Portal de Periódicos da CAPES

Trimming it short: PNLDC1 is required for piRNA maturation during mouse spermatogenesis

2018; Springer Nature; Volume: 19; Issue: 3 Linguagem: Inglês

10.15252/embr.201845824

1469-3178

Alfred W. Bronkhorst, René F. Ketting,

RNA Interference and Gene Delivery

News & Views19 February 2018free access Trimming it short: PNLDC1 is required for piRNA maturation during mouse spermatogenesis Alfred W Bronkhorst Institute of Molecular Biology, Biology of non-coding RNA, Mainz, Germany Search for more papers by this author René F Ketting [email protected] orcid.org/0000-0001-6161-5621 Institute of Molecular Biology, Biology of non-coding RNA, Mainz, Germany Search for more papers by this author Alfred W Bronkhorst Institute of Molecular Biology, Biology of non-coding RNA, Mainz, Germany Search for more papers by this author René F Ketting [email protected] orcid.org/0000-0001-6161-5621 Institute of Molecular Biology, Biology of non-coding RNA, Mainz, Germany Search for more papers by this author Author Information Alfred W Bronkhorst1 and René F Ketting1 1Institute of Molecular Biology, Biology of non-coding RNA, Mainz, Germany EMBO Rep (2018)19:e45824https://doi.org/10.15252/embr.201845824 See also: T Nishimura et al (March 2018) PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Transposon silencing within the germ line requires the proper processing of piRNA intermediates. However, the enzyme that is required for piRNA 3′ end maturation in vertebrates remained enigmatic. Nishimura et al 1 in this issue of EMBO Reports and two independent studies 23 now identified PNLDC1 as the exonuclease that is responsible for piRNA 3′ end processing and transposon silencing during mouse spermatogenesis. Together, these studies establish PNLDC1 as the piRNA 3′ end trimmer in mouse. Transposon silencing is essential for proper germ line development in animals. De-silencing of transposons affects germ cell genome integrity, which is associated with developmental defects during gametogenesis and often results in fertility defects. In the past two decades, it has become clear that PIWI-interacting RNAs (piRNAs) are at the core of transposon repression in germ cells, and for many piRNA factors, sterility phenotypes were observed in mutant animals. piRNAs are a class of small RNAs that associate with PIWI proteins, a specific subclade of the Argonaute protein family that is mostly expressed in the animal gonad. piRNAs serve as sequence-specific guides for PIWI proteins to find their target RNA based on sequence complementarity with the piRNA. Typically, multiple Piwi-paralogs act together in order to initiate transposon silencing, which can happen both at the transcriptional and post-transcriptional levels. In mouse, three different PIWI proteins are expressed: Miwi, Mili, and Miwi2. Mili and Miwi2 are expressed during early development in the gonocytes, where Miwi2 enters the nucleus to drive transcriptional gene silencing. Mili, together with Miwi, is also expressed at later stages during mouse spermatogenesis to silence mRNA targets post-transcriptionally. During piRNA biogenesis, long single-stranded piRNA precursor transcripts (pre-piRNAs) that are often derived from specific genomic locations, called piRNA clusters, need to undergo multiple maturation steps to form a mature piRNA that is bound by a PIWI protein. In recent years, several studies revealed how piRNA maturation takes place in a consecutive manner, involving various specific endo- and exonucleases. Initially, piRNA precursor transcripts are cleaved by an endonuclease that liberates the 5′ end of the to-be-formed piRNA. Interestingly, two of such endonucleolytic activities have been identified. First, the mitochondria-anchored endonuclease Zucchini/mitoPLD can cleave upstream of uridine. The resulting piRNA intermediates therefore harbor a U-nucleotide at their 5′ end, a characteristic seen for the majority of mature piRNAs in virtually all species studied. Interestingly, Zucchini can cleave pre-piRNA consecutively, generating a phased series of piRNA intermediates, all carrying a 5′U, that can each be converted into a mature piRNA. In mice, the fragments generated as such by Zucchini are larger than the mature piRNAs 456. Both Mili and Miwi can be loaded with pre-piRNAs derived from Zucchini (Fig 1, Zucchini-dependent). A second 5′ end generating mechanism, that seems independent of Zucchini, is driven by the so-called slicer activity of PIWI proteins themselves (Fig 1, slicer-dependent). In mice, Miwi2 is a Piwi protein that primarily accepts such slicer-dependent piRNAs. More specifically, Miwi2 is loaded with piRNAs that stem from Mili-cleaved target transcripts. Additionally, this mechanism potentially allows the amplification of piRNA species when both sense and anti-sense transcripts of a piRNA target are present. Notably, pre-piRNAs whose 5′ end is generated through slicer activity are, just like Zucchini-derived pre-piRNAs, still longer than the mature piRNA. Figure 1. piRNA biogenesis during mouse spermatogenesisZucchini-driven and slicer-driven endonuclease activities represent two different modes for piRNA 5′ end generation. This step of piRNA biogenesis may be spatially and/or mechanistically coupled to loading into Piwi proteins (Mili, Miwi, Miwi2). The resulting piRNA intermediates need to be further trimmed at their 3′ end, which requires TdrKH and PNLDC1. The mature piRNA is methylated at its 3′ end by the methyltransferase Hen1. The Miwi2/piRNA complex acts in the nucleus to repress retrotransposon transcription. TGS: transcriptional gene silencing. Download figure Download PowerPoint Following 5′ end generation the piRNA intermediate is bound by a PIWI protein. At this stage the bound piRNA intermediate can be trimmed from the 3′ end. However, the trimmer exonuclease has remained unknown in mammals thus far. Finally, piRNA biogenesis is finished off by Hen1-mediated 2′O-methylation of the most 3′ nucleoside. Initial studies in mouse and in silkworm-derived cells indicated that TdrKH/BmPapi was required for the 3′ end maturation of pre-piRNAs 78. However, mechanistic insights into TdrKH-mediated trimming remained unclear. TdrKH does not contain a nuclease domain and, hence, the identity of the 3′–5′ exonuclease, already then referred to as trimmer, remained mysterious. Recently, an elegant biochemical study identified the poly(A)-specific ribonuclease (PARN)-like domain containing 1 (PNLDC1) as the 3′–5′ exonuclease that is responsible for pre-piRNA 3′ end maturation in silkworms 9. The authors further showed that PNLDC1 interacts with TdrKH and that both factors cooperate in piRNA maturation. Interestingly, PNLDC1 association with TdrKH was found to be required for its trimming activity. Notably, the exo- and endonuclease factors that are involved in these piRNA maturation steps are mitochondria-associated proteins. Thus far, the enzyme that is required for piRNA 3′ end maturation during mouse spermatogenesis remained unknown. Three recent studies now show that PNLDC1 is required for piRNA 3′ end formation in mouse testis at different developmental stages. Nishimura et al 1, Ding et al 2, and Zhang et al 3 independently revealed that PNLDC1 is required for 3′ end maturation for all (Mili-, Miwi2-, and Miwi-bound) piRNAs. Given that TdrKH was previously shown to affect 3′ end trimming in mice, a comparison between TdrKH and PNLDC1 mutants is warranted. Interestingly, the overall phenotypes of the germ cells, both in terms of their development as well as in piRNA generation, are significantly weaker in pnldc1 mutants compared to tdrkh mutants 8. Nevertheless, the effects on piRNA 3′ end trimming seem very similar, if not identical in both mutants. What then makes the difference between these mutants? Obviously, redundancy of enzymatic activity should be considered. A close homolog of PNLDC1 is PARN, and this enzyme has been shown to drive piRNA 3′ end processing in Caenorhabditis elegans (which does not have PNLDC1) 10. PARN could therefore compensate at least partially for the loss of PNLDC1 function. However, if PARN would indeed be involved, one would expect to see at least some more trimming of 3′ end extended pre-piRNAs in the pnldc1 mutants compared to tdrkh mutants, which is not the case. It seems more probable that loss of TdrKH affects more than just the 3′ end trimming. TdrKH interacts physically with both PIWI proteins and the PNLDC1 enzyme, and like Zucchini, is present on the mitochondria. This constellation may ensure that Zucchini-generated transcripts are promptly loaded into a PIWI protein, after which they can be 3′ end