The Cecere Laboratory is based at Institut Pasteur In Paris in the Department of Developmental and Stem Cell Biology. Our main focus is to characterize the role of RNA-based mechanisms of epigenetic inheritance during animal development and upon environmental stresses.
Heritable traits are predominantly encoded within genomic DNA, but it is now appreciated that epigenetic information is also inherited through DNA methylation, histone modifications and RNA. Indeed, examples of transgenerational epigenetic inheritance via the germline has been documented in plants and animals (1).
The capacity of RNA to alter trait inheritance became evident with the discovery of RNA interference (RNAi) over 20 years ago (2). Pioneering experiments by Mello and Fire showed that Caenorhabditis elegans treated with double-stranded RNAs (dsRNAs) displayed gene silencing that persisted in descendants not originally exposed to the dsRNAs (2). Many of the players and mechanisms that underlie the epigenetic transmission of RNAi have been elucidated (3).
RNA interference is an epigenetic phenomenon
The initial steps of the RNAi process involve the production of primary small interfering RNAs (siRNAs), which are cleaved from the initial dsRNA trigger by the endoribonucleases Dicer and its cofactor RDE-4 (5). The primary siRNAs, which are loaded into the Argonaute protein RDE-1, derive from both strands and carry a 5' monophosphate terminus (6). RDE-1 is not directly cleaving the target mRNAs and its catalytic activity is only required for generating a single-stranded small RNAs (7). The interaction between the primary siRNAs loaded into RDE-1 and the complementary target mRNAs lead to the recruitment of the endonuclease RDE-8, which cleaves the mRNA target (8). These cleaved mRNAs become the substrate for RNA-dependent RNA Polymerases (RdRPs) to produce secondary siRNAs that are single-stranded and carry 5' triphosphate terminus (9,10).
In addition to the cleavage by RDE-8, the ribonucleotidyltransferase RDE-3 add stretches of poly(UG) at the 3’end of the cleaved target mRNA and are essential for the production of secondary siRNAs by the RdRP (11). The secondary siRNAs are all antisense to the target transcript and are loaded into downstream Worm-specific Argonaute proteins (WAGOs). Once secondary siRNAs are triggered, they can be produced even in the absence of the initial dsRNA triggers. Therefore, the inheritance of secondary siRNAs can maintain the multigenerational silencing even without the initial dsRNA trigger (12). The nuclear Argonaute HRDE-1, which load secondary siRNAs, is required to maintain the silencing across generations in absence of the initial dsRNA trigger (13). HRDE-1 repress transcription and initiate the chromatin silencing by promoting the deposition of H3K9me3 and H3K27me3 on target loci (12,14).
Despite this knowledge, the mechanism by which nuclear small RNAs initiate and maintain the chromatin silencing across generations is currently unknown.
Germline transmission of epigenetic information
More recently, endogenous small RNAs have been shown to control epigenetic inheritance. For instance, we recently discovered how the inheritance of endogenous small RNAs antisense to histone genes impairs fertility across generations (4).Two main endogenous small RNA pathways regulate heritable epigenetic processes in the C. elegans germline (15): (i) thousands of PIWI-interacting RNAs (piRNAs) target foreign “non-self” mRNAs, by a mechanism similar to the dsRNA-induced RNAi response (16), and (ii) endogenous antisense small RNAs target active “self” mRNAs (17). These small RNAs are loaded into the argonaute CSR-1 and are essential for fertility and embryonic development (18–20), and can protect their targets from piRNA-mediated silencing (21,22). Therefore, the C. elegans germline coordinates the transmission of active and silent gene expression programs across generations through maternally inherited endogenous small RNAs.
The mechanism by which piRNAs initiate endogenous silencing is similar to the one induced by dsRNAs during the RNAi phenomenon.
Despite this knowledge, fundamental questions surrounding epigenetic inheritance remain unresolved:
Which developmental processes in the progeny are epigenetically regulated by these pathways?
