Communication
Periodic Epi-organization of the Yeast Genome Revealed by the Distribution of Promoter Sites

https://doi.org/10.1016/S0022-2836(03)00535-7Get rights and content

Abstract

The organization of transcription within the eukaryotic nucleus may be expected to both depend on and determine the structure of the chromosomes. This study shows that, in yeast, genes that are controlled by the same sequence-specific transcription factor tend to be regularly spaced along the chromosome arms; a similar period characterizes the spacing of origins of replication, although periodicity is less pronounced. The same period is found for most transcription factors within a chromosome arm. However, different periods are observed for different chromosome arms, making it unlikely that periodicity is caused by dedicated scaffolding proteins. Such regularities are consistent with a genome-wide loop model of chromosomes, in which coregulated genes tend to dynamically colocalize in 3D. This colocalization may also involve co-regulated genes belonging to different chromosomes, as suggested by partial conservation of the respective positioning of different transcription factors around the loops. Thus, binding at genuine regulatory sites on DNA would be optimized by locally increasing the concentration of multimeric transcription factors. In this model, self-organization of transcriptional initiation plays a major role in the functional nuclear architecture.

Introduction

Several specialized regions have been discovered within the eukaryotic nucleus, including discrete chomosome territories, peripheral telomeres, and the association of replication- and transcription-active chromosomal sites into foci. The best-characterized nuclear compartment and the only one that has a clearly identified function is the nucleolus. Nucleoli bring together many chromosomes carrying rRNA encoding genes and organize them into a transcriptionally-efficient organelle that also processes and assembles ribosomal components.1., 2., 3. Other, less well-characterized nuclear compartments such as Cajal bodies, speckles, gems and promyelocytic leukemia bodies are engaged in various aspects of acid nucleic metabolism in metazoans (reviewed in Ref. 4). These long-lived compartments are essentially dynamic structures composed of mobile elements.5., 6. Recent studies of chromatin structure have revealed that chromosomes are compartmentalized into discrete territories and it has been suggested that gene positioning in a territory or an interchromatin compartment influences its regulation.7 RNA polymerase II, the enzyme responsible for transcribing protein-coding genes, is found with its associated factors and transcripts in several thousand nuclear foci with diameters 40–80 nm, that together occupy <0.5% nuclear volume, much less than that of active chromatin.8 Each focus, sometimes called a “transcription factory”, typically contains tens of enzymes and transcripts.8., 9.

Some regional specialization is also apparent in the much smaller yeast nucleus. In yeast, centromeres cluster near one pole of the interphasic nucleus in a microtubule-dependent fashion, while chromosome arms extend outwards from this pole, and the nucleolus is opposite to this pole.10 The nuclear periphery harbors spatially confined telomeric domains that are actively involved in transcriptional silencing11 (reviewed in Refs 12., 13.). Additional evidence supports an interrelationship between nuclear architecture and transcriptional activity. For instance, silencing defects14 or transcriptional response to mating pheromone15 were found to induce alterations of the nuclear morphology. Actively transcribed tRNA genes repress adjacent protein-encoding genes, probably through a subnuclear localization effect.16

Eukaryotes share a conserved mechanism of transcriptional regulation.17 RNA polymerase II and the general transcription factors assemble in a regulated fashion. However, the predominant level of control involves dedicated transcription factors, or “regulators”, that bind to short DNA sequence motifs in the cis-regulatory region of a protein-encoding gene and activate or repress its transcription. The transcription of many genes is cooperatively regulated via one or several regulators.17 Multivalent regulators bind to closely spaced sites in the regulatory region of a gene and this close spacing is in part responsible for increasing the local concentration of the regulators such that binding to one site facilitates binding to a second site.18 Similarly, DNA can be used as a scaffold to build unique nucleoprotein complexes that can establish long-range interactions with similarly unique complexes.18., 19.

Here, I analyse the transcriptional organization of the yeast, Saccharomyces cerevisiae, and reveal a remarkable periodicity. This periodicity can be parsimoniously explained by invoking a role for local concentrations and transcriptional dynamics in the optimization of nuclear functions via the generation of spatial chromosomal patterns. It suggests a functional architecture of the nucleus in which coregulated genes tend to colocalize in 3D.

