Transcriptional control at regulatory checkpoints by ... - Oxford Academic

© 2001 Oxford University Press

Human Molecular Genetics, 2001, Vol. 10, No. 7

693–698

Transcriptional control at regulatory checkpoints by histone deacetylases: molecular connections between cancer and chromatin Paul A. Wade+ Emory University School of Medicine, Department of Pathology and Laboratory Medicine, Woodruff Memorial Research Building—Room 7105B, 1639 Pierce Drive, Atlanta, GA 30322, USA Received 5 January 2001; Revised and Accepted 19 January 2001

Cancer cells exhibit a set of unique properties that distinguish them from their normal counterparts. Among these features are increased growth rates, loss of differentiation, escape from cell death pathways, evasion of anti-proliferative signals, a decreased reliance on exogenous growth factors and escape from replicative senescence. Acquisition of these features by malignant cells requires impairment of normal cellular control mechanisms. Over the past few years, it has become increasingly apparent that an important subset of the molecular changes commonly found in cancer cells involves inappropriate regulation of gene expression. This review will address regulatory pathways whose disruption contributes to the malignant phenotype. The failure to deacetylate and thus repress transcription by the Class I histone deacetylases HDAC1 and HDAC2 due to disruption of the Rb family of proteins has been firmly established as a mechanism leading to increases in growth rate and cellular proliferation. Recent data suggest that this regulatory circuit also executes G1 checkpoint arrest downstream of DNA damage, cellular senescence and contact inhibition. In contrast to this failure to deacetylate, it now seems probable that changes in differentiation status may result in part from inappropriate deacetylation and concomitant transcriptional repression mediated by the Class II histone deacetylases. This inappropriate deacetylation by HDAC4, HDAC5 and HDAC6 follows their relocalization from the cytoplasm to the nucleus. Thus, multiple classical features of cancer cells can be manifested by improper histone deacetylation. Transcription in eukaryotic cells is profoundly influenced by the manner in which DNA is packaged. Local chromatin architecture is now generally recognized as playing a crucial and active role in regulation of gene expression. This architecture is strongly influenced by post-translational modifications of the core histones, the building blocks of the fundamental structural unit of chromatin (1–3). Core histone acetylation is probably the best appreciated of these modifications. It occurs at the epsilon amino groups of evolutionarily conserved lysine residues located in the N-termini of the core histones (1–4). All the core histones are acetylated in vivo; however, modifications of histones H3 and H4 are much more extensively characterized than those of H2A and H2B (1–4). Acetylation of histone H3 is primarily associated with transcription whereas H4 acetylation is associated with both transcription and with chromatin assembly (1–4). Steady-state levels of acetylation in the core histones result from the balance of the antagonistic activities of histone acetyltransferases and histone deacetylases (2–4). In general, increased levels of histone acetylation are associated with transcriptional activity (Fig.1) whereas decreased acetylation levels are associated with repression (2,4). Histone deacetylases fall into four major categories. The human Class I histone deacetylases include HDAC1, HDAC2, HDAC3 and HDAC8: enzymes that are similar to the yeast +Tel:

+1 404 712 9666; Fax: +1 404 727 8540; Email: [email protected]

transcriptional regulator Rpd3p (5–8). These proteins have been associated with classical transcriptional corepressors including Sin3, NCoR and methyl CpG binding proteins (5,9– 11). The human Class II deacetylases HDAC4, HDAC5, HDAC6 and HDAC7 are more similar to yeast Hda1p than to Rpd3 (12–14). Little is currently known about the biochemistry of this category of enzymes. The remaining two classes of deacetylase, the Sir2 family and maize HD2, are beyond the scope of this review. Both Class I and Class II HDACs share a common catalytic motif (Fig. 2); the two classes are differentiated by unique N-terminal sequences found only in the Class II enzymes (12–14). The classical paradigm for transcriptional regulation by histone deacetylases invokes recruitment to a gene regulatory region through a series of protein–protein interactions (Fig. 1). The DNA binding properties of transcriptional repressors such as unliganded nuclear hormone receptors, the Mad/Max heterodimer, YY1, methyl CpG binding proteins or the yeast Ume6 protein dictate the specificity of repression (2,5,9,15). These proteins interact either directly with histone deacetylase enzymes or indirectly with them through the bridging functions of transcriptional corepressors such as Sin3 and NCoR/ SMRT (2,5,9,15). Increased local concentrations of deacetylase lead to a region of hypoacetylated chromatin (16,17), which exerts a dominant repressive effect on transcription. The

