Thompson Group - Oikopleura dioica Chromatin
Results obtained from multiple knockouts of key mammalian cell cycle regulators combined with efforts at decomposing the complex cell cycle regulatory network into interlaced, yet somewhat independent oscillatory modules, are presenting new research opportunities in understanding evolution of the cell cycle itself and defining its links to developmental processes and differentiation. For example, experiments using mice and cell lines demonstrate cell division in the absence of S-phase CDK2, G1-phase CDK4/CDK6 and even upon simultaneous deletion of all interphase CDKs. Instead, CDK1 assumed these functions through association with interphase cyclins. Similar results have been obtained upon inactivation of cyclin E or deletion of all cyclin Ds, indicative of significant functional plasticity of cyclins and CDKs.
Fig 1. Histone variant deployments throughout O. dioica development. Developmental times are indicated with reference to predominance of different cell cycle types (bottom). Cartoons (top) illustrate potential combinatorial possibilities of histone core (wedges: H4s grey, H3s yellow, H2As blue, H2Bs green), and linker H1 (rectangles: orange to red) variants through development. Histone transcripts are represented as lines with thickened portions indicating significant increases either through greater transcriptional activity of a given gene or contributions from additional genes encoding the same variant. The latter was observed for canonical histones of all types during organogenesis with activation of additional genes. All core and linker histones are represented by at least one male-specific isoform (at right). Larger view
Among universal histone variants, interesting findings were that no OdH2As contained the PIKK substrate SQE/Dɸ phosphorylation motif characteristic of all H2AXs and O. dioica uses alternative splicing to control the number of acetylatable K residues in the essential N-terminal charge patch of H2A.Z. The absence of H2AX raises interesting questions about genome integrity and the double-strand break (DSB)-repair pathway in O. dioica. Rapid phosphorylation of H2AX is central to each alternative DNA repair pathway, nonhomologous end-joining (NHEJ) and homologous recombination (HR). HR efficiency is influenced by template accessibility, causing an up regulation of HR during S and G2 phases of the cell cycle when sister chromatids are available. The challenge of locating a homologous template for HR repair may not be an obstacle in the compact genome of O. dioica, where up to several hundred copies of each locus are available for recombination in endocycling cells. It is therefore possible that DSBs arising in endocycling cells that do not traverse mitosis, could be repaired by an H2AX-independent HR pathway. In support of this, the O. dioica genome lacks many key components of the NHEJ pathway, including DNA-PK, which facilitates alignment of non-complementary ends and regulates end-processing. The observation that both H2AX and DNA-PK are present in the sister class ascidians, indicates secondary loss of H2AX in the appendicularian lineage as opposed to urochordates as a whole having failed to evolve an H2AX variant.
Of particular note among O. dioica lineage-specific variants was diversity in all core and linker histone families associated with differentiation of the male germline including the first described histone H4 variant, raising intriguing questions regarding the testes as an evolutionary playground for the innovation of histone variants in the fast evolving appendicularian and other lineages.
We have now begun with ChIP-chip assessment in an attempt to determine Oikopleura`s genome-wide chromatin landscape in based on sixteen histone modifications, comparable to the modENCODE effort on Drosophila cell lines. Our focus is principally on histone H3 modifications to explore a number of questions. What are the effects of secondary genome compaction in the chordate lineage? Extreme compaction of Oikopleura genome means that regulatory regions have been strongly compressed. Often these are on the order of a single nucleosome or less. A significant portion of the genome is transcribed as operons and there is also complex interlacing of transcriptional units with different expression profiles. In such a context, are promoters and enhancers defined by similar modification combinations as described elsewhere and how prevalent are enhancers compared to vertebrate genomes? What are the H3 modifications in transcriptionally active polyploid germ nuclei? Polyploid nurse nuclei have to co-regulate 100s of copies of each gene locus as compared to two in diploid cells and in Oikopleura we do not observe polytene alignment of these loci. Polyploid nuclei will not traverse mitosis. How does this impact the modification combinations and breadth of epigenetic memory domains? Finally, how do H3 modification combinations evolve in meaning in different evolutionary lineages?