Embryonic development in multicellular eukaryotic organisms is an intricate dance of signals that determine when and where genes are expressed, allowing the zygote to produce the cells that will ultimately differentiate in to the tissues and organs of the adult. Some of this gene expression is regulated by maternal and zygotic transcription factors, but much of this regulation is epigenetic: caused by inherited changes in gene activity that do not involve alternations to the primary DNA sequence, but rather modifications to the DNA molecule itself or modifications to the chromatin structure.
Chromatin structure can influence gene expression by limiting or allowing transcriptional machinery access to DNA. Histone proteins are major players in determining chromatin structure, and histone methylation is one of several modifications that influences the interaction between histone proteins and DNA.
A recent paper by Furuhashi and colleagues in Epigenetics & Chromatin asks how epigenetic memory is passed from one generation to the next. The authors used epigenetic repression of transcription of the developing germline in C. elegans as the model for their investigations.
During embryogenesis gene transcription is repressed in the developing germline in C. elegans (and other organisms including fly and mouse). In early stages, this repression occurs by a mechanism in which the maternal protein, PIE-1, prevents hyperphosphorylation of the c-terminal domain (CTD) of RNA pol II (and thus activation of RNA pol II). However, in the primordial germ cells (PCGs) Z2 and Z3, hyperphospholyation of the CTD occurs simultaneously with loss of histone modifications that are associated with gene expression. The net effect is repression of gene expression, although by some mechanism other than preventing phosphorylation of the CTD of RNA pol II.
The authors of this paper investigated whether the continued transcriptional quiescence in Z2 and Z3 might be related to histone modifications. They examined two C. elegans strains containing maternal-effect sterile (MES) mutations. The mutations mes-2 and mes-4 both affected histone modifying enzymes. The mes-2 gene encodes an H3K27 methyltransferase, and the mes-4 gene encodes the H3K36 methyltransferase.
In mes-4 mutant embryos, staining for the hyperphosphorylated RNA pol II CTD was quite prominent in late state primordial germ cells, but was lost in wild type embryos. In mes-2 mutant embryos, staining for the hyperphosphorylated RNA pol II CTD was similar to wild type cells. These staining patterns suggest that mes-4 is required to maintain the hypophosphorylated (inactive) RNA pol II state in the Z2 and Z3 germline precursor cells.
H3K4 methylation (H3K4me2) correlates with increased transcriptional activity. In both mes-2 and mes-4 mutant embryos, “erasure” of H3H4me2 occurred around the 90 cell stage (as in wild type embryos); however, in mes-4 embryos, the germline erasure of H3K4me3 was not maintained, with H3K4 methylation increasing at the 550-cell stage. This suggests that MES-4 is not required for the initial epigenetic reprogramming event at the 90-cell stage but that it is required to maintain this reprogramming (transcriptional quiescence of the germline) later in embryogenesis.
The MET-1 methyltransferase in C. elegans is responsible for H3K36me3 methylation (associated with gene activity) apparent in all nuclei at all stages of the wild type embryo. Observations of mes-4 mutants reveals active RNA pol II, H3K4 methylation and MET-1-dependent H3K36 methylation in the PCGs, indicating these cells are actively transcribing genes when, in a wild type embryo, transcription in would be quiescent. This and other evidence indicate that mes-4 is necessary to maintain H3K36 methylation, and that the maintainence of this epigenetic information is important for preventing aberrant transcriptional activity within the PCGs.
The mes-4 phenotype (maternal effect sterile) itself hints at its requirement for the generational memory. In mes-4 embryos in which there is maternally provided MES-4, MES-4 is localized to the PGCs and maintains methylation status, and about 1,000 gametes are produced and develop normally. However, in the next generation, in which there is no maternal or zygotic MES-4, the information is lost, and the few embryos produced die early.
During the cell cycle nucleosomes, the structures formed by histones, are assembled and disassembled, and imagining how histone modifications could be remembered from cell cycle to cell cycle, much less inherited from generation to generation is difficult. It’s possible that MES-4 somehow recognizes and binds to methylated portions of histones, and it is this binding that is the “memory”, and the position of the mutations in the mes-4 alleles may provide some hints. But, we don’t really understand how these epigenetic memories are passed down from one generation to the next. However, like our own generational stories, once the “memory” is lost the cost is great.
Furuhashi, H., Takasaki, T., Rechtsteiner, A., Li, T., Kimura, H., Checchi, P., Strome, S., & Kelly, W. (2010). Trans-generational epigenetic regulation of C. elegans primordial germ cells Epigenetics & Chromatin, 3 (1) DOI: 10.1186/1756-8935-3-15