694


inappropriate redistribution of Class II enzymes through perturbation of cellular signaling pathways will be discussed. Although quite distinct mechanistically, both pathways lead ultimately to inappropriate gene expression by improper localization of histone deacetylase enzymes, a phenomenon known to be causal in some leukemias (25). THE HISTONE DEACETYLASE/RB CONNECTION AND G1 CHECKPOINT CONTROL

Figure 1. Transcriptional repression and activation in chromatin. The diagram depicts various aspects of the transcription process and its regulation by histone modification. Blue circles represent core histone octamers, in the lower panel acetylated histone tails are depicted emerging from the octamers. DNA is in orange and the solid black arrow represents active transcription. Sequence-specific DNA binding proteins can recruit histone deacetylases (in the case of repressors) or histone acetyltransferases (in the case of activators) to regulatory sequences. HDAC activity results in histone hypoacetylation, chromatin condensation and transcriptional repression. HAT activity results in hyperacetylated histones, accessible chromatin structure and transcriptional activity.

most thoroughly characterized examples of regulation of this type involve the Class I histone deacetylase enzymes. The Class II deacetylase enzymes are also thought to be recruited to distinct regions of the genome by sequence-specific DNA binding proteins (18–20). However, in the case of the Class II deacetylases, there is clearly an additional level of regulation. These enzymes actively shuttle between the nucleus and cytoplasm and their subcellular distribution appears to be under the control of classic cellular signaling pathways (19,21–24). This review will address how cancer cells manipulate the regulatory functions of histone deacetylases resulting in inappropriate gene expression through the use of two examples. First I will describe the loss of targeting of Class I histone deacetylases through disruption of a transcriptional corepressor. Second, the

The Rb protein and its close relatives, p107 and p130, are critical regulators of cell growth (26–28). They share common properties including induction of cell-cycle arrest when expressed exogenously (29) and binding to E2F transcription factors (30,31). The E2F family of transcription factors, in turn, binds to the promoters of cell-cycle regulators such as cyclin E and of genes required for DNA synthesis (31,32). The binding of Rb family members to E2F family members is regulated by phosphorylation status, hypophosphorylated Rb binds to E2F while hyperphosphorylated Rb is released from this interaction (26–31). Binding of Rb, p107 and p130 to E2F family members requires a common sequence motif termed the ‘pocket’ (26–31). Pocket protein phosphorylation is mediated by cyclin-dependent kinases whose activity is in turn regulated by cyclin-dependent kinase inhibitors such as p16INK4A (26– 32). A primary mechanism for active repression of E2F responsive genes by Rb family members is through the recruitment of histone deacetylases (5,33). HDAC1 binds preferentially to the repressive, hypophosphorylated form of Rb (34). In vitro, this interaction appears to be direct (35,36), and is mediated by a sequence motif (LXCXE) shared with several viral oncoproteins known to bind Rb directly (36,37). Artificial recruitment of Rb through a heterologous DNA binding domain results in transcriptional repression sensitive to inhibitors of histone deacetylase (34–36). Further, Rb-dependent repression of a subset of E2F responsive genes is also sensitive to inhibitors of histone deacetylase (34–36). These data firmly place histone deacetylases as important executors of Rb-dependent repression of E2F responsive genes (Fig. 3). How is this repressive mechanism disrupted in cancer? The Rb/E2F pathway is impaired in many human tumors. Loss of function can result from mutation of Rb, a frequently observed genetic change in malignant cells (26) although mutation of p107 and p130 is relatively infrequent (38,39). However, additional mechanisms exist for cancer cells to bypass normal Rb/ E2F function. Loss of p16INK4A activity results in net increases

Figure 2. Two classes of histone deacetylase. The diagram illustrates the human Class I and Class II histone deacetylases. The red box represents a conserved motif presumed to be the catalytic site conserved in all the Class I and Class II enzymes. Note the presence of two catalytic motifs in HDAC6. A conserved sequence motif found in HDAC4 and HDAC5 is indicated in these two proteins as a yellow bar.