Section snippets

Periodicity of coregulated gene positions

Rap1p is the dedicated regulator in yeast whose gene targets are most numerous and most expressed.20 This was determined by chromatin immunoprecipitation (ChIP) which measures in vivo occupancy of DNA sites by a chosen protein, using DNA/protein crosslinking. Regularities were sought in the distances between pairs of Rap1p targets from the same chromosome. The distribution of such distances has a peak for the shortest distances (Figure 1(a), top). This indicates that coregulated genes tend to

Diversity of chromosome arms

Table 1 shows that the spacing of coregulated genes vary between the chromosome arms. These periods were separately determined as in Figure 1(b) for each of the 32 yeast chromosome arms, corresponding to rows in Table 1. They were independently determined from three datasets,20., 23., 26., 27. corresponding to the three last columns of Table 1. The first set comprises 124 regulators studied by classical genetic and biochemical means,27 the other sets comprise 320., 26. and 10623 regulators

Periodicity of replication origins

To test whether this distinctive organization pervades other aspects of DNA metabolism, replication origins, termed autonomously replicating sequences (ARS), were analysed in a similar fashion, using recent ChIP mapping data.28 The ARS of yeast chromosome IX tend to be spaced by 31 kbp or multiples thereof (Figure 1(c), top), like coregulated genes, albeit with a 15.5 kbp shift of the peaks after distance 124 kbp. Regularities are not observed when gene positions are attributed at random (Figure

Coregulation or colocalization

In ChIP experiments, DNA/regulator crosslinking is solely based on spatial proximity. In view of the above results, a “backwards” interpretation of the ChIP data could be that genes are generally crosslinked to a regulator, not because they are its targets, but because they colocalize with a true target due to some unidentified long-range structure. If the backwards interpretation were correct, assessment of the crosslinking data with decreasing stringency should produce increasingly long lists

Full-genome transcriptional scheme

The function served by the regular spacing of coregulated genes should not be restricted to intra-chromosomal use, as chromosomes appear as rather arbitrary boundaries to gene distribution when close species are compared. However, any inter-chromosomal regularity would imply enormous topological constraints, since each chromosome arm is already constrained to a different period which holds for most, if not all, regulators (Table 1). The periodicity of target positions suggests to represent a

Molecular and evolutionary basis

Spatial confinement of target genes synergistically increases the local concentration of their multimeric regulator just where it is required. It thus favors binding at specific DNA consensus sites through elevated local concentration, thereby sequestering the regulator away from spurious DNA binding sites. One way to achieve this local concentration effect is to cluster target genes linearly along the chromosome. Contiguity indeed occurs in yeast coregulated genes (Figure 1).21., 22. However,

Conclusion

The concept of a genome-level transcriptional scheme readily explains the specificity, responsiveness and versatility of transcriptional and replicational regulations, despite the overwhelming abundance of unused protein binding motifs.20., 25., 28., 29. It suggests a guideline for discriminating genuine protein binding sites among those detected through sequence analysis or ChIP. This concept also accounts for the optimization of gene control by several regulators, despite combinatorial

Supplementary Files

Acknowledgements

I am grateful to Vic Norris, Benno Müller-Hill and Eric H. Davidson for critically reading this paper. Supported by funding from CNRS, genopole® and Conseil Régional d'Ile-de-France.

References (34)

  • R.D. Phair et al.

    High mobility of proteins in the mammalian cell nucleus

    Nature

    (2000)
  • T. Cremer et al.

    Chromosome territories, nuclear architecture and gene regulation in mammalian cells

    Nature Rev. Genet.

    (2001)
  • D.A. Jackson et al.

    Numbers and organization of RNA polymerases, nascent transcripts, and transcription units in HeLa nuclei

    Mol. Biol. Cell

    (1998)
  • A. Pombo et al.

    Bridging the resolution gap: imaging the same transcription factories in cryosections by light and electron microscopy

    J. Histochem. Cytochem.

    (1999)
  • Q.W. Jin et al.

    Centromere clustering is a major determinant of yeast interphase nuclear organization

    J. Cell Sci.

    (2000)
  • F. Feuerbach et al.

    Nuclear architecture and spatial positioning help establish transcriptional states of telomeres in yeast

    Nature Cell Biol.

    (2002)
  • J.P. Cooper

    Telomere transitions in yeast: the end of the chromosome as we know it

    Curr. Opin. Genet. Dev.

    (2000)
  • Cited by (33)

    • The layout of a bacterial genome

      2012, FEBS Letters
      Citation Excerpt :

      On the one hand, chromosomal proximity has been found to be effectively limited to 20 successive genes in enterobacteria [70]. On the other hand, periodic positioning has been observed for microbial genes that are either co-regulated [71–73], co-expressed [74], evolutionarily correlated [75,70], or highly codon-biased [76]. Képès has postulated the existence of a positive feedback loop connecting periodic genome layout to the cellular conformation of chromosomes onto transcriptional control [71].

    • Synthetic and integrative biology

      2011, Comptes Rendus Chimie
    • A thermodynamic model for the agglomeration of DNA-looping proteins

      2008, Journal of Statistical Mechanics: Theory and Experiment
    View all citing articles on Scopus
    View full text