University of Southern California Dissertations and Theses

Regulation of Caenorhabditis elegans small RNA pathways: an examination of Argonaute protein RNA binding and post-translational modifications in C. elegans germline

(USC Thesis Other)

Regulation of Caenorhabditis elegans small RNA pathways: an examination of Argonaute protein RNA binding and post-translational modifications in C. elegans germline

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Regulation of Caenorhabditis elegans small RNA pathways An examination of Argonaute protein RNA binding and post-translational modifications in C. elegans germline by Dieu An Hoang Nguyen A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (MOLECULAR BIOLOGY) December 2021 Copyright 2021 Dieu An Hoang Nguyen i DEDICATION To the village that raised me. ii ACKNOWLEDGMENTS A Ph.D. is an inherently solitary journey, but without the support of the following people, it is not certain that I would have finished this odyssey. First, I would like to thank my advisor, Carolyn M. Phillips, for her supervision these past few years. I do not suppose she knows how profound of an impact she has on me, how deeply I value her integrity, discipline, patience, and generosity, and how liberating she has set me for trusting that I can become more than I thought I was capable of. Second, I would like to thank the people I have had the privilege of teaching and mentoring – Diego, Dylan, Jenny, and Diala, all of whom contributed to the making of this dissertation. Without your patience and forgiveness, I would never have learned how to teach properly, and without that, my education here at USC would only be half complete. Third, I would like to thank the friends that I’ve made in RRI, Caleb, Yunke, Soo Lim, for too many good memories, kbbq, late-night tacos run, and endless rants about the humdrums of graduate school. Also, a shoutout to Gui for lending me $5 cash in the middle of the night so I could have tacos from the across the street before running back to experiments. That random act of kindness meant the world to me. Then there is Yael whose friendship had forever changed me for the better. I was ready to believe in the worst of man when your friendship saved me from the precipice. Your selflessness, kindness, and generosity towards me had never gone unappreciated; whatever goodness I have is much owed to you. To Dr. Klaskin for saving my sanity during the most tumultuous part of my graduate career, and for mending and nurturing the child within me with love and empathy. By far the greatest thank goes to my family, which if accounting for members from both maternal and paternal sides, the number goes into the hundreds, all of whom had given me happy memories over the years. But within the nuclear circle, I am indebted to my siblings, Dan, San, Lan, Uyen, for being in the trenches with me since day one. Nobody tolerates my haughtiness, arrogance, and the occasional bad sense of humor more than they did and do. I must especially shoutout to my little sister for unwittingly accepting me as her third parent, I dare say I was her strictest and least fun guardian, which I sometimes regret. Finally, I bow to my parents, to whom I owe most if not all my good fortunes in life, including this Ph.D. The 12-year-old me could not comprehend the sacrifice my parents had made, to leave their country in their forties and enter a new society, not knowing the language, system, not having a trade, but only with the shirt on their back and their two young, helpless children. It is still hard for me to fully grasp the hardship that our family had endured for almost two decades, and the strength of character my parents possess to make it through those hard times. You’ve taught me to hold my head up high during abject poverty, to be kind when the world is cruel, and to work hard even when no one is looking. Who would I have become if I weren’t born your daughter? I’ve picked up in my earlier years some bad habits of self-importance and a self- prescribed grandiosity, only to realize that only a third of my success could be ascribed iii to my merits (by a liberal estimation). Without my parents, family, and friends who provide an environment in which I can pursuit my scholastic interests, and without a tremendous luck which I have nothing to do with, I would never have been able to get very far. Perhaps this was the most meaningful lesson from this Ph.D. I stand on the shoulders of giants. iv TABLE OF CONTENTS Dedication .................................................................................................................. i Acknowledgements .................................................................................................... ii List of Figures ............................................................................................................. vi Introduction ............................................................................................................. 1 1 | Overview of C. elegans RNAi ..................................................................... 1 2 | PZM granule and the germline environment .............................................. 5 2.1 | P granules, small RNAs, and the basic principles of phase separation ............................................................................................. 5 2.2 | Z granules, Tudor proteins, and other intermediaries ................... 10 2.3 | mutator complex and 22G-RNA biogenesis ................................. 14 3 | Concluding remarks .................................................................................... 16 Chapter 1: Literature Review: Chromosome Segregation and RNAi Defective, CSR-1, from embryos to adult germline .......................................................... 18 1.1 | Overview on CSR-1 ...................................................................... 19 1.2 | Oocytes to embryos ...................................................................... 21 1.3 | One-cell embryos .......................................................................... 23 1.4 | One-cell embryos and beyond ...................................................... 25 1.5 | Attenuation of translation elongation ............................................ 28 1.6 | CSR-1 and the enlarged P-granule phenotypes ........................... 31 1.7 | Concluding remarks ...................................................................... 32 1.8 | Figures .......................................................................................... 35 Chapter 2: Arginine methylation promotes siRNA-binding specificity for a spermatogenesis-specific isoform of the Argonaute protein CSR-1 ................ 36 2.1 | Abstract ......................................................................................... 37 2.2 | Introduction ................................................................................... 37 2.3 | The long isoform of CSR-1 is selectively expressed during spermatogenesis .................................................................................. 41 2.4 | CSR-1A and CSR-1B are expressed from independent promoters .............................................................................................. 42 2.5 | CSR-1A localizes to the P granules during spermatogenesis ...... 43 2.6 | CSR-1A is required for optimal sperm fertility ............................... 44 2.7 | CSR-1A and CSR-1B are expressed independently of one another.................................................................................................. 46 2.8 | CSR-1A and CSR-1B target distinct groups of genes .................. 47 2.9 | CSR-1A does not require ALG-3/4 to bind spermatogenic small RNAs .................................................................................................... 50 2.10 | CSR-1A expression is positively correlated with the expression of CSR-1A target genes ........................................................................... 51 2.11 | The first exon of CSR-1A is unstructured and contains RG v motifs .................................................................................................... 54 2.12 | The RG motifs in the first exon of CSR-1A are demethylated .... 55 2.13 | The RG motifs in CSR-1A promote specificity for binding small RNAs targeting spermatogenic genes .................................................. 57 2.14 | Discussion .................................................................................. 58 2.15 | Materials and Methods ............................................................... 63 2.16 | Figures ................................................................................... 76-86 Chapter 3: WAGO-10 is a spermatogenesis-specific secondary Argonaute Protein ............................................................................................................. 92 3.1 | Abstract ......................................................................................... 93 3.2 | Introduction ................................................................................... 93 3.3 | Worm-specific Argonaute protein WAGO-10 is expressed during spermatogenesis .................................................................................. 97 3.4 | WAGO-10 localizes to the P granules in the spermatogenesis region .................................................................................................... 98 3.5 | WAGO-10 is essential for optimizing male sperm fitness ............ 99 3.6 | WAGO-10 binds to a subset of 22G-RNAs that do not target spermatogenic genes .......................................................................... 101 3.7 | WAGO-10 binds to small RNAs that recognize the same targets as ALG-3 ................................................................................................... 102 3.8 | Discussion .................................................................................... 103 3.9 | Materials and Methods ................................................................. 105 3.10 | Figures ................................................................................. 113-117 Chapter 4: Localization and regulation of Argonaute proteins .................................. 118 4.1 | Introduction ................................................................................... 119 4.2 | Creating binding deficient AGO mutants ..................................... 122 4.3 | Most binding-defective Argonaute proteins are stably expressed ............................................................................................. 124 4.4 | Binding-defective Argonaute proteins fail to localize to the P granule .............................................................................................. 125 4.5 | alg-3[YK-to-AA] mutants phenocopy null alg-3 mutants ............... 126 4.6 | csr-1[HK-to-AA] mutants are sterile and form misshapen P granules............................................................................................. 127 4.7 | The RG motifs in multiple C. elegans Argonaute proteins are dimethylated ......................................................................................... 128 4.8 | Summary ...................................................................................... 129 4.9 | Materials and Methods ................................................................. 130 4.10 | Figures ................................................................................. 139-142 Appendix A: Knockdown of CSR-1 results enlarged germline nuclei ...................... 143 Reference .................................................................................................................. 146 vi LIST OF FIGURES Chapter 1: Fig 1 | CSR-1 is largely cytoplasmic in oocytes. ............................................. 35 Chapter 2: Fig 1 | CSR-1 isoforms have distinct spatial and temporal expression patterns............................................................................................................ 76 Fig 2 | CSR-1A is expressed in the spermatogenesis region of the germline . 77 Fig 3 | CSR-1 isoforms function independently of one another ....................... 78 Fig 4 | CSR-1A and CSR-1B target distinct groups of genes .......................... 80 Fig 5 | CSR-1A targets spermatogenesis-expressed genes ........................... 82 Fig 6 | Methylation of CSR-1A N-terminal exon promotes binding preference for spermatogenic siRNAs ............................................................................... 84 Fig S1 | CSR-1A is expressed in the spermatogenesis region of the male germline ........................................................................................................... 8 Fig S2 | CSR-1A is excluded from secondary spermatocytes ........................ 86 Fig S3 | CSR-1A is required for optimal male fertility ...................................... 87 Fig S4 | Both CSR-1 isoforms bind germline-enriched 22G-RNAs ................. 88 Fig S5 | Clustal Omega alignment of CSR-1 in Caenorhabditis species ........ 90 Fig S6 | The N-terminal exon of CSR-1 is disordered across Caenorhabditis species ............................................................................................................ 91 Chapter 3: Fig 1 | WAGO-10 is selectively expressed during the fourth larval stage ....... 113 Fig 2 | WAGO-10 is expressed in the spermatogenesis region of the germline ........................................................................................................... 114 Fig 3 | WAGO-10 is essential for optimal male sperm fitness ......................... 115 Fig 4 | WAGO-10 binds to 22G-RNAs that do not specifically target spermatogenic genes ...................................................................................... 116 Fig 5 | WAGO-10 binds small RNAs that target the same genes as ALG-3 ... 117 Chapter 4: Fig 1 | Modifying Argonaute proteins binding pockets to render binding-deficient mutants ............................................................................................................ 139 Fig 2 | Binding-deficient mutants fail to localize to P granules ........................ 140 Fig 3 | csr-1[HK-to-AA] mutants are sterile and form misshapen P granules . 141 Fig 4 | Argonaute proteins are dimethylated at the RG motifs ........................ 141 Appendix A: Appendix A | Absence of CSR-1 results in enlarged mitotic nuclei in the germline ........................................................................................................... 145 1 INTRODUCTION 1 | Overview of C. elegans RNAi: RNA interference or RNAi is an extremely well-conserved pathway, evolved to combat exogenous or endogenous harmful genetic elements. The pathway is ubiquitous beyond the eukaryotic kingdom, from heat-thriving archaeal Archaeoglobus fulgidus to Homo sapiens, and can even be found in some prokaryotic species. The core factors for RNAi are a family of RNA-binding proteins called Argonaute proteins and a distinct population of RNAs that range 18-30 nucleotides, collectively known as small RNAs. Once coupled with an Argonaute protein, small RNAs provide sequence specificity to the ribonucleoprotein complex to surveil for complementarity in target mRNA transcripts. Despite the simple set up, small RNAs vastly diversify in their biogenesis, modifications, and selection of a proper Argonaute co-factor. For example, there are three flavors of small RNAs: micro-RNA (miRNA), piwi-interacting RNA (piRNA), and short-interfering RNA (siRNA). The diversification becomes more impressive with the selection of Argonaute proteins: Humans have eight Argonaute-coding genes, Arabidopsis thaliana have 10, Caenorhabditis elegans have 27, and Schizosaccharomyces pombe has only one (Carmell et al., 2002; Meister et al., 2004; Sasaki et al., 2003; Yigit et al., 2006). First observed in plants, then in C. elegans, the interference of RNA was initially thought to be the work of antisense RNA hybridizing with endogenous mRNA to suppress gene expression (Guo & Kemphues, 1995). This model proves to be incorrect when in 1998, Andrew Fire and Craig Mello injected C. elegans with double-stranded RNA and observed a profoundly more effective gene silencing than when injected with 2 purified sense or antisense single-stranded RNA (Fire et al., 1998). The silencing was so robust even with very few molecules of injected dsRNA that Fire and Mello inferred there must be an amplification process; that inference proved correct. In C. elegans, foreign dsRNA, upon being introduced via a viral host or microinjection, is first recognized and cleaved by the RNase III enzyme, Dicer or DCR-1 (Duchaine et al., 2006; Yigit et al., 2006). Following cleavage, a single strand of the duplex, either the sense or antisense strand, is selected and loaded on to the Argonaute protein RDE-1. The silencing signal is then amplified – small RNA-bound RDE-1 searches for mRNA with complementarity, triggering the synthesis of secondary siRNAs by RNA-dependent RNA polymerases (RdRPs) and the secondary siRNA pathway (Pak & Fire, 2007; Yigit et al., 2006). Harmful endogenous transcripts can also trigger production of small RNAs, bound by a distinct Argonaute, ERGO-1. These small RNAs are synthesized by the RdRP RRF-3 and are typically 26 nucleotides in length with a strong bias for guanine at the 5’ end (26G-RNA), targeting poorly conserved and recently duplicated genes (Fischer et al., 2011; Han et al., 2009; Vasale et al., 2010). Similar to the exogenous RNAi pathway, ERGO-1 26G-RNAs exist in low abundance and triggers synthesis of secondary siRNA for signal amplification. Another pathway that triggers siRNA amplification is the piRNA pathway. In C. elegans, Pol II transcribes piRNA precursors from the piRNA clusters, which are then processed into a 21- nucleotide long RNA with a strong bias for uracil at the 5’ prime end (21U-RNA) to be bound by PRG-1 to target transposons and initiate secondary siRNA amplification (Bagijn et al., 2012; Batista et al., 2008). The extremely well-conserved piwi Argonaute 3 protein PRG-1 has homology in many higher eukaryotes, from Drosophila, mouse, to human, where they are required for fertility by preventing transposon mobilization (Ghildiyal & Zamore, 2009; Izumi & Tomari, 2014; Teixeira et al., 2017). Both ERGO-1 and PRG-1 belong to the PIWI clade of Argonaute protein, and both ERGO-class 26G- RNA and 21U-RNA are subjected to methylation at their 3’ end (Montgomery et al., 2012). PRG-1 and its associated 21U-RNAs recognize endogenous transcripts with imperfect complementarity, capable of recognizing foreign non-self sequences like transgenes and transposons as well as germline-expressed transcripts and, in C. elegans, triggers amplification by the secondary siRNA pathway, thus converging with the exogenous siRNA and ERGO-class 26G-RNA pathways (Ashe et al., 2012; Bagijn et al., 2012; Lee et al., 2012; Shirayama et al., 2012). The amplification step of the secondary siRNA pathway requires the RdRPs RRF-1 and EGO-1. These RdRPs synthesize a subclass of siRNA that are typically 22 nucleotide long with a bias for a guanine at the 5’ end (22G-RNA). 22G-RNAs are bound by the worm-specific Argonaute, or WAGO, and comprise two major small RNA pathways, the WAGO 22G-RNAs and the CSR-1 22G-RNAs. In the soma, RRF-1 primarily synthesizes 22G-RNAs that are downstream of ERGO-1 and are subsequently bound by WAGO (Gent et al., 2010). In the germline, the WAGO 22G-RNA amplification components localize to the mutator complex, a perinuclear cytoplasmic compartment adjacent to the RNA-processing germ granule, the P granule (see more in 2.1 and 2.3) (Phillips et al., 2012). The complex is nucleated by the scaffolding protein MUT-16 and is a congregation hub for WAGO 22G-RNA biogenesis factors including the 4 nucleotidyltransferase MUT-2/RDE-3, the DEAD-box helicases SMUT-1/MUT-14, and the NYN-domain endoribonuclease RDE-8 (Phillips et al., 2012; Uebel et al., 2018). Ablation of the mutator complex leads to a massive loss of WAGO 22G-RNA targeting pseudogenes, intergenic regions, cryptic loci, transposons, and protein-coding, initially recognized by the ERGO-1, PRG-1, and RDE-1 pathways (W. Gu et al., 2009; Lee et al., 2012; Phillips et al., 2012; Vasale et al., 2010; C. Zhang et al., 2011). Thus, genes that are targeted by mutator-dependent WAGO 22G-RNAs are also known as mutator targets. Bound to mutator-dependent 22G-RNAs, cytoplasmic Argonaute WAGO-1 silences transcripts in the cytoplasm, whereas the nuclear, germline-expressed Argonaute HRDE-1 and its somatic variant NRDE-3 translocate into the nucleus to direct heterochromatinization of the target genomic region (Buckley et al., 2012; W. Gu et al., 2009; Guang et al., 2008). Along with the amplified 22G-RNA signals, the nuclear RNAi pathway serves as a memory of silencing, depositing repressive histone markers at genomic sites where nascent transcripts are freshly synthesized (Buckley et al., 2012; Burkhart et al., 2011; S. G. Gu et al., 2012). Thus, a single trigger recognized by the primary Argonaute protein sets off an efficient and thorough response – a systemic repression of harmful transcripts and an epigenetic memory to guard off the same harm in future generations. The RdRP EGO-1 similarly produces 22G-RNAs, but localizes to the P granules, distinctly away from the mutator complex (Claycomb et al., 2009). EGO-1 alone synthesizes 22G-RNAs that are complementary to germline-coding genes that are 5 bound by the essential Argonaute CSR-1, or Chromosome Segregation and RNAi deficient (Claycomb et al., 2009; W. Gu et al., 2009). In the absence of RRF-1, EGO-1 can compensate for the missing RdRP and synthesizes WAGO 22G-RNAs (Sijen et al., 2001). Both RRF-1 and EGO-1 complex with the helicase DRH-3 and the Tudor protein EKL-1, though the WAGO 22G-RNA RdRP module involves an additional cohort of mutator proteins in its production (Claycomb et al., 2009; W. Gu et al., 2009; Phillips et al., 2012; C. Zhang et al., 2011). Despite being partially redundant with RRF-1 and sharing cofactors, the function of EGO-1 is tightly linked with the CSR-1 22G-RNA pathway and the modulation of germline gene expression. Loss of any of the CSR-1 22G-RNA biogenesis factor, ego-1, drh-3, ekl-1, or csr-1 results sterility and/or embryonically lethality (Duchaine et al., 2006; Rocheleau et al., 2008; Smardon et al., 2000). Unlike WAGO 22G-RNA targets, CSR-1 22G-RNA targets are thought to be protected from silencing, thus “licensed”. The balance between the silencing WAGO 22G-RNAs and the licensing CSR-1 22G-RNAs, the former targeting transposons and foreign sequences (non-self) and the latter recognizing most germline protein-encoding genes (self) stands at the core of small RNA-mediated gene regulation, to both protect the genome from invasive harmful genetic elements and promote the expression of functional and essential genes. 