Covalent modification of all members of human cullin family proteins by NEDD8


Recently we found that NEDD8, a ubiquitin-like protein, was linked covalently to human cullin-4A (abbreviated Cul-4A) by a new ubiquitin-related pathway that is analogous to but distinct from the ligating system for SUMO1, another ubiquitin-like protein. However, it remained unknown whether the other five members of the family of human cullin/Cdc53 proteins are modified by NEDD8. Here we report that all Hs-Cul family proteins, such as Cul-1, Cul-2, Cul-3, Cul-4B, and Cul-5, in addition to Cul-4A, were modified by covalent attachment of NEDD8 in rabbit reticulocyte lysates. Moreover, by comprehensive Northern-blot analyses, we examined multiple tissue distributions of the messages for all Cul-family proteins, NEDD8, and the NEDD8-ligating system consisting of APP-BP1/hUba3, and hUbc12, which function as E1- and E2-like enzymes, respectively. The expressions of Cul-1, Cul-2, and Cul-3 resembled each other and were apparently correlated to those of NEDD8 and the NEDD8-ligating system in various human cells and tissues. However, the mRNA levels of Cul-4A, Cul-4B, and Cul-5 differed considerably from each other as well as from other Cul-family proteins. The enhanced expression of all Cul-family proteins except Cul-5 was observed in a variety of tumor cell lines.


Ubiquitin (Ub) is a landmark molecule in a post-translational protein-modification system that plays a central role in intracellular protein breakdown (reviewed in Hochstrasser, 1996; Hershko and Ciechanover, 1998). It is attached covalently to target proteins via an isopeptide linkage between the C-terminal Gly of Ub and the ε-NH2 group of the Lys residue of the acceptor substrate. The covalent attachment of Ub to substrate proteins is known to be mediated by a cascade of three enzymes, termed E1 (Ub-activating), E2 (Ub-conjugating), and E3 (Ub-ligating) enzymes. During the past decade the Ub-dependent proteolytic pathway has been shown to be involved in various biologically important processes, such as the cell cycle, signal transduction, and cancer (reviewed in Hershko and Ciechanover, 1998; Elledge and Harper, 1998).

The progress in the study of the Ub system has revealed the existence of multiple molecules that are structurally related to Ub (reviewed in Tanaka et al., 1998). These proteins are called Ub-like proteins (Ubl), but their biological roles were largely unknown until quite recently. Of them, SUMO1 (small Ub-related modifier: human homolog of yeast Smt3) is capable of modifying a nuclear pore complex protein, RanGAP1, indicating its role in regulating the translocation of various nuclear proteins (reviewed in Saitoh et al., 1997). In addition, it is of particular interest that SUMO1 can be conjugated to two Lys residues of I-κB where Ub is linked for proteasomal degradation, indicating that SUMO1 may contribute to negative regulation of the NF-κB signaling pathway (Desterro et al., 1998). Other Ubl proteins, yeast Rub1 and its mammalian homolog NEDD8 that was identified as one of the multiple neural precursor cell-expressed developmentally down-regulated genes (Kumar et al., 1993) were found to be ligated to yeast Cdc53 and human Cul-4A, respectively (Liakopoulos et al., 1998; Osaka et al., 1998), which are members of the family of cullin/Cdc53 proteins functioning as an essential component of a multifunctional Ub-protein ligase E3 complex called SCF (Skp1-Cdc53/Cul-1-F-box protein) (Patton et al., 1998a; Elledge and Harper, 1998; Harper and Elledge, 1999). These findings suggest that Rub1 and NEDD8 are functionally equivalent and that the Rub1/NEDD8-modifying pathway plays a key role in the Ub-mediated pathway with respect to cell-cycle regulation (reviewed in Hochstrasser, 1998).

