Advertisement

Cyclin-dependent Kinase 5 (Cdk5) Activation Domain of Neuronal Cdk5 Activator

EVIDENCE OF THE EXISTENCE OF CYCLIN FOLD IN NEURONAL Cdk5a ACTIVATOR*
  • Damu Tang
    Affiliations
    Department of Biochemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
    Search for articles by this author
  • Abel C.S. Chun
    Affiliations
    Department of Biochemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
    Search for articles by this author
  • Mingjie Zhang
    Affiliations
    Department of Biochemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
    Search for articles by this author
  • Jerry H. Wang
    Correspondence
    To whom correspondence should be addressed:
    Affiliations
    Department of Biochemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
    Search for articles by this author
  • Author Footnotes
    * This work was supported by the Research Grant Council of Hong Kong, a Hong Kong University of Science and Technology Infrastructure Grant, and by the Biotechnology Research Institute.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:May 09, 1997DOI:https://doi.org/10.1074/jbc.272.19.12318
      Neuronal Cdk5 activator (Nck5a) differs from other cyclin-dependent kinase (Cdk) activators in that its amino acid sequence is only marginally similar to the cyclin consensus sequence. Nevertheless, computer modeling has suggested that Nck5a contains the cyclin-fold motif recently identified in the crystal structure of cyclin A. In the present study, a number of truncation mutants and substitution mutants of the Nck5a were produced and tested for the Cdk5 activation and Cdk5 binding activity. The active domain of Nck5a determined by using the truncation mutants consists of the region spanning residues 150 to 291. The size of Nck5a active domain is essentially the same as that of cyclin A required for Cdk2 activation (Lees, E. M., and Harlow, E. (1993) Mol. Cell. Biol. 13, 1194–1201). The change, or the lack of change, in Cdk5 activation activity observed with a number of substitution mutants may be understood on the basis of structure and function relationship of cyclin A. These results provide support to the previous suggestion (Brown, N. R., Noble, M. E. M., Endicott, J. A., Garman, E. F., Wakatsuki, S., Mitchell, E., Rasmussen, B., Hunt, T., and Johnson, L. N. (1995) Structure 3, 1235–1247) that the activation domain of Nck5a adopts a conformation similar to that of cyclin A. They also provide a partial answer to the question of how Nck5a, a non-cyclin, activates a cyclin-dependent kinase.
      Progress through animal cell cycle depends on the coordinated actions of a family of cdc2-like kinases (
      • Hunter T.
      • Pines J.
      ,
      • Morgan D.O.
      ,
      • Sherr C.J.
      ,
      • Pines J.
      • Hunter T.
      ,
      • Norbury C.
      • Nurse P.
      ) that are heterodimers of a cdc2-homologous catalytic subunit, called cyclin-dependent kinases (Cdks),
      The abbreviations used are: Cdk, cyclin-dependent kinase; Nck5a, neuronal Cdk5 activator; Nck5ai, Nck5a isoform; PCR, polymerase chain reaction; BSA, bovine serum albumin; DTT, dithiothreitol; MOPS, 4-morpholinepropanesulfonic acid; GST, glutathione S-transferase.
      1The abbreviations used are: Cdk, cyclin-dependent kinase; Nck5a, neuronal Cdk5 activator; Nck5ai, Nck5a isoform; PCR, polymerase chain reaction; BSA, bovine serum albumin; DTT, dithiothreitol; MOPS, 4-morpholinepropanesulfonic acid; GST, glutathione S-transferase.
      and an essential regulatory subunit belonging to the cyclin family (
      • Hunt T.
      ,
      • Draetta G.
      ). Cyclins are molecules of diverse molecular masses, but they share the characteristics of containing a homologous region of approximately 100 residues, called cyclin box (
      • Nugent J.A.
      • Alfa C.E.
      • Young T.
      • Hyams J.S.
      ). The recently elucidated crystallographic structure of an active fragment of cyclin A has shown that the cyclin box (of cyclin A) displays a uniquely folded structure comprising five alpha helices. This structure, cyclin fold, repeats itself in a region extending from the C terminus of the cyclin box. There is little amino acid sequence similarity between the two cyclin folds (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Jeffrey P.D.
      • Russo A.A.
      • Polyak K.
      • Gibbs E.