2 | PMZ granules and the germline environment 2.1. P granules, small RNAs, and the basic principles of phase separation 6 The P in P granules signifies the key cell lineage that P granules are found in, more specifically in P4, the primordial germ cells (PGC). After the first three divisions, the constituents of P granules are robustly collected at the periphery of the nucleus of the PGCs, such that all subsequent division products will become germ cells. Constitutively expressed P granules members such as GLH-1/2/4 of the Vasa Helicase family and PGL-1/3, the RNA-binding disordered proteins, assemble into membrane- less electron-dense granules that sit atop multiple nuclear pores, to capture and process mRNA transcripts exiting the nucleus (Pitt et al., 2000). The association between P granules to the nuclear pore occurs via the FG repeats found in GLH proteins, as opposed to the RGG repeats in the PGLs, which are associated with RNA binding (Marnik et al., 2019; D. L. Updike et al., 2011; D. L. Updike & Strome, 2009). Despite having no membrane, selectivity for substrates is P granule modus operandi. Similar to the nuclear pore complex keeping passively diffused proteins > 45kDa out of the nucleus by the hydrophobic mesh created by the FG-domains of the complex, P granules extend this size exclusion into the cytoplasm (Marnik et al., 2019; D. L. Updike et al., 2011). Gatekeeping for proteins somewhere between 40kDa – 150kDa, P granules allow for passive diffusion of individual ribosome proteins less than 30kDa, but exclude assembled ribosomes from the cytoplasmic substructure (Marnik et al., 2019; D. L. Updike et al., 2011) During embryogenesis, the maternally inherited transcripts sequestered in the P granules are protected from degradation, while those expressed in somatic daughter cells are selectively degraded (G Seydoux & Fire, 1994). Within the germ line 7 blastomeres, P granules pose strict selectivity on admittance, enriching for developmentally regulated transcripts like pos-1, mex-1, gld-1 mRNA while excluding other elements like actin, tubulin, and other cytoplasmic components such as microtubules, centrioles, and mitochondria (Pitt et al., 2000; Schisa et al., 2001; D. Updike & Strome, 2010). Those that are admitted to the P granule range widely in functions, though most hinge around RNA processing, such as components of the spliceosomes, the poly(A) binding protein PAB-1, and the mRNA-cap binding protein eIF4E, recruited by PGL-1 (Amiri et al., 2001; D. L. Updike & Strome, 2009). Importantly, many components of RNAi are also found in P granules, including the piRNA Argonaute PRG-1 and the piRNA genesis complex, PARN-1 and PETISCO, the primary Argonaute proteins ALG-3/4, the secondary Argonaute WAGO-1, WAGO-3, WAGO-10, and both isoforms of CSR-1, and finally the small RNA biogenesis proteins, EGO-1, EKL-1, and DRH-3 (Claycomb et al., 2009; W. Gu et al., 2009; Nguyen & Phillips, 2021; Rodrigues et al., 2019; Schreier et al., 2020; Tang et al., 2016). The presence of RNAi proteins at the P granules is hardly coincidental – P granule morphology and subsequent germline propagation rely on the presence of some Argonaute proteins. One constitutive member of P granules, DEPS-1, physically interacts with the piRNA Argonaute PRG-1, anchoring PRG-1 at P granules to facilitate downstream piRNA activity (Spike et al., 2008; Suen et al., 2020). In the null or the PRG-1 binding mutant deps-1 mutants, downstream 22G-RNAs of piRNA targets reduce in abundance, though the level of 21U-RNAs is undisturbed (Suen et al., 2020). In this background, PRG-1 perinuclear foci, which has been characterized to localize to 8 P granules, become more densely packed, and the opposite is true for the DEPS-1 foci in the null prg-1 background, implying a mutual anchor between two proteins at the perinuclear substructure (Suen et al., 2020). The essential Argonaute protein CSR-1 even more tightly regulates P granules, that in its absence, P granules dramatically enlarge and fall off the nuclear periphery – csr-1 mutant animals are sterile and/or embryonically lethal (Claycomb et al., 2009; D. L. Updike & Strome, 2009). Compromising other CSR-1 co-factors like ego-1 and drh-3 also results in similarly drastic change in P granule morphology (see more in Chapter 1 Section 6). Thus, the health and robustness of P granules is tightly linked to siRNA activity, such that compromising either pathway results in severe disruption of the other. As kinetic as the internal composition of P granules turns out to be, the interface of P granules with the cytoplasm is just as dynamic, due to a phenomenon called phase separation. A dissected GFP::PGL-1 germline nucleus subjected to shearing forces shows that P granules behave like liquids, as the granules are capable of fusing, dripping, and wetting (Brangwynne et al., 2009). In vitro expression of PGL-3 is sufficient to reconstitute P granule droplets with similar biophysical properties to P granules in vivo, with mRNA binding to PGL-3 RGG repeats to facilitate droplet formation (Saha et al., 2016). The abundance of mRNA and small RNAs at P granules, intermixed with intrinsically disordered proteins, create a liquid-like phase-separated environment, a concept that has emerged in the past decade as a prevalent alternative for membrane-bound organelles (Elbaum-Garfinkle et al., 2015; J. Smith et al., 2016; Wei et al., 2017). Lacking the physical demarcation between compartments, the 9 membrane-less organelles rely on different biophysical properties such as the inverse capillary velocity, or the ratio between viscosity to surface tension, between subcompartments to create higher ordered organization (Feric et al., 2016). The molecular composition of the participating proteins dictates these properties. In Xenopus laevis nucleolus, the arginine/glycine-rich intrinsically disordered region of the protein fibrillarin FIB1, a member of the dense fibrillar component of the nucleolus, is sufficient to phase separate in vitro, while the structured region of FIB1 is not. However, the FIB1 structured region confers immiscibility with the protein NPM1 from the granular component of the nucleolus, such that without this region, the separation between these distinct compartments would be nonexistent (Feric et al., 2016). Conversely, in mammalian cell lines, the intrinsically disordered regions of the SMN protein cannot drive condensation, but the functional Tudor domain alone is sufficient for such task (Courchaine et al., 2021). This latter example points to yet another mode of higher organization in membrane-less organelles, one that relies on the Tudor domain competing for either asymmetrically dimethylated or symmetrically dimethylated substrates for either distinct or merged subcompartments, respectively (Courchaine et al., 2021) The role of Tudor domain in higher organization is further discussed in 2.2. Since the highlight on unstructured proteins becomes prominent, many more liquid-like granules emerge onto the scene, including the Cajal bodies, stress granules, heterochromatin and even more RNA-processing granules, namely the mutator foci and Z granules (Larson et al., 2017; Molliex et al., 2015; Strom et al., 2017; Uebel et al., 2018; Wan et al., 2018) 10 2.2. Z granules, Tudor proteins, and other intermediaries The Z granule derives its name from one of its primary constituents, the well- conserved helicase ZNFX-1. ZNFX-1 co-immunoprecipitates with WAGO-4, and together they form perinuclear granules that colocalize with P granules until the Z2/Z3 cell stage when they demix into adjacent granules (Wan et al., 2018). Both ZNFX-1 and WAGO-4 are required for RNAi inheritance and both znfx-1 and wago-4 mutants exhibit mortal germline (Mrt) phenotype at 25°C. Additionally, ZNFX-1 co-immunoprecipitates with CSR-1, PRG-1, EGO-1, and WAGO-1, all of which are known to stably localize at the P granules (Ishidate et al., 2018). Both ZNFX-1 and WAGO-4 share a role in maintaining RNAi inheritance across multiple generations, in a distinct inheritance pathway from that of the nuclear HRDE-1(Wan et al., 2018; Xu et al., 2018). One model proposing how Z granule would achieve this task was deduced from the abnormal distribution of 22G-RNAs on WAGO and CSR-1 targets in the znfx-1 mutant, where 22G-RNAs are shifted towards the 5’ end of the target mRNA. The model proposes that ZNFX-1 helicase activity is required to distribute the RdRP to the 3’ end of the targeting mRNA by unwinding the small RNA-mRNA duplex and displacing the Argonaute proteins, which would ensure the stable and balanced load of 22G-RNAs on the silencing target over time (Ishidate et al., 2018). Interesting new insights into the internal regulation of Z granules recently emerged, with the identification of the proteins that contribute to the regulation of Z granules, including the intrinsically disordered protein ZSP-1/PID-2, and the LOTUS domain protein LOTR-1 (Marnik et al., 2021; Placentino et al., 2021; Wan et al., 2021). 11 ZSP-1/PID-2 came out of two genetic screens looking for either 1) regulators of Z granule morphology (hence its name Z-granule surface protein, ZSP) or 2) genes required for 21U-RNA driven silencing (hence its other name piRNA-induced silencing defective, or PID) (Placentino et al., 2021; Wan et al., 2021). Both studies found that ZSP-1/PID-2 localizes distinctly away from the P-granules, and under stimulated emission depletion (STED) microscopy, ZSP-1/PID-2 was shown to be even distinct from ZNFX-1 in smaller puncta, surrounding and caging the much larger ZNFX-1 foci (Placentino et al., 2021; Wan et al., 2021). In fact, ZSP-1/PID-2 provides the Z granule the liquid like property that allows the Z granule to be its own phase-separated granules (Wan et al., 2021). Yet how do proteins localize to their respective condensates, when there is no shortage of membrane-less granules decorating the perinuclear estate? LOTR-1 provides a simple and elegant answer – via domain-specific interaction. The Lotus-and- Tudor-domain-containing protein LOTR-1 localizes to the Z granule via the interaction between its Tudor domain and unidentified constituents of the Z granules (Marnik et al., 2021). Similarly, the Tudor protein SIMR-1 l localizes to its own perinuclear granules via its Tudor domain to facilitate piRNA-mediated silencing (Manage et al., 2020). Some notable characteristics of SIMR-1 include: 1) SIMR-1 came out of an IP/mass spectrometry identifying interactors with MUT-16; 2) SIMR-1 does not localize to the P granule or has been demonstrated to physically interact with PRG-1; and 3) loss of SIMR-1 affects piRNA-dependent 22G-RNAs but not the upstream 21U-RNAs. Together these observations lead to the hypothesis of SIMR-1 being the mediator that 12 physically connects the piRNA to the mutator-dependent secondary siRNA pathways. This hypothesis echoes a similar proposal for the role of Tudor proteins in Drosophila germline, where they are thought to be the scaffolds of the piRNA amplification center nuage (Pek et al., 2012). Characterized by a β-barrel core and in some cases an aromatic cage, Tudor proteins belong to the “royal family” proteins that facilitate protein-protein interactions, via reading and interacting with post-translational modified peptides, namely methylated arginine or lysine (Adams-Cioaba & Min, 2009; J. Kim et al., 2006). One particularly celebrated Tudor protein is the mammalian 53BP1, or p53 binding protein, which recognizes the methylated lysine, double-stranded-break-specific histone code via its tandem Tudor domain (Panier & Boulton, 2014). Loss of 53BP1 abrogates the DSB DNA damage checkpoint response; its reduced expression is found in BRCA- associated breast cancers (Bouwman et al., 2010). While Tudor proteins that regulated histone and DNA modifications tend to interact with methylated lysine, Tudor proteins implicated in RNA metabolism are thought to interact with methylated arginine, which are often flanked by glycine and presented as the RG/RGG repeat motif (Pek et al., 2012). Structural analysis demonstrates a higher affinity of Tudor proteins with peptides containing symmetrically dimethylated arginine (sDMA), catalyzed by the protein arginine methyltransferase PRMT-5, than with asymmetrically dimethylated arginine (aDMA), catalyzed by PRMT-1 (Dhar et al., 2013; K. Liu et al., 2010). This preference for sDMA peptides is accommodated by an addition of an a-helix and two b-strands at 13 the N-terminus of the existing canonical Tudor, concisely known as an extended Tudor domain, capturing the methylated arginine in the aromatic cage (Jin et al., 2009; K. Liu et al., 2010). Germline Tudor proteins typically possess this modification, accommodating for the prevalent sDMA on Piwi protein family, a PTM event well- conserved in mammals, Drosophila, and Xenopus (Kirino et al., 2009; Kirino, Vourekas, Sayed, et al., 2010; Wang et al., 2009). Recently more Tudor proteins have been identified in the C. elegans piRNA pathway, including but not limited to PID-4, PID-5, and TOFU-6 (Goh et al., 2014; Placentino et al., 2021; Rodrigues et al., 2019). Mutations in these mutants often result in defective piRNA biogenesis, and presumably loss of downstream piRNA-dependent 22G-RNAs. In the siRNA pathway, the Tudor protein ERI-5 tethers the RdRP RRF-3 to DCR- 1, and thus bifurcates the role of DCR-1 in the exo- and endo-RNAi branches of RNAi regulations (Ketting et al., 2001; Pavelec et al., 2009; Thivierge et al., 2011). This tethering activity is crucial for the endo-siRNA activity, as knockdown of eri-5 followed by RNAi of the compensatory Tudor ekl-1 obliterates all 26G-RNA production (Thivierge et al., 2011). It is currently unclear if Tudor proteins involved in the siRNA pathway have preference for sDMA or aDMA substrate, but a recent study demonstrates that Tudor proteins can compete for either sDMA or aDMA substrates (Courchaine et al., 2021). Depending on the flavor of the methylated ligands, the Tudor protein can sequester its substrates to distinct docking sites within a granule, generating another mode of organization in an otherwise disordered environment of phase-separated condensates (Courchaine et al., 2021). 14 2.3. Mutator complex and the initiation and maintenance of 22G-RNA biogenesis The mutator complex is comprised of almost a dozen proteins, most of which are required for the synthesis of WAGO 22G-RNAs, scaffolded and nucleated by the intrinsically disordered protein MUT-16 (Phillips et al., 2012; Uebel et al., 2018). The complex situates adjacent to the P and Z granules at the perinuclear region. Like P and Z granules, the mutator complex also behaves like a liquid, phase separates, and sequesters its own microenvironment where catalytic proteins like the ribonucleotidyltransferase RDE-3/MUT-2, the RdRP RRF-1, the endoribonuclease RDE- 8, and the DEAD-box helicases SMUT-1/MUT-14 exclusively localize (Uebel et al., 2018). RDE-8 probably complexes with RDE-3/MUT-2 to first cleave the mRNA transcript targeted by RNAi, presumably to add a sequence of alternating uracil and guanidine to the 3’ end of the upstream fragments, a process called pUGylation (Tsai et al., 2015). A string of 14 to 18 UG repeats appended to the 3’ end of any transcript is sufficient as a platform for the RdRP RRF-1 to use as a template for siRNA synthesis. The current model proposes pUGylation to be a mechanism of maintaining cytoplasmic silencing across generations, by cycling siRNA production from pUG RNAs and siRNA- directed mRNA pUGylation (Shukla et al., 2020). MUT-14 and its paralog SMUT-1, two partially redundant DEAD-box helicases closely related to Drosophila Vasa, are thought to be essential for the initiation of siRNA amplification, as loss of smut-1; mut-14 causes a greater loss of endo-siRNA than other mutator mutants, though not more than mut-16 (Phillips et al., 2014). Interestingly, siRNA synthesis in vitro does not require MUT-14 catalytic activity (Phillips et al., 2014). 15 Both RDE-3/MUT-2 and RDE-8 are recruited to the relatively structured N- terminus of MUT-16, with the former directly interacting with MUT-16 and the latter indirectly tethered to MUT-16 via its interaction with NYN-1/2 which was, in turn, recruited to the Mutator complex by MUT-15 (Uebel et al., 2018). SMUT-1/MUT-14 also interacts with MUT-16 in this region. Finally, the RdRP RRF-1 localizes to MUT-16 disordered region. For a transcript to be targeted by RRF-1 for amplification, it would probably first be identified by an Argonaute protein via complementary base-pairing with its small RNA, cleaved by RDE-8 then pUGylated by RDE-3MUT-2 to prime for RRF-1. RRF-1 then complexes with the Dicer-related helicase (DRH-3) at the pUGylated RNA primer and replicates a population of 22 nt, triphosphorylated small RNAs (W. Gu et al., 2009; Pak & Fire, 2007; Sijen et al., 2007). These 22G-RNAs are subsequently bound by WAGO-1 to silent transposons, pseudogenes, and repetitive elements in the cytoplasm and by HRDE-1 and enter the nucleus. The potency and heritability of the original silencing signal hinge on RRF-1and its redundant counterpart EGO-1, for they are responsible for the total pool of de novo 22G-RNAs in C. elegans. In the soma, RRF-1 is directly downstream of the RdRP RRF-3, DCR-1, and the double-strand-RNA binding protein RDE-4, which composite the complex responsible for synthesizing the 26 nt monophosphorylated small RNA, to be bound by ERGO-1 (Gent et al., 2010). RRF-1 products in the soma are the binding substrates for the nuclear Argonaute NRDE-3 (Gent et al., 2010). Unlike RRF-1 and its dual role in somatic and germline endo-siRNA, EGO-1 is first and foremost a germline-specific RdRP, and second closely linked with the CSR-1/DRH-3/EKL-1 complex in modulating 16 germline-expressed transcripts (Maniar & Fire, 2011). EGO-1 has been reproducibly demonstrated to interact with the CSR-1 complex, via either co-IP or IP followed by mass spectrometry, and CSR-1 distinctly localizes to either the P granules or the cytoplasmic syncytium (Barucci et al., 2020; Claycomb et al., 2009). The implied locality of EGO-1 at the P-granule poses an interesting question of how the two RdRPs that 1) synthesize small RNAs from different populations of target transcripts (ERGO-1 and PRG-1 derived transcripts by RRF-1 and germline protein-encoding transcripts by EGO- 1) 2) physically localize to two distinct granules can be partially redundant with one another. In fact, in the absence of rrf-1, EGO-1 can weakly interact with MUT-16 and synthesize some of the RRF-1 targets, though the reverse does not happen (Phillips et al., 2014; Sijen et al., 2001). 3 | Concluding remarks Since the last decade, the many moving parts of the small RNA machinery have been discovered and characterized, revealing a complex interwoven relationship with the germline precursor P granules. The emergence of novel granules and the concept of liquid-liquid phase separation have allowed for researchers to tackle finer questions regarding the specialization within these granules. The granule liquid-like properties pose interesting challenges in addressing the mobility and kinetic of proteins trafficking within and in between granules (inter- vs. intra-), how the proteins are retained within each granule and by what mechanism, and how the concentration and/or properties of the involved RNA contribute to the biophysical state of each individual granules. 17 Simultaneously, despite many advancements in understanding the mechanism of C. elegans RNAi, many glaring questions regarding the small RNA pathways remain – what signals CSR-1 22G-RNAs amplification? How do WAGO-1 and CSR-1 recognize their respective triphosphorylated 22G-RNAs? More generally, how do Argonaute proteins recognize their binding substrates? Or where does siRNA-mediated silencing primarily happen, within the granules or outside of them, i.e., in the cytoplasm? 