Yeast Smt3 (SUMO1 homolog) is covalently ligated to substrate proteins by a novel pathway related to the Ub-ligating system, being activated by a new heterodimeric complex of Aos1/yUba2 and conjugated by yUbc9 as the E1- and E2-like enzymes, respectively, in the ubiquitinylation pathway (Johnson et al., 1997; Schwarz et al., 1998). On the other hand, recently we showed that the NEDD8-ligating pathway (abbreviated NEDD8-ylation) resembles that of Smt3/SUMO1, consisting of the APP-BP1/hUba3 complex and hUbc12, which are related to E1 and E2 enzymes, respectively (Osaka et al., 1998). Intriguingly, Rub1, a presumptive yeast homolog of mammalian NEDD8 displaying 59% amino acid identity to human NEDD8, was found to be ligated to the target protein through Ula1/yUba3 and yUbc12 as the E1- and E2-like enzymes, respectively (Liakopoulos et al., 1998). It is of interest that Ub is conjugated by multiple species of E2, whereas Smt3/SUMO1 and Rub1/NEDD8 each use a single conjugating enzyme, Ubc9 and Ubc12, respectively. Taken together, it becomes clear that three distinct pathways for modification of Ub and Ub-like proteins exist in both yeast and mammalian cells to date.

Quite recently, we reported that a major target of NEDD8-ylation is Hs-Cul-4A (Osaka et al., 1998), but it has been reported so far that human cullins consist of six family proteins termed Cul-1, Cul-2, Cul-3, Cul-4A, Cul-4B, and Cul-5 (Kipreos et al., 1996). It remains obscure whether NEDD8 is also ligated to other Hs-Cul-family proteins in addition to Cul-4A. We, accordingly, examined whether five other members of the Hs-Cul-family proteins are also targets for the NEDD8-ligating system. As shown in Figure 1, all six [35S]Cul-family proteins were found to be modified by GST-NEDD8. This linkage between GST-NEDD8 and these cullins was resistant to treatment with 50 mM dithiothreitol (DTT). Note that two NEDD8-ligated bands were observed for Cul-5, but this may be due to the presence of two in vitro-translated [35S]Cul-5. Previously, as we reported the C-terminal 171 amino acid residues of Cul-4A are sufficient for modification of NEDD8 (Osaka et al., 1998), it is likely that both mature [35S]Cul-5 and the N-terminal shorter [35S]Cul-5 were ligated by NEDD8. No complexes with these [35S]cullins were formed when the mutated GST-NEDD8(Δ76G) (the C-terminal Gly residue of NEDD8 had been deleted) was used instead of GST-NEDD8, implying that NEDD8 is conjugated to all cullins via the C-terminal Gly residue in a manner analogous to the NEDD8-ylation of Cul-4A.

Figure 1

Linkage of GST-NEDD8 to various Hs-Cul-family proteins in a DTT-insensitive fashion. Samples (0.5 μg) of cDNAs encoding six Cul-proteins ligated in a Bluescript SKII vector were incubated for 60 min at 30°C in 5 μl of a TNT® coupled reticulocyte lysate transcription/translation system containing [35S]methionine (Promega) in the presence of 0.35 μg of recombinant GST-NEDD8 or GST-NEDD8(Δ76G) lacking the C-terminal Gly residue. The reaction was terminated by addition of equal volume of SDS-sample buffer containing 100 mM DTT and boiled for 3 min. The translational products (5 μl) were subjected directly to SDS – PAGE (7.5% gel). The gel was dried and the signals were detected by autoradiography, as described previously (Osaka et al., 1998). Asterisks on the right indicate the 35S-labeled various cullins complexed with GST-NEDD8

Recently, it was reported that Skp1, a component of a SCF Ub-ligase complex, is needed for the modification of Cdc53 by Rub1 in Saccaromyces cerevisiae (Lammer et al., 1998) and human Skp1 is specifically associated with Cul-1 without interacting with five other Cul-family proteins (Michel and Xiong, 1998). Therefore, we examined the effect of human Skp1 on the NEDD8-ylation of human Cul-proteins, but no appreciable effect was observed for not only Cul-1 but also the other five Cul-proteins (data not shown). This apparently conflicting result may be due to the presence of excess amounts of Skp1 in the reticulocyte lysates used. Alternatively, the possibility that Skp1 is not required for NEDD8-ylation in humans cannot be ruled out completely. However, this possibility is perhaps unlikely, judging from the high evolutional conservation of the post-translational protein-modifying system for Rub1 and NEDD8 (Liakopoulos et al., 1998; Osaka et al., 1998). Further study is required to clarify this issue.