      • Hurwitz J.
      • Massague J.
      • Pavletich N.P.
      ).
      Among Cdks, Cdk5 is unique in several respects. Although Cdk5 is present in all mammalian tissues and cell extracts examined (
      • Hayes T.E.
      • Valtz N.L.M.
      • McKay R.D.G.
      ), brain is the only source where Cdk5 kinase activity has been demonstrated (
      • Ishiguro K.
      • Takamatsu M.
      • Tomizawa K.
      • Omori A.
      • Takahashi M.
      • Arioka M.
      • Uchida T.
      • Imahori K.
      ,
      • Lew J.
      • Beaudette K.
      • Litwin C.M.E.
      • Wang J.H.
      ). The protein is highly expressed in neurons of the central nervous system (
      • Hellmich M.R.
      • Pant H.C.
      • Wada E.
      • Battey J.F.
      ,
      • Tsai L.-H.
      • Takahashi T.
      • Caviness Jr., V.S.
      • Harlow E.
      ), whereas most of the other Cdks such as Cdk1 and Cdk2 are essentially undetectable in post-mitotic neurons (
      • Hayes T.E.
      • Valtz N.L.M.
      • McKay R.D.G.
      ). In parallel with these observations is the finding in the neurons of mammalian brains of two highly homologous Cdk5 activator proteins called neuronal Cdk5 activator (Nck5a) and neuronal Cdk5 activator isoform (Nck5ai) (
      • Ishiguro K.
      • Kobayashi S.
      • Omori A.
      • Takamatsu M.
      • Yonekura S.
      • Anzai K.
      • Imahori K.
      • Uchita T.
      ,
      • Tsai L.-H.
      • Delalle I.
      • Caviness Jr., V.S.
      • Chae T.
      • Harlow E.
      ,
      • Lew J.
      • Huang Q.-Q.
      • Qi Z.
      • Winkfein R.J.
      • Aebersold R.
      • Hunt T.
      • Wang J.H.
      ,
      • Tang D.
      • Yeung J.
      • Lee K.-Y.
      • Matsushita M.
      • Matsui H.
      • Tomizawa K.
      • Hatase O.
      • Wang J.H.
      ). Curiously, despite their ability in activating a cyclin-dependent kinase, these two Cdk5 activators show little sequence similarity to members of cyclin family (
      • Lew J.
      • Huang Q.-Q.
      • Qi Z.
      • Winkfein R.J.
      • Aebersold R.
      • Hunt T.
      • Wang J.H.
      ,
      • Tang D.
      • Yeung J.
      • Lee K.-Y.
      • Matsushita M.
      • Matsui H.
      • Tomizawa K.
      • Hatase O.
      • Wang J.H.
      ). The mechanism of activation of Cdk5 by Nck5a and Nck5ai appears to differ from that of Cdk activation by cyclins. While the activation of Cdk1, Cdk2, or Cdk4 by the respective cyclin has been shown to depend on the phosphorylation of the Cdk subunit on a specific threonine residue by an activating kinase, CAK (
      • Solomon M.J.
      • Harper J.W.
      • Shuttleworth J.
      ,
      • Poon R.Y.C.
      • Yamashita K.
      • Adamczewski J.P.
      • Hunt T.
      • Shuttleworth J.
      ,
      • Kato J.-Y.
      • Matsuoka M.
      • Strom D.K.
      • Sherr C.J.
      ,
      • Fesquet D.
      • Labbe J.C.
      • Derancourt J.
      • Capony J.P.
      • Galas S.
      • Girard F.
      • Lorca T.
      • Shuttleworth J.
      • Doree M.
      • Cavadore J.C.
      ,
      • Fisher R.P.
      • Morgan D.O.
      ), Cdk5 activation by Nck5a or Nck5ai is independent of the phosphorylation of Cdk5 (
      • Qi Z.
      • Huang Q.-Q.
      • Lee K.-Y.
      • Lew J.
      • Wang J.H.
      ).
      The present study was initiated to probe the structural domain of Nck5a essential for the activation of Cdk5. We find that the size of the active domain of Nck5a is essentially identical to that of cyclin A. By site-directed mutagenesis, a few residues of the protein important for Cdk5 binding and Cdk5 activation were located. The results are compatible with a previous suggestion from computer modeling that Nck5a may assume a tertiary structure similar to that of cyclin A (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Bazan J.F.
