Zucchini: the key ingredient to unveil piRNA precursor processing

Piwi-interacting RNAs (piRNAs) are a class of small non-coding RNAs that play pivotal roles in protecting the germline genome and promoting animal fertility (1,2). The central players in the piRNA pathway are PIWI proteins and their associated 24–31 nt piRNAs. Guided by piRNAs, PIWI proteins act as cytoplasmic endonucleases (slicer) or nuclear epigenetic regulators to silence complementary RNA targets in germ cells. The most conserved function of piRNAs is to repress harmful transposons that invade the germline genome at both posttranscriptional and transcriptional levels. In addition to transposon silencing, piRNAs are implicated in the regulation of germline mRNA transcripts in a sequence-specific or sequence non-specific manner (3–5). It is crucial to maintain a functional piRNA pathway in germ cells because any defects will cause transposon misregulation and disruption in gametogenesis, leading to infertility in animals (1).

Since the discovery of piRNAs almost two decades ago, significant progress has been made in our understanding of the biogenesis and function of piRNAs. piRNAs are derived from long precursor RNAs that undergo multi-step fragmentation, which are eventually processed into mature 24–31 nt piRNAs bound to PIWI proteins. piRNA precursor processing occurs primarily at a unique site, on the mitochondrial surface, where numerous piRNA biogenesis factors reside. A piRNA precursor RNA is first cleaved by PIWI slicing to generate a free 5′-monophosphate end, which allows PIWI proteins to load the cleaved precursor (pre–pre-piRNA) onto their 5′-monophosphate-binding sites (Figure ure 1). The pre–pre-piRNA is then cleaved by an endonuclease at a site downstream of the PIWI-bound region to generate a shorter piRNA intermediate fragment termed pre-piRNA. The 3′ end of the pre-piRNA is then trimmed by the exonuclease trimmer (PNLDC1 in mice) to the mature length and further 2′-O-methylated by a methyltransferase Hen1 (HENMT1 in mice) (6) (Figure 1). Despite the recent discovery of pre-piRNA 3′-trimming enzymes (7–10), the understanding of the generation of pre-piRNAs from pre–pre-piRNAs remains obscure and speculative. The mitochondrial outer membrane protein Zucchini (Zuc, also known as MitoPLD or PLD6 in mice) has long been assumed to be a major endonuclease to cleave pre–pre-piRNAs to form pre-piRNAs, however, direct evidence has been lacking due to the lack of a faithful in vitro system to recapitulate Zuc-mediated pre-pre-piRNA cleavage.

Figure 1.

A proposed model of general steps in piRNA biogenesis. piRNA precursors are cleaved by PIWI slicing to generate pre–pre-piRNAs. After PIWI loading at the 5′ ends, pre–pre-piRNAs are cleaved in a Zucchini (Zuc)-dependent or Zuc-independent manner to form pre-piRNAs. Pre-piRNAs are further processed by exonuclease trimmer and methylated by HEN1 to form mature piRNAs.

In this report, Yukihide Tomari’s group established an elegant in vitro system faithfully reconstituting Zuc’s pre–pre-piRNA cleavage activity in the test tube (11). It not only revealed the identity of Zuc as a bona fide pre–pre-piRNA cleaving enzyme but also uncovered two parallel pathways (Zuc-dependent and Zuc-independent) for pre-piRNA production in animal germ cells (Figure ure 1). Previously, genetic studies implied that Zuc cuts a sequence on a piRNA precursor immediately before uridine (U) in vivo (12,13). However, purified recombinant Zuc does not show a U preference, but rather an overall nonspecific endonuclease activity in vitro (14–16). This discrepancy has long been puzzling as to whether Zuc is indeed the enzyme that generates the 5′-U of pre-piRNAs in vivo. Here, Izumi et al. devised an unconventional biochemical trick that successfully solved this puzzle.