Next, we determined the mRNA levels of NEDD8 and both E1 and E2 for the NEDD8-modifying system in various adult tissues and cultured tumor cells in human and embryonic development in mice. In this work, RNA blot analysis was carried out simultaneously using four different plates for individual probe, As shown in Figure 2, the expression of NEDD8 was high in various tissues, such as heart, skeletal muscle, spleen, thymus, prostate, testis, ovary, colon, and leukocytes, but low in pancreas, lung, liver, kidney, and small intestine. The elevation of NEDD8 expression was observed in various human tumor cell lines, such as leukemia, HeLa, K562, and MOLT-4 cells but was low in Raji, SW480, and G361 cells. No appreciable change in the NEDD8 expression was observed during embryonic development of mice, although it was reported to be down-regulated developmentally (Kumar et al., 1993; Kamitani et al., 1997).

Figure 2

Expression of mRNAs encoding NEDD8 and members of the ligating system determined by Northern blot analysis. For RNA blot hybridization analysis, multiple tissue Northern blot filters containing 2 μg of poly(A)+RNAs from various human tissues and mouse embryos (Clontech) were hybridized with various 32P-labeled probes encoding NEDD8, APP-BP1, hUba3, and hUbc12 (Osaka et al., 1998), as indicated. The β-actin probe (Clontech) was used as an internal standard. The probes used were full-length cDNAs for various proteins labeled by the `Ready to Go DNA Labelling Beads (−dCTP)' kit (Amersham Pharmacia Biotech Inc.). Hybridization was performed at 42°C for 14 h in the solution containing 5×SSPE, 50% formamide, 5×Denhardt's solution, 0.5% SDS, and labeled cDNA (4×105 c.p.m./ml) which had a specific activity of about 109 c.p.m./μg. After hybridization, the nylon membrane was washed successively with 2×SSPE, 1×SSPE (twice), and 0.1×SSPE containing 0.1% SDS for 10, 20 and 20 min, respectively, at 65°C. The membrane was exposed to Kodak XAR-5 film at −80°C with an intensifying screen. The positions of RNA size markers are shown (in kilobases) on the left

The messages for the NEDD8 ligating system, APP-BP1, hUba3, and hUbc12, were high in muscular tissues, such as skeletal muscle and heart, spleen, and testis in adult human tissues but were low in lung, liver, and small intestine. The expression pattern of both NEDD8 and three genes catalyzing its ligation resembled each other, except that their expression levels appear to differ somewhat and that the expression of NEDD8 was high in the thymus, whereas those of the NEDD8-ligating system were expressed conversely in the tissue. The expressions of APP-BP1, hUba3, and hUbc12 were high in all established cell lines, except that hUbc12 was low in Raji. The NEDD8-ligating system showed no significant alteration during embryonic development in mice, except high expression at 7 days post-gestation. The reason why the RNA-blot intensity of the NEDD8-ligating system genes was relatively high compared to that of NEDD8 is unknown.

Subsequently, we examined the mRNA levels of Hs-Cul-family proteins, which are major targets in the modification of NEDD8. As shown in Figure 3, the mRNA levels of Cul-1, Cul-2, and Cul-3 were similar in various normal human tissues; i.e., high in heart, skeletal muscle, and testis, except that Cul-3 was specifically high in ovary, colon, and leukocytes but was low in lung, liver, thymus, and small intestine. Their expressions were high in all tumor cells examined, except Raji cells and were not dramatically changed in various stages of post-gestation in mice. The expression pattern of these three cullins appears to resemble those of NEDD8 and the NEDD8-ligating system (Figure 2).

