; 150:3058-66. 53. Lu Y, Yi Y, Liu P, Wen W, James M, Wang D, et al. Common Human Cancer Genes Discovered by Integrated Gene-Expression Analysis. PLoS One 2007; 2:1149.

Cell Cycle

54. Sperger JM, Chen X, Draper JS, Antosiewicz JE, Chon CH, Jones SB, et al. Gene expression patterns in human embryonic stem cells and human pluripotent germ cell tumors. Proceedings of the National Academy of Sciences of the United States of America 2003; 100:13350-5. 55. Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucl Acids Res 2002; 30:207-10. 56. Murai-Takebe R, Noguchi T, Ogura T, Mikami T, Yanagi K, Inagaki K, et al. Ubiquitination-mediated regulation of biosynthesis of the adhesion receptor SHPS-1 in response to endoplasmic reticulum stress. J Biol Chem 2004; 279:11616-25. 57. Ogura T, Noguchi T, Murai-Takebe R, Hosooka T, Honma N, Kasuga M. Resistance of B16 melanoma cells to CD47-induced negative regulation of motility as a result of aberrant N-glycosylation of SHPS-1. J Biol Chem 2004; 279:13711-20. 58. Oshima K, Machida K, Ichigotani Y, Nimura Y, Shirafuji N, Hamaguchi M, et al. SHPS-1: a budding molecule against cancer dissemination. Cancer Res 2002; 62:3929-33. 59. Gonzalez-Alegre P, Paulson HL. Aberrant cellular behavior of mutant torsinA implicates nuclear envelope dysfunction in DYT1 dystonia. J Neurosci 2004; 24:2593-601. 60. Erhardt JA, Hynicka W, DiBenedetto A, Shen N, Stone N, Paulson H, et al. A Novel F Box Protein, NFB42, Is Highly Enriched in Neurons and Induces Growth Arrest. J Biol Chem 1998; 273:35222-7. 61. Zheng N, Schulman BA, Song L, Miller JJ, Jeffrey PD, Wang P, et al. Structure of the Cul1-Rbx1-Skp1-F boxSkp2 SCF ubiquitin ligase complex. Nature 2002; 416:703-9.

4517