695

in Rb phosphorylation and loss of transcriptional repression function (40). Amplification of cyclin D1 or CDK4 (or increases in its activity via mutation) ultimately results in loss of repression by Rb via phosphorylation (41). A precise definition of the subset of all the genes regulated by the E2F and Rb families remains an important experimental goal. Not surprisingly, the biology of this pathway is quite complicated. The E2F family consists of a minimum of five different proteins and the Rb family consists of three (29,30). These factors exhibit selectivity in the interaction of a specific pocket protein with specific E2F family members, in the point in the cell cycle during which they are active, and in expression patterns in vivo (27,29–31). An additional complication is the inability to attribute a specific functional outcome to a single Rb or E2F family member due to the possibility of functional redundancy. Recently, however, cell lines have been described that ablate all three Rb family members (42,43). These cells exhibit multiple defects centering on the transition from G1 to S phase including a shortened cell cycle (42,43). In addition, the triple knock out cells fail to arrest in G 1 following DNA damage, serum deprivation or contact inhibition, and have a limited capacity to differentiate (42,43). These lines also exhibit some characteristics of immortalized cells including an escape from replicative senescence (42,43). The multiple checkpoint defects exhibited by these cell lines potentially implicate aberrant histone deacetylation patterns in the generation of several of the characteristics common to cancer cells. While pocket proteins clearly possess the capacity to repress by mechanisms other than recruitment of histone deacetylase (27–33), the multiple checkpoint pathways lost in the triple knockout cells suggest that loss of deacetylation represents an important mechanism contributing to multiple aspects of the malignant phenotype. NUCLEAR LOCALIZATION OF CLASS II HDACS, SIGNAL TRANSDUCTION AND DIFFERENTIATION The Class II histone deacetylases were initially characterized as sequence relatives of yeast Hda1p (12–14,19,20). From the time of their initial discovery these enzymes were predicted to be involved in cellular differentiation pathways (13,19,20). In fact, the mRNAs for HDAC4 and HDAC5 are expressed at highest levels in skeletal muscle, heart and brain (12,13), tissues with low levels of mitotic activity. The biochemistry of this class of histone deacetylases, however, has not been explored to the same extent as the Class I enzymes. A paradigm rapidly emerged implicating two Class II deacetylases, HDAC4 and HDAC5, in regulation of genes involved in muscle differentiation (Fig. 4). A key transcriptional regulator in muscle differentiation is the MEF2 transcription factor family (44). The four MEF2 family members are MADS box [MCM1, AGAMOUS, DEFICIENS and SRF (serum response factor)] transcriptional regulators that heterodimerize with basic helix–loop–helix (bHLH) transcription factors such as MyoD (44). Surprisingly, MEF2 interacts directly with a conserved sequence present in both HDAC4 and HDAC5 (Fig. 2) through the MADS box motif (18–20,45). This interaction leads to transcriptional repression of genes involved in muscle differentiation (18,19,45). Activation of these genes requires modification of the HDACs by calcium, calmodulin-dependent protein kinase resulting in phosphoryla-

Figure 3. The Rb/E2F pathway. A series of regulatory interactions involved in E2F/Rb-dependent regulation of transcription. The E2F family of transcription factors regulates genes whose products are required for passage from G1 to S phase of the cell cycle at the transcriptional level. The Rb protein, through recruitment of histone deacetylases, represses E2F activity. The interaction of Rb with E2F is in turn regulated by cyclin-dependent kinases and cdk inhibitors.

tion of residues in their N-termini (22,24). This phosphorylation leads to binding of the HDACs by 14-3-3 proteins and their active export from the nucleus to the cytoplasm (20,22,24,46). Subsequent activation of muscle-specific genes is facilitated by activation of the MEF2 factors by MAP kinase signaling pathways (44). There are currently a number of unanswered questions regarding the Class II histone deacetylases. It is unclear precisely what spectrum of genes they regulate and it is likewise unclear precisely which repressors recruit them to regulatory sequences. However, the MEF2 paradigm makes some interesting predictions regarding this family. The active shuttling of HDAC4 and HDAC5 between compartments takes place in diverse cell types including many unrelated to muscle (18–20,22). Cellular localization probably represents a fundamental regulatory mechanism for these deacetylases. Further, physiologic states resulting in terminal differentiation and loss

