ChIP-ping the branches of the tree: functional genomics and the evolution of eukaryotic gene regulation

Abstract

Advances in the methods for detecting protein-DNA interactions have played a key role in determining the directions of research into the mechanisms of transcriptional regulation. The most recent major technological transformation happened a decade ago, with the move from using tiling arrays [chromatin immunoprecipitation (ChIP)-on-Chip] to high-throughput sequencing (ChIP-seq) as a readout for ChIP assays. In addition to the numerous other ways in which it is superior to arrays, by eliminating the need to design and manufacture them, sequencing also opened the door to carrying out comparative analyses of genome-wide transcription factor occupancy across species and studying chromatin biology in previously less accessible model and nonmodel organisms, thus allowing us to understand the evolution and diversity of regulatory mechanisms in unprecedented detail. Here, we review the biological insights obtained from such studies in recent years and discuss anticipated future developments in the field.

Figure 1.

Patterns and mechanisms of regulatory element, transcription factor occupancy and gene expression evolution. (A) Patterns of conservation and divergence. Functional conservation of gene expression (top row) can be achieved through conservation of both regulatory elements and transcription factor occupancy (left), but it can also be maintained in the presence of significant turnover, both of occupancy and at the sequence level, either through replacement of individual transcription factors occupying an orthologous regulatory region (right) or by the evolution of nonorthologous regulatory regions (middle). Loss of regulatory elements and alterations in transcription factor occupancy can also lead to changes in gene expression (bottom row). (B and C) Mechanisms of transcription factor occupancy and regulatory element evolution. (B) Mechanisms of loss of transcription factor binding: loss of cognate motif (top), loss of cofactor binding (middle), loss of pioneer factor binding (bottom); (C) Mechanisms of gain of transcription factor binding and de novo evolution of regulatory elements: duplication and/or sub/neofunctionalization, direct de novo exaptation of existing intergenic space, insertion of TEs, exaptation of ancestral TE sequences.

Figure 2.

Theoretical expectations regarding the evolutionary dynamics of transcription factor occupancy and interpretations of existing data sets. (A). Population genetic theoretical considerations lead to an expectation of faster turnover of regulatory sites in organisms with low effective population sizes (N_e) than in species with large N_e. (B) Distribution of effective population size values in some of the main model systems [68]. Shown is the product $N_e\mu$ of N_e and the mutation rate μ, which can be most directly estimated empirically, unlike N_e alone [68, 69]. (C) The compilation of individual studies summarized in [47] suggested higher rates of regulatory site turnover in mammals than in flies. (D). In contrast, a reanalysis of several data sets in flies and mammals using a uniform data processing pipeline found similar rates of regulatory divergence within the two groups [70].

Figure 3.

Examples of using ChIP-seq to map GRNs in development and evolution. (A). Cataloging enhancers involved in patterning the morphology of bat limbs. The forelimb bud of bats develops into elongated webbed wings, while the hindlimb bud produces much shorter legs. H3K27ac and H3K27me3 ChIP-seq and RNA-seq were applied to developing limb buds to chart the developmental enhancer landscape involved in the specification of these structures [82]. (B) Understanding the role of limb-specific enhancers in snake evolution. Snakes lack legs, but H3K27ac ChIP-seq in mouse and lizard embryos and comparative genomics reveal that a substantial portion of limb enhancers are in fact conserved in snakes, one major reason for which is their role during phallus development, where a similar developmental program is deployed [83].

Figure 4.

Major eukaryotic clades, their epigenomic characterization and the origins of multicellularity. The tree shown follows previously published topologies [112], but it should be noted that the precise deep branching is still to be fully resolved by future phylogenomic studies. Red rounded rectangles are placed next to the lineages in which chromatin and transcriptional biology have been studied in considerable detail. Pink rounded rectangles are placed next to clades for some representatives of which initial epigenomic studies have been carried out in some detail. Clades in which multicellularity has evolved are indicated with three circles where multicellularity results from cell division, and by three triangles where multicellularity is aggregative [113]. The lineages containing nucleomorphs (the chlorarachniyophytes and the cryptophytes) are indicated with ‘NM’.

Figure 5.

Relationships between metazoan phyla and their closest relatives and the extent of their epigenomic characterization. Taxons for which the correct phylogenetic relationships is still to be conclusively established (such as the position of Ctenophora relative to other metazoans [201–205], the monophyly of poriferans and the overall relationships between non-bilaterian clades [206] and the placement of Chaetognatha [207], Acoelomorpha [208, 209], Xenoturbellida [209, 210] and others) are incorporated as unresolved branches on the cladogram, with Orthonectida and Dicyemida (tentatively placed in Lophotrochozoa [211]) being omitted. Phyla in which at least some epigenomic studies have been carried out are indicated on the right-hand side with the corresponding genus names for the main taxa studied. Lineages without published sequenced genomes have been left blank. Note that CTCF has been lost in some nematodes such as C. elegans but is present in other members of the phylum.