trimmed immediately. In the absence of PNLDC1, the prompt loading may still take place, whereas this may be additionally disrupted in TdrKH mutants. However, no direct evidence for an effect of TdrKH on pre-piRNA loading into Piwi protein exists to date. We note that such a coupling between loading and Zucchini-driven 5′ end generation may very well also affect slicer-driven piRNA biogenesis, since it has been shown that cleavage by Mili can in fact feed a transcript into further Zucchini-mediated digestion 456. This suggests that slicer activity can take place in the vicinity of Zucchini, and thereby of TdrKH. Indeed, the perinuclear aggregates known to be home to Mili and many other piRNA pathway factors are rich in mitochondria. This also couples to the open question of how Zucchini recognizes and initiates the 5′ end processing of transcripts that derive from piRNA clusters in the first place. All three studies report that basically all piRNAs in the mouse are affected by loss of PNLDC1, but that Miwi is much more sensitive to loss of this enzyme than Mili. Almost no Miwi-associated piRNA could be detected, and Miwi protein levels were very low. What makes Miwi so hypersensitive to the absence of 3′ end trimming? One option is that Miwi can be ubiquitylated upon loading with piRNAs, leading to its degradation, as also suggested by Nishimura et al 1. While this seems to be rather counterproductive (Piwi proteins need a piRNA for target recognition), such a mechanism could be in place to specifically take out Miwi loaded with untrimmed piRNAs, as Piwi proteins with 3′ end extended piRNAs will be significantly slowed down in the release of their bound target RNA, potentially posing a threat to germ cell development or function. However, Mili would face similar problems with 3′ untrimmed piRNAs and is not so strongly affected. Alternatively, Mili, which starts to be expressed before Miwi, may somehow stimulate the processing of piRNA precursor transcripts for loading into Miwi, resulting in a double effect on Miwi upon loss of PNLDC1. With the 3′ end processing enzymes now clearly defined also in mice, the stage is set for further biochemical assessment of Piwi protein biochemistry. Such studies will be essential to further deepen our insights, since the strong developmental defects of many of the mutants in these pathways severely compromise the mechanistic insights that can be obtained from pure in vivo studies, as also indicated by the unresolved issues mentioned above. References 1. Nishimura T, Nagamori I, Nakatani T et al (2018) EMBO Rep 19: e44957Google Scholar 2. Ding D, Liu J, Dong K et al (2017) Nat Commun 8: 819Google Scholar 3. Zhang Y, Guo R, Cui Y et al (2017) Cell Res 27: 1392–1396CrossrefGoogle Scholar 4. Han BW, Wang W, Li C et al (2015) Science 348: 817–821CrossrefCASPubMedWeb of Science®Google Scholar 5. Mohn F, Handler D, Brennecke J (2015) Science 348: 812–817CrossrefCASPubMedWeb of Science®Google Scholar 6. Yang Z, Chen KM, Pandey RR et al (2016) Mol Cell 61: 138–152CrossrefPubMedWeb of Science®Google Scholar 7. Honda S, Kirino Y, Maragkakis M et al (2013) RNA 19: 1405–1418CrossrefCASPubMedWeb of Science®Google Scholar 8. Saxe JP, Chen M, Zhao H et al (2013) EMBO J 32: 1869–1885Wiley Online LibraryCASPubMedWeb of Science®Google Scholar 9. Izumi N, Shoji K, Sakaguchi Y et al (2016) Cell 164: 962–973CrossrefCASPubMedWeb of Science®Google Scholar 10. Tang W, Tu S, Lee HC et al (2016) Cell 164: 974–984CrossrefCASPubMedWeb of Science®Google Scholar Previous ArticleNext Article Read MoreAbout the coverClose modalView large imageVolume 19,Issue 3,March 2018Cover: Target of rapamycin complex 1 (TORC1) regulates cell growth in response to environmental changes. In the fission yeast Schizosaccharomyces pombe, TORC1 promotes vegetative growth and suppresses sexual differentiation under nutrient‐rich conditions. Otsubo, et al. demonstrate that tRNA precursors sense nutrient availability to regulate TORC1 activity. From Yoko Otsubo, Akira Yamashita and colleagues: tRNA production links nutrient conditions to the onset of sexual differentiation through the TORC1 pathway. For detail, see Article on page e44867. Scientific image by Akira Yamashita. (Cover design by Uta Mackensen) Volume 19Issue 31 March 2018In this issue FiguresReferencesRelatedDetailsLoading ...