696


Figure 4. Cellular localization of Class II histone deacetylases. The Class II deacetylases, as exemplified by HDAC4, are localized in either the cytoplasm or in the nucleus. Phosphorylation of HDAC4 at residues on its N-terminus by calcium, calmodulin-dependent kinase (CaMK) leads to sequestration of the enzyme by 14-3-3 proteins and active transport out of the nucleus. Inactive HDAC4 is a target of the Ras signal transduction cascade. Phosphorylation of HDAC4, presumably on different residues than utilized by CaMK, leads to either active translocation into the nucleus or a release from sequestration by 14-3-3. Nuclear HDAC4 is targeted to specific loci by interaction with transcription factors (such as MEF2) leading to active repression.

of proliferative capacity result in the rapid exclusion of these deacetylases from the nucleus. The observation that HDAC6 also actively shuttles (21) implies that cellular localization is also important for transcriptional regulation by this enzyme. However, in the case of HDAC6, signals leading to import and/ or export remain unknown. The export of HDAC4 and HDAC5 from the nucleus follows a differentiation signal. What happens when cells encounter proliferation signals? The experimental data on Class II histone deacetylase enzymes is still in its infancy, but a recent set of experiments suggests exciting possibilities. HDAC4 was found associated with active versions of the ERK1/ERK2 protein kinases following immunoprecipitation from 293 cells (23). These kinases are components of the Ras-MAPK signal transduction pathway that mediates communication between the cell surface and the nucleus. Activation of this pathway is initiated by binding of ligands, such as growth factors, to cell surface receptors, which leads to Ras activation. Active Ras converts the Raf protein kinase to an active form leading to sequential phosphorylation and activation of the downstream kinases MEK and ERK. The eventual targets of ERK include factors such as cytoskeletal proteins, kinases, phosphatases, enzymes and transcription factors whose modification is crucial to cell growth and proliferation (47). Is cellular localization of HDAC4 modulated by receptor based signaling pathways? The preliminary answer is yes. Transfected HDAC4 fused to green fluorescent protein clearly concentrates in nuclei following transfection with an oncogenic, constitutively active form of Ras (23). Further, similar effects were observed with a constitutively active MEK1 kinase, an intermediate in

the Ras pathway, but not with a dominant negative Ras (23). This clearly suggests that activation of growth promoting pathways can lead to nuclear localization of Class II HDACs. While these results clearly call for further investigation to delineate the effects of activation of the Ras signaling pathway on endogenous histone deacetylases and to identify phosphorylation sites, they lead immediately to an eminently testable hypothesis. The nuclear/cytoplasmic distribution of Class II histone deacetylases might result from a net balance of two opposing activities. Calcium/calmodulin-dependent pathways could lead to active export of these species from the nucleus leading to activation of their target genes. Conversely, a subset of receptor-mediated events could lead to activation of Ras signaling (or potentially other signal transduction pathways), culminating in nuclear localization of deacetylases and transcriptional repression. As Class II deacetylases apparently repress genes crucial for differentiation, their nuclear accumulation could lead to transcriptional silencing of those same genes and a loss of the differentiated phenotype. What can this model add to current pictures of cancer? The hallmarks of the malignant phenotype include loss of differentiated status and decreased reliance on exogenous growth factors. Mutations resulting in constitutive activation of signal transduction pathways, such as the Ras pathway, are among the most frequent genetic changes in cancer cells (48). One can easily rationalize how these mutations lead to decreased reliance on growth factors. Potentially, mislocalization of Class II histone deacetylases may be an unanticipated result of these mutations, leading ultimately to a loss of differentiation through inappropriate deacetylation of their target loci. The differentiationpromoting and anti-proliferative properties of histone deacetylase inhibitors are entirely consistent with this possibility (49). CONCLUSIONS AND FUTURE PERSPECTIVES It is now clear that histone deacetylases belong to the rather large family of factors that regulate transcription. Current models stipulate that HDACs are recruited to target sequences through protein–protein interactions. Increased local concentrations lead to hypoacetylation of core histones, ultimately resulting in transcriptional repression. Well-documented examples exist throughout biology to support these models. However the current models cannot explain all the biology associated with histone deacetylases. For example, recent evidence from Drosophila clearly demonstrates that the distribution of histone deacetylases and transcriptional corepressors is much more widespread in the genome than anticipated (50). In fact, some now propose a requirement for a general monitoring of acetylation status in actively growing cells (51). Emerging data depict histone deacetylases as primary agents for the execution of changes in gene expression. Undoubtedly, these enzymes will prove to be regulated at multiple levels including their abundance, cellular compartmentalization, association with other factors, activity states and genome-wide distribution depending on the physiologic needs of the cell. Clearly, each cell type requires a unique pattern of gene expression, resulting in cell-type-dependent differences in the status of histone deacetylases. As cellular transformation and oncogenesis result in major changes in cell physiology and patterns of gene expression, it is not at all surprising that histone deacetylase biology will also change. It will be both