      ).

      DISCUSSION

      One of the salient features of Nclk that distinguishes the enzyme from other cdc2-like kinases is its unique regulatory subunit, the neuronal Cdk5 activator. In addition to its tissue- and cell-type-specific expression, Nck5a does not have sufficient amino acid sequence similarity to the consensus cyclin box sequence to be classified as a cyclin. Although Cdk5 shares a high degree of sequence similarity with other Cdks (
      • Meyerson M.
      • Enders G.H.
      • Wu C.-L.
      • Su L.-K.
      • Gorka C.
      • Nelson C.
      • Harlow E.
      • Tsai L.-H.
      ), no cyclin has been demonstrated to activate Cdk5. The observation has raised the question as to how a non-cyclin protein carries out Cdk activation activity, such as the Nck5a action on Cdk5 activation.
      The crystallographic structure of a truncated and active cyclin A, both in its free and Cdk2-bound states, has been elucidated recently (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Jeffrey P.D.
      • Russo A.A.
      • Polyak K.
      • Gibbs E.
      • Hurwitz J.
      • Massague J.
      • Pavletich N.P.
      ). The characteristic structure comprises two repeating subdomains, called cyclin fold, each containing five helices. The two repeating subdomains are sandwiched by two helices, an N-terminal and a C-terminal α-helix. Although there is little sequence similarity between the first and the second subdomains, the tertiary structures of the two repeats are almost superimposable, thus suggesting that the amino acid sequence requirement of the “cyclin fold” is not highly rigid. This suggestion is supported by the observation that proteins such as Rb and TFIIB, which are not related to cyclins either structurally or functionally, appear to contain the cyclin-fold structure (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Bazan J.F.
      ,
      • Bagby S.
      • Kim S.
      • Maldonado E.
      • Tong K.I.
      • Reinberg D.
      • Ikura M.
      ,
      • Gibson T.J.
      • Thompson J.D.
      • Blocker A.
      • Kouzarides T.
      ,
      • Nikolov D.B.
      • Chen H.
      • Halay E.D.
      • Usheva A.A.
      • Hisatake K.
      • Lee D.K.
      • Roeder R.G.
      • Burley S.K.
      ). Furthermore, Nck5a has been suggested to adopt the cyclin-fold structure on the basis of computer modeling (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Bazan J.F.
      ). By analysis of the Cdk5 activation and Cdk5 binding activities of a large number of deletion mutants and a few amino acid substitution mutants of Nck5a, the present study provides an experimental support for the suggestion that Nck5a may assume a cyclin-like tertiary structure.
      By systematic truncation in combination with mutation of the terminal residues of the truncated forms of the protein, the minimal size of Nck5a capable of fully activating Cdk5 has been determined to be 142 amino acid residues, residue 150 to 291. Sequence alignment of this protein region with the active domain of cyclin A (
      • Lees E.M.
      • Harlow E.
      ), using the Multalin program (
      • Corpent F.
      ) (Fig. 10), shows that the active domains of the two proteins are essentially of the same size. As expected, this region of cyclin A includes all the amino acid residues that make contacts with Cdk2, as revealed in the crystal structure of the cyclin A-Cdk2 complex (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Jeffrey P.D.
      • Russo A.A.
      • Polyak K.
      • Gibbs E.
      • Hurwitz J.
      • Massague J.
      • Pavletich N.P.
      ). Furthermore, the secondary structure of Nck5a predicted by using a neural network algorithm (
      • Rust B.
      • Sander C.
      ) shows that the predicted α-helices of the active domain of Nck5a are located at the regions well matched with those of α-helices in cyclin A. Although the overall sequence similarity between cyclin A and Nck5a is very low, a small region of approximately 20 amino acids has a significant number of identical residues between the two sequences ( Fig. 10, open box region; Refs.
      • Lew J.
      • Huang Q.-Q.
      • Qi Z.
      • Winkfein R.J.
      • Aebersold R.
      • Hunt T.
      • Wang J.H.
      and
      • Tang D.
      • Yeung J.
      • Lee K.-Y.
      • Matsushita M.
      • Matsui H.
      • Tomizawa K.
      • Hatase O.
