233As alterations in gene expression underlie a considerableproportion of human diseases, correcting such aberranttranscription in vivo is expected to provide therapeutic benefitto the patient. Attempts to control endogenous mammaliangenes, however, face a significant obstacle in the form ofchromatin. Aberrant gene repression can be alleviated by usingsmall-molecule inhibitors that exert nucleus-wide effects onchromatin-based repressors. Genome-wide chromatinremodeling also occurs during cloning via nuclear transfer, andcauses the deregulation of epigenetically controlled genes.Regulation of genes in vivo can be accomplished via the useof designed transcription factors — these result from a fusion ofa designed DNA-binding domain based on the zinc fingerprotein motif to a functional domain of choice.AddressesSangamo Biosciences, Pt Richmond Tech Center, 501 Canal Blvd,Suite A100, Richmond, California 94804, USACorrespondence: Fyodor D Urnov; e-mail: [email protected] Opinion in Genetics & Development 2002, 12:233–2420959-437X/02/$ — see front matter© 2002 Elsevier Science Ltd. All rights reserved.AbbreviationsAza-C 5-azacytidineDNMT DNA methyltransferaseES embryonic stemHAT histone acetyltransferaseHDAC hstone deacetylaseKAP-1 KRAB repressor domain associated protein-1LOS large offspring syndrome MBD methylated DNA-binding proteinMEL murine erythroid leukemiaVEGF A vascular endothelial growth factor AZFP zinc-finger proteinIntroductionEverything that happens to the eukaryotic genome in vivooccurs on a chromatin template [1]. Given the relativeobscurity to which the histones were relegated for the first~25 years since the discovery by Williamson and byHewish and Burgoyne that chromatin had a ‘sub-structure’[1], their rise to dominance over genome biology docu-mented by the many articles in this issue of Current Opinionin Genetics & Development offers a case study in paradigmreversal. Concomitant with an exponential increase in ourunderstanding of genome-control pathways came a grow-ing realization that the etiology of human disease can, in avery large number of cases, be traced to the misregulationof particular genes. This gave rise to the notion of ‘transcriptional therapy’ [2], in which the genome is controlled for therapeutic purposes.However exciting our growing knowledge about thegenome may be, its failure to be broadly useful to clinicalpractice was well described by Lewontin in 1991: “… None of the [advances of 20thcentury medicine]depend on a deep knowledge of cellular processes or onany discoveries of molecular biology. Cancer is still treatedby gross physical and chemical assaults on the offendingtissue. Cardiovascular disease is treated by surgery whoseanatomical bases go back to the 19th century … Of course,intimate knowledge of the living cell and of basic molecularprocesses may be useful eventually” [3]. A decade later,molecular biology can claim very few successes for drugs inclinical use that were designed ab initio to control a specificcomponent of a pathway linked to disease: these includethe monoclonal antibody Herceptin®(trastuzumab) [4]and the kinase inhibitor STI571 (Gleevec™) [5].Nevertheless, it is likely that the paucity of available molecular biology therapies may soon end. Such optimismis founded on an unprecedented array of technicaladvances in whole-genome analysis (see article by Wyrickand Young, this issue [pp 130–136]), high-throughput drugselection and screening, and the development of noveltechnologies to regulate genes in vivo that is based on agrowing understanding of the interplay between chromatinand genome control.A treatment for repressionThe burden of repetitive DNA carried by genomes ofmetazoa, and selective pressure to evolve mechanisms foracute gene regulation have yielded a complex machineryfor transcriptional repression (see articles by Berger[pp 142–148], Grewal and Elgin [pp 178–187], Kouzarides[pp 198–209], and Simon and Tamkun [pp 210–218] in thisissue). It includes the histone deacetylases (HDACs) andmethyltransferases, ATPases such as the Mi-2/NRD com-plex [6], the Polycomb proteins, and various components ofrepressive chromatin such as HP1. In addition, higher vertebrates possess a DNA methylation-based host genomedefense system [7] that includes a number of DNA methyl-transferases (DNMTs) [8] and methylated DNA-bindingproteins (MBDs) [9]. The majority, if not all, of the afore-mentioned components of the chromatin-based repressionmachinery collude with the DNMTs and the MBDs toeffect gene silencing over methylated DNA loci [10]. Aberrant gene repression underlies a variety of human disease, most notably cancer. Fortunately, various steps inthis repression pathway can be inhibited by potent small-molecule agents such as the HDAC inhibitors [11] andDNMT inhibitors [12] — both have been used in clinicalpractice to achieve a therapeutic alleviation of diseasesymptoms. Interestingly, the DNMT inhibitor 5-azacytidine(Aza-C) was first used to treat leukemia in 1971 [12] —that is, well before the discovery that cell-cycle misregula-tion in cancer is at least partly effected via aberrant DNAmethylation and silencing of key regulators such asBiotechnologies and therapeutics: chromatin as a targetAndreas Reik, Philip D Gregory and Fyodor D Urnovp16INK4A[13]. In general, there appears to be a significantgap between the finding that compound X is clinicallyeffective in the treatment of a given disease, and ourunderstanding of how the biological properties of the molecules targeted by that compound contribute to theetiology of that disease. Because a number of highly effec-tive inhibitors of the metazoan repression machinery havebeen developed and are being tested in clinical practice[14–16], there is a pressing need to evaluate their utility inthe context of well-studied disease models to determinewhether their efficacy is as a result of the presumed modeof their function, or some unanticipated, albeit beneficial,side effect. An example of this approach is provided bywork studying the influence of the human protein huntingtin on Drosophila neural development [17•]. Earlierfindings suggested that huntingtin interacts
View Full Document
Unlocking...