fascinating and informative to unravel the gene regulatory networks controlled by histone deacetylase and to learn how cancer cells manipulate these pathways for their own purposes. ACKNOWLEDGEMENTS I thank Anne Whalen and Fyodor Urnov for comments on this manuscript. I apologize to my colleagues whose work was not cited here due to space limitations. Work in my laboratory is supported by grants from the National Institute of Child Health and Human Development (1 K22 HD01238-01) and from the Rett Syndrome Research Foundation. I gratefully acknowledge the Massachusetts Rett Syndrome Association for financial support. REFERENCES 1. Annunziato, A.T. and Hansen, J.C. (2000) Role of histone acetylation in the assembly and modulation of chromatin structures. Gene Expr., 9, 37– 61. 2. Grunstein, M. (1997) Histone acetylation in chromatin structure and transcription. Nature, 389, 349–352. 3. Strahl, B.D. and Allis, C.D. (2000) The language of covalent histone modifications. Nature, 403, 41–45. 4. Struhl, K. (1998) Histone acetylation and transcriptional regulatory mechanisms. Genes Dev., 12, 599–606. 5. Cress, W.D. and Seto, E. (2000) Histone deacetylases, transcriptional control and cancer. J. Cell Physiol., 184, 1–16. 6. Buggy, J.J., Sideris, M.L., Mak, P., Lorimer, D.D., McIntosh, B. and Clark, J.M. (2000) Cloning and characterization of a novel human histone deacetylase, HDAC8. Biochem. J., 350, 199–205. 7. Hu, E., Chen, Z., Fredrickson, T., Zhu, Y., Kirkpatrick, R., Zhang, G.F., Johanson, K., Sung, C.M., Liu, R. and Winkler, J. (2000) Cloning and characterization of a novel human class I histone deacetylase that functions as a transcription repressor. J. Biol. Chem., 275, 15254–15264. 8. Van den Wyngaert, I., de Vries, W., Kremer, A., Neefs, J., Verhasselt, P., Luyten, W.H. and Kass, S.U. (2000) Cloning and characterization of human histone deacetylase 8. FEBS Lett., 478, 77–83. 9. Ng, H.H. and Bird, A. (2000) Histone deacetylases: silencers for hire. Trends Biochem. Sci., 25, 121–126. 10. Wen, Y.D., Perissi, V., Staszewski, L.M., Yang, W.M., Krones, A., Glass, C.K., Rosenfeld, M.G. and Seto, E. (2000) The histone deacetylase-3 complex contains nuclear receptor corepressors. Proc. Natl Acad. Sci. USA, 97, 7202–7207. 11. Li, J., Wang, J., Nawaz, Z., Liu, J.M., Qin, J. and Wong, J. (2000) Both corepressor proteins SMRT and N-CoR exist in large protein complexes containing HDAC3. EMBO J., 19, 4342–4350. 12. Grozinger, C.M., Hassig, C.A. and Schreiber, S.L. (1999) Three proteins define a class of human histone deacetylases related to yeast Hda1p. Proc. Natl Acad. Sci. USA, 96, 4868–4873. 13. Verdel, A. and Khochbin, S. (1999) Identification of a new family of higher eukaryotic histone deacetylases. Coordinate expression of differentiation-dependent chromatin modifiers. J. Biol. Chem., 274, 2440–2445. 14. Kao, H.Y., Downes, M., Ordentlich, P. and Evans, R.M. (2000) Isolation of a novel histone deacetylase reveals that class I and class II deacetylases promote SMRT-mediated repression. Genes Dev., 14, 55–66. 15. Knoepfler, P.S. and Eisenman, R.N. (1999) Sin meets NuRD and other tails of repression. Cell, 99, 447–450. 16. Rundlett, S.E., Carmen, A.A., Suka, N., Turner, B.M. and Grunstein, M. (1998) Transcriptional repression by UME6 involves deacetylation of lysine 5 of histone H4 by RPD3. Nature, 392, 831–835. 17. Kadosh, D. and Struhl, K. (1998) Targeted recruitment of the Sin3-Rpd3 histone deacetylase complex generates a highly localized domain of repressed chromatin in vivo. Mol. Cell. Biol., 18, 5121–5127. 18. Lemercier, C., Verdel, A., Galloo, B., Curtet, S., Brocard, M.P. and Khochbin, S. (2000) mHDA1/HDAC5 histone deacetylase interacts with and represses MEF2A transcriptional activity. J. Biol. Chem., 275, 15594– 15599.