      • Wang J.H.
      ). This region correlates approximately with the α3-helix, which appears to play a pivotal role in maintaining the cyclin-fold structure of cyclin A. The α3-helix of cyclin A serves as the core upon which the other four α-helices of the first of the two repeating subdomains are packed. In addition, the N-terminal α-helix is also packed closely with the α3-helix (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Jeffrey P.D.
      • Russo A.A.
      • Polyak K.
      • Gibbs E.
      • Hurwitz J.
      • Massague J.
      • Pavletich N.P.
      ). Thus, it seems that the general structural features of the active domain of Nck5a are compatible with the notion that Nck5a adopts a conformation similar to that of a cyclin fold.
      Figure thumbnail gr10
      Figure 10Sequence alignment of the conserved core region of Nck5a with that of cyclin A. The black barsabove the cyclin A sequences represent the α-helices observed in the crystal structure of cyclin A (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ,
      • Jeffrey P.D.
      • Russo A.A.
      • Polyak K.
      • Gibbs E.
      • Hurwitz J.
      • Massague J.
      • Pavletich N.P.
      ). The dotted regionsbelow the Nck5a sequences typify the α-helix predicted for Nck5a. Theshaded regions represent the Nck5a residues that have been subjected for site-directed mutagenesis studies.
      Although many of the amino acid residues of cyclin A identified to be involved in direct contact with Cdk2 or directly involved in intramolecular interactions are not conserved in Nck5a on the basis of the sequence alignment of Fig. 10, analysis of a few amino acid substitution mutants of Nck5a has revealed potential common structural basis of Nck5a and cyclin A in the protein folding and in Cdk activation. For example, Leu-151 and Leu-152, which contribute to the Cdk5 activation by participating in an essential hydrophobic interaction (see Fig. 4 and Table I), are located in the region corresponding to isoleucine 182 of cyclin A (Fig. 10). From the crystal structure, Ile-182 is seen to be in the N-terminal α-helix of cyclin A. The N-terminal helix is packed intimately with the α3-helix and the residue Ile-182 interacts directly with a phenylalanine residue, Phe-152 of Cdk2. This phenylalanine and its neighboring residues are strongly conserved throughout the Cdk family of kinases, including Cdk5. Thus, it may be suggested that the residues Leu-151 and Leu-152 are involved in the interaction with the phenylalanine residue of Cdk5 that is equivalent to Phe-152 of Cdk2. On the other hand, the observation that substitution of Arg-153 of Nck5a to alanine has little effect on the kinase activation activity of Nck5a can also be understood by assuming that Nck5a and cyclin A have similar structure-function relationships in kinase activation. The equivalence of Arg-153 in cyclin A is a threonine residue, Thr-184. Crystal structure of cyclin A shows that Thr-184 has no direct contact with any residue in Cdk2.
      While the critical importance of the residues at the N-terminal boundary of the active domain of Nck5a may be understood by the participation of these residues in the kinase binding and activation, residues at the C-terminal boundary appear to correspond to a region of cyclin A important for the protein folding. For example, the observation that mutation of Leu-289 had a large effect on the kinase activation activity of Nck5a may be understood by assuming that this residue is important for the protein folding. The Leu-289 equivalent residue in cyclin A is Tyr-318, which is buried deeply in the hydrophobic core of cyclin A, suggesting an involvement in the maintenance of the tertiary structure of the protein. Thus, in addition to defining the active domain of Nck5a, the functional roles of the boundary regions of the active domain are elucidated in the present study. In addition, a number of substitution mutants with mutations within or in proximity to the putative α3-helix region were examined. The results also support the suggestion that Nck5a may have similar general conformation as cyclin A. Two aspartate residues, Glu-221 and Glu-240, which correspond to Gly-232 and Glu-269 of cyclin A, respectively, have been substituted individually by alanine (Fig. 10). It was observed that while substitution of Glu-221 had only little effect, substitution of Glu-240 markedly decreased the kinase activation activity of Nck5a (Fig. 8). Presumably, like Glu-269 in cyclin A, Glu-240 of Nck5a participates in the Cdk5 activation by interacting with Arg-149, equivalent to Arg-150 of Cdk2 (
      • Jeffrey P.D.
      • Russo A.A.
      • Polyak K.