18 CHAPTER 1 Literature Review: Chromosome Segregation and RNAi Defective, CSR-1, from embryos to adult germline Dieu An H. Nguyen 1 19 1.1 Overview of Chromosome Segregation and RNAi Defective, CSR-1 Characterized by Claycomb et al., 2009, CSR-1, or Chromosome Segregation and RNAi Defective -1, was first implicated in chromosome segregation. Though CSR-1 direct role in segregation has since been significantly challenged, this first study provided several important insights that paved ways to future study which are still upheld since their first observation. One such observation is that CSR-1 and its 22G- RNA biogenesis cofactors, DRH-3, EKL-1, and EGO-1 are expressed in adult germline and embryo. Once bound to a 22G-RNA, the CSR-1 22G-RNA complex targets protein- coding germline transcripts, which was later deduced to protect them from silencing by the WAGO 22G-RNA, and in effect, licensing the transcripts for expression. So far, this licensing capacity is the most widely acknowledged function of CSR-1, working antagonistically against the silencing WAGO 22G-RNA branch that target foreign sequences in the genome. Yet CSR-1 is the only Worm-specific Argonaute to possess slicing activity both in vivo and in vitro (Aoki et al., 2007; Fassnacht et al., 2018; Gerson- Gurwitz et al., 2016). To add to the puzzle, CSR-1 target transcripts do not experience a drastic change in abundance in null csr-1 mutants; the animals, instead, go sterile and experience embryonic lethality. Of the 27 annotated Argonaute protein, CSR-1 is the only essential protein. Once expressed, CSR-1 localizes to the P granules as early as at the 4-cell stage and continues this localization until gravid adult. DRH-3, EKL-1, and EGO-1 also localize to P granules, prompting the assumption that P granule is where CSR-1 22G- RNAs are synthesized. Notably, immunofluorescence staining for the P granule marker 20 PGL-1 in the hypomorphic csr-1(tm892) mutant germline shows detachment of P granules from the perinuclear region, hinting at a unique relationship between CSR-1 and the P granules. Coming from an almost directly opposite angle, D. L. Updike & Strome, 2009 performed a genome-wide RNAi screen for genes that affect P-granules, using GFP::PGL-1 as a reporter. In csr-1(RNAi) animals, GFP::PGL-1 accumulate at a higher level compared to wild-type, form into large aggregates in both P0 germline and F1 embryos, and are typically detached from the nucleus. Interestingly, both drh-3 and ego-1(RNAi) animals also have his enlarged granule phenotype, suggesting that the entire CSR-1 22G-RNA pathway is responsible for maintaining the structure and integrity of P granules, not just CSR-1. Thus, both Claycomb et al., 2009, and Updike & Strome, 2009 catapulted CSR-1 onto the center stage as one of the essential regulators of the germline, specifically in its regulation of the germline structure, P granules. Since then, a plethora of papers dissecting CSR-1 in multiple cellular contexts at different developmental points had been published at a dizzying pace. The range of processes that CSR-1 is implicated in diversifies as the sheer volume of CSR-1 literature increases and molecular techniques and technologies advance. Some notable proposed roles for CSR-1 include maturation of core histone mRNAs, promotion of sense-oriented RNA polymerase II transcription, attenuation of translation elongation, prevention of premature activation of embryonic transcripts in oocytes, alternative splicing, and paternal inheritance (Avgousti et al., 2012; Barberán-Soler et al., 2014; Cecere et al., 21 2014; Conine et al., 2013; Fassnacht et al., 2018; Friend et al., 2012). Yet among this heap of data lacks unifying insights for how one protein can be so multi-faceted. In recent years, new studies have emerged to challenge the basic assumption of CSR-1 protecting its target transcripts, as csr-1 catalytically dead mutants phenocopies csr-1 null mutants. These papers further challenge the assumption that CSR-1 is mostly active in P granules, and new hypothesis have emerged to tie CSR-1 to the translational machinery in the cytoplasm. This chapter sets out to review these recent findings, in hope of finding a unifying model for CSR-1 and its multiple implicated roles. 1.2 Oocytes to embryos Fassnacht, C., Tocchini, C., Kumari, P., Gaidatzis, D., Stadler, M. B. & Ciosk, R. (2018). The CSR-1 endogenous RNAi pathway ensures accurate transcriptional reprogramming during the oocyte-to-embryo transition in Caenorhabditis elegans. PLoS Genetics, 14(3), e1007252. CSR-1 was first reported to be robustly expressed in the adult germline and embryos, where it localizes to P granules. Live imaging of GFP::CSR-1 shows a clear deposition of CSR-1 into oocytes, where P granules had effectively dissolved. For an oocyte to become an embryo, it must first undergo maturation, followed by ovulation and fertilization. In C. elegans, the most proximal oocyte closest to the spermatheca undergoes maturation, which is characterized by nuclear envelop breakdown (NEBD), cortical rearrangement, and meiotic chromosome which was arrested at diakinesis aligning to enter metaphase (McCarter et al., 1999). Notably, the oocyte-to-embryo transition is marked by a global repression of Pol II transcription, starting from late stages of meiosis I in oocytes to the embryonic genome activation (EGA) in embryos 22 (Edgar et al., 1994; Kelly et al., 2002; Geraldine Seydoux et al., 1996). Interestingly, two of the three CSR-1 22G-RNA biogenesis factors, ego-1 and drh-3, were identified in a genetic screen for precocious EGA, marked by gonadal expression of an EGA-GFP reporter. csr-1 (tm892) shares a similar phenotype, but not MAGO12, where all 12 WAGO genes are mutated (W. Gu et al., 2009). By examining the upregulation of several embryonic transcripts by RT-qPCR in csr-1(tm892) mutants (and not MAGO12), Fassnacht et al., 2018 concluded that CSR-1 is required to prevent precocious EGA. The premature EGA phenotype found in all three players of the CSR-1 22G-RNA pathway, csr-1, ego-1, and drh-3 suggests that it’s the entire CSR-1 22G-RNA pathway is involved in preventing this mishap. These silencing activities are contingent upon the slicing activity in CSR-1 PIWI domain and was presumably guided by its 22G-RNAs. Thus, as early as oocytes, CSR-1 is already implicated in silencing transcripts, not licensing. It’s worth noting that in oocytes, P granules, where CSR-1 localizes in adult germline, detach from the nuclear periphery, and take form as small aggregates in the cytoplasm (D. Updike & Strome, 2010). It is unclear whether at this stage CSR-1 still co-localizes with P granules, as CSR-1 appears largely cytoplasmic (Fig 1). CSR-1 slicing activity is thus more likely to happen in the cytoplasm. When the oocyte is fertilized and become an embryo, CSR-1 is diffusely expressed in the cytoplasm of the one-cell embryo, where its catalytic activity becomes essential. How exactly is CSR-1 critical in the cytoplasm? Gerson-Gurwitz et al., 2016 shed light on the role of CSR-1 in one-cell embryos, where its first role of chromatin segregation is challenged. 23 1.3 One-cell embryo Gerson-Gurwitz, A., Wang, S., Sathe, S., Green, R., Yeo, G. W., Oegema, K. & Desai, A. (2016). A Small RNA-Catalytic Argonaute Pathway Tunes Germline Transcript Levels to Ensure Embryonic Divisions. Cell, 165(2), 396–409. CSR-1 and its associated 22G-RNAs are abundantly deposited into the embryos, alongside many other transcripts and proteins. In one-cell embryos, CSR-1 is evenly distributed in both anterior and posterior ends, much like all the maternal RNAs, in contrast with the germline progenitor P granules, which migrate to the posterior end of the embryos (Gerson-Gurwitz et al., 2016; Quarato et al., 2021; G Seydoux & Fire, 1994). At this stage, compromised CSR-1 and its associated 22G-RNAs results in severe defect in microtubule assembly at the first mitotic event, a phenotype that was first attributed as defect in chromosome segregation. At first glance, csr-1(RNAi) mutants fail to properly segregate its chromosomes during the first mitotic division, which was first proposed after a fixed immunostained analysis. However, by filming the progression of the division, Gerson-Gurwtiz et al., 2016 found that the defects can be seen in the unstable positioning of the mitotic spindle along the A-P axis and unreliable spread of metaphase spindle angles relative to said axis in csr-1 (RNAi) mutants, which is due to reduction of microtubule assembly. Unlike the previous immunostaining of CSR-1 in wild-type embryos that show CSR-1 physically localizes to the mitotically dividing chromosomes, Gerson-Gurwitz et al., 2016 by live imaging a single-copy transgene of GFP::CSR-1, reported no enrichment of GFP signal on chromosomes in oocytes or early mitotic embryos. Furthermore, inhibition of CSR-1 does not alter the pattern of the centromeric histone 24 variant CENP-A along the length of the holocentric chromosomes, indicating that CSR-1 is not required for organizing either the chromosome or the centromere structures, as was previously proposed. Thus, it appears that the defects in chromosome segregation detected in csr-1 mutant from earlier study are due to the reduction of microtubule assembly, and not to CSR-1 direct role in patterning the holocentric chromosome organization. In agreement with the said early study, Gerson-Gurwitz et al., 2016 also proposed that the role of CSR-1 in the embryos require its 22G-RNAs, as RNAi on the CSR-1 22G-RNA biogenesis cofactors EKL-1/EGO-1/DRH-3 yields similar defective phenotype in microtubule positioning. How does inhibiting CSR-1 reduce microtubule assembly? MCAK KLP-7 , a microtubule depolymerase, is a target of CSR-1 slicing. Under wild-type condition, CSR- 1 binds to a 22G-RNA antisense to the maternally loaded MCAK KLP-7 mRNA, base-pairs and cleaves the transcripts to a correct concentration. When CSR-1 is catalytically dead, excess MCAK KLP-7 builds up, and microtubules rapidly disassemble. This modulation of MCAK KLP-7 by CSR-1 catalytic activities underlies the microtubule assembly defect when CSR-1 is inhibited, highlighting the role of CSR-1 in maintaining proper stoichiometry of maternally loaded transcripts. Importantly, MCAK KLP-7 is not CSR-1 only target – many proteins involved in embryonic cell division are modulated by CSR-1 slicing activity. In the absence of CSR-1, however, the effects in CSR-1 targets range from upregulation to downregulation of protein levels, some of which can be attributed to CSR-1 slicing activity in translation (see more in 1.4). 25 One last important proposal that Gerson-Gurwitz et al. 2016 brought forth is the relationship between the density of the 22G-RNAs and mRNA levels of CSR-1 targets. Target transcripts with the highest 22G-RNA density have the most increased mRNA levels in catalytically dead CSR-1, meaning CSR-1 slicing activity positively correlates with the abundance of 22G-RNAs. Thus, there are two active processes that need to happen at a given CSR-1 target mRNA: 1) CSR-1 slicing activity keeps the transcripts at a proper level, and 2) an RdRP, presumably EGO-1, concomitantly synthesizes 22G- RNAs off these transcripts to be bound by CSR-1 and continues the slicing activity, completing the cycle. Yet, which of the two processes needs to happen first to kickstart this cycle of slicing and synthesizing? This challenge proves to be a fertile ground for future studies to solidify and broaden CSR-1 role in embryos beyond the one-cell stage, a torch taken up by Quarato et al., 2021. 1.4. One-cell embryos and beyond Quarato, P., Singh, M., Cornes, E., Li, B., Bourdon, L., Mueller, F., Didier, C. & Cecere, G. (2021). Germline inherited small RNAs facilitate the clearance of untranslated maternal mRNAs in C. elegans embryos. Nature Communications, 12(1), 1441–14. In the study, Quarato et al. 2021 reaffirms the observation of CSR-1 being evenly distributed in the one-cell embryos and furthers the observation into later staged embryos. At the 4-cell and 8-cell staged embryo, CSR-1 is present in both somatic and germline precursor cells, localizing both in the cytoplasmic and at the P granules. At the 100-cell stage, CSR-1 expression is confined to the P granules, suggesting a role of CSR-1 in the somatic blastomeres in early embryos. By immunoprecipitating CSR-1 in 26 embryos and adult worms, subjecting the associated 22G-RNAs to sequencing, and classifying the targets based on the abundance of complementary 22G-RNAs made to them, the group observes that CSR-1 targets with the highest abundance of complementary 22G-RNA are mostly unique to either embryos or adults. This observation is important, as it indicates that CSR-1 targets two subsets of genes in adults and embryos. The mRNA transcripts with the higher abundance of 22G-RNAs are inherited at a much higher level than those with less complementary 22G-RNAs and have the most significant drop in transcript level by the late-stage embryos. In agreement with Gerson- Gurwitz et al., 2016, Quarato et al., 2021 found that CSR-1 catalytic activity is essential for cleaving transcripts with the highest abundance of corresponding 22G- RNA. Thus, the fate of the transcript is determined by the amount of 22G-RNAs made to it, the more 22G-RNA antisense to the transcript, the more likely the transcript would be cleaved. Such revelation reveals two contingent truths: 1) transcripts that are meant to be silenced are solely maternally loaded and thus maternally controlled and 2) the 22G- RNAs made to these transcripts were synthesized before being deposited into the embryos. In other words, CSR-1 cleaving its transcripts in the embryos is not an event carried out by embryonic programming, but entirely by what was dictated by mother. The follow-up question is how the maternal germline decides which mRNA transcripts to be synthesized at the highest level. Quarato et al., 2021 hypothesized that ribosome occupancy could be a feature that would exclude a transcript from CSR-1 targeting. Using Ribo-seq, which arrests ribosome with the associated transcripts, the 27 group found that mRNAs that are more prone to be targeted by CSR-1 are less likely to be decorated with ribosome. Furthermore, the translational efficiency of CSR-1 targeted transcripts is lower than those not targeted by CSR-1. This mechanism begs one essential question: what determines which transcripts to be bound by ribosome? Gerson-Gurwitz et al., 2016 proposed that CSR-1 is responsible for maintaining the correct stoichiometry of the maternally loaded transcripts – can multiple copies of the same transcript get stochastically decorated by ribosomes, creating a range of high ribosome occupancy to low ribosome occupancy, with the those with low ribosome occupancy being targeted by CSR-1? In a smFISH experiment, Quarato et al., 2021 demonstrated that a CSR-1 target C01G8.1 is cleared from somatic blastomeres and retained in the PGC in a control stain but is present in both somatic blastomeres and PGC when CSR-1 is depleted. What role would ribosome occupancy play for this transcript? Perhaps ribosome occupancy might dictate where the transcript can localize: C01G8.1 transcripts with high ribosome occupancy would localize to somatic and cytoplasmic blastomere, thus cleaved by cytoplasmic CSR-1. C01G8.1 transcripts with low ribosome occupancy would localize to the P granules at the PGC, where assembled ribosomes are typically excluded, and CSR-1 cleaving activity is significantly lower compared to its activity on transcripts in the cytoplasm. If the hypothesis on a single transcript having both high and low ribosome occupancy can localize to different substructures is correct, then CSR-1 targets can be both cleaved and protected, not by the merit of CSR-1 activity, but by the environment that each transcript finds itself it. 28 1.5 Attenuation of translation elongation Singh, M., Cornes, E., Li, B., Quarato, P., Bourdon, L., Dingli, F., Loew, D., Proccacia, S. & Cecere, G. (2021). Translation and codon usage regulate Argonaute slicer activity to trigger small RNA biogenesis. Nature Communications, 12(1), 3492. In recent years, CSR-1 catalytic activity receives the proper spotlight and diligent characterization, highlighting its role in maintaining proper stoichiometry of the maternally deposited transcripts in embryos. Yet CSR-1 was first observed to be robustly expressed in the adult germline, and many phenotypes attributed to CSR-1 converged at this developmental stage. Considering the essentiality of CSR-1 slicing activity in embryogenesis, it is thus worth examining if the catalytic activity is also essential in the adult germline. In 2021, Singh et al., 2021 performed these experiments and carefully staged and sorted the developmental stage of their animals to be in late L4s, which is a short timeframe when spermatogenesis begins to shut down and oogenesis begins to initiate. In CSR-1 catalytic mutants, CSR-1 targets with the most abundant 22G-RNAs suffer the most drastic loss in 22G-RNAs, and mRNA transcripts with the most 22G-RNAs mapped to them have the most increased expression. Interestingly, when CSR-1 catalytic activity is inhibited, CSR-1 target transcripts that have the most 22G-RNA mapped experience an increase in ribosome association, suggesting that in wild-type condition, CSR-1 catalytic activity is responsible for keeping a subpopulation of CSR-1 targets from being translated and expressed. Upon mapping the 22G-RNA reads to its targets, Singh et al., 2021 observes an enrichment of reads being mapped to the Transcription End Site compared to wild-type 29 and a depletion of reads at the start site. This suggests that 1) CSR-1 complementary base-pairing with the downstream transcripts starts at the 3’UTR 2) some amount of CSR-1 22G-RNAs are synthesized at this locus regardless of CSR-1 slicing activity, suggesting that CSR-1 22G-RNA RdRP EGO-1 might also localize to 3’UTR of the target transcripts and 3) CSR-1 slicing activity helps to advance the synthesis of the antisense 22G-RNAs towards the 5’ end of the transcripts, such that without it, CSR-1 22G-RNAs are stuck and accumulate at the 3’UTR. In support of the first implication that CSR-1 begins targeting transcripts at the 3’ UTR, a study by Friend et al., 2012 described a direct interaction between CSR-1 and a regulator of mRNA 3’UTR, the PUF RNA-binding protein FBF-1. CSR-1 functions with FBF-1 to repress mRNA targets through FBF binding elements at the 3’UTR. This unique interaction presumably recruites CSR-1 to the 3’UTR locus to repress translation (Crittenden et al., 2002; Friend et al., 2012). FBF-1 forms a ternary complex with CSR-1 and EFT-3, the C. elegans homolog for the elongation factor eEF1A. Upon being sequestered into the complex, EFT-3, a GTPase required to release aminoacyl-tRNAs from the ribosome, is inhibited, effectively repressing translation elongation. Perhaps the inhibition of translation elongation caused by the FBF-1/CSR-1/EFT-3 complex might serve as a stalling mechanism for EGO-1 to scan towards the 5’ end of mRNA and synthesize 22G-RNAs, without having to bypass ribosomes. EGO-1 had come out of an IP/MS analysis of CSR-1 to physically interact with CSR-1, mildly supporting the implication that EGO-1 might localize with CSR-1 to the 3’ UTR to synthesize 22G-RNAs (Barucci et al., 2020). As pointed out by Singh et al., 30 mRNA transcripts that are most effected by CSR-1 catalytic activity have the most 22G- RNAs made to them, suggesting of a positive feedback loop whereby CSR-1 slicing activity begets synthesis of new 22G-RNAs. Thus, in adult germline, CSR-1 catalytic activity is robust, despite being characterized as a protective Argonaute protein of its targeting transcripts. Notably, transcripts with the highest abundance of 22G-RNAs are cytoplasmic, and in the absence of P granules the biogenesis of CSR-1 22G-RNAs is unaffected, indicating that most of the activities regarding CSR-1 targeting transcripts with high slicing activities and 22G-RNA amplifications are excluded from the P granules. It is worth noting that the population of mRNA with high abundance of 22G-RNA constitutes only about 6.2% of total identified CSR-1 targets; almost 70% of CSR-1 targets have low abundance of 22G-RNA made to them. Do these transcripts also undergo active slicing like those in with high abundance of 22G-RNA? Or are the transcripts with low abundance of 22G-RNA somehow protected? If CSR-1 catalysis is required for generating abundant 22G-RNAs, then that would imply CSR-1 transcripts with low abundance of 22G-RNA do not experience high level of cleaving. Considering most biogenesis of CSR-1 22G-RNAs happen in the cytosol, it is thus presumable that CSR-1 targets with low abundance of 22G-RNAs are trapped in the P granules, where neither actively slicing of transcripts or the subsequent generation of 22G-RNAs are particularly robust. However, this does not preclude the possibility that CSR-1 slicing does not occur in the P granule at a lower level. In fact, there’s a body of evidence to support that this is well within the realm of possibility. 31 1.6 CSR-1 and the enlarged P-granule phenotypes Over the course of a decade of productive research on CSR-1, one phenotype emerges to be consistently robust and reproducible: enlarged P granules. First noted in Updike & Strome, 2009 in their genome-wide RNAi screen, then Claycomb et al, 2009 in their immunostaining in the hypomorph csr-1(tm892) mutant, the misshapen P granules phenotype keeps popping up in CSR-1-related mutants. The granules observed in these mutants are much larger compared to wild-type P granules, typically lose uniformity in size, and in some cases, form shapeless aggregates that fall off the nuclear periphery and suspend in the rachis. Furthermore, there are typically fewer P granules decorating the nuclear periphery, as if the granules swallow one another, undergoing Ostwald ripening. RNA staining shows accumulation of excess RNA also in the rachis of csr-1 (RNAi) adult germline. Updike & Strome, 2009 hypothesized that the accumulation of cytoplasmic RNA and of enlarged P granules in csr-1(RNAi) is due to reduced slicing and degradation of mRNAs. If this hypothesis is to be true, compromising any step of the CSR-1 regulating its target transcripts would result in the enlarged P granule phenotype. These steps would include 1) CSR-1 22G-RNA synthesis by EGO-1 and DRH-3, 2) CSR-1 binding to 22G- RNAs, and 3) CSR-1 cleaving downstream targets after complementary base-pairing. So far, there are observations of the phenotype in all three steps. First, RNAi of drh-3 and ego-1, and not rrf-3, rrf-1 or rde-3, results in enlarged P granule phenotype and accumulation of cytoplasmic RNA (D. L. Updike & Strome, 2009). Second, we have generated CSR-1 binding mutants using CRISPR, where two of the four conserved 32 residues responsible for anchoring the 5’ of small RNA into the MID domain are mutated to alanine and observed the same phenotype (Chapter 4, Fig 3). Third, catalytically dead CSR-1, though still localize to the large perinuclear granule, was reported to also form aggregates in the germline cytoplasm (Gerson-Gurwitz et al., 2016). Thus far, evidence of both genetic and molecular mutants converges to support the hypothesis that CSR-1 small RNA binding and catalytic activities modulate transcript levels in both the P granules and in the cytoplasm. Considering that most mRNA transcripts in the P granules under wild-type condition do not experience cleaving to the extent as those in the cytoplasm yet CSR-1 slicing activity is essential for the maintenance of P granule morphology, CSR-1 activity in the P granules can be inferred to maintain the proper stoichiometry of transcripts retained in each P granule (Singh et al., 2021). As previous studies have shown that the liquid droplet property of P granules is modulated by protein and RNA interactions (Brangwynne et al., 2009; Elbaum-Garfinkle et al., 2015), disturbances in the balance of these interactions can alter the fundamental molecular dynamics within the granules. Importantly, the disfiguration of the granules is not developmental-stage specific – the observation was detected in embryos, L4 and adult germlines, suggesting a shared underlying mechanism of properly maintaining the size of the P granules. 