697

19. Miska, E.A., Karlsson, C., Langley, E., Nielsen, S.J., Pines, J. and Kouzarides, T. (1999) HDAC4 deacetylase associates with and represses the MEF2 transcription factor. EMBO J., 18, 5099–5107. 20. Wang, A.H., Bertos, N.R., Vezmar, M., Pelletier, N., Crosato, M., Heng, H.H., Th′ng, J., Han, J. and Yang, X.J. (1999) HDAC4, a human histone deacetylase related to yeast HDA1, is a transcriptional corepressor. Mol. Cell. Biol., 19, 7816–7827. 21. Verdel, A., Curtet, S., Brocard, M.P., Rousseaux, S., Lemercier, C., Yoshida, M. and Khochbin, S. (2000) Active maintenance of mHDA2/ mHDAC6 histone-deacetylase in the cytoplasm. Curr. Biol., 10, 747–749. 22. McKinsey, T.A., Zhang, C.L., Lu, J. and Olson, E.N. (2000) Signaldependent nuclear export of a histone deacetylase regulates muscle differentiation. Nature, 408, 106–111. 23. Zhou, X., Richon, V.M., Wang, A.H., Yang, X.J., Rifkind, R.A. and Marks, P.A. (2000) Histone deacetylase 4 associates with extracellular signal-regulated kinases 1 and 2 and its cellular localization is regulated by oncogenic Ras. Proc. Natl Acad. Sci. USA, 97, 14329–14333. 24. Grozinger, C.M. and Schreiber, S.L. (2000) Regulation of histone deacetylase 4 and 5 and transcriptional activity by 14-3-3-dependent cellular localization. Proc. Natl Acad. Sci. USA, 97, 7835–7840. 25. Lin, R.J., Egan, D.A. and Evans, R.M. (1999) Molecular genetics of acute promyelocytic leukemia. Trends Genet., 15, 179–184. 26. Weinberg, R.A. (1995) The retinoblastoma protein and cell cycle control. Cell, 81, 323–330. 27. Mulligan, G. and Jacks, T. (1998) The retinoblastoma gene family: cousins with overlapping interests. Trends Genet., 14, 223–229. 28. Sellers, W.R. and Kaelin, W.G., Jr (1997) Role of the retinoblastoma protein in the pathogenesis of human cancer. J. Clin. Oncol., 15, 3301–3312. 29. Harbour, J.W. and Dean, D.C. (2000) The Rb/E2F pathway: expanding roles and emerging paradigms. Genes Dev., 14, 2393–2409. 30. Dyson, N. (1998) The regulation of E2F by pRB-family proteins. Genes Dev., 12, 2245–2262. 31. Nevins, J.R. (1998) Toward an understanding of the functional complexity of the E2F and retinoblastoma families. Cell Growth Differ., 9, 585– 593. 32. Johnson, D.G. and Schneider-Broussard, R. (1998) Role of E2F in cell cycle control and cancer. Front. Biosci., 3, 447–448. 33. Brehm, A. and Kouzarides, T. (1999) Retinoblastoma protein meets chromatin. Trends Biochem. Sci., 24, 142–145. 34. Luo, R.X., Postigo, A.A. and Dean, D.C. (1998) Rb interacts with histone deacetylase to repress transcription. Cell, 92, 463–473. 35. Brehm, A., Miska, E.A., McCance, D.J., Reid, J.L., Bannister, A.J. and Kouzarides, T. (1998) Retinoblastoma protein recruits histone deacetylase to repress transcription. Nature, 391, 597–601. 36. Magnaghi-Jaulin, L., Groisman, R., Naguibneva, I., Robin, P., Lorain, S., Le Villain, J.P., Troalen, F., Trouche, D. and Harel-Bellan, A. (1998) Retinoblastoma protein represses transcription by recruiting a histone deacetylase. Nature, 391, 601–605. 37. Nevins, J.R. (1994) Cell cycle targets of the DNA tumor viruses. Curr. Opin. Genet. Dev., 4, 130–134. 38. Takimoto, H., Tsukuda, K., Ichimura, K., Hanafusa, H., Nakamura, A., Oda, M., Harada, M. and Shimizu, K. (1998) Genetic alterations in the retinoblastoma protein-related p107 gene in human hematologic malignancies. Biochem. Biophys. Res. Commun., 251, 264–268. 39. Claudio, P.P., Howard, C.M., Pacilio, C., Cinti, C., Romano, G., Minimo, C., Maraldi, N.M., Minna, J.D., Gelbert, L., Leoncini, L. et al. (2000) Mutations in the retinoblastoma-related gene RB2/p130 in lung tumors and suppression of tumor growth in vivo by retrovirus-mediated gene transfer. Cancer Res., 60, 372–382. 40. Sherr, C.J. (1996) Cancer cell cycles. Science, 274, 1672–1677. 41. Hall, M. and Peters, G. (1996) Genetic alterations of cyclins, cyclindependent kinases and Cdk inhibitors in human cancer. Adv. Cancer Res., 68, 67–108. 42. Dannenberg, J.H., van Rossum, A., Schuijff, L. and te Riele, H. (2000) Ablation of the retinoblastoma gene family deregulates G(1) control causing immortalization and increased cell turnover under growth-restricting conditions. Genes Dev., 14, 3051–3064. 43. Sage, J., Mulligan, G.J., Attardi, L.D., Miller, A., Chen, S., Williams, B., Theodorou, E. and Jacks, T. (2000) Targeted disruption of the three Rbrelated genes leads to loss of G(1) control and immortalization. Genes Dev., 14, 3037–3050. 44. Black, B.L. and Olson, E.N. (1998) Transcriptional control of muscle development by myocyte enhancer factor-2 (MEF2) proteins. Annu. Rev. Cell. Dev. Biol., 14, 167–196.