      • Gibbs E.
      • Hurwitz J.
      • Massague J.
      • Pavletich N.P.
      ,
      • Russo A.A.
      • Jeffrey P.D.
      • Pavletich N.P.
      ). Together, these results provide strong support to the suggestion that Nck5a assumes a conformation similar to that of cyclin A (
      • Brown N.R.
      • Noble M.E.M.
      • Endicott J.A.
      • Garman E.F.
      • Wakatsuki S.
      • Mitchell E.
      • Rasmussen B.
      • Hunt T.
      • Johnson L.N.
      ). Such a suggestion is further supported by the observation that Cdk2 could be activated by p25nck5a and some of the derivatives of Nck5a (
      • Poon R.Y.C.
      • Lew J.
      • Hunter T.
      ).
      While the full activation of Cdks by their respective cyclins typically depends on the phosphorylation of the Cdk by the activating kinase, CAK (
      • Solomon M.J.
      • Harper J.W.
      • Shuttleworth J.
      ,
      • Poon R.Y.C.
      • Yamashita K.
      • Adamczewski J.P.
      • Hunt T.
      • Shuttleworth J.
      ,
      • Kato J.-Y.
      • Matsuoka M.
      • Strom D.K.
      • Sherr C.J.
      ,
      • Fesquet D.
      • Labbe J.C.
      • Derancourt J.
      • Capony J.P.
      • Galas S.
      • Girard F.
      • Lorca T.
      • Shuttleworth J.
      • Doree M.
      • Cavadore J.C.
      ,
      • Fisher R.P.
      • Morgan D.O.
      ), Cdk5 is maximally activated by Nck5a in the absence of Cdk5 phosphorylation (
      • Qi Z.
      • Huang Q.-Q.
      • Lee K.-Y.
      • Lew J.
      • Wang J.H.
      ). The observation has raised the question of whether the unique mechanism of the Cdk5 activation by Nck5a is attributable to Cdk5 or to Nck5a (
      • Qi Z.
      • Huang Q.-Q.
      • Lee K.-Y.
      • Lew J.
      • Wang J.H.
      ). The observation that Cdk2 activation by Nck5a is also independent of CAK suggests the phosphorylation-independent activation of the Cdks is determined by the activator protein. In addition to its phosphorylation-independent activation, Nclk does not appear to be significantly inhibited by the common Cdk inhibitory kinase Wee1 kinase (
      • Poon R.Y.C.
      • Lew J.
      • Hunter T.
      ,
      • Tang D.
      • Lee K.-Y.
      • Qi Z.
      • Matsuura I.
      • Wang J.H.
      ), nor by the inhibitor proteins of Cdks, p21Cip, and p27kip1 (
      • Harper J.W.
      • Elledge S.J.
      • Keyomarsi K.
      • Dynlacht B.
      • Tsai L.-H.
      • Zhang P.
      • Dobrowolski S.
      • Bai C.
      • Connell-Crowley L.
      • Swindell E.
      • Fox M.P.
      • Wei N.
      ,
      • Lee M.-H.
      • Nikollic M.
      • Baptista C.A.
      • Lai E.
      • Tsai L.-H.
      • Massague J.
      ). The question of what structural differences between cyclins and Nck5a contribute to the unique regulatory properties of Nck5a is therefore raised. To address this and other related questions, a more in depth characterization of the structure and function of Nck5a is required.
      Poon et al. (
      • Poon R.Y.C.
      • Lew J.
      • Hunter T.
      ), independently, have carried out a study on the Cdk5 active domain of Nck5a and obtained similar results. The minimally sized Nck5a derivative capable of activating Cdk5 determined in their study is the truncated form of residues 150 to 292, essentially the same as that obtained in this study. However, they have found that truncation derivatives with up to 168 residues deleted from the C terminus of Nck5a can still bind Cdk5, a finding significantly different from our observation that only a small region of C terminus deletion can be tolerated in terms of Cdk5 binding. This difference may be attributed to the different experimental approaches used in the two studies. The method used to detect Cdk5 binding in this study was designed mainly to reveal protein derivatives of Nck5a with high affinity Cdk5 binding activity.
      In conclusion, although the amino acid sequence of Nck5a has little similarity to those of cyclins, results of the present study support the previous suggestion (
      • Tang D.