Figure 3

Expression of mRNAs encoding Cul-family proteins determined by Northern blot analysis. Northern blot analysis was carried out as in Figure 2. The human cDNAs for Cul-1 and Cul-2 were purchased from ATCC. The cDNAs of Cul-3 and Cul-4B were kindly given by Kazusa DNA Research Institute (Chiba, Japan). The cDNA of Cul-4A and Cul-5 were synthesized by PCR amplification using human testis Marathon Ready cDNA (Clontech)

For the expressions of Cul-4A and Cul-4B, they were high in testis and spleen and low not only in lung, liver, thymus, and small intestine but also in muscular tissues, such as skeletal muscle and heart, differing from the expressions of Cul-1, Cul-2, and Cul-3. The expression of Cul-4A and Cul-4B with a high homology, displaying an identity of 83%, differed considerably, revealing high expression of Cul-4A in heart and skeletal muscle whereas high expression of Cul-4B in prostate, ovary, colon, and leukocytes. The Cul-4A and Cul-4B mRNAs were elevated greatly in various tumor cells, except Raji. The expressions of these two homologous cullins were not altered in various embryonic developmental stages, except that Cul-4B was expressed specifically high at 11 days post-gestation. On the other hand, the expression of Cul-5 was relatively low in all human tissues and embryonic mice, but it was found that considerable expressions were observed in muscular tissues similar to those of Cul-1, Cul-2, Cul-3 and Cul-4A. Interestingly, only Cul-5 was very low in testis and spleen, nonetheless the other five Cul-family proteins were expressed at quite high levels in such tissues. Moreover, no significant enhancement of the Cul-5 mRNA was observed in various tumor cells. Thus, the expression of Cul-5 was quite different from those of the other five Cul-members.

According to the present study, the expression profiles of the NEDD8-ligation system resemble those of Cul-1, Cul-2, and Cul-3, implying that they may have redundant functions. It is of note that the expressions of Cul-4A and Cul-4B resemble each other but differ from those of all other Cul-family proteins and that Cul-5 showed very unique tissue expression, differing considerably from those of all other Cul-family proteins, indicating that they may not be functionally equivalent to other cullins. Intriguingly, the expressions of all cullins except Cul-5 were very high in various tumor cells, except Burkit lymphoma Raji. However, the high expression of these Cul-proteins does not appears to be simply correlated with cell proliferation, because their expressions were very low with small intestine having very active proliferating cells, implying that these expressions may be related to tumorigenesis of the cells. This finding was consistent with the fact that human Cul-3 and Cul-4A were recently identified as showing high expressions in cultured colon cancer cells (Du et al., 1998) and primary breast cancer (Chen et al., 1998), respectively.

Evidence is increasing to address the biological importance of Cul-family proteins. For example, human Cul-1 was found to be a component of an SCF Ub-ligase that regulates multiple cellular functions, such as the G1/S progression of the cell cycle (Lisztwan et al., 1998; Lyapina et al., 1998). Indeed, its yeast homolog Cdc53 was found to be involved not only in the degradation of the CDK (cyclin-dependent kinase) inhibitor Sic1 (Skowyra et al., 1997; Feldman et al., 1997) and the S phase initiator Cdc6 (Drury et al., 1997) but also in gene expression (Li and Johnston, 1997) and methionine biosynthesis (Patton et al., 1998b). Moreover, mammalian SCF has been shown to function in ubiquitinylation for CD4 (Margottin et al., 1998), cyclin D (Russell et al., 1999), and E2F-1 (Marti et al., 1999). There has been growing evidence to demonstrate that the SCF complex, consisting of Skp1, Cul-1, and βTrCP, acts as a Ub-ligase for IκBα (Yaron et al., 1998; Winston et al., 1999; Spencer et al., 1999; Suzuki et al., 1999; Hatakeyama et al., 1999) and β-catenin (Winston et al., 1999; Latres et al., 1999; Hart et al., 1999; Kitagawa et al., 1999). Intriguingly, we recently found that Cul-1 of SCFβTrCP bound to phosphorylated IκBα was preferentially modified by NEDD8 in HeLa cells (Chiba et al., unpublished data).

