B RIEFINGS IN FUNC TIONAL GENOMICS . VOL 14. NO 1. 1^2
doi:10.1093/bfgp/elu050
Editorial
Recent progress in understanding transcription factor binding specificity Methodologies that profile the DNA binding specificity of a transcription factor in vitro have also been put to excellent use in recent years. Siggers and colleagues (3) review the widely used protein binding microarray (PBM) technology. They highlight recent PBM studies of protein variants and multi-protein complexes. Affinity-based selection of random pools of DNA molecules followed by massively parallel sequencing (HT-SELEX or SELEX-seq) has become a powerful alternative to microarray-based approaches. A paper from the Stormo lab (4) describes a new variant of this technology and puts it in context. Purely computational strategies have also been explored. Havranek, Bradley and colleagues (5) review how supercomputers can be used to simulate how amino-acid side chains interact with DNA nucleotides, and thereby predict binding free energies. They focus on the use of accurate potentials, backbone flexibility and the role of water molecules at the protein–DNA interface. Genomic targeting by transcription factors is notoriously context dependent. Even the best models of in vitro protein–DNA interaction have limited power to predict in vivo occupancy. Competitive interactions with nucleosomes are an important aspect of this context dependence. Morozov and coworkers (6) summarize our current understanding of how nucleosomes alter transcription factor binding in vivo. Epigenetic modifications of DNA are often associated with gene expression changes. However, the mechanisms underlying these associations mostly remain obscure. Rohs, Bussemaker and coworkers (7) review our current understanding of how cytosine methylation affects protein–DNA interaction. They propose that subtle methylation-induced
ß The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://bfg.oxfordjournals.org/ at University of California, San Francisco on April 10, 2015
Gene expression levels can vary greatly from gene to gene and between individuals. To understand how these differences arise, and be able to predict and manipulate them, we need to dissect the molecular mechanisms by which the regulatory programs embedded in the genome are interpreted by the cellular machinery. This issue of Briefings in Functional Genomics provides an overview of the available approaches for quantifying the nucleotide binding specificity of trans-acting factors, a prerequisite for understanding and predicting gene regulatory network function. Transcription factors typically belong to a structural family containing many other proteins with a similar amino acid sequence. Even when the difference in nucleotide sequence preference between such proteins is subtle, their target genes, and thus the processes that they control, can be quite distinct. In a pioneering microfluidics study, Maerkl and Quake (1) showed that it is possible to predict such functional differences using a quantitative, purely sequence-based, thermodynamic model. Comprehensive and accurate quantification of the DNA binding specificity of all transcription factors encoded in the genome may therefore transform our ability to make functional predictions about the regulatory network of the cell. A first review by the Noyes laboratory (2) describes the bacterial one-hybrid (B1H) approach, in which protein–DNA interaction strength is measured via expression of a reporter gene. This technology has been used to perform in-depth analyses of the variation in DNA binding specificity within the homeodomain and zinc finger families. A unique advantage of the B1H approach is the ease with which a large number of different protein sequences can be assayed in parallel.
2
Editorial interactions have been studied in detail will require many more years of experimental and computational effort, but we are off to a good start. Harmen J. Bussemaker Department of Biological Sciences, Columbia University, 1212 Amsterdam Ave, MC 2441, NewYork, NY 10027, USA [email protected]
References 1.
2.
3.
4.
5.
6.
7.
8.