1.7 Concluding remarks This chapter sets out to review recent findings about CSR-1, specifically regarding its catalytic activity in cleaving target transcripts, which challenges the 33 protective capacity of CSR-1 on germline transcripts in the embryos and adult germline. While it is still true that germline transcripts that are targeted by CSR-1 are still expressed, CSR-1 also cleaves them to maintain the proper stoichiometry of the transcript. As has been demonstrated by Gerson-Gurwitz et al. 2016, inhibiting CSR-1 catalytic activity results in increased expressed of the tubulin depolymerase MCAK KLP-7 , which results in excessive disassembly of microtubule and consequently, defective microtubule positioning and failed chromosome segregation. The protective capacity of CSR-1, perhaps, is a reductive view of CSR-1 function on its targets. Transcripts that are recognized by CSR-1 are being fine-tuned by CSR-1 catalytic action; the outcome of this fine-tuning is a proper transcript and therefore protein expression level. These studies argue against the binary view of either the on/off state of transcripts being targeted by CSR-1 and support CSR-1 role as a gradient modulator in maintaining a balance of expression. The obvious question for this model is thus what stops CSR-1 from cleaving all transcripts that have antisense 22G-RNAs made to them? Correlation between transcripts with the most abundance of 22G-RNAs and low ribosome occupancy suggests that CSR-1 catalytic activity keeps a subpopulation of transcripts from being translated while leaving the population of transcripts with low abundance 22G-RNA seemingly unregulated. Those with high abundance of 22G-RNAs are actively cleaved in the cytosol, though it is unclear if CSR-1 targets in the P granules experience the same level of cleaving. The revelation of CSR-1 being present in both the cytoplasm and P granules, and that cytoplasmic CSR-1 is catalytically active is crucial in understanding the duality of CSR-1. To say that cytoplasmic CSR-1 is 34 catalytically active and P granule CSR-1 is not active is, again, reductive. Inhibiting CSR-1 catalytic activity results in an unusual enlargement of P granules, a phenomenon which was hypothesized to be the result of accumulated mRNAs, presumably due to lack of cleaving by P granule CSR-1. Taken together, CSR-1 appears to be a catalytically active Argnonaute protein throughout development and in both cytoplasm and P granules. Perhaps it comes down to either the free environment of the cytoplasm or the unique environment of the P granule that dictates what level of activity CSR-1 could exercise on its target transcripts and it is this imposition of the environment on the same protein that has yielded the seemingly antagonistic role of CSR-1 in both protecting and cleaving its transcripts. 35 Fig 1 | CSR-1 is largely cytoplasmic in the oocytes. Gravid adult germline, 68 hours post L1 arrest. Arrows indicate oocytes. Scale bar, 25µM. 36 CHAPTER 2: Arginine methylation promotes siRNA-binding specificity for a spermatogenesis-specific isoform of the Argonaute protein CSR-1 Dieu An H. Nguyen 1 , Carolyn M. Phillips 1 37 2.1 Abstract CSR-1 is an essential Argonaute protein that binds to a subclass of 22G- RNAs targeting most germline-expressed genes. Here we show that the two isoforms of CSR-1 have distinct expression patterns; CSR-1B is ubiquitously expressed throughout the germline and during all stages of development while CSR-1A expression is restricted to germ cells undergoing spermatogenesis. Furthermore, CSR-1A associates preferentially with 22G-RNAs mapping to spermatogenesis-specific genes whereas CSR-1B-bound small RNAs map predominantly to oogenesis-specific genes. Interestingly, the exon unique to CSR-1A contains multiple dimethylarginine modifications, which are necessary for the preferential binding of CSR-1A to spermatogenesis-specific 22G-RNAs. Thus, we have discovered a regulatory mechanism for C. elegans Argonaute proteins that allows for specificity of small RNA binding between similar Argonaute proteins with overlapping temporal and spatial localization. 2.2 Introduction In the race for evolutionary fitness, sexual selection is a strong selective force. Multicellular eukaryotes have long favored sexual divergence of the male and female genders to increase genetic diversity, often resulting in sexual dimorphism. In C. elegans, a protandrous nematode, sexual reproduction can occur in a single germline where spermatogenesis and oogenesis take place sequentially. Both eggs and sperm are derived from the same pool of mitotically replicating nuclei at the distal tip of each 38 gonad arm, which subsequently undergo meiosis as they progress through the germline toward the proximal end. During the last larval stage (L4), ~40 meiotic nuclei in each gonad arm begin the process of spermatogenesis, where they differentiate into mature sperm (Klass et al., 1976). Subsequently, oogenesis begins as the animal enters the adult stage, and the germline continues to produce oocytes for the rest of the reproductive cycle (Riddle et al., 1997). Any given germ cell has the potential to differentiate into either spermatocytes or oocytes, though once committed to a sexual fate, the cells undergo distinct meiotic programs that vary remarkably in rate and mechanics (Shakes et al., 2009). Regardless of these differences, meiosis occurs exclusively in germ cells, which are surrounded by an environment inducive to the expression of germline-specific transcripts and suppression of somatic transcripts. One mechanism by which C. elegans germ cells promote this selective environment is by way of a perinuclear structure called the P granule 4 . These germline granules, which help to ensure pluripotency and germ cell identity, form a phase-separated condensate outside of the nucleus and contiguous with nuclear pores, where they regulate newly synthesized germline transcripts (Brangwynne et al., 2009; Knutson et al., 2017; Pitt et al., 2000; Sheth et al., 2010). One key regulator of P granule morphology and C. elegans germline development is the Argonaute protein, CSR-1 (Chromosome Segregation and RNAi deficient) (Claycomb et al., 2009; D. L. Updike & Strome, 2009). Argonaute proteins are the core effectors of all small RNA-related pathways. These proteins bind small guide RNAs and regulate complementary transcripts by either 39 directly cleaving target RNAs or recruiting cofactors that promote transcriptional or post- transcriptional gene regulation (Ghildiyal & Zamore, 2009; Hutvagner & Simard, 2008). Of the ~27 Argonaute proteins annotated in the C. elegans genome, only CSR-1 is essential for fertility, and it is reported to regulate over 4,000 germline-expressed genes (Claycomb et al., 2009; Yigit et al., 2006). CSR-1 binds to a class of antisense 22- nucleotide small interfering RNAs with a 5’ guanosine (22G-RNAs) whose target transcripts are thought to be protected from silencing by other small RNA-mediated pathways, and are thus “licensed” for germline expression (Claycomb et al., 2009; Seth et al., 2013; Wedeles et al., 2013). In contrast, foreign DNA such as transposons and transgenes containing non-C. elegans sequences are recognized by piwi-interacting RNAs (piRNAs, also known as 21U-RNAs in C. elegans), which, along with their Argonaute co-factor PRG-1, can initiate a multigenerational epigenetic silencing signal that depends on chromatin factors and 22G-RNAs bound by the worm-specific Argonaute proteins (WAGOs) (Ashe et al., 2012; Bagijn et al., 2012; Shirayama et al., 2012). The balance of gene licensing and gene silencing by CSR-1 and the WAGO proteins, respectively, is critical to promote expression of essential germline genes; and a disrupted balance severely compromises germline development and fertility (de Albuquerque et al., 2015; Phillips et al., 2015). Studies of CSR-1 have reported several severe phenotypes that lead to defects in viability and fertility. In csr-1 mutant embryos, chromosomes fail to properly align on the metaphase plate, resulting in aberrant mitotic division and anaphase bridging (Claycomb et al., 2009; Gerson-Gurwitz et al., 2016; Yigit et al., 2006). In the 40 adult germline, loss of CSR-1 leads to fewer germ cells, defects in meiotic progression, and a delayed spermatogenesis to oogenesis switch (She et al., 2009). Additionally, CSR-1 has been implicated in a myriad of other cellular processes, including maturation of core histone mRNAs, promotion of sense-oriented RNA polymerase II transcription, attenuation of translation elongation, prevention of premature activation of embryonic transcripts in oocytes, alternative splicing, and paternal inheritance (Avgousti et al., 2012; Barberán-Soler et al., 2014; Cecere et al., 2014; Conine et al., 2013; Fassnacht et al., 2018). Yet how one Argonaute protein can regulate so many seemingly distinct processes remains unanswered. Here we demonstrate that CSR-1 has two isoforms – CSR-1B, which is present throughout the germline, and CSR-1A, which is specific to spermatogenic germ cells. CSR-1A and CSR-1B associate with distinct subsets of small RNAs to regulate spermatogenic or oogenic transcripts, respectively. The specificity of the two CSR-1 isoforms is interesting, considering they share nearly complete sequence homology and co-localize at the P granule in L4 larval and male germlines. We found that the first exon of CSR-1A is modified at arginine/glycine (RG) motifs by dimethylarginine. Here we show that loss of the dimethylarginine results in the loss of CSR-1A specificity for its preferred spermatogenic small RNA partners, resulting in CSR-1A indiscriminately binding to both spermatogenic and oogenic siRNAs. Thus, in this study, we have identified the first instance of methylarginine modification of a C. elegans Argonaute protein and demonstrated a previously unappreciated mechanism by which Argonaute proteins can acquire small RNA specificity. 41 2.3 The long isoform of CSR-1 is selectively expressed during spermatogenesis The Argonaute protein, CSR-1, has two isoforms, though previously little was known about their distinct functions. The longer isoform, referred to as CSR-1A, and the shorter isoform, CSR-1B, share complete sequence homology, except for a unique exon at the 5’ end of CSR-1A (Fig. 1a). Both isoforms are expressed, though at differential levels (Fig. 1b)(Claycomb et al., 2009; Gerson-Gurwitz et al., 2016). Furthermore, transcriptome analysis suggests that CSR-1A is expressed in spermatogenic gonads, but excluded from oogenic gonads (Ortiz et al., 2014). To explore the distinct functions of the two CSR-1 isoforms, we used CRISPR/Cas9 to endogenously tag CSR-1A at its N-terminus with a 2xHA/mCherry tag. We also tagged both CSR-1A and CSR-1B in the same strain by adding a 3xFLAG/GFP tag to the N- terminus of CSR-1B (Fig. 1a). When both isoforms are tagged, CSR-1A+B is strongly expressed throughout the germline during both the fourth larval (L4) and adult stages, in agreement with previous studies (Fig. 1a) (Claycomb et al., 2009). Interestingly, when only CSR-1A is tagged, expression is restricted to the spermatogenesis region of the germline in L4 hermaphrodites and males (Fig. 1a, Extended Data Fig. 1). We confirmed this observation by western blot analysis of all larval stage and adult animals, detecting CSR-1A expression robustly during the L4 stage and very weakly in gravid adults, coinciding temporally with spermatogenesis in C. elegans (Fig. 1c). Taken together, these results demonstrate that though both CSR-1 isoforms are expressed in the germline, CSR-1A expression is limited to germ cells undergoing spermatogenesis while CSR-1B is expressed throughout the germline in larval and adult animals. 42 2.4 CSR-1A and CSR-1B are expressed from independent promoters To address how CSR-1A and CSR-1B have such distinct expression patterns, we sought to determine whether they are expressed from independent promoters. We first fused ~1.5kb of DNA preceding the csr-1a start codon to an mCherry reporter and exogenously expressed it in the germline using the MosSCI system (Fig. 1d) (Frøkjær- Jensen et al., 2008). The region preceding the csr-1a start codon, the putative csr-1a promoter, drives mCherry expression only during the L4 stage and, during this time, expression is restricted to the spermatogenesis region of the germline. In gravid adults, residual mCherry expression can be detected inside the spermatheca, but is completely absent from the rest of the germline (Fig. 1d). To determine whether any regulatory sequences reside in the intron between the unique csr-1a exon and the start codon for csr-1b, we fused ~0.5kb of DNA from this intron to a GFP reporter and expressed it in the germline using the MosSCI system (Fig. 1d). This region, the putative csr-1b promoter, drives GFP expression throughout the germline cytoplasm, in all developmental stages. The GFP expression can be observed in oocytes and fertilized embryos but is excluded from the spermatheca (Fig. 1d). These promoter fusion experiments corroborate our western blot analysis and indicate that CSR-1A is expressed predominantly during spermatogenesis, hinting at a unique role for CSR-1A during this developmental time point. Furthermore, the promoter fusion transgenes were able to recapitulate the expression pattern of the isoform-specific translation fusion proteins (Fig. 1a), indicating that the promoters are sufficient to drive the distinct expression patterns. Together, these data indicate that the intronic region that sits 43 between the first csr-1a exon and the start codon for csr-1b contains the promoter sequence for CSR-1B, and that their distinct promoters independently establish the differential expression patterns of CSR-1A and CSR-1B. 2.5 CSR-1A localizes to the P granules during spermatogenesis To further characterize the expression pattern of CSR-1A, we immunostained for CSR-1A and DAPI-stained for germline nuclei in dissected L4 hermaphrodite gonads. As the germline progresses towards the proximal end, germ cells undergo early stages of meiosis up until pachytene in a non-sex-specific manner (Shakes et al., 2009). In post-pachytene, however, spermatocytes and oocytes commit to their respective sexual fates, and spermatocyte chromosomes exhibit an extensive condensation phase (Shakes et al., 2009). During L4, when spermatogenesis begins, CSR-1A expression is first observed in cells that are exiting pachytene, namely cells transitioning into diplotene and karyosome, indicating that CSR-1A is selectively present in spermatocytes (Fig. 2a). This expression is reminiscent of the sperm-specific Argonaute protein ALG-3, which is also present in spermatocytes during spermatogenesis (Conine et al., 2010). To further investigate the role of CSR-1A during spermatogenesis, we carefully examined protein localization at different time points. As ALG-3 is another Argonaute protein that is expressed during spermatogenesis and is required for proper sperm development, we created a double-transgenic strain labeling both CSR-1A and ALG-3, using CRISPR. At early L4, or 45 hours post hatching, both CSR-1A and ALG-3 form perinuclear foci in the spermatogenesis region that colocalize with the P granule 44 marker PGL-1 (Fig. 2b,c). In contrast, CSR-1A+B forms perinuclear foci throughout the germline (Extended Data Fig. 2a). As the animals reach the L4 to adult transition, 52 hours post hatching, CSR-1A and ALG-3 still colocalize during early spermatogenesis. However, by the time germ cells reach the secondary spermatocyte stage, CSR-1A expression becomes undetectable, while ALG-3 expression persists (Fig. 2b, lower panel, Extended Data Fig. 2b). PGL-1 similarly disappears from P granules in primary spermatocytes and is subsequently cleared from secondary spermatocytes (Amiri et al., 2001). Together, these data demonstrate that like ALG-3, CSR-1A localizes to P granules during spermatogenesis, but unlike ALG-3, CSR-1A expression is restricted to the early stages of spermatogenesis and is not found in secondary spermatocytes. 2.6 CSR-1A is required for optimal sperm fertility We next sought to determine the role of CSR-1A in fertility. We first created two mutant alleles of csr-1a using CRISPR (Fig. 3a). csr-1a(cmp135) removes the first seven amino acids including the start codon and csr-1a(cmp143) removes the same region plus ~500 bp of the csr-1a promoter region. We confirmed that the mutants have severely reduced csr-1a expression via RT-qPCR and western blot analysis (Extended Data Fig. 3a,b). We first examined the fertility of csr-1a mutants by comparing the brood size of csr-1a(cmp135) hermaphrodites to wild-type hermaphrodites at 20ºC and 25ºC. We did not observe a significant reduction in brood size in csr-1a(cmp135) mutants compared to wild-type animals at 20ºC. However, at 25ºC, the brood size of csr- 1a(cmp135) mutant animals was significantly reduced compared to wild-type animals 45 (Fig. 3b). Because CSR-1A is expressed during spermatogenesis, we next sought to determine if CSR-1A is required for optimal male fertility. One aspect of sperm quality that can be tested in vitro is sperm activation. Sperm activation is the process by which round spermatids are transformed into mature and motile spermatozoa (H. E. Smith, 2014). This process can be recapitulated in vitro using Pronase E, a cocktail of proteases that induces sperm maturation and pseudopod formation (Samuel Ward et al., 1983). Because C. elegans fertility is reduced at 25°C, we raised wild-type and csr-1a mutant males at 25°C to assess sperm activation under compromised conditions. Wild-type spermatids from males that were raised at 25°C for one generation were activated at a rate of 97%, and after two generations at 25°C, they were activated at a rate of 96% (Fig. 3c). In csr-1a males, we saw a modest reduction in the percentage of activated spermatids after one and two generations at 25°C (69% for cmp135 and 70% for cmp143 after one generation, and 64% for cmp135 and 66% for cmp143 after two generations) (Fig. 3c). Because the effects we observed on sperm activation were modest yet significant, we generated two additional alleles of csr-1a to confirm our results. These new alleles remove the majority of the unique csr-1a exon (Extended Data Fig. 3c). Similar to what we observed with csr-1a(cmp135) and csr- 1a(cmp143), the two new alleles, csr-1a(cmp253) and csr-1a(cmp254) reduced the percentage of activated spermatids by approximately 25-30% (Extended Data Fig. 3d). Together, these data suggest that CSR-1A contributes to optimizing fertility and is necessary to promote sperm activation. 46 2.7 CSR-1A and CSR-1B are expressed independently of one another Disruption of CSR-1 expression results in severe infertility and embryonic lethality (Claycomb et al., 2009; Yigit et al., 2006). To determine if our csr-1a alleles also affect the expression of CSR-1B protein, we introduced a 2xFLAG tag immediately following the start codon of CSR-1B in the csr-1a(cmp135) mutant background using CRISPR (Fig. 3d). Similar to wild-type, CSR-1B in the csr-1a(cmp135) mutant localized to perinuclear foci associated with the P granule marker, PGL-1 (Fig. 3e). Furthermore, by western blot, the CSR-1B protein was expressed in wild-type and csr-1a(cmp135) mutant animals, whereas the CSR-1A was undetectable in the csr-1a(cmp135) mutant but present in the wild-type strain (Fig. 3f). Interestingly, we observed that CSR-1B appears to accumulate at a modestly higher level in the csr-1a(cmp135) mutant, but not in the csr-1a(cmp143) mutant (Extended Data Fig. 3b). To determine if CSR-1B is required for CSR-1A expression, we generated an effectively null allele of csr-1b, where we mutated the csr-1b start codon to isoleucine, in the mCherry::CSR-1A strain using CRISPR (Fig. 3g). In this csr-1b(cmp258) mutant, mCherry::CSR-1A is expressed at perinuclear foci in the spermatogenesis region of L4 animals, similar to wild-type expression of CSR-1A (Fig. 3h). These animals, however, are sterile and need to be maintained over a balancer, a phenotype previously associated with csr-1 mutants (Yigit et al., 2006). Because CSR-1B is expressed during spermatogenesis and, like ALG-3, is present in the secondary spermatocytes, (Extended Data Fig. 2a-b), we next asked if CSR-1B contributes to optimizing sperm health. To this end, we performed in vitro 47 sperm activation on csr-1b(cmp258) mutants at 25°C. We observed that 39% of csr-1b mutant sperm are activated after one generation raised at 25°C, in contrast with 97% activation rate in the control animals (Fig. 3c). This rate of spermatid activation drops further to 22% after two generations at 25°C while the rate remains at 97% in the control animals. Furthermore, among the population of activated sperm in the csr-1b mutant, we occasionally observed a striking phenotype of spiky pseudopods, reminiscent of the compromised sperm of the alg-3/4 mutants (Extended Data Fig. 3e) (Conine et al., 2010). This phenotype was never observed in wild-type or csr-1a mutants. Thus, these data demonstrate that both CSR-1 isoforms are required for robust sperm activation. Together, these data demonstrate that CSR-1A and CSR-1B do not depend on one another for localization or expression, and thus they behave as independent proteins with seemingly specialized functions in the C. elegans germline. More importantly, the data suggest that both CSR-1 isoforms are intimately involved in optimizing sperm health, though the mechanism used by each isoform to achieve this end still needs to be determined. 