Recently, Lammer et al. (1998) reported that modification of yeast Cdc53 by Rub1 may affect the optimal assembly or function of the SCF complex. Moreover, a deletion of ENR2 (equivalent to ULA1) is synthetic lethal in temperaturesensitive alleles of cdc34 (that is Ubc3) and enhances the phenotypes of cdc4, cdc53, and skp1, all of which are components of the SCF Ub-ligase complex, implying that the Rub1 ligation pathway is closely linked to cell-cycle regulation. Consistent with this notion, the mutation of hamster SMC, encoding a protein nearly identical to APP-BP1, is responsible for cell-cycle defects in the ts41 cell line (Handeli and Weintraub, 1992; Hochstrasser, 1998). Currently we found that the NEDD8-modifying pathway is essential for cell viability and function of Pcu1 (equivalent to Cul-1) in fission yeast and that Pcu1 complexed to SCF was absolutely modified by NEDD8 in vivo and Pcu1K713R (Lys at position 713 in the Pcu1 was replaced by Arg) which is defective for modification by NEDD8 resulted in impairment of S. pombe SCF Ub-ligase activity (Osaka et al., submitted). Taken together, it is suggested that the NEDD8/Rub1-ligating pathway may play a critical role in the function of the SCF Ub-ligase complex containing Cul-1 and Skp1.

However, the biological significance of why NEDD8 modifies other 5 Hs-Cul-family proteins is unclear at present. During preparation of this manuscript, Liakopoulos et al. (1999) reported that NEDD8 was linked to human Cul-2 which assembles with the von Hippel-Lindau tumor suppressor protein pVHL and elongin B/C to form an SCF-like protein complex, CBCpVHL. The formation of the NEDD8-Cul-2 conjugate is stimulated by the pVHL, but not by a tumorigenic pVHL variant, suggesting that ligation of NEDD8 to Cul-2 presumably is important for pVHL tumor suppression function via perhaps CBCpVHL. However, it is still unknown whether all Cul-family proteins form a SCF-like complex to function as a Ub-protein ligase. We would favor the hypothesis that the NEDD8-ylation for all members of the family of human Cul proteins is indispensable for their functions as an essential component of a multifunctional Ub-protein ligase E3 complex that plays a critical role in Ub-mediated proteolysis. Whether the NEDD8-ylation of these Cul-family proteins is critically responsible for their functions awaits further study.


  1. Chen LC, Manjeshwar S, Lu Y, Moore D, Ljung BM, Kuo WL, Dairkee SH, Wernick M, Collins C and Smith HS. . 1998 Cancer Res. 58: 3677–3683.

  2. Desterro JM, Rodriguez MS and Hay RT. . 1998 Mol. Cell. 2: 233–239.

  3. Drury LS, Perkins G and Diffley JF. . 1997 EMBO J. 16: 5966–5976.

  4. Du M, Sansores-Garcia L, Zu Z and Wu KK. . 1998 J. Biol. Chem. 273: 24289–24292.

  5. Elledge SJ and Harper JW. . 1998 Biochim. Biophys. Acta 1377: M61–M70.

  6. Feldman RM, Correll CC, Kaplan KB and Deshaies RJ. . 1997 Cell 91: 221–230.

  7. Handeli S and Weintraub H. . 1992 Cell 71: 599–611.

  8. Harper JW and Elledge SJ. . 1999 Nature Cell Biol. 1: E5–E7.

  9. Hart M, Concordet JP, Lassot I, Albert I, del los Santos R, Durand H, Perret C, Rubinfeld B, Margottin F, Benarous R and Polakis P. . 1999 Curr. Biol. 9: 207–210.

  10. Hatakeyama S, Kitagawa M, Nakayama K, Shirane M, Matsumoto M, Hattori K, Higashi H, Nakano H, Okumura K, Onoe K, Good RA and Nakayama Ki. . 1999 Proc. Natl. Acad. Sci. USA 96: 3859–3863.

  11. Hershko A and Ciechanover A. . 1998 Annu. Rev. Biochem. 67: 425479.

  12. Hochstrasser M. . 1996 Annu. Rev. Genet. 30: 405–439.

  13. Hochstrasser M. . 1998 Genes Dev. 12: 901–907.

  14. Johnson ES, Schwienhorst I, Dohmen RJ and Blobel G. . 1997 EMBO J. 16: 5509–5519.

  15. Kamitani T, Kito K, Nguyen HP and Yeh EH. . 1997 J. Biol. Chem. 272: 28557–28562.

  16. Kipreos ET, Lander LE, Wing JP, He WW and Hedgecock EM. . 1996 Cell 85: 829–839.

  17. Kitagawa M, Hatakeyama S, Shirane M, Matsumoto M, Ishida N, Hattori K, Nakamichi I, Kikuchi A, Nakayama Ki and Nakayama K. . 1999 EMBO J. 18: 2401–2410.