      • Yeung J.
      • Lee K.-Y.
      • Matsushita M.
      • Matsui H.
      • Tomizawa K.
      • Hatase O.
      • Wang J.H.
      ,
      • Bazan J.F.
      ) that Nck5a may adopt a conformation containing the cyclin-fold structure. In addition, a number of amino acid residues in Nck5a have been identified as playing important roles in the kinase activation or protein folding in Nck5a. The success in assigning functions to specific residues on the basis of cyclin A structure has greatly strengthened the suggestion. More importantly, it partly answers the question of how a non-cyclin may activate a cyclin-dependent kinase. The present study, however, does not address the question about the structure-function relationship of Nck5a concerning the unique regulatory properties of the protein. With the knowledge of the crystal structure of T-160 phosphorylated cyclin A-Cdk2 at hand (
      • Russo A.A.
      • Jeffrey P.D.
      • Pavletich N.P.
      ), further studies using specially designed Nck5a mutants may be constructed to address the question of why Nck5a activation of Cdk5 is phosphorylation-independent.

      REFERENCES

        • Hunter T.
        • Pines J.
        Cell. 1994; 79: 573-582
        • Morgan D.O.
        Nature. 1995; 374: 131-134
        • Sherr C.J.
        Trends Biochem. Sci. 1995; 20: 187-190
        • Pines J.
        • Hunter T.
        Trends Cell Biol. 1991; 1: 117-121
        • Norbury C.
        • Nurse P.
        Annu. Rev. Biochem. 1992; 61: 441-470
        • Hunt T.
        Semin. Cell. Biol. 1991; 2: 213-222
        • Draetta G.
        Trends Cell Biol. 1993; 3: 287-289
        • Nugent J.A.
        • Alfa C.E.
        • Young T.
        • Hyams J.S.
        J. Cell Sci. 1991; 99: 674-699
        • Brown N.R.
        • Noble M.E.M.
        • Endicott J.A.
        • Garman E.F.
        • Wakatsuki S.
        • Mitchell E.
        • Rasmussen B.
        • Hunt T.
        • Johnson L.N.
        Structure ( Lond. ). 1995; 3: 1235-1247
        • Jeffrey P.D.
        • Russo A.A.
        • Polyak K.
        • Gibbs E.
        • Hurwitz J.
        • Massague J.
        • Pavletich N.P.
        Nature. 1996; 376: 313-320
        • Hayes T.E.
        • Valtz N.L.M.
        • McKay R.D.G.
        New Biol. 1991; 3: 259-269
        • Ishiguro K.
        • Takamatsu M.
        • Tomizawa K.
        • Omori A.
        • Takahashi M.
        • Arioka M.
        • Uchida T.
        • Imahori K.
        J. Biol. Chem. 1992; 267: 10897-10901
        • Lew J.
        • Beaudette K.
        • Litwin C.M.E.
        • Wang J.H.
        J. Biol. Chem. 1992; 267: 13383-13390
        • Hellmich M.R.
        • Pant H.C.
        • Wada E.
        • Battey J.F.
        Proc. Natl. Acad Sci. U. S. A. 1992; 89: 10867-10871
        • Tsai L.-H.
        • Takahashi T.
        • Caviness Jr., V.S.
        • Harlow E.
        Development. 1993; 119: 1029-1040
        • Ishiguro K.
        • Kobayashi S.
        • Omori A.
        • Takamatsu M.
        • Yonekura S.
        • Anzai K.
        • Imahori K.
        • Uchita T.
        FEBS Lett. 1994; 342: 203-208
        • Tsai L.-H.
        • Delalle I.
        • Caviness Jr., V.S.
        • Chae T.
        • Harlow E.
        Nature. 1994; 371: 419-423
        • Lew J.
        • Huang Q.-Q.
        • Qi Z.
        • Winkfein R.J.
        • Aebersold R.
        • Hunt T.
        • Wang J.H.
        Nature. 1994; 371: 423-426
        • Tang D.
        • Yeung J.
        • Lee K.-Y.
        • Matsushita M.
        • Matsui H.
        • Tomizawa K.
        • Hatase O.
        • Wang J.H.
        J. Biol. Chem. 1995; 270: 26897-26903
        • Solomon M.J.
        • Harper J.W.