2.8 CSR-1A and CSR-1B target distinct groups of genes To investigate the small RNAs bound by each CSR-1 isoform and thus determine whether CSR-1A and CSR-1B have distinct small RNA partners, we immunoprecipitated CSR-1A (from the 2xHA::csr-1a strain), CSR-1B (from the csr- 1a(cmp135); 2xFLAG::csr-1b strain), and CSR-1A+B (from the 2xFLAG::csr-1b strain), and sequenced the associated small RNAs. CSR-1A, CSR-1B, and CSR-1A+B were 48 immunoprecipitated from L4 stage animals, and CSR-1B and CSR-1A+B were additionally immunoprecipitated from the adult stage, for comparison. We found that the small RNAs that immunoprecipitate with CSR-1A, CSR-1B, and CSR-1A+B are enriched for germline genes at both L4 and adult stages (Fig. 4a and Extended Data Fig. 4a). However, CSR-1A preferentially associates with small RNAs mapping to spermatogenic genes at the L4 stage while CSR-1B preferentially associates with small RNAs mapping to oogenic genes both at L4 and adult stages (Fig. 4a and Extended Data Fig. 4a). CSR-1B is expressed at much higher levels than CSR-1A at the L4 stage (Figs. 1b and 3f), therefore we would expect that when we tag both isoforms together (CSR-1A+B) the majority of the small RNAs immunoprecipitated would associate with CSR-1B. Indeed, we observed that the majority of the small RNAs immunoprecipitated with CSR-1A+B map to oogenic genes at both L4 and adult stages (Fig. 4a and Extended Data Fig. 4a). Regardless of isoform, the majority of the small RNAs associated with CSR-1 are 22-nucleotide long with a strong bias for guanine as the 5’ nucleotide (Extended Data Fig. 4b-d). These data indicate that, while both isoforms of CSR-1 bind 22G-RNAs mapping to germline genes, CSR-1A is enriched for siRNAs mapping to spermatogenesis-specific genes and CSR-1B is enriched for siRNAs mapping to oogenesis-specific genes. We next sought to define a list of genes targeted by CSR-1A and CSR-1B. We defined the CSR-1A, CSR-1B, and CSR-1A+B target genes at both L4 and adult stages as those with complementary small RNAs at least two-fold enriched in the IP compared to input, with at least 10 RPM in the IP samples and a DESeq2 adjusted p- 49 value of ≤0.05 (Supplementary Table 1). Comparing these gene lists with one another, we found that the majority of CSR-1B target genes are also CSR-1A+B target genes at both L4 (94.5% overlap) and adult stages (81.4% overlap) (Fig. 4b,c). Furthermore, the CSR-1B target genes at L4 stage overlap significantly with the CSR-1B target genes at adult stage (94.0% overlap), and similarly the CSR-1A+B target genes at L4 stage overlap significantly with the CSR-1A+B target genes at adult stage (78.0% overlap) (Extended Data Fig. 4e-f). In contrast, only 28.8% of CSR-1A target genes at L4 stage overlap with CSR-1B target genes at L4 stage, and only 46.6% of CSR-1A target genes at L4 stage overlap with CSR-1A+B target genes at L4 stage, despite the CSR-1A+B IP at L4 stage immunoprecipitating both isoforms (Fig. 4b). Thus, these data demonstrate that CSR-1A targets a distinct set of genes and that due to the much higher expression of CSR-1B, immunoprecipitation of the two isoforms together at the L4 stage enriches for small RNAs that predominantly map to CSR-1B target genes. To further our understanding of which genes are targeted by CSR-1A and CSR-1B, we compared the CSR-1A, CSR-1B, and CSR-1A+B target gene lists to previously described target gene lists for other small RNA pathways (Conine et al., 2013; W. Gu et al., 2009; Lee et al., 2012; Ortiz et al., 2014; Phillips et al., 2014; Reinke et al., 2004). As expected, CSR-1B and CSR-1A+B are strongly enriched for hermaphrodite and male CSR-1 target genes, and more modestly enriched for ALG-3/4 target genes (Fig. 4d). In contrast, CSR-1A is only modestly enriched for hermaphrodite CSR-1 target genes, and strongly enriched for ALG-3/4 target genes and male CSR-1 target genes (Fig. 4d). It has previously been shown that male CSR-1 target genes, 50 identified by immunoprecipitating both isoforms of CSR-1 in males, include both oogenesis-specific genes identified as CSR-1 targets in hermaphrodites, and male- specific genes that are also ALG-3/4 targets (Conine et al., 2013). Thus, it is not surprising that both CSR-1B, which preferentially binds hermaphrodite CSR-1 target genes, and CSR-1A, which preferentially binds ALG-3/4 target genes, are enriched for male CSR-1 target genes (Fig. 4d). In further agreement with these data, the CSR-1A target gene list is strongly enriched for spermatogenic genes, and the CSR-1B and CSR-1A+B target gene lists are strongly enriched for oogenic genes (Fig. 4d). In contrast, neither CSR-1A, CSR-1B, nor CSR-1A+B are enriched for mutator target genes, which include pseudogenes, repetitive elements, transposons, and other germline-repressed genes (Fig. 4d). Together, these data further show that CSR-1A and CSR-1B bind small RNAs targeting germline genes associated with distinct small RNA pathways, with CSR-1A preferentially targeting spermatogenic genes and ALG-3/4 targets and CSR-1B preferentially targeting oogenic genes and hermaphrodite CSR-1 targets, 2.9 CSR-1A does not require ALG-3/4 to bind spermatogenic small RNAs Because CSR-1A shares a similar expression pattern to ALG-3 and targets spermatogenic genes, we next sought to determine whether the small RNAs bound to CSR-1A require ALG-3/4 for their production. To this end, we immunoprecipitated CSR- 1A in an alg-3/4 mutant background and sequenced the associated small RNAs. Interestingly, in the absence of alg-3 and alg-4, CSR-1A was still preferentially loaded 51 with small RNAs targeting spermatogenic genes (Fig. 4a,d). In addition, 82.4% of CSR- 1A target genes in the wild-type background were also defined as targets of CSR-1A in the alg-3; alg-4 double mutant background (Fig. 4e). Furthermore, CSR-1A target genes in the alg-3; alg-4 double mutant background were strongly enriched for ALG-3/4 target genes and male CSR-1 target genes, and modestly enriched for hermaphrodite CSR-1 target genes, similar to CSR-1A target genes in a wild-type background (Fig. 4d). Thus, these data show, despite CSR-1A and ALG-3/4 targeting many of the same genes, CSR-1A is not directly downstream of ALG-3/4 in regulating spermatogenic gene expression and ALG-3 and ALG-4 are dispensable for the production of CSR-1A-bound spermatogenic small RNAs. 2.10 CSR-1A expression is positively correlated with the expression of CSR-1A target genes It is well-established that CSR-1 targets germline-expressed genes, and that CSR-1 can either license and protect these transcripts from silencing by the piRNA pathway in the adult germline or can modestly tune the transcript level in developing oocytes and embryos (Claycomb et al., 2009; Gerson-Gurwitz et al., 2016; Seth et al., 2013). To determine whether CSR-1A is similarly protecting or tuning its target transcripts, we generated mRNA high-throughput sequencing libraries from wild-type, csr-1a(cmp135) mutants, and csr-1a(cmp143) mutants at 20°C (Supplementary Table 2). In both csr-1a mutant alleles, CSR-1A targets are modestly, but significantly, downregulated compared to wild-type (Fig. 5a). Similarly, spermatogenic genes are 52 significantly downregulated in both csr-1a mutants compared to wild-type (Fig. 5a). In contrast, mutator target genes were unchanged in both csr-1a mutants (Fig. 5a). To confirm that CSR-1A target transcripts are indeed reduced in the csr-1a mutants, we performed RT-qPCR on three spermatogenic genes that were strongly enriched for mapped small RNAs in the CSR-1A immunoprecipitation. We determined that all three genes showed a consistent and significant decrease in transcript levels in both csr-1a mutants compared to wild-type animals (Fig. 5b). In contrast, the csr-1a mutants had no effect on morc-1, a germline gene that was only enriched in CSR-1B and CSR1A+B immunoprecipitation, and not in CSR-1A (Fig. 5b). Therefore, CSR-1A appears to have modest but significant positive effects on its target transcripts, such that in its absence, these genes have reduced expression. Recent studies have examined the relationship between the levels of siRNAs targeting each transcript in correlation with mRNA expression (Bezler et al., 2019; Gerson-Gurwitz et al., 2016). In general, mRNAs expression tends to be lower for genes with very high levels of siRNAs and for genes that are targets of the Argonaute protein, WAGO-1 (Bezler et al., 2019). In contrast, mRNA expression tends to be higher for genes that are targets of CSR-1 (Bezler et al., 2019). To determine whether this trend holds true for CSR-1A target genes, for comparison, we generated a list of WAGO-1 target genes at the L4 stage by immunoprecipitating WAGO-1 and defining targets as those at least two-fold enriched in a WAGO-1 IP compared to input, with at least 10 RPM in the IP samples, and a DESeq2 adjusted p-value of ≤0.05. Then using mRNA and small RNA sequencing libraries from wild-type animals at L4 stage, we 53 found that WAGO-1 targets have significantly higher levels of mapping complementary small RNAs compared to CSR-1A and CSR-1B targets (Fig. 5c-e). Moreover, mRNA expression of WAGO-1 targets is significantly lower than mRNA expression of CSR-1A and CSR-1B target genes (Fig. 5c-e). Therefore, the CSR-1A bound small RNAs and target mRNAs are more similar in expression level to those of CSR-1B compared to WAGO-1, further supporting the argument that CSR-1A may function similarly to CSR- 1B to promote expression of its target genes. We next grouped the CSR-1A target genes into seven bins based on their small RNA expression in wild-type animals. We found that higher levels of CSR-1A siRNAs (>15 RPM in wild-type animals) correlates with reduced mRNA expression in csr-1a mutants, while the expression of genes with lower levels of CSR-1A siRNAs tends to be unchanged in csr-1a mutants (Fig. 5f). This result is consistent with both the very modest effect of the csr-1a mutants on transcript level of all CSR-1A target genes (Fig. 5a) and the much more significant effect of the csr-1a mutants on the transcript level of the three CSR-1A target genes tested by qPCR (Fig. 5b), which were selected based on the strong enrichment for their mapped small RNAs in the CSR-1A immunoprecipitation. Together, these data suggest that CSR-1A may share a similar licensing mechanism to what has been previously proposed for CSR-1, and that in csr-1a mutants, the spermatogenic transcripts licensed by CSR-1A are no longer protected and may be subjected to degradation. 54 2.11 The first exon of CSR-1A is unstructured and contains RG motifs CSR-1A differs from its shorter counterpart by a single N-terminal exon. The exon is arginine/glycine-rich (containing RG motifs), and we sought to determine whether this unique exon is conserved across the Caenorhabditis genus. We first identified CSR-1 orthologs in several closely related nematode species, including C. brenneri, C. briggsae, C. japonica, C. latens, C. nigoni, C. remanei, C. sinica, and C. tropicalis. The first exon of CSR-1 in each species is the least conserved portion of the protein; however, like C. elegans CSR-1A, each ortholog had enrichment of arginines and glycines in this region (Extended Data Fig. 5). Furthermore, while only a single isoform is annotated in most of the species, all orthologs possessed a conserved methionine at the position of the C. elegans CSR-1B start codon (Extended Data Fig. 5). Additionally, all eight orthologs have a large intron between the first exon and the rest of the protein, with a median size of 617 bp and the smallest being 503 bp in C. tropicalis. These data suggest that the regulatory elements driving CSR-1B expression in C. elegans, which are found in this intron, may also be conserved. Because of the repetitive nature of the RG motif region and the lack of otherwise strong sequence conservation, we next asked whether the first exon of CSR- 1 contains regions of intrinsic disorder. Using IUPred2, we determined that the first exon of CSR-1A in C. elegans is highly disordered while the rest of the protein is predicted to be structured (Extended Data Fig. 6a). We then examined the CSR-1 orthologs in related Caenorhabditis species and found that, like in C. elegans, the first exon of CSR- 1 is highly disordered in all related nematode species (Extended Data Fig. 6b), 55 demonstrating that the disordered nature of the first exon is a conserved feature of the CSR-1 protein. 2.12 The RG motifs in the first exon of CSR-1A are dimethylated It has been shown previously that RG motifs are often targets of Protein Arginine Methyltransferases (PRMTs), which catalyze the methylation of arginine residues (Kirino et al., 2009; Kirino, Vourekas, Kim, et al., 2010; Nishida et al., 2009; Vagin et al., 2009). To determine whether CSR-1A contains methylated arginines, we immunoprecipitated 2xHA::CSR-1A and subjected the sample to mass spectrometry analysis. We identified six dimethylated arginines in the first exon of CSR-1A, which constituted 100% of the RG motifs captured by mass spectrometry (Fig. 6a and Supplementary Table 3). The remaining RG motifs were not captured, methylated or unmethylated, by our mass spectrometry experiment. We did not identify any dimethylarginines in the portion of the CSR-1 protein found in both CSR-1A and CSR- 1B isoforms, however the conserved RGRG sequence found near the N terminus of CSR-1B, a good candidate region for dimethylation, was also not captured by our mass spectrometry experiment. We were also unable to determine whether these arginines were symmetrically or asymmetrically dimethylated, which will be an important distinction for future studies. Together, these data indicate that the first exon of CSR-1A is heavily methylated, and, given the conservation of RG motifs in this region across Caenorhabditis species, this methylation is likely also conserved. 56 We next sought to study the function of the dimethylarginine modification. To this end, we mutated all arginines found in RG motifs within the first exon of CSR-1A to alanine using CRISPR, thus rendering the exon unmethylatable. Because we generated the mutations sequentially from the N-terminus of the protein, three-to-four arginines at a time, we created a series of proteins with four, seven, 11 or all 15 RG motifs mutated to AG – hereafter referred to as CSR-1A[4xAG], CSR-1A[7xAG], CSR-1A[11xAG] and CSR-1A[15xAG] (Fig. 6b). We first asked if dimethylation of arginines affects the expression or stability of the CSR-1A protein. By western blot, we determined that the CSR-1A[4xAG], CSR-1A[7xAG] and CSR-1A[15xAG] proteins are expressed at levels comparable to wild-type (Fig. 6c). Because dimethylation has been shown previously to contribute to Argonaute protein localization (Webster et al., 2015), we next tested whether dimethylation is necessary for CSR-1A localization to P granules. We performed immunostaining on CSR-1A[wild-type], CSR-1A[4xAG], and CSR-1A[15xAG] L4 stage hermaphrodites and detected no discernable differences in localization in the RG-to-AG mutant strains (Extended Data Fig. 6c). Finally, to determine whether the dimethylated RG motifs are required for fertility, we performed brood size assays on CSR-1A[wild-type], CSR-1A[7xAG], and CSR-1A[15xAG] animals. We found that CSR- 1A[7xAG] and CSR-1A[15xAG] animals have a modest but significant reduction in brood size compared to the control strain, indicating that csr-1a RG motif mutants phenocopy a null csr-1a mutant (Extended Data Fig. 6d). Therefore, our data demonstrates that the first exon of CSR-1A contains the dimethylarginine modification, however this modification is not required for expression or localization of the CSR-1A 57 protein. Furthermore, since the brood sizes were similar between CSR-1A[7xAG] and CSR-1A[15xAG], this data may suggest that the seven most N-terminal RG sites are more crucial or that a minimum number of methylated sites is required for optimal CSR- 1A function. 2.13 The RG motifs in CSR-1A promote specificity for binding small RNAs targeting spermatogenic genes CSR-1A and CSR-1B both localize to P granules in the spermatogenesis region of the germline in L4 stage hermaphrodites and males, yet each isoform has distinct small RNA targets (Figs. 1a and 4a,b). Because the isoforms only differ by the presence of the RG-motif-containing exon in CSR-1A, we asked if the RG-motifs and the dimethylarginine modification could be a mechanism by which CSR-1A preferentially associates with spermatogenic small RNAs. To test this hypothesis, we immunoprecipitated 2xHA::CSR-1A[15xAG] from L4 stage animals and subjected the associated small RNAs to high-throughput sequencing. We found that the small RNAs that immunoprecipitate with CSR-1A[15xAG] map to both spermatogenic and oogenic genes, with no discernable preference for either (Fig. 6d). We next defined the list of genes targeted by CSR-1A[15xAG] as those at least two-fold enriched in the IP compared to input, with at least 10 RPM in the IP samples and a DESeq2 adjusted p- value of ≤0.05 (Supplementary Table 1). Comparing this gene list to the wild-type CSR- 1A and CSR-1B target gene lists at L4 stage, we found that 34.9% (978 genes) of the genes in the CSR-1A[15xAG] list overlapped with the wild-type CSR-1A gene list and 58 55.5% (1555 genes) of the genes in the CSR-1A[15xAG] list overlapped with the CSR- 1B gene list (Fig. 6e). Furthermore, when we compared the CSR-1A[15xAG] target gene list to previously described target genes lists for other small RNA pathways (Conine et al., 2013; W. Gu et al., 2009; Lee et al., 2012; Ortiz et al., 2014; Phillips et al., 2014; Reinke et al., 2004; Tsai et al., 2015; C. Zhang et al., 2011), we found that the CSR-1A[15xAG] target gene list is strongly enriched for hermaphrodite and male CSR-1 target genes, and for ALG-3/4 target genes (Fig. 6f). Additionally, the CSR-1A[15xAG] target gene list is strongly enriched for both spermatogenic and oogenic genes (Fig. 6f). These results are in contrast to the wild-type CSR-1A target gene list, which is strongly enriched for spermatogenic genes but not oogenic genes, and to the CSR-1B target gene list, which is strongly enriched for oogenic genes but not spermatogenic genes (Fig. 6f). However, like wild-type CSR-1A and CSR-1B, CSR-1A[15xAG] is not enriched for mutator target genes (Fig. 6f). Together, these data indicate that loss of the RG motifs, and therefore loss of the dimethylarginine modification, is associated with a reduced specificity of CSR-1A for small RNAs targeting spermatogenesis genes and an increased binding of oogenic small RNAs. 2.14 Discussion The Argonaute protein CSR-1 is critical to promote germline gene expression and fertility. Here we demonstrate that, in addition to the short isoform CSR-1B which is expressed throughout germline development and during gametogenesis of both sexes, a longer isoform, CSR-1A, is selectively expressed during spermatogenesis. During L4, 59 where both isoforms are expressed in the spermatogenesis region, CSR-1A and CSR- 1B co-localize at P granules. Yet despite localizing to the same subcellular structure and sharing nearly complete sequence and presumably structural identity, CSR-1A and CSR-1B associate with distinct subsets of small RNAs targeting germline expressed genes – CSR-1A preferentially binds to small RNAs targeting spermatogenic genes, whereas CSR-1B binds to small RNAs targeting oogenic genes. The single exon unique to CSR-1A contains RG motifs that are modified with dimethylarginine and these methylated RG motifs are critical to CSR-1A specificity for spermatogenic small RNA substrates. CSR-1A promotes spermatogenic gene expression Previous work has shown that CSR-1 can license targeted transcripts for expression in the adult germline, while its catalytic domain is necessary to tune embryonic gene expression and clear maternal transcripts (Gerson-Gurwitz et al., 2016; Quarato et al., 2021; Seth et al., 2013; Wedeles et al., 2013). Though these studies did not make a distinction between CSR-1A and CSR-1B, based on the expression of only CSR-1B in the adult germline and embryos, we can attribute these CSR-1 activities to the CSR-1B isoform. Here we show that CSR-1A similarly does not down-regulate its target genes during spermatogenesis and in fact, seems to modestly promote spermatogenic gene expression. The Argonaute proteins ALG-3 and ALG-4 similarly promote the expression of many spermatogenic target genes, and trigger the biogenesis of ALG-3/4-dependent 60 22G-RNAs as part of a feed-forward loop to promote paternal inheritance (Conine et al., 2010, 2013). With ALG-3 and CSR-1A sharing nearly identical expression patterns and CSR-1A being a spermatogenic Argonaute with a preference for spermatogenic 22G- RNAs, an obvious hypothesis would be that CSR-1A acts downstream of ALG-3 and ALG-4. However, the CSR-1A