Maerkl SJ, Quake SR. A systems approach to measuring the binding energy landscapes of transcription factors. Science 2007;315(5809):233–7. Xu DJ, Noyes MB. Understanding DNA-binding specificity by bacteria hybrid selection. Brief Funct Genomics 2015; 14:3–16. Andrilenas KK, Penvose A, Siggers T. Using proteinbinding microarrays to study transcription factor specificity: homologs, isoforms and complexes. Brief Funct Genomics 2015;14:17–29. Stormo GD, Zuo Z, Chang YK. Spec-seq: determining protein-DNA-binding specificity by sequencing. Brief Funct Genomics 2015;14:30–8. Joyce AP, Zhang C, Bradley P, Havranek JJ. Structurebased modeling of protein: DNA specificity. Brief Funct Genomics 2015;14:39–49. Chereji RV, Morozov AV. Functional roles of nucleosome stability and dynamics. Brief Funct Genomics 2015;14: 50–60. Dantas Machado AC, Zhou T, Rao S, et al. Evolving insights on how cytosine methylation affects protein-DNA binding. Brief Funct Genomics 2015;14:61–73. Cook KB, Hughes TR, Morris QD. High-throughput characterization of protein-RNA interactions. Brief Funct Genomics 2015;14:74–89.
Downloaded from http://bfg.oxfordjournals.org/ at University of California, San Francisco on April 10, 2015
changes in the geometry of DNA can modulate the interaction with proteins via the minor groove in a highly context-specific manner. Control of local chromatin state and transcription rates by DNA binding factors is only one side of the coin of gene expression regulation. Equally important are the sequence-specific RNA binding factors that interpret the signals embedded in RNA transcripts, which control splicing, transport and turnover. The complexity of this posttranscriptional regulatory system rivals that of the transcriptional machinery. Morris, Hughes and colleagues (8) review the use of high-throughput technologies for quantifying the sequence specificity of RNA binding proteins, paying particular attention to the role of secondary structure. In conclusion, thanks to the efforts of various groups, binding specificity information is now available for a significant fraction of DNA and RNA binding factors in human and other eukaryotes. Nevertheless, we have only scratched the surface of what we need to know to predict expression from sequence. Much more experimental and computational work will be required to extend coverage to all transcription factors and increase the accuracy and precision of the sequence-to-affinity models that can currently be derived from the data to a level where functional differences between closely related family members can be resolved. It has also become clear that transcription factors can greatly influence each other’s interaction with DNA when they bind as a complex. Extending these insights beyond the small number of specific cases where such complex
doi:10.1093/bfgp/elu050
Editorial
Recent progress in understanding transcription factor binding specificity Methodologies that profile the DNA binding specificity of a transcription factor in vitro have also been put to excellent use in recent years. Siggers and colleagues (3) review the widely used protein binding microarray (PBM) technology. They highlight recent PBM studies of protein variants and multi-protein complexes. Affinity-based selection of random pools of DNA molecules followed by massively parallel sequencing (HT-SELEX or SELEX-seq) has become a powerful alternative to microarray-based approaches. A paper from the Stormo lab (4) describes a new variant of this technology and puts it in context. Purely computational strategies have also been explored. Havranek, Bradley and colleagues (5) review how supercomputers can be used to simulate how amino-acid side chains interact with DNA nucleotides, and thereby predict binding free energies. They focus on the use of accurate potentials, backbone flexibility and the role of water molecules at the protein–DNA interface. Genomic targeting by transcription factors is notoriously context dependent. Even the best models of in vitro protein–DNA interaction have limited power to predict in vivo occupancy. Competitive interactions with nucleosomes are an important aspect of this context dependence. Morozov and coworkers (6) summarize our current understanding of how nucleosomes alter transcription factor binding in vivo. Epigenetic modifications of DNA are often associated with gene expression changes. However, the mechanisms underlying these associations mostly remain obscure. Rohs, Bussemaker and coworkers (7) review our current understanding of how cytosine methylation affects protein–DNA interaction. They propose that subtle methylation-induced
ß The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://bfg.oxfordjournals.org/ at University of California, San Francisco on April 10, 2015
Gene expression levels can vary greatly from gene to gene and between individuals. To understand how these differences arise, and be able to predict and manipulate them, we need to dissect the molecular mechanisms by which the regulatory programs embedded in the genome are interpreted by the cellular machinery. This issue of Briefings in Functional Genomics provides an overview of the available approaches for quantifying the nucleotide binding specificity of trans-acting factors, a prerequisite for understanding and predicting gene regulatory network function. Transcription factors typically belong to a structural family containing many other proteins with a similar amino acid sequence. Even when the difference in nucleotide sequence preference between such proteins is subtle, their target genes, and thus the processes that they control, can be quite distinct. In a pioneering microfluidics study, Maerkl and Quake (1) showed that it is possible to predict such functional differences using a quantitative, purely sequence-based, thermodynamic model. Comprehensive and accurate quantification of the DNA binding specificity of all transcription factors encoded in the genome may therefore transform our ability to make functional predictions about the regulatory network of the cell. A first review by the Noyes laboratory (2) describes the bacterial one-hybrid (B1H) approach, in which protein–DNA interaction strength is measured via expression of a reporter gene. This technology has been used to perform in-depth analyses of the variation in DNA binding specificity within the homeodomain and zinc finger families. A unique advantage of the B1H approach is the ease with which a large number of different protein sequences can be assayed in parallel.