To test whether Zuc converts pre–pre-piRNAs to pre-piRNAs, Tomari’s group first developed a physiological in vitro Zuc cleavage assay for analyzing Zuc’s endonuclease activity (11). This innovative method relies on the use of a well-studied silkworm cell line Bombyx mori BmN4 in which the piRNA biogenesis pathway is biologically active. One big challenge in studying Zuc in vitro is that it is a mitochondrial outer membrane protein and its endonuclease activity might not function fully without the facilitation of other mitochondrial cofactors. So instead of using purified Zucchini, the authors cleverly used a “crude cell pellet” containing the mitochondrial faction that includes Zuc and its associated factors. This 1000 g centrifugal pellet, normally discarded when preparing cell lysates, is the key to the success because it theoretically harbors all the key components for piRNA precursor processing. Indeed, when incubating the 1000 g pellet from naïve BmN4 cells with PIWI-loaded pre–pre-piRNAs, it efficiently processes pre–pre-piRNAs into mature piRNAs. To precisely understand how Zuc is involved in pre–pre-piRNA cleavage, Izumi et al. used crude pellets isolated from the trimmer-knockout (Tri-KO) BmN4 cells in the cleavage array. In this way, pre-piRNAs generated from pre–pre-piRNA cleavage are not further processed by trimmer. This allows for the identification of intact Zuc-cleaved fragments without further trimming. Using in vitro Zuc cleavage assay in the Tri-KO background with Zuc knockdown/overexpression or Zuc catalytic inactivation mutation, the authors demonstrate that Zuc is indeed the endonuclease for cleaving pre–pre-piRNAs and the cleavage sites have a moderate 5′-U preference. In addition, they reveal that Zuc does not act alone, but requires several other mitochondrial factors (Armi, GPAT1 and Gasz) for pre–pre-piRNA cleavage. In particular, RNA helicase Armi is the rate limiting factor for Zuc-mediated pre–pre-piRNA cleavage (17,18). These new results highlight a previously underappreciated action of Zuc that requires the coordinate actions with mitochondrial cofactors for efficient processing of piRNA precursors. The authors then took a step further to identify a new Zuc consensus motif (–10 ANNNNNNNAUUUNNC +4) that dictates the sequence preference for Zuc cleavage in silkworm. Together, these findings provide novel insights into how Zuc, together with other factors, preferentially selects target sequences for pre–pre-piRNA cleavage in a physiological setting.

During the course of the investigation, Izumi et al. also discovered two populations of untrimmed pre-piRNAs in silkworm using Tri-KO BmN4 cells. One population with extended lengths (≥31 nt, peak at 35 nt) is named type-E pre-piRNAs; the other population with non-extended lengths (<31 nt, peak at 27 nt) is termed type-N pre-piRNAs. Further experiments revealed that type-E pre-piRNAs are primarily derived from Zuc cleavage, while type-N pre-piRNAs are generated in a Zuc-independent manner, likely from the cleavage by PIWI slicer activities (6). Furthermore, the 3′ ends of type-E, but not type-N pre-piRNAs are efficiently 2′-O-methylated, corresponding to their differences in endonucleolytic origin. These data underscore the simultaneous operation of both Zuc-dependent and Zuc-independent pre–pre-piRNA cleavage pathways for piRNA biogenesis in silkworm.

How does this new information in silkworm translate into the mammalian system? In mice, the Zuc ortholog MitoPLD is a mitochondrial protein essential for piRNA biogenesis, spermatogenesis and male fertility (19,20). Like in silkworm, MitoPLD is assumed to mediate the cleavage of pre–pre-piRNAs and has a strong preference for a 5′-U cleavage site in vivo. However, whether MitoPLD has a similar consensus motif or cooperates with a similar set of mitochondrial cofactors is unclear. In addition, the existence of MitoPLD-independent pre–pre-piRNA cleavage pathway in mice remains unknown and requires further investigation. Thus, it would be interesting to understand how the Zuc/MitoPLD protein complex forms in multiple piRNA model systems, its components and interrelationships among various cofactors. Ultimately, at the structural level, future investigation is needed to understand how Zuc collaborates with its cofactors to recognize target sites to efficiently catalyze the cleavage reaction of pre–pre-piRNAs.

In summary, Izumi et al. used a novel in vitro system to pinpoint the mechanism of endonuclease Zuc in processing pre–pre-piRNAs. It opens up exciting new avenues to study piRNA processing mechanisms in various model organisms to better understand this essential class of small RNAs in reproduction and development.