spermatogenic 22G-RNAs are produced independently of ALG-3/4 indicating that CSR-1A is likely part of a distinct pathway that targets the same spermatogenic genes as ALG-3/4 and another still-unidentified Argonaute binds the ALG-3/4-dependent 22G-RNAs. Importantly, these findings do not exclude the possibility of CSR-1B being an essential player in paternal epigenetic inheritance and acting downstream of ALG-3/4, as CSR-1B is still present in secondary spermatocytes with ALG-3, when CSR-1A has already disappeared (Conine et al., 2013). At the L4 stage, when both isoforms are expressed, our immunoprecipitation and sequencing of CSR-1B-bound small RNAs indicate that CSR-1B targets gender-neutral and oogenic mRNAs. It is important to note that, during the L4 stage, CSR-1B is expressed throughout the germline while CSR-1A and ALG-3 are restricted to the spermatogenic region (Fig. 2b, Extended Data Fig. S2a) (Conine et al., 2010). Thus, when sequencing small RNAs immunoprecipitated from whole animals, we are unable to determine whether CSR-1B is binding oogenic siRNAs throughout the germline or in specific germline regions, and whether CSR-1B could bind low levels of spermatogenic siRNAs during spermatogenesis that fall below our cutoff for enrichment. Furthermore, since the majority of CSR-1 studies have examined adults or embryos (Claycomb et al., 2009; Gerson-Gurwitz et al., 2016; Seth et al., 2013; Wedeles et al., 2013), it is unclear 61 whether, during the L4 stage, CSR-1B is licensing or repressing its target mRNAs, or even protecting them from ALG-3/4 or CSR-1A targeting. Or perhaps oogenic mRNAs are at such low levels throughout development that most CSR-1B Argonaute proteins remain unassociated with mRNA until the transition to adulthood and the onset of oogenesis. If CSR-1A does not act downstream of ALG-3/4 for paternal inheritance, then how might it function to promote spermatogenic gene expression? If we look to CSR-1B as a guide, then perhaps CSR-1A acts to license spermatogenic transcripts for expression only in the L4 and male germlines by protecting them from targeting by piRNAs and the WAGO clade of Argonaute proteins. CSR-1A could further tune spermatogenic gene expression or clear spermatogenic transcripts at the spermatogenesis to oogenesis transition. Alternatively, the role for CSR-1A may be to sequester the abundant sperm transcripts and spermatogenic small RNAs away from CSR-1B, so that the dominant isoform can appropriately bind to its oogenic small RNA targets. Through this lens, CSR-1A serves almost exclusively to titrate CSR-1B siRNA levels targeting oogenic genes and might explain the modest effect on the sperm transcripts upon removing CSR-1A. Further experiments will be needed to sort out these possibilities. Dimethylarginine promotes isoform-specific small RNA loading Post-translational modification has been shown to play a key role for Argonaute function in C. elegans and in other systems. The C. elegans miRNA 62 Argonaute protein, ALG-1, contains a cluster of phosphorylation sites that is conserved in human Argonaute Ago2, demonstrating that these modification sites are highly conserved between species. The function of this phosphorylation also appears to be conserved – the phosphorylated Argonaute proteins cannot associate with mRNAs, suggesting the role of the modification may be in mediating release of mRNA-Argonaute complexes for recycling (Huberdeau et al., 2017). PIWI proteins from mouse, Xenopus laevis, and Drosophila melanogaster contain symmetrical dimethylarginines (sDMA), and in Drosophila this modification is mediated by protein arginine methyltransferase PRMT5 (Kirino et al., 2009; Reuter et al., 2009; Vagin et al., 2009). The dimethylarginine modification allows the Argonaute protein to interact with members of the Tudor domain protein family (H. Liu et al., 2010; Nishida et al., 2009; Reuter et al., 2009; Vagin et al., 2009; Webster et al., 2015). Tudor domains, which are protein- protein interaction modules, recognize methylated arginines and thus can mediate protein-protein interactions in a methylation-specific manner. Furthermore, for some PIWI proteins, subcellular localization to nuage is dependent on its interaction with Tudor proteins (Vagin et al., 2009; Webster et al., 2015). Here, we have found the first exon of CSR-1A to be highly modified with dimethylarginine. Interestingly, dimethylarginine is not required for CSR-1A localization to the P granule, but instead for small RNA specificity of CSR-1A, and thus recognition of the correct target transcripts. We hypothesize that CSR-1A interacts with an unknown Tudor domain protein through its dimethylated RG motifs, to promote engagement with a distinct small RNA biogenesis complex. In this scenario, dimethylarginine provides a 63 new binding platform for proteins that cannot associate with CSR-1B, allowing CSR-1A to make distinct protein-protein interactions and ultimately to target a unique subset of genes. It is interesting to note that the majority of the RG motifs, including all methylated sites captured in our mass spectrometry experiment, are found in the exon unique to CSR-1A and absent from CSR-1B, making this highly modified region isoform-specific. There are several other C. elegans Argonaute proteins with multiple splice variants, including the miRNA Argonaute proteins ALG-1 and ALG-2, the oogenesis-specific primary Argonaute protein ERGO-1, and the WAGO-clade Argonaute protein PPW-1. It is currently unknown whether the isoforms of any of these genes have distinct functions or unique protein modifications, or if CSR-1 is unique in this regard. 2.15 Methods C. elegans strains Strains were maintained at 20°C on NGM plates seeded with OP50 E. coli according to standard conditions unless otherwise stated (Brenner, 1974). Plasmid and strain construction Plasmid-based CRISPR: All fluorescent and epitope tags were integrated at the endogenous loci by CRISPR genome editing (Arribere et al., 2014; Dickinson et al., 2013, 2015; Friedland et al., 2013; J. D. Ward, 2015). For all CRISPR insertions of fluorescent tags, we generated homologous repair templates. Design of the 2xHA::mCherry plasmid was described previously (Uebel et al., 2018). The 64 2xHA::mCherry[w/internal Floxed Cbr-unc-119(+)] was amplified by PCR and assembled by isothermal cloning with ~1.5kb of sequence from either side of the csr-1a start codon (Gibson et al., 2009). gfp::3xFLAG::wago-1, gfp::3xFLAG::csr-1a+b and gfp::3xFLAG::alg-3 were assembled into pDD282 (Addgene #66823) by isothermal assembly according to published protocols (Dickinson et al., 2015; Gibson et al., 2009). To protect the repair template from cleavage, we introduced silent mutations at the site of guide RNA targeting by incorporating these mutations into one of the homology arm primers or, if necessary, by performing site-directed mutagenesis (Dickinson et al., 2013). All guide RNA plasmids were generated by ligating oligos containing the guide RNA sequence into BsaI-digested pRB1017 (Addgene #59936) (Arribere et al., 2014). GFP/mcherry CRISPR injection mixes included 25-50 ng/μl repair template, 50 ng/μl guide RNA plasmid, 50 ng/μl eft-3p::cas9-SV40_NLS::tbb-2 3’UTR (Addgene #46168), 2.5–10 ng/μl GFP or mCherry co-injection markers. The 2xHA::mCherry::csr-1a construct was injected into USC868 (mut-16(cmp3[mut-16::gfp + loxP]) I ; unc-119(ed3) III), the gfp::3xFLAG::wago-1 construct was injected into USC896 (mut-16(cmp41[mut- 16::mCherry::2xHA + loxP]) I), the gfp::3xFLAG::csr-1a+b construct was injected into the wild-type strain, and the gfp::3xFLAG::alg-3 construct was injected into both the wild-type strain and USC1066 (csr-1a(cmp90[(2xHA + mCherry + loxP Cbr-unc-119(+) loxP)::csr-1a]) IV) (Dickinson et al., 2013, 2015). For csr-1a deletions (cmp135 and cmp143), 2xHA::csr-1a, and 2xFLAG::csr-1a+b, the injection mixes included 50 ng/μl repair oligo, 25 ng/μl guide RNA plasmid, 50 ng/µl pha-1 repair template, and eft- 3p::Cas9 + pha-1 guide (pJW1285, Addgene #61252). GE24 (pha-1(e2123) III) mutant 65 animals were injected and subsequently shifted to restrictive temperature (25°C). Surviving F1 progeny were genotyped by PCR to identify the deletions and insertions of interest (J. D. Ward, 2015). For 2xFLAG::csr-1b, the injection mix included 50 ng/μl csr- 1b repair oligo, 25 ng/μl csr-1b guide RNA plasmid, 20 ng/µl dpy-10 repair template, 25 ng/µl dpy-10 guide RNA (pJA58, Addgene #59933), and 50 ng/µl eft-3p::Cas9 (pJW1259, Addgene #61251). Mixture was injected into USC1065 (csr-1a(cmp135) IV) mutant animals. F1 animals with the Rol phenotype were isolated and genotyped by PCR to identify animals with the 2xFLAG insertion (Arribere et al., 2014). The csr-1a RG-to-AG mutants were created sequentially, starting with csr-1a[7xAG]. csr-1a[4xAG] was an incomplete repair event identified from the csr-1a[7xAG] injections. csr- 1a[11xAG] was injected into the csr-1a[7xAG] mutant and csr-1a[15xAG] was subsequently injected into the csr-1a[11xAG] mutant. The csr-1a RGG injection mixes include 50 ng/µl repair template, 25 ng/µl each of two guide RNA plasmids, 25 ng/µl rol- 6 guide RNA (pJA42, Addgene #59930), 20 ng/µl rol-6 repair template, and 50 ng/µl eft- 3p::Cas9 (pJW1259, Addgene #61251). F1 animals with the Rol phenotype were isolated and genotyped by PCR to identify animals with the RG-to-AG mutations (Arribere et al., 2014). Protein-based CRISPR: For For csr-1a deletions (cmp253 and cmp254), csr- 1b(cmp258), and 2xHA::csr-1a in the alg-3; alg-4 mutant, we used an oligo repair template and RNA guide. The injection mixes for csr-1b(cmp258) and 2xHA::csr-1a included 0.25 μg/μl Cas9 protein (IDT), 100 ng/μl tracrRNA (IDT), 14 ng/μl dpy-10 66 crRNA, 42 ng/μl gene-specific crRNA, and 110 ng/μl of each repair template, and were injected into USC1066 (csr-1a(cmp90[(2xHA + mCherry + loxP Cbr-unc-119(+) loxP)::csr-1a]) IV) and WM200 (alg-4(ok1041) III; alg-3(tm1155) IV), respectively. The injection mix for csr-1a(cmp253) and csr-1a(cmp254) included 0.25 µg/µl Cas9 protein (IDT), 100 ng/μl tracrRNA (IDT), 14 ng/μl dpy-10 crRNA, 21 ng/μl each gene-specific crRNA, and 110 ng/µl of each repair template, and was injected into the wild-type strain. The repair template was designed to create the cmp254 mutation; the larger cmp253 deletion was an incorrect repair event. Following injection, F1 animals with the Rol phenotype were isolated and genotyped by PCR to identify animals with the insertions and deletions of interest (Dokshin et al., 2018; Paix et al., 2015). MosSCI: csr-1 mCherry and GFP promoter fusions were integrated by Mos-mediated single-copy transgene insertion (MosSCI) (Frøkjær-Jensen et al., 2008). For the MosSCI insertions of promoter-fused GFP and mCherry, we amplified csr-1a and csr-1b endogenous promoters, the mCherry and GFP genes, and the csr-1 3’UTR. Plasmids were assembled by isothermal cloning (Gibson et al., 2009). For MosSCI injections, we integrated transgenes into the ttTi5605 MosI site in strain EG4322 (Ch. II) following a published MosSCI protocol (Frøkjær-Jensen et al., 2008). Injection mixes contained 50 ng/µl MosSCI-targeting vector, 50 ng/µl eft-3p::Mos1 transposase (pCFJ601, Addgene #34874), 10 ng/µl rab-3p::mCherry (pGH8, Addgene #19359), 2.5 ng/µl myo- 2p::mCherry (pCFJ90, Addgene #19327), 5 ng/µl myo-3p::mCherry (pCFJ104, Addgene 67 #19328), and 10 ng/µl hsp-16.1::peel-1negative selection (pMA122, Addgene #34873) (Frøkjær-Jensen et al., 2008). Antibody staining and imaging Live imaging of C. elegans was performed in M9 buffer containing sodium azide to prevent movement. For immunofluorescence, C. elegans were dissected in egg buffer containing 0.1% Tween-20 and fixed in 1% formaldehyde in egg buffer as described (Phillips et al., 2009). Samples were immunostained with mouse anti-FLAG 1:500 (Sigma Aldrich, F1804), mouse anti-PGL-1 1:100 (DSHB K76) (Strome & Wood, 1983), and rat anti-HA 1:500 (Roche, 11867423001). Alexa-Fluor secondary antibodies were purchased from Thermo Fisher. Animals were dissected at the L4 (48 hours post hatching) or young adult stage (52 hours post hatching). Imaging was performed on a DeltaVision Elite microscope (GE Healthcare) using a 60x N.A. 1.42 oil-immersion objective. When data stacks were collected, three-dimensional images are presented as maximum intensity projections. Images were pseudocolored using the SoftWoRx package or Adobe Photoshop. Western blots C. elegans were synchronized at 20°C by bleaching gravid adult animals and maintaining starved L1 larvae for at least 24 hours before plating on OP50. For sample collection, animals were harvested after 2 hours (L1), 12 hours (L2), 32 hours (L3), 50 hours (L4), and 72 hours (gravid adult) on OP50. Number of animals loaded per lane 68 was normalized for actin – approximately 1000 L1s, 800 L2s, 600 L3s, 400 L4s, and 200 gravid adults. Proteins were resolved on 4-12% Bis-Tris polyacrylamide gels (Thermo Fisher), transferred to nitrocellulose membranes (Thermo Fisher), and probed with rat anti-HA-peroxidase 1:1,000 (Roche 12013819001), mouse anti-FLAG 1:1,000 (Sigma, F1804), mouse anti-actin 1:10,000 (Abcam ab3280), or rabbit anti-CSR-1 1:2,000 antibodies (Claycomb et al., 2009). Secondary HRP antibodies were purchased from Thermo Fisher. RNA Isolation and qRT-PCR Synchronized animals were collected at L1, L2, L3, L4, and gravid adult stages, after two, 12, 32, 48, and 68 hours of feeding following L1 arrest, respectively. For mRNA libraries, synchronized L4 stage animals (~48 hours at 20°C and ~34 hours at 25°C after L1 arrest) were collected. RNA was isolated using Trizol reagent (Thermo Fisher), followed by chloroform extraction and isopropanol precipitation. RNA samples were normalized to 10µg/µL prior to DNase treatment (TURBO DNA-free kit, Thermo Fisher AM1907) and reverse transcribed with SuperScript III Reverse Transcriptase (Thermo Fisher 18080-051). All Real time PCR reactions were performed using the 2x iTaq Universal SYBR Green Supermix (Bio-Rad 1725121), following manufacturer’s protocols, and run in the CFX96 Touch Real-Time PCR System (Bio-Rad 1855195). Samples were run with three technical replicates and three biological replicates, and normalized to rpl-32. 69 Brood size analysis Wild-type and mutant C. elegans strains were maintained at 20°C prior to temperature- shift experiments. Animals were either maintained at 20°C or shifted to 25°C as L4 larvae and ten of their progeny were picked to individual plates for 25°C brood size analysis. To score the complete brood, each animal was moved to a fresh plate every day until egg-laying was complete. After allowing the progeny 2-3 days to develop, the total number of animals on each plate was counted. In vitro sperm activation assay Virgin L4 males were isolated 24 hours before assay. 10-15 males were dissected in 30uL of 500µg/mL Pronase E (Millipore Sigma), which was dissolved in sperm medium (50 mM HEPES, 50 mM NaCl, 25 mM KCl, 5 mM CaCl2, 1 mM MgSO4, 1 mg/ml BSA), as described previously(Samuel Ward et al., 1983; Yen et al., 2020). Spermatids were incubated for 15 minutes at room temperature in a humid chamber before mounting and imaging on a DeltaVision Elite microscope (GE Healthcare) using a 60x N.A. 1.42 oil- immersion objective. Immunoprecipitations and mass spectrometry For immunoprecipitation experiments followed by small RNA library preparation, ~100,000 synchronized L4 animals (~49 hours at 20°C after L1 arrest) or adult animals ~68 hours at 20°C after L1 arrest) were collected in IP Buffer (50 mM Tris-Cl pH 7.4, 100 mM KCl, 2.5 mM MgCl2, 0.1% Igapal CA-630, 0.5 mM PMSF, cOmplete Protease 70 Inhibitor Cocktail (Roche 04693159001), and RNaseOUT Ribonuclease Inhibitor (Thermo Fisher 10777019)), frozen in liquid nitrogen, and homogenized using a mortar and pestle. After further dilution into IP buffer (1:10 packed worms:buffer), insoluble particulate was removed by centrifugation and 10% of sample was taken as “input.” The remaining lysate was used for the immunoprecipitation. Immunoprecipitation was performed at 4°C for 1 hour with pre-conjugated anti-HA affinity matrix (Roche 11815016001) or anti-FLAG affinity matrix (Sigma Aldrich A22220), then washed at least 3 times in immunoprecipitation buffer. A fraction of each sample was analyzed by western blot to confirm efficacy of immunoprecipitation. Trizol reagent (Thermo Fisher) was added to the remainder of each sample, followed by chloroform extraction, isopropanol precipitation, and small RNA library preparation. For mass spectrometry experiments to identify post-translational modifications, immunoprecipitation was performed as described above, starting with ~1.25 million synchronized USC1110 (csr-1a(cmp165[2xHA::csr-1a])) L4 stage animals (~48 hours at 20°C after L1 arrest). Wild-type animals were prepped alongside as a negative control. Immunoprecipitation was performed using anti-HA affinity matrix (Roche 11815016001). After immunoprecipitation, a fraction of each sample was analyzed by western blot to confirm efficacy of immunoprecipitation. 2x sample buffer was added to the remainder of each sample, followed by gel electrophoresis (4-12% Bis-Tris polyacrylamide gels, Thermo Fisher) and overnight colloidal Coomassie staining. Bands containing immunoprecipitated protein were excised from gel and cut into approximately 1 mm 3 pieces. Gel pieces were then subjected to a modified in-gel 71 chymotrypsin digestion procedure (Shevchenko et al., 1996). Gel pieces were washed and dehydrated with acetonitrile for 10 min. followed by removal of acetonitrile. Pieces were then completely dried in a speed-vac. Rehydration of the gel pieces was with 50 mM ammonium bicarbonate solution containing 12.5 ng/µl modified sequencing-grade chymotrypsin (Promega) at 4ºC. After 45 min., the excess chymotrypsin solution was removed and replaced with 50 mM ammonium bicarbonate solution to just cover the gel pieces. Samples were then incubated at room temperature overnight. Peptides were later extracted by removing the ammonium bicarbonate solution, followed by one wash with a solution containing 50% acetonitrile and 1% formic acid. The extracts were dried in a speed-vac (~1 hr) and stored at 4ºC until analysis. On the day of analysis, the samples were reconstituted in 5 - 10 µl of HPLC solvent A (2.5% acetonitrile, 0.1% formic acid). A nano-scale reverse-phase HPLC capillary column was created by packing 2.6 µm C18 spherical silica beads into a fused silica capillary (100 µm inner diameter x ~30 cm length) with a flame-drawn tip (Peng & Gygi, 2001). After equilibrating the column each sample was loaded via a Famos auto sampler (LC Packings) onto the column. A gradient was formed and peptides were eluted with increasing concentrations of solvent B (97.5% acetonitrile, 0.1% formic acid). As peptides eluted, they were subjected to electrospray ionization and then entered into an LTQ Orbitrap Velos Pro ion-trap mass spectrometer (Thermo Fisher). Peptides were detected, isolated, and fragmented by collision-induced dissociation to produce a tandem mass spectrum of specific fragment ions for each peptide. Peptide sequences 72 (and hence protein identity) were determined by matching protein databases with the acquired fragmentation pattern by the software program, Sequest (Thermo Fisher) (Eng et al., 1994). All databases include a reversed version of all the sequences and the data was filtered to between a one and two percent peptide false discovery rate. Small and mRNA library preparation Small RNAs (18 to 30-nt) were size selected on denaturing 15% polyacrylamide gels (Bio-Rad 3450091) from total RNA samples. Small RNAs were treated with 5’ RNA polyphosphatase (Epicentre RP8092H) and ligated to 3’ pre-adenylated adapter with Truncated T4 RNA ligase (NEB M0373L). Small RNAs were then hybridized to the reverse transcription primer, ligated to the 5’ adapter with T4 RNA ligase (NEB M0204L), and reverse transcribed with Superscript III (Thermo Fisher 18080-051). Small RNA libraries were amplified using Q5 High-Fidelity DNA polymerase (NEB M0491L) and size selected on a 10% polyacrylamide gel (Bio-Rad 3450051). Library concentration was determined using the Qubit 1X dsDNA HS Assay kit (Thermo Fisher Q33231) and quality was assessed using the Agilent BioAnalyzer. Libraries were sequenced on the Illumina NextSeq500 (SE 75-bp reads) platform. For mRNA sequencing, total RNA samples were submitted in triplicate to Novogene Genome Sequencing Company for library preparation. Libraries were sequenced on the Illumina (PE 150-bp reads) platform. 