Soma-to-germline transfer of epigenetic information
Another exciting possibility is that small RNAs in somatic cells can be transferred to germ cells to provide heritable epigenetic information of acquired traits. Indeed, the initial RNAi experiments (2) demonstrated that heritable RNAi response in worms was induced not only by the direct injection of dsRNA into their germline but also by the accumulation of dsRNA in intestinal cells from ingested bacteria. Endogenous small RNAs in C. elegans underlie the transgenerational inheritance of a response to infection (23) and learned behaviour (24,25). These findings suggest that somatic cells communicate with germ cells to transmit epigenetic information to the next generation, potentially via small RNAs. However, evidence for inter-organ transfer of endogenous dsRNA and/or small RNA is lacking. Thus, a fundamental unresolved question is:
Germ granules in heritable RNAi responses
Image: Eric Cornes
Recent research on the RNAi phenomenon has revealed the involvement of phase-separated germ granules, which are membrane less organelles, in the inheritance of the RNA silencing response (26-30). Most of the enzymes responsible for the production of secondary germline siRNAs, including the RdRPs and their cofactors, and some Argonaute proteins localize to germ granules. Most importantly, structural components of these membrane-less organelles are also involved in the transmission of the silencing signal across generations (26,27).
Moreover, embryos that fail to assemble germ granules do not transmit the silencing signal and inherit an improper pool of secondary small RNAs (28–30).
1. Heard, E. & Martienssen, R. A. Cell (2014). doi:10.1016/j.cell.2014.02.045
2. Fire, A. et al. Nature (1998). doi:10.1038/35888
3. Youngman, E. M. & Claycomb, J. M. Frontiers in Genetics (2014). doi:10.3389/fgene.2014.00416
4. Barucci, G. et al. Nat. Cell Biol. (2020). doi:10.1038/s41556-020-0462-7
5. Tabara, H. et al., Cell (2002). doi:10.1016/S0092-8674(02)00793-6
6. Yigit, E. et al. Cell (2006). doi:10.1016/j.cell.2006.09.033
7. Steiner, F. A. et al., Nat. Struct. Mol. Biol. (2009). doi:10.1038/nsmb.1541
8. Tsai, H. Y. et al. Cell (2015). doi:10.1016/j.cell.2015.01.010
9. Pak, J. & Fire, A. Science (2007). doi:10.1126/science.1132839
10. Sijen, T. et al., Science (2007). doi:10.1126/science.1136699
11. Shukla, A. et al. Nature (2020). doi.org/10.1038/s41586-020-2323-8
12. Gu, S. G. et al. Nat. Genet. (2012). doi:10.1038/ng.1039
13. Buckley, B. A. et al. Nature (2012). doi:10.1038/nature11352
14. Mao, H. et al. Curr. Biol. (2015). doi:10.1016/j.cub.2015.07.051
15. Claycomb, J. M. Current Biology (2014). doi:10.1016/j.cub.2014.06.009
16. Luteijn, M. J. & Ketting, R. F. Nature Reviews Genetics (2013). doi:10.1038/nrg3495
17. Claycomb, J. M. et al. Cell (2009). doi:10.1016/j.cell.2009.09.014
18. Yigit, E. et al. Cell (2006). doi:10.1016/j.cell.2006.09.033
19. Gerson-Gurwitz, A. et al. Cell (2016). doi:10.1016/j.cell.2016.02.040
20. Quarato, P. et al. Nature Communications (2021). doi:/10.1038/s41467-021-21691-6
21. Wedeles, C. J., et al., Dev. Cell (2013). doi:10.1016/j.devcel.2013.11.016
22. Seth, M. et al. Cell (2013). doi:10.1016/j.devcel.2013.11.014
23. Gammon, D. B. et al. Curr. Biol. (2017). doi:10.1016/j.cub.2017.02.004
24. Posner, R. et al. Cell (2019). doi:10.1016/j.cell.2019.04.029
25. Moore, R. S., Kaletsky, R. & Murphy, C. T. Cell (2019). doi:10.1016/j.cell.2019.05.024
26. Wan, G. et al. Nature (2018). doi:10.1038/s41586-018-0132-0
27. Spracklin, G. et al. Genetics (2017). doi:10.1534/genetics.116.198812
28. Dodson, A. E. & Kennedy, S. Dev. Cell (2019). doi:10.1016/j.devcel.2019.07.025
29. Ouyang, J. P. T. et al. Dev. Cell (2019). doi:10.1016/j.devcel.2019.07.026
30. Lev, I. et al. Curr. Biol. (2019). doi:10.1016/j.cub.2019.07.054