2
Editorial interactions have been studied in detail will require many more years of experimental and computational effort, but we are off to a good start. Harmen J. Bussemaker Department of Biological Sciences, Columbia University, 1212 Amsterdam Ave, MC 2441, NewYork, NY 10027, USA [email protected]
References 1.
2.
3.
4.
5.
6.
7.
8.
Maerkl SJ, Quake SR. A systems approach to measuring the binding energy landscapes of transcription factors. Science 2007;315(5809):233–7. Xu DJ, Noyes MB. Understanding DNA-binding specificity by bacteria hybrid selection. Brief Funct Genomics 2015; 14:3–16. Andrilenas KK, Penvose A, Siggers T. Using proteinbinding microarrays to study transcription factor specificity: homologs, isoforms and complexes. Brief Funct Genomics 2015;14:17–29. Stormo GD, Zuo Z, Chang YK. Spec-seq: determining protein-DNA-binding specificity by sequencing. Brief Funct Genomics 2015;14:30–8. Joyce AP, Zhang C, Bradley P, Havranek JJ. Structurebased modeling of protein: DNA specificity. Brief Funct Genomics 2015;14:39–49. Chereji RV, Morozov AV. Functional roles of nucleosome stability and dynamics. Brief Funct Genomics 2015;14: 50–60. Dantas Machado AC, Zhou T, Rao S, et al. Evolving insights on how cytosine methylation affects protein-DNA binding. Brief Funct Genomics 2015;14:61–73. Cook KB, Hughes TR, Morris QD. High-throughput characterization of protein-RNA interactions. Brief Funct Genomics 2015;14:74–89.
Downloaded from http://bfg.oxfordjournals.org/ at University of California, San Francisco on April 10, 2015
changes in the geometry of DNA can modulate the interaction with proteins via the minor groove in a highly context-specific manner. Control of local chromatin state and transcription rates by DNA binding factors is only one side of the coin of gene expression regulation. Equally important are the sequence-specific RNA binding factors that interpret the signals embedded in RNA transcripts, which control splicing, transport and turnover. The complexity of this posttranscriptional regulatory system rivals that of the transcriptional machinery. Morris, Hughes and colleagues (8) review the use of high-throughput technologies for quantifying the sequence specificity of RNA binding proteins, paying particular attention to the role of secondary structure. In conclusion, thanks to the efforts of various groups, binding specificity information is now available for a significant fraction of DNA and RNA binding factors in human and other eukaryotes. Nevertheless, we have only scratched the surface of what we need to know to predict expression from sequence. Much more experimental and computational work will be required to extend coverage to all transcription factors and increase the accuracy and precision of the sequence-to-affinity models that can currently be derived from the data to a level where functional differences between closely related family members can be resolved. It has also become clear that transcription factors can greatly influence each other’s interaction with DNA when they bind as a complex. Extending these insights beyond the small number of specific cases where such complex