73 Clustal Omega alignments and IUPred Disorder prediction Clustal Omega alignment was performed using protein sequences for CSR-1 orthologs available on Wormbase (www.wormbase.org)(Sievers et al., 2011). When more than one CSR-1 ortholog was present in a given species, a single protein sequence was selected for analysis. Proteins are C. brenneri CBN29996 (WBGene00191961), C. briggsae CSR-1 (WBGene00037276), C. elegans CSR-1, isoform a (WBGene00017641), C. japonica CSR-1 (WBGene00126657), C. latens PRJNA248912_FL83_15994, C. nigoni CSR-1 (PRJNA384657_Cni-csr-1), C. sinica PRJNA194557_Csp5_scaffold_00781.g15416.t2, C. remanei PRJNA248911_FL82_23103, C. tropicalis PRJNA53597_Csp11.Scaffold629.g12789.t1. For C. japonica, exon 1 of CSR-1A was manually annotated from the CJA07453.1 transcript. For disorder prediction, we used IUPred2A (https://iupred2a.elte.hu/) with long disorder parameters and the same protein sequences as were used for Clustal Omega alignment (Dosztányi et al., 2005; Mészáros et al., 2018). Bioinformatic Analysis For small RNA libraries, sequences were parsed from adapters using FASTQ/A Clipper (options: -Q33 -l 17 -c -n -a TGGAATTCTCGGGTGCCAAGG) and quality filtered using the FASTQ Quality Filter (options: -Q33 -q 27 -p 65) from the FASTX-Toolkit (http://hannonlab.cshl.edu/fastx_toolkit/), mapped to the C. elegans genome WS258 using Bowtie2 v. 2.2.2 (default parameters) (Langmead & Salzberg, 2012), and reads were assigned to genomic features using FeatureCounts (options: -t exon -g gene_id -O 74 --fraction –largestOverlap) which is part of the Subread v. 1.5.1 package (Liao et al., 2013, 2014). Differential expression analysis was done using DESeq2 v. 1.22.2 72 . To define gene lists from IP experiments, a 2-fold-change cutoff, a DESeq2 adjusted p- value of ≤0.05, and at least 10 RPM in the IP libraries were required to identify genes with significant changes in small RNA levels. Additionally, any genes identified as having differentially enriched small RNAs from control samples (HA or FLAG immunoprecipitations from wild-type animals), were removed from further analysis. For mRNA libraries, sequences were parsed from adapters using Trimmomatic v. 0.36 (options: PE -phred33 ILLUMINACLIP:<fasta with adaptor sequences>:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:30 MINLEN:30) (Bolger et al., 2014) and mapped to the C. elegans genome WS258 using HISAT2 v. 2.1.0 (options: --dta- cufflinks --known-splicesite-infile <path to file of known splice sites>) (D. Kim et al., 2015). Differential expression analysis was done using Cuffdiff v. 2.1.1 (default parameters) (Love et al., 2014). CSR-1 target genes, male CSR-1 target genes, ALG-3/4 target genes, mutator target genes, germline-enriched genes, spermatogenesis-enriched genes, and oogenesis- enriched genes were previously described (Bezler et al., 2019; W. Gu et al., 2009; Lee et al., 2012; Manage et al., 2020; Ortiz et al., 2014; Phillips et al., 2014; Reinke et al., 2004; Tsai et al., 2015; C. Zhang et al., 2011). Additional data analysis was done using R, Excel, and Python. Venn diagrams were generated using BioVenn (Hulsen et al., 2008) and modified in Adobe Illustrator. 75 Data Availability High-throughput sequencing data for RNA-sequencing libraries generated during this study are available through Gene Expression Omnibus (GSE151828). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository 78 with the dataset identifier PXD021227 and 10.6019/PXD021227. 76 Fig 1 | CSR-1 isoforms have distinct spatial and temporal expression patterns. a, Live imaging of CSR-1A (top) and CSR-1A+B (bottom) in an L4 hermaphrodite germline. Scale bars, 25µM. b, Western blot detecting CSR-1 expression, using α-CSR-1 antibodies, in wild-type animals at L4 stage. c, Western blot detecting for CSR-1A at each larval stage and gravid adult in 2xHA/mCherry::CSR-1A strain using α-HA antibody. Actin is shown as a loading control. d, Schematic representation (top) of putative csr-1a and csr-1b promoters driving mCherry and GFP. The csr-1a promoter comprises ~1.5kb of sequence preceding the first csr-1a exon, and the csr-1b promoter is the entire 544-bp intron between the unique csr-1a exon and the start codon of csr-1b. Schematic of the promoter reporter constructs created using the MosSCI system for csr-1a (middle) and csr-1b (bottom) are shown above images of gonads from L4 and gravid adult stages expressing the reporter constructs. Asterisk indicates intestinal autofluorescence visible near the bend in adult csr-1a::mCherry gonad. Scale bars, 25µM. 77 Fig 2 | CSR-1A is expressed in the spermatogenesis region of the germline. a, Immunofluorescence staining for HA::CSR-1A in dissected L4 hermaphrodite germline, using α-HA. Asterisk indicates distal end. Inset indicates region of cells exiting pachytene and entering post-pachytene where CSR-1A expression is detected. Dashed circles indicate karyosomes. DNA is stained with DAPI, blue. Scale bar, 15µM. b, 3xFLAG/GFP::ALG-3 (left), 3xHA/mCherry::CSR-1A (middle), and merge (right) at 45 hrs (early L4 stage) or 52 hrs (young adult stage) post-L1 arrest. White arrows indicate region of secondary spermatocytes. Scale bars, 25 µM. c, Immunofluorescent staining of a 2xHA::mCherry::CSR-1A; GFP::3xFLAG::ALG-3 dissected L4 hermaphrodite germline, using antibodies against HA, FLAG, and PGL-1.Scale bars represent 5µM. 78 79 Fig 3 | CSR-1 isoforms function independently of one another. a, Schematic representation of two deletion alleles of csr-1a. cmp135 removes 20 bp of coding sequence, including the start codon. cmp143 removes ~500 bp of promoter sequence and 20 bp of coding sequence. b, Brood-size assay performed on wild-type and csr-1a mutant animals at 20ºC and 25ºC. Three replicates were performed for each genotype, with 10 animals per replicate. Error bars indicate standard deviation. c, An in vitro sperm activation assay was performed on spermatids from wild-type, csr-1a(cmp135), csr- 1a(cmp143), mCherry::CSR-1A, and mCherry::CSR-1A csr-1b(cmp258) mutants. Animals were raised at 25°C for either one or two generations and each experiment was performed in duplicate. At least 100 spermatids were counted for each replicate for each condition, for a total of at least 300 spermatids per condition. Error bars indicate standard deviation. d, Schematic representation of 2xFLAG::CSR-1B in a csr-1a(cmp135) mutant. The 2xFLAG tag was introduced immediately after the CSR-1B start codon in a csr-1a(cmp135) mutant animal. e, Immunofluorescence staining for 2xFLAG::CSR-1B and P granules (PGL-1) in a csr-1a(cmp135) mutant at L4 stage. Scale bar, 5µM. f, Western blot for CSR-1 expression levels, in wild-type and csr-1a(cmp135) mutant animals at L4 stage. Actin is shown as a loading control. g, Schematic representation of disrupting csr-1b expression in a 2xHA/mCherry::CSR-1A strain. The csr-1b start codon was mutated from methionine to isoleucine (M1I) using CRISPR, which also introduces a M164I point mutation in the CSR-1A protein. h, Live imaging of mCherry::CSR-1A in a wildtype and csr-1b(cmp258) mutant L4 hermaphrodite. Scale bar, 25 µM. Two-tail T-tests were performed to determine statistical significance. n.s. denotes not significant and indicates a p-value > 0.05, * indicates a p-value ≤ 0.05, ** indicates a p-value ≤ 0.01 and, *** indicates a p-value ≤ 0.001. 80 81 Fig 4 | CSR-1A and CSR-1B target distinct groups of genes. a, Normalized reads for oogenesis and spermatogenesis genes at L4 stage from HA::CSR-1A IP, FLAG::CSR-1B IP, FLAG::CSR-1A+B IP, and HA::CSR-1A; alg-3/4 IP compared to input. Oogenesis and spermatogenesis genes are indicated in blue and red respectively. Grey dotted line indicates 2-fold enrichment in IP relative to input. b, Venn diagram indicates overlap of CSR-1A, CSR-1B, and CSR-1A+B target genes at L4 stage. Target genes are defined are at least 2-fold enriched in the IP, with at least 10 RPM in IP samples and a DESeq2 adjusted p-value ≤ 0.05. c, Venn diagram indicates overlap of CSR-1B and CSR-1A+B gene lists at adult stage. d, Enrichment analysis (log2(fold enrichment)) examining the overlap of CSR-1A, CSR-1B, and CSR-1A+B target genes with known targets of the CSR-1, male CSR-1, ALG-3/4, and mutator small RNA pathways and oogenesis and spermatogenesis-enriched genes. See Materials and Methods for gene list information. Statistical significance for enrichment was calculated using the Fisher’s Exact Test function in R. n.s. denotes not significant and indicates a p-value > 0.05 and **** indicates a p-value ≤ 0.0001. e, Venn diagram indicates overlap of CSR-1A target genes in the presence or absence of alg-4; alg-3. 82 83 Fig 5 | CSR-1A targets spermatogenesis-expressed genes. a, Box plots depicting log2(fold change mRNA abundance) for L4 stage csr-1a(cmp135) (left) and csr- 1a(cmp143) (right) mutants relative to wild-type animals at 20ºC. Statistical significance was calculated in R using the Student’s t-test. b, RT-qPCR from wild-type, csr-1a (cmp135), and csr-1a (cmp143) L4 animals raised at 20ºC for R05D7.2, Y41C4A.7, and F44D12.6, genes that are highly enriched in HA::CSR-1A IPs, and morc-1, which is enriched in FLAG::CSR-1B and FLAG::CSR-1A+B IPs, but not HA::CSR-1A IPs. Relative expression was normalized over rpl-32 and calculated relative to wild-type animals. Two-tail T-tests were performed to determine statistical significance. Error bars indicate SEM. c-d, Small RNA expression (RPKM) plotted against mRNA expression (FPKM) in wild-type animals for CSR-1B (c) and CSR-1A (d) target genes compared against WAGO-1 target genes. CSR-1B, CSR-1A, and WAGO-1 gene lists come from FLAG::CSR-1B IP, HA::CSR-1A IP, and FLAG::WAGO-1 IP in L4 animals, respectively. e, Box plot depicting small RNA reads per kilobase per million (RPKM, left) and mRNA fragments per kilobase per million (FPKM, right) from L4 wild-type animals, raised at 20ºC, 48 hours post L1. Small RNAs and mRNAs transcripts were grouped based on their mapping to CSR-1B, CSR-1A, and WAGO-1 target genes, identified from the RNA-IP experiments. f, Box plot depicting log2(fold change mRNA abundance) of CSR-1A target genes for L4 stage csr- 1a(cmp135) mutants relative to wild-type animals at 20ºC where the CSR-1A target genes are binned based on the mapped small RNA reads per million (RPM) in wild-type animals. Statistical significance was calculated using the Student’s t-test in R. n.s denotes not significant and indicates a p-value > 0.05, * indicates a p-value ≤ 0.05, ** indicates a p-value ≤ 0.01, *** indicates a p-value ≤ 0.001, and **** indicates a p-value ≤ 0.0001. 84 85 Fig 6 | Methylation of CSR-1A N-terminal exon promotes binding preference for spermatogenic siRNAs. a, Graphical display of dimethylation modifications detected on CSR-1A by mass spectrometry following IP. b, Schematic representation of the series of arginine to alanine 928 mutations created by CRISPR to generate the unmethylatable CSR-1A[15xAG]. c, Western blot for HA::CSR-1A in the RG-to-AG mutants. Actin is shown as a loading control. d, Normalized reads for spermatogenesis and oogenesis genes from HA::CSR-1A[15xAG] IP compared to input. Spermatogenesis and oogenesis are highlighted in red and blue respectively. Grey dotted line indicates 2-fold enrichment in IP relative to input. e, Venn diagram indicates overlap of CSR-1A, CSR-1B, and CSR-1A[15xAG] target genes at L4 stage. f, Enrichment analysis (log2(fold enrichment)) examining the overlap of CSR-1A, CSR-1B, and CSR- 1A[15xAG] target genes with known targets of the CSR-1, male CSR-1, ALG-3/4, and mutator small RNA pathways and oogenesis and spermatogenesis enriched genes. See Materials and Methods for gene list information. Statistical significance for enrichment was calculated using the Fisher’s Exact Test function in R. n.s denotes not significant and indicates a p-value > 0.05 and **** indicates a p-value ≤ 0.0001. See 86 Fig S1 | CSR-1A is expressed in the spermatogenesis region of the male germline. Immunofluorescence staining of CSR-1A in a dissected L4 male germline carrying the 2xHA::mCherry::CSR-1A transgene. HA antibodies are used to recognize CSR-1A and DNA is stained with DAPI. Scale bar, 20µM. Fig S2 | CSR-1A is excluded from secondary spermatocytes. a, 3xFLAG/GFP::CSR-1A+B at 45 hrs (early L4 stage) or 52 hrs (young adult stage) post-L1 arrest. White arrows indicate region of secondary spermatocytes. b, Live imaging of double-transgenic animal labelled for both GFP::ALG-3 and mCherry::CSR-1A at the young adult stage. Scale bars, 25µM. 87 Fig S3 | CSR-1A is required for optimal male fertility. a, RT-qPCR on csr-1a mutants and wild-type animals at L4. Relative expression was normalized to rpl-32 and calculated relative to wild-type. Error bars indicate SEM. b, Western blot detecting for both CSR-1 isoforms expression in wild-type and csr-1a mutants L4 hermaphrodites at 100% and 66.7% loading volumes, using CSR-1 antibody. Actin is shown as a loading control. c, Schematic representation of two new csr-1a mutants (cmp253 and cmp254) d, in vitro sperm activation assay on additional csr-1a alleles and wild-type. Animals were raised at 25ºC for either one or two generations. Each experiment was performed in triplicate. At least 100 spermatids were counted for each replicate for each condition, for a total of at least 400 spermatids per condition. Error bars indicate standard deviation. Two-tail T-tests were performed to determine statistical significance. n.s. denotes not significant and indicates a p-value > 0.05, ** indicates a p-value ≤ 0.01, and **** indicates a p- value ≤ 0.0001. e, Representative images from the in vitro sperm activation assay of unactivated spermatid, activated spermatids with wild-type pseudopods, and activated spermatids with spiky pseudopods observed only in csr-1b(cmp258) mutants. 88 89 Fig S4 | Both CSR-1 isoforms bind germline-enriched 22G-RNAs. a, Normalized reads for oogenesis and spermatogenesis genes at adult stage from FLAG::CSR-1B IP and FLAG::CSR-1A+B IP compared to input. Oogenesis and spermatogenesis genes are indicated in blue and red respectively. b-d, 5’ length and nucleotide distribution in representative input and IP libraries from L4 stage HA::CSR-1A (b), FLAG::CSR-1B (c), and FLAG::CSR-1A+B (d). e-f, Venn diagram indicates overlap of L4 and adult target genes from CSR-1B (e) and CSR-1A+B (f) immunoprecipitations. g, Enrichment analysis (log2(fold enrichment)) examining the overlap of adult stage CSR-1B and CSR- 1A+B target genes with known targets of the CSR-1, male CSR-1, ALG-3/4, and mutator small RNA pathways and oogenesis and spermatogenesis-enriched genes. See Materials and Methods for gene list information. Statistical significance for enrichment was calculated using the Fisher’s Exact Test function in R. n.s. denotes not significant and indicates a p-value > 0.05 and **** indicates a p-value ≤ 0.0001. 90 Fig S5 | Clustal Omega alignment of CSR-1 in Caenorhabditis species. Arginine/glycine motifs are highlighted in yellow and start codons for both CSR-1A and CSR-1B are highlighted in blue. See Materials and Methods for protein sequence information. 91 Fig S6 | The N-terminal exon of CSR-1 is disordered across Caenorhabditis species. a-b, Graphs examining the disorder tendencies of CSR-1A in C. elegans (a) and CSR-1 in related nematode species (b). Residue position is plotted on the x-axes and disorder tendency scores (y-axes) above 0.5 indicate disorder. Red bar above each plot marks region of first exon and green bars mark all other exons. c, Immunofluorescence staining of dissected L4 hermaphrodite germlines in HA::CSR-1A, HA::CSR-1A[4xAG], and HA::CSR-1A[15xAG], using antibodies against HA and PGL-1. Scale bars, 5µM. d, Brood size assay on wild-type and mutant animals raised at 25ºC (n=10). Two-tail T- tests were performed to determined statistical significance. ** indicates p-value ≤ 0.01, *** indicates a p-value ≤ 0.001. 92 CHAPTER 3 WAGO-10 is a spermatogenesis-specific secondary Argonaute Protein Dieu An H. Nguyen 1 , Dylan Wallis 1 , Diego Cervantes 1 , Carolyn M. Phillips 1 93 3.1 Abstract Small RNAs target viral RNAs and harmful endogenous elements to silencing them, via RNAi. Small RNAs, typically 18-30 nt, first bind to a protein co-factor, called Argonaute protein, before surveilling downstream target transcripts for silencing. In C. elegans, there are 27 annotated Argonaute proteins, most of which are cytoplasmic when expressed, while only two have defined nuclear localization. Of the cytoplasmic Argonaute proteins, the majority localize to germ granules to regulate their target mRNAs post-transcriptionally. Nuclear Argonaute proteins translocate into the nucleus, where they find complementary mRNA that are still being transcribed, and deposit epigenetic silencing marks to establish a more permanent and heritable silencing memory. So far, only a handful of C. elegans Argonaute proteins have been characterized. In this work, we demonstrated that previously uncharacterized WAGO-10 (T22H9.3) shares the unique temporal and spatial expression pattern in the germline as the spermatogenesis-specific Argonaute ALG-3 and CSR-1A. Furthermore, we identify the small RNAs to which WAGO-10 binds, which are 22G-RNAs, and identify a list of WAGO-10 target genes. Thus, we demonstrate that WAGO-10 is a spermatogenesis- specific secondary Argonaute protein. 3.2 Introduction Small RNAs protect the genome from harmful foreign viral RNAs and aberrant endogenous elements by RNA interference, or RNAi. RNAi is astoundingly well- conserved across the eukaryotic kingdom, and its ubiquity reflects the fierce 94 evolutionary arms race between the selfish deleterious elements, such as transposons, and their hosts. small RNAs come in three flavors, micro-RNAs (miRNAs), piwi- interacting RNAs (piRNAs), and small-interfering RNAs (siRNAs), which can be distinguished from one another by their physical characteristics, modes of biogenesis, and binding cofactors – Argonaute proteins (Wilson & Doudna, 2013). The aptly named small RNAs, typically 18-30 nt, provide the sequence specificity to surveil downstream complementary target mRNA transcripts. Argonaute proteins, on the other hand, provide stability and function, and are thus the determinants which pathway the bound small RNAs are implicated in. In C. elegans, there are ~27 annotated Argonaute proteins, defined by the PAZ and MID domains, which anchor the 3’ and 5’ ends of the small RNA, respectively(Ma et al., 2004, 2005; Yigit et al., 2006). The robust diversity of C. elegans Argonaute proteins allows for small RNAs to be implicated in myriad of cellular processes. For example, the Argonaute protein RDE- 1 binds small RNAs processed from exogenous double-stranded RNAs to downregulate harmful foreign elements (Yigit et al., 2006). Concurrently, ERGO-1 binds to a subclass of small RNAs that are typically 26 nt in length with a strong bias for guanine at the 5’ end (26G-RNA) to target poorly conserved and recently duplicated genes (Fischer et al., 2011; Han et al., 2009). Unlike RDE-1, ERGO-1 binds to 26G-RNA made from endogenous genomic loci, by the RNA-dependent RNA polymerase (RdRP) RRF-3 (Han et al., 2009; Vasale et al., 2010). Other AGOs that bind to 26G-RNA include the partially redundant ALG-3 and ALG-4, which are required for spermatogenesis and specifically target spermatogenesis genes (Conine et al., 2010). Once paired with a 95 complementary target mRNA, Argonaute proteins can promote gene silencing by various mechanisms: translational repression, destabilization of target mRNA via decapping and deadenylation, RNA cleavage, or amplification of the original silencing signal by RNA-dependent RNA polymerases (RdRPs) (Jackson & Standart, 2007; Sijen et al., 2001). In C. elegans, the amplification step produces a subclass of small RNAs that are typically 22 nt long with a strong bias for guanine at the 5’ end (22G-RNA)(Pak & Fire, 2007; Sijen et al., 2007). By using an mRNA transcript as a template, the RdRPs EGO- 1 and RRF-1 synthesize an abundance of anti-sense 22G-RNAs, which ensures the silencing potency and prolongs the silencing effects and the heritability of the original small RNA signal (Maniar & Fire, 2011; Sijen et al., 2001). Despite synthesizing similar siRNAs, the two RdRPs. EGO-1 and RRF-1, localize to two distinct cytoplasmic substructures. EGO-1 localizes to the germline-specific P granules, an RNA processing body that hosts many RNA processing proteins, Tudor domain proteins, and Argonaute proteins (Claycomb et al., 2009; D. Updike & Strome, 2010) and many of its siRNA products are bound by CSR-1, a unique Argonaute protein. In contrast, RRF-1 localizes to a structure adjacent to P granules called the Mutator focus, to synthesize 22G-RNAs that are bound by the worm-specific Argonaute proteins (WAGOs) (W. Gu et al., 2009; Phillips et al., 2012). WAGO 22G-RNAs, also known as Mutator siRNAs, recognize foreign elements initially targeted by the exogenous RNAi pathway, transposons and transgenes from the piwi-interacting RNA (piRNA, also known as 21U-RNA) pathway, and pseudogenes and genes of recently duplicated events from the endogenous siRNA 96 pathway (Aoki et al., 2007; Bagijn et al., 2012; Fischer et al., 2011; Phillips et al., 2012; Shirayama et al., 2012). Once coupled with an Argonaute protein, WAGO 22G-RNAs either silence target transcripts in the cytoplasm or translocate into the nucleus to facilitate heterochromatinization of the genomic region from which their target mRNAs are transcribed, creating a heritable memory of the original silencing event(Buckley et al., 2012; Burkhart et al., 2011; W. Gu et al., 2009; Guang et al., 2008; Mao et al., 2015). Of the twelve annotated WAGOs, only WAGO-9/HRDE-1 and WAGO-12/NRDE- 3 localize to the nucleus, both belonging to Clade III of the WAGO clade, alongside the uncharacterized WAGO-10 and WAGO-11 (Buckley et al., 2012; W. Gu et al., 2009; Guang et al., 2008). In this work, we sought to characterized WAGO-10. Based on its phylogenetic relationship to NRDE-3 and HRDE-1, we expected WAGO-10 to be another nuclear Argonaute protein. However, to our surprise, WAGO-10 instead localizes to P granules in the cytoplasm. Furthermore, WAGO-10 is expressed only during the last larval stage of development, and its expression is spatially restricted to only the proximal region where spermatogenesis occurs. When we immunoprecipitated WAGO-10 to identify the associated small RNAs, we find that WAGO-10 binds to 22G- RNAs that mapped to ALG-3 targets, prompting us to hypothesize WAGO-10 to be the downstream effector of the spermatogenesis-specific ALG-3. Thus, in this study we have provided the basic framework for the previously uncharacterized WAGO-10, firmly demonstrated that WAGO-10 is a cytoplasmic Argonaute protein, and identified the genes that are targeted by WAGO-10 in the cytoplasm. 