  18. Kumar S, Yoshida Y and Noda M. . 1993 Biochem. Biophys. Res. Commun. 195: 393–399.

  19. Lammer D, Mathias N, Laplaza JM, Jiang W, Liu Y, Callis J, Goebl M and Estelle M. . 1998 Genes Dev. 12: 914–926.

  20. Latres E, Chiaur DS and Pagano M. . 1999 Oncogene 18: 849–854.

  21. Li FN and Johnston M. . 1997 EMBO J. 16: 5629–5638.

  22. Liakopoulos D, Büsgen T, Brychzy A, Jentsch S. and Pause A. . 1999 Proc. Natl. Acad. Sci. USA 96: 5510–5515.

  23. Liakopoulos D, Doenges G, Matuschewski K and Jentsch S. . 1998 EMBO J. 17: 2208–2214.

  24. Lisztwan J, Marti A, Sutterluty H, Gstaiger M, Wirbelauer C and Krek W. . 1998 EMBO J. 17: 368–383.

  25. Lyapina SA, Correll CC, Kipreos ET and Deshaies RJ. . 1998 Proc. Natl. Acad. Sci. USA 95: 7451–7456.

  26. Margottin F, Bour SP, Durand H, Selig L, Benichou S, Richard V, Thomas D, Strebel K and Benarous R. . 1998 Mol. Cell 1: 565–574.

  27. Marti A, Wirbelauer C, Scheffner M and Krek W. . 1999 Nature Cell Biol. 1: 14–19.

  28. Michel JJ and Xiong Y. . 1998 Cell Growth Differ. 9: 435–449.

  29. Osaka F, Kawasaki H, Aida N, Saeki M, Chiba T, Kawashima S, Tanaka K and Kato S. . 1998 Genes Dev. 12: 2263–2268.

  30. Patton EE, Willems AR and Tyers M. . 1998a Trends Genet. 14: 236–243.

  31. Patton EE, Willems AR, Sa D, Kuras L, Thomas D, Craig KL and Tyers M. . 1998b Genes Dev. 12: 692–705.

  32. Russell A, Thompson MA, Hendley J, Trute L, Armes J and Germain D. . 1999 Oncogene 18: 1983–1991.

  33. Schwarz SE, Matuschewski K, Liakopoulos D, Scheffner M and Jentsch S. . 1998 Proc. Natl. Acad. Sci. USA 95: 560–564.

  34. Saitoh H, Pu RT and Dasso M. . 1997 Trends Biochem. Sci. 22: 374–376.

  35. Skowyra D, Craig KL, Tyers M, Elledge SJ and Harper JW. . 1997 Cell 91: 209–219.

  36. Spencer E, Jiang J and Chen ZJ. . 1999 Genes Dev. 13: 284–294.

  37. Suzuki H, Chiba T, Kobayashi M, Takeuchi M, Suzuki T, Ichiyama A, Ikenoue T, Omata M, Furuichi K and Tanaka K. . 1999 Biochem. Biophys. Res. Commun. 256: 127–132.

  38. Tanaka K, Suzuki T and Chiba T. . 1998 Mol. Cells 8: 503–512.

  39. Winston JT, Strack P, Beer-Romero P, Chu CY, Elledge SJ and Harper JW. . 1999 Genes Dev. 13: 270–283.

  40. Yaron A, Hatzubai A, Davis M, Lavon I, Amit S, Manning AM, Anderson JS, Mann M, Mercurio F and Ben-Neriah Y. . 1998 Nature 396: 590–594.

Download references


This work was supported in part by grants from the program Grants-in-aid for Scientific Research on Priority Areas (Intracellular Proteolysis) from the Ministry of Education, Science, Sports, and Culture of Japan, and the Human Frontier Science Promotion Organization (K Tanaka).

Author information

Correspondence to Keiji Tanaka.

Rights and permissions

Reprints and Permissions

About this article


  • cullin
  • NEDD8
  • SCF
  • ubiquitin
  • ubiquitin-like protein

Further reading