        • Shuttleworth J.
        EMBO J. 1993; 12: 3133-3142
        • Poon R.Y.C.
        • Yamashita K.
        • Adamczewski J.P.
        • Hunt T.
        • Shuttleworth J.
        EMBO J. 1993; 12: 3123-3132
        • Kato J.-Y.
        • Matsuoka M.
        • Strom D.K.
        • Sherr C.J.
        Mol. Cell. Biol. 1994; 14: 2713-2721
        • Fesquet D.
        • Labbe J.C.
        • Derancourt J.
        • Capony J.P.
        • Galas S.
        • Girard F.
        • Lorca T.
        • Shuttleworth J.
        • Doree M.
        • Cavadore J.C.
        EMBO J. 1993; 12: 3111-3121
        • Fisher R.P.
        • Morgan D.O.
        Cell. 1994; 78: 713-724
        • Qi Z.
        • Huang Q.-Q.
        • Lee K.-Y.
        • Lew J.
        • Wang J.H.
        J. Biol. Chem. 1995; 270: 10847-10854
        • Bazan J.F.
        Proteins Struct. Funct. Genet. 1996; 24: 1-17
        • Lees E.M.
        • Harlow E.
        Mol. Cell. Biol. 1993; 13: 1194-1201
      1. Sambrook, J., Fritsch, E. F., Maniatis, T., Molecular Cloning: A Laboratory Manual, 2nd Ed., 1989, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

        • Ausubel F.M.
        • Brent R.
        • Kingston R.E.
        • Moore D.D.
        • Seidman J.G.
        • Smith J.A.
        • Struhl K.
        Current Protocols in Molecular Biology. John Wiley and Sons, Inc., 1994: 3.5.11-3.5.12
        • Mikaelian I.
        • Sergeant A.
        Nucleic Acids Res. 1992; 20: 376
        • Smith D.B.
        • Johnson K.S.
        Gene ( Amst. ). 1988; 67: 31-40
        • Laemmli U.K.
        Nature. 1970; 227: 680-685
        • Lee K.-Y.
        • Rosales J.L.
        • Tang D.
        • Wang J.H.
        J. Biol. Chem. 1996; 271: 1538-1543
        • Meyerson M.
        • Enders G.H.
        • Wu C.-L.
        • Su L.-K.
        • Gorka C.
        • Nelson C.
        • Harlow E.
        • Tsai L.-H.
        EMBO J. 1992; 11: 2909-2919
        • Poon R.Y.C.
        • Lew J.
        • Hunter T.
        J. Biol. Chem. 1996; 272: 5703-5708
        • Harper J.W.
        • Elledge S.J.
        • Keyomarsi K.
        • Dynlacht B.
        • Tsai L.-H.
        • Zhang P.
        • Dobrowolski S.
        • Bai C.
        • Connell-Crowley L.
        • Swindell E.
        • Fox M.P.
        • Wei N.
        Mol. Biol. Cell. 1995; 6: 387-400
        • Lee M.-H.
        • Nikollic M.
        • Baptista C.A.
        • Lai E.
        • Tsai L.-H.
        • Massague J.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 3259-3263
        • Bagby S.
        • Kim S.
        • Maldonado E.
        • Tong K.I.
        • Reinberg D.
        • Ikura M.
        Cell. 1995; 82: 857-867
        • Gibson T.J.
        • Thompson J.D.
        • Blocker A.
        • Kouzarides T.
        Nucleic Acids Res. 1994; 22: 946-952
        • Nikolov D.B.
        • Chen H.
        • Halay E.D.
        • Usheva A.A.
        • Hisatake K.
        • Lee D.K.
        • Roeder R.G.
        • Burley S.K.
        Nature. 1995; 377: 119-128
        • Corpent F.
        Nucleic Acids Res. 1988; 16: 10881-10890
        • Rust B.
        • Sander C.
        J. Mol. Biol. 1993; 232: 584-599
        • Russo A.A.
        • Jeffrey P.D.
        • Pavletich N.P.
        Nat. Struct. Biol. 1996; 3: 696-700
        • Tang D.
        • Lee K.-Y.
        • Qi Z.
        • Matsuura I.
        • Wang J.H.
        Biochem. Cell Biol. 1996; 74: 419-429