698


45. Lu, J., McKinsey, T.A., Nicol, R.L. and Olson, E.N. (2000) Signaldependent activation of the MEF2 transcription factor by dissociation from histone deacetylases. Proc. Natl Acad. Sci. USA, 97, 4070–4075. 46. McKinsey, T.A., Zhang, C.L. and Olson, E.N. (2000) Activation of the myocyte enhancer factor-2 transcription factor by calcium/calmodulindependent protein kinase-stimulated binding of 14-3-3 to histone deacetylase 5. Proc. Natl Acad. Sci. USA, 97, 14400–14405. 47. Seger, R. and Krebs, E.G. (1995) The MAPK signaling cascade. FASEB J., 9, 726–735.

48. Hanahan, D. and Weinberg, R.A. (2000) The hallmarks of cancer. Cell, 100, 57–70. 49. Marks, P.A., Richon, V.M. and Rifkind, R.A. (2000) Histone deacetylase inhibitors: Inducers of differentiation or apoptosis of transformed cells. J. Natl Cancer Inst., 92, 1210–1216. 50. Pile, L.A. and Wassarman, D.A. (2000) Chromosomal localization links the SIN3-RPD3 complex to the regulation of chromatin condensation, histone acetylation and gene expression. EMBO J., 19, 6131–6140. 51. Vogelauer, M., Wu, J., Suka, N. and Grunstein, M. (2000) Global histone acetylation and deacetylation in yeast. Nature, 408, 495–498.