97 3.3 Worm-specific Argonaute protein WAGO-10 is expressed during spermatogenesis Of the 27 annotated Argonaute proteins, there are 12 Worm-specific Argonaute genes, or WAGOs (W. Gu et al., 2009; Youngman & Claycomb, 2014). The 12 WAGOs are further classified in three subclades, I, II, and III. Of the 12 WAGOs, only five genes have been characterized, and among those, three WAGOs localize to the perinuclear germ granules in the cytoplasm. The cytoplasmic WAGOs (WAGO-1, WAGO-3, and WAGO-4) are non-redundant, localizing to distinct perinuclear substructures, interacting with different protein complexes and are implicated different pathways (W. Gu et al., 2009; Schreier et al., 2020; Wan et al., 2018; Xu et al., 2018). The remaining two WAGOs (HRDE-1 and NRDE-3) localize to the nucleus, in either the germline or the soma, to establish heterochromatinization of the target loci and transgenerational heritable silencing(Buckley et al., 2012; Burkhart et al., 2011; Guang et al., 2008). Considering the wide array of functions attributed to the characterized WAGOs has been associated with, we sought to understand the molecular function of other uncharacterized WAGOs. To do this, we first looked for unique WAGOs transcription pattern on publicly available dataset and found that wago-10 (T22H9.3) shares a similar pattern to the spermatogenesis-specific AGOs alg-3 and alg-4 (Wormbase.org, Fig 1a). To confirm this observation, we performed RT-qPCR for wago-10 mRNA across all larval and adult developmental stages in wild-type animals. We found that wago-10 transcripts increase by approximately 50-fold during the fourth larval stage, and sharply drop off in gravid adult animals (Fig 1b). This transcriptional profile is similar to that of 98 the spermatogenesis-specific Argonaute alg-3, whereas expression of the Argonaute wago-4 peaks at the adult stage (Fig 1b). We next performed a western blot to detect endogenously tagged WAGO-10 across all larval and adult developmental stages. In agreement with the transcriptional data, WAGO-10 protein is expressed only during the fourth larval stage and is nearly undetectable in gravid adult (Fig 1c). Together these data indicate that WAGO-10 is uniquely expressed during the fourth larval stage, coinciding with the timing of spermatogenesis in C. elegans germline. 3.4 WAGO-10 localizes to the P granules in the spermatogenesis region WAGO-10 is found in the same clade of the worm-specific Argonaute proteins as HRDE-1 and NRDE-3, both of which localize to the nucleus and regulate target RNAs transcriptionally. Therefore, we next asked if WAGO-10 is expressed in germline nuclei (Fig 2a). By immunostaining L4 germlines, we found that WAGO-10 is expressed only at the most proximal region of the germline, an expression we previously observed in the spermatogenesis-specific isoform of CSR-1, CSR-1A (Fig 2b) (Nguyen & Phillips, 2021). Unexpectedly, WAGO-10 was excluded from germline nuclei, and instead is found in the cytoplasm and associated with P granules, similar to other spermatogenic Argonaute proteins CSR-1A and ALG-3 (Fig 2c) (Conine et al., 2010; Nguyen & Phillips, 2021). Thus we demonstrate that WAGO-10 is expressed in the germline at the spermatogenesis region, and that WAGO-10 is not a nuclear Argonaute protein, but rather localizes to P granules. 99 3.5 WAGO-10 is essential for optimizing male sperm fitness So far, we have demonstrated that WAGO-10 expression and localization coincide with the temporal and spatial timeframe of spermatogenesis in the germline. We next sought to characterize if or how WAGO-10 is involved in the overall fertility of the animals. To this end, we created a null wago-10 mutants using CRISPR, removing almost the entire wago-10 open reading frame (wago-10(cmp217)) (Fig 3a). To determine if WAGO-10 plays a role in fertility at permissive temperature, we performed brood size assays on wild-type and wago-10 mutant animals at 20°C. We did not observe any significant reduction in brood between the wild-type and wago-10 mutants at the permissive temperature (Fig 3b, left). We next asked if WAGO-10 is responsible for maintaining optimal fertility in a stressful environment, at the elevated temperature of 25°C (Fig 3b, right). Under this condition, we observed a slight reduction in brood in wago-10 mutant animals, though the change is not significant. Thus far, wago-10 appears to not be essential for fertility but may be necessary for optimal fertility at elevated temperature, though repeated experiment will need to be carried out to confirm this observation. Because WAGO-10 is expressed during spermatogenesis, we next sought to determine if WAGO-10 is required for optimal male fertility. To this end, we assayed the ability of wild-type and wago-10 mutant male sperm to compete with wild-type hermaphroditic sperm by crossing males of the indicated genotypes to movement- defective, unc-3, hermaphrodites. We counted the number of progeny with the uncoordinated (Unc) phenotype, indicating self-progeny from hermaphrodite sperm, and 100 with wild-type movement, indicating cross-progeny from male sperm. As expected, the wild-type male sperm outcompeted the hermaphroditic sperm for fertilization, due to their competitive advantages such as larger size and faster speed (Fig 3c) (LaMunyon & Ward, 1998; S Ward & Carrel, 1979). alg-3;alg-4 mutant male sperm, which has previously been shown to be defective, compete extremely poorly against the unc-3 hermaphroditic sperm. Similarly, wago-10 mutant male sperm were outcompeted by the unc-3 hermaphroditic sperm, however the defect was not as severe as with the alg-3/4 mutant sperm. Thus, despite not being essential for overall fertility, WAGO-10 appears to be essential for optimal male sperm fitness in competition, as the wago-10 mutant male sperm are less fit to compete with the typically smaller and slower hermaphroditic sperm (Fig 3c). We further dissected how WAGO-10 contributes to optimal male sperm fitness, by examining the rate of which wago-10 male sperm can be activated in vitro. In C. elegans, immotile round spermatids must first be activated to transform into mature and motile spermatozoa after the formation of pseudopods, which allows the sperm to physically latch on and subsequently fertilize oocytes (H. E. Smith, 2014). The process of sperm activation can be induced in vitro by using Pronase E, a cocktail of proteases that induces sperm to mature by forming pseudopods (Samuel Ward et al., 1983). Because we had previously observed a modest reduction in brood size at 25°C, we performed the in vitro sperm assay at 25°C to exacerbate phenotypes. After one generation at 25°C, wild-type spermatids are activated at a rate of 92%, while wago-10 mutant spermatids are activated at a rate of 78.7% (Fig 3d). While this reduction in 101 sperm activation is modest, it was observed in all three biological replicates. After a second generation at 25°C, wild-type spermatids are activated at a rate of 93%, and wago-10 mutant spermatids are activated at a rate of 47.3% (Fig 3d). The data suggests that one aspect of sperm fitness that is compromised in wago-10 male mutants is its less effective activation rate compared to wild-type male sperm. Together the data indicates that, while WAGO-10 has little to no effect on the fertility of hermaphrodites, WAGO-10 is required for optimizing male sperm fitness, seemingly to provide the male sperm the physiological advantages to outcompete the self-producing hermaphroditic sperm. 3.6 WAGO-10 binds to a subset of 22G-RNAs that do not target spermatogenic genes Being an Argonaute protein, WAGO-10 is predicted to bind to small RNA. To investigate WAGO-10 small RNA-binding partners, we immunoprecipitated WAGO-10, along with the spermatogenesis-specific ALG-3 from L4 stage animals and sequenced the associated small RNAs. As expected, ALG-3 has a strong preference for small RNAs that are 26 nucleotides in length, with a bias for guanine at the 5’ end (26G-RNA) (Fig 4a). In contrast, WAGO-10 binds to small RNAs that are, on average, 22 nucleotides in length with a bias for guanine at the 5’ end (22G-RNAs), characteristic of secondary siRNAs (Fig 4b). To identify the small RNAs that are enriched in the WAGO- 10 IP, we plotted the mapped small RNAs in the input against the small RNAs from the IP samples. Because WAGO-10 is expressed in the spermatogenesis region of the germline and is involved in optimizing male sperm health, we asked if WAGO-10 binds 102 to small RNAs that target spermatogenesis genes. Using published sex-specific gene lists (Ortiz et al., 2014), we found that WAGO-10 target genes are not specifically enriched for spermatogenenic or oogenic genes (Fig 4c). We then defined a list of 176 WAGO-10 target genes, as those with complementary small RNAs at least two-fold enriched in the WAGO-10 IP compared to input, with at least 10 RPM in the IP samples, and a DESeq2 adjusted p-value of ≤ 0.05. Of these WAGO-10 target genes, less than 20% overlap with the list of spermatogenic genes (Fig 4d). Thus, the data suggests that despite WAGO-10 having the temporal and spatial expression coinciding with spermatogenesis, WAGO-10 does not appear to broadly target and regulate genes associated with spermatogenesis. 3.7 WAGO-10 binds to small RNAs that recognize the same targets as ALG-3 We next defined a list of ALG-3 target genes at the L4 stage as those with complementary small RNAs at least two-fold enriched in the ALG-3 IP compared to input, with at least 10 RPM in the IP samples, and a DESeq2 adjusted p-value of ≤ 0.05. When we compared the ALG-3 gene list to the spermatogenesis list, we found that 47.9% of ALG-3 genes overlap with spermatogenic genes, indicating that more than half of ALG-3 targets are not directly involved in spermatogenesis (Fig 4e). We sought to determine whether WAGO-10 targets some of the same genes as ALG-3. Indeed, we find that small RNAs that are enriched in the WAGO-10 IP mapped to ALG-3 target genes and not targets of the Mutator pathway (Fig 5a). This small RNA profile is in contrast with WAGO-1, which binds small RNAs that target exclusively Mutator genes, 103 and is depleted of ALG-3 genes (Fig 5b). When compared between the WAGO-10 and ALG-3 gene lists, we found 47.7% of WAGO-10 genes overlap with ALG-3 genes (Fig 5c). This overlap is significantly higher than the overlap between WAGO-10 and the spermatogenic genes, but still constitutes less than half of the WAGO-10 genes. We next asked if the WAGO-10 gene list is enriched for ALG-3 targets, that is to see if the WAGO-10 genes is over-represented in ALG-3 targets. Indeed, we found that WAGO- 10 target genes are enriched for ALG-3 targets but not for Mutator targets. In contrast, WAGO-1 genes are enriched for Mutator targets and are depleted of the ALG-3 targets (Fig 5d-e). Thus, the data indicates that even though WAGO-10 small RNAs do not target spermatogenesis transcripts, WAGO-10 binds to a subset of small RNAs that target the same genes as ALG-3, a property unique to WAGO-10 as WAGO-1, which is also expressed during L4 and localize to the P granules, selectively binds to small RNAs that recognize only Mutator targets. 3.8 Discussion In this study, we have characterized a novel secondary Argonaute protein, WAGO-10 which is involved in promoting optimal male sperm health. In our earlier work, we have characterized CSR-1A to be a spermatogenesis-isoform and that it binds to 22G-RNAs that target spermatogenic genes. Despite sharing a similar spatiotemporal pattern and both localizing to the P granules, WAGO-10 does not bind preferentially to spermatogenic small RNAs as CSR-1A does. Instead, WAGO-10 binds to small RNAs that target the same genes as ALG-3, a spermatogenesis-specific primary Argonaute 104 protein. We have demonstrated that CSR-1A does not bind secondary siRNAs downstream of ALG-3, which left the obvious hypothesis that WAGO-10 could be the downstream secondary effector of the ALG-3 pathway. With the current data this scenario seems unlikely. Null wago-10 mutants have little to no effect on the overall fertility of the animals, regardless of permissive or elevated temperature. Null alg-3, on the other hand, along with its paralog alg-4 mutants, show dramatic reduction in brood at both 20ºC and complete sterility at 25ºC. What Argonaute protein, then, would be downstream effectors of ALG-3/4? Even though WAGO-10 appears to target the same genes as ALG-3, the complete gene list of WAGO-10 is modest, comprises of 176 genes, whereas ALG-3 gene list comprises of 3485 genes. This seems to eliminate the possibility of WAGO-10 to be the sole downstream effector of ALG-3. To either confirm or deny this possibility, we will perform an RNA-IP on FLAG::WAGO-10 and FLAG::WAGO-10; alg-3/4 and subject the associated small RNAs for sequencing. Changes or the lack thereof in associated WAGO-10 small RNAs would inform as to whether ALG-3 is required for biogenesis of WAGO-10 siRNAs. With the small gene list and a mild reduction in fertility, it is tempting to propose for WAGO-10 to be accidental relic of a gene duplication event, one that did not arise out of necessity. Nevertheless, the reduction in fertility, however mild between the mutants and wild-type, is reproducible and thus warrants careful examination. Because WAGO-10 belongs in the same clade as HRDE-1, which is essential for RNAi memory and inheritance, and recent work has reported the cytoplasmic Argonaute WAGO-4 105 being involved in RNAi inheritance, it would be worth examining WAGO-10 under this light. One possibility of the role of WAGO-10 is maintaining a paternal epigenetic memory. A simple assay that can be done is a transgenerational assay at 25°C to test if wago-10 mutants have a Mrt phenotype. Perhaps there is not one single secondary WAGO downstream of ALG-3/4, but a cohort of WAGOs, each binding to a subset of the ALG-3 targets. Recently WAGO-3 has been characterized to be involved in the small RNA paternal epigenetic inheritance pathway, making it an ideal candidate. Unlike ALG-3, which is discarded into the residual bodies of the budding spermatids, WAGO-3 remains in the sperm and is passed onto the progeny. So far, it is the only WAGO that has been observed to be passed onto progeny via sperm. Alternatively, SAGO-2 (WAGO-6) or PPW-1 (WAGO- 7), two relatively uncharacterized proteins that show high transcript expression in males, could be additional downstream candidates of ALG-3/4. 3.9 Methods C. elegans strains Strains were maintained at 20°C on NGM plates seeded with OP50 E. coli according to standard conditions unless otherwise stated (Brenner, 1974). All strains used in this project are listed in Supplementary Table 1. 106 Plasmid and strain construction The fluorescent and epitope tags were integrated at wago-10 endogenous loci by CRISPR genome editing (Arribere et al., 2014; Dickinson et al., 2013, 2015; Friedland et al., 2013; J. D. Ward, 2015). For GFP/3xFLAG::WAGO-10, we generated homologous repair templates. gfp::3xFLAG::wago-10 was assembled into pDD282 (Addgene #66823) by isothermal assembly according to published protocols (Dickinson et al., 2015; Gibson et al., 2009). To protect the repair template from cleavage, we introduced silent mutations at the site of guide RNA targeting by incorporating these mutations into one of the homology arm primers or, if necessary, by performing site-directed mutagenesis (Dickinson et al., 2013). All guide RNA plasmids were generated by ligating oligos containing the guide RNA sequence into BsaI-digested pRB1017 (Addgene #59936) (Arribere et al., 2014). GFP CRISPR injection mixes included 25 ng/μl repair template, 50 ng/μl guide RNA plasmid, 50 ng/μl eft-3p::cas9 (pJW1259, Addgene #61251), 2.5–10 ng/μl GFP or mCherry co-injection markers. The gfp::3xFLAG::wago-10 construct was injected into the wild-type strain (Dickinson et al., 2013, 2015). For wago-10 deletions (cmp217) and 2xFLAG::wago-10, the injection mix included 50 ng/μl wago-10 deletion repair and FLAG repair oligos, respectively, 25 ng/μl wago-10 guide RNA plasmids, 20 ng/µl rol-6 repair template, 25 ng/µl rol-6 guide RNA (pJA42, Addgene #59930), and 50 ng/µl eft-3p::Cas9 (pJW1259, Addgene #61251). Mixture was injected into wild-type animals. F1 animals with the Rol phenotype were isolated and genotyped by PCR to identify animals with the 2xFLAG insertion (Arribere et al., 2014). 107 Antibody staining and imaging For immunofluorescence, C. elegans were dissected in egg buffer containing 0.1% Tween-20 and fixed in 1% formaldehyde in egg buffer as described (Phillips et al., 2009). Samples were immunostained with mouse anti-FLAG 1:500 (Sigma Aldrich, F1804) and mouse anti-PGL-1 1:100 (DSHB K76) (Strome & Wood, 1983). Alexa-Fluor secondary antibodies were purchased from Thermo Fisher. Animals were dissected at the L4 (48 hours post hatching). Imaging was performed on a DeltaVision Elite microscope (GE Healthcare) using a 60x N.A. 1.42 oil-immersion objective. When data stacks were collected, three-dimensional images are presented as maximum intensity projections. Images were pseudocolored using the SoftWoRx package or Adobe Photoshop. Western blots C. elegans were synchronized at 20°C by bleaching gravid adult animals and maintaining starved L1 larvae for at least 24 hours before plating on OP50. For sample collection, animals were harvested after 2 hours (L1), 12 hours (L2), 32 hours (L3), 50 hours (L4), and 72 hours (gravid adult) on OP50. Number of animals loaded per lane was normalized for actin – approximately 1000 L1s, 800 L2s, 600 L3s, 400 L4s, and 200 gravid adults. Proteins were resolved on 4-12% Bis-Tris polyacrylamide gels (Thermo Fisher), transferred to nitrocellulose membranes (Thermo Fisher), and mouse anti-FLAG 1:1,000 (Sigma, F1804), or mouse anti-actin 1:10,000 (Abcam ab3280). Secondary HRP antibodies were purchased from Thermo Fisher. 108 RNA Isolation and qRT-PCR Synchronized animals were collected at L1, L2, L3, L4, and gravid adult stages, after two, 12, 32, 48, and 68 hours of feeding following L1 arrest, respectively. For mRNA libraries, synchronized L4 stage animals (~48 hours at 20°C and ~34 hours at 25°C after L1 arrest) were collected. RNA was isolated using Trizol reagent (Thermo Fisher), followed by chloroform extraction and isopropanol precipitation. RNA samples were normalized to 10µg/µL prior to DNase treatment (TURBO DNA-free kit, Thermo Fisher AM1907) and reverse transcribed with SuperScript III Reverse Transcriptase (Thermo Fisher 18080- 051). All Real time PCR reactions were performed using the 2x iTaq Universal SYBR Green Supermix (Bio-Rad 1725121), following manufacturer’s protocols, and run in the CFX96 Touch Real-Time PCR System (Bio-Rad 1855195). Samples were run with three technical replicates and two biological replicates, normalized to Y45F10D.4 and rpl-32. Brood size analysis Wild-type and mutant C. elegans strains were maintained at 20°C prior to temperature- shift experiments. Animals were shifted to 25°C as L4 larvae and ten of their progeny were picked to individual plates for 25°C brood size analysis. To score the complete brood, each animal was moved to a fresh plate every day until egg-laying was complete. After allowing the progeny 2-3 days to develop, the total number of animals on each plate was counted. 109 In vitro sperm activation assay Virgin L4 males were isolated 24 hours before assay. 10-15 males were dissected in 30uL of 500µg/mL Pronase E (Millipore Sigma), which was dissolved in sperm medium (50 mM HEPES, 50 mM NaCl, 25 mM KCl, 5 mM CaCl2, 1 mM MgSO4, 1 mg/ml BSA), as described previously (Samuel Ward et al., 1983; Yen et al., 2020). Spermatids were incubated for 15 minutes at room temperature in a humid chamber before mounting and imaging on a DeltaVision Elite microscope (GE Healthcare) using a 60x N.A. 1.42 oil- immersion objective. Male fertility assay Virgin L4 males were isolated 24 hours before assay and incubated at 20°C. L4 unc-3 hermaphrodites were also selected 24 hours before assay and incubated at 20°C to allow for the first pass of egg-laying. On the day of the assay, four males were crossed to one hermaphrodite, 10 crosses per genotype. Crosses were incubated at 25°C and transferred to fresh plates every day for three days. Progeny were scored when they reached L3-L4 stage and the Unc phenotype was observable. Unc progeny indicated hermaphrodite sperm and wild-type progeny indicated male sperm (S Ward & Carrel, 1979). Immunoprecipitations For immunoprecipitation experiments followed by small RNA library preparation, ~100,000 synchronized L4 animals (~49 hours at 20°C after L1 arrest) were collected in 110 IP Buffer (50 mM Tris-Cl pH 7.4, 100 mM KCl, 2.5 mM MgCl2, 0.1% Igapal CA-630, 0.5 mM PMSF, cOmplete Protease Inhibitor Cocktail (Roche 04693159001), and RNaseOUT Ribonuclease Inhibitor (Thermo Fisher 10777019)), frozen in liquid nitrogen, and homogenized using a mortar and pestle. After further dilution into IP buffer (1:10 packed worms:buffer), insoluble particulate was removed by centrifugation and 10% of sample was taken as “input.” The remaining lysate was used for the immunoprecipitation. Immunoprecipitation was performed at 4°C for 1 hour with pre-conjugated anti-HA affinity matrix (Roche 11815016001) or anti-FLAG affinity matrix (Sigma Aldrich A22220), then washed at least 3 times in immunoprecipitation buffer. A fraction of each sample was analyzed by western blot to confirm efficacy of immunoprecipitation. Trizol reagent (Thermo Fisher) was added to the remainder of each sample, followed by chloroform extraction, isopropanol precipitation, and small RNA library preparation. Small and mRNA library preparation Small RNAs (18 to 30-nt) were size selected on denaturing 15% polyacrylamide gels (Bio- Rad 3450091) from total RNA samples. Small RNAs were treated with 5’ RNA polyphosphatase (Epicentre RP8092H) and ligated to 3’ pre-adenylated adapter with Truncated T4 RNA ligase (NEB M0373L). Small RNAs were then hybridized to the reverse transcription primer, ligated to the 5’ adapter with T4 RNA ligase (NEB M0204L), and reverse transcribed with Superscript III (Thermo Fisher 18080-051). Small RNA libraries were amplified using Q5 High-Fidelity DNA polymerase (NEB M0491L) and size selected on a 10% polyacrylamide gel (Bio-Rad 3450051). Library concentration was 111 determined using the Qubit 1X dsDNA HS Assay kit (Thermo Fisher Q33231) and quality was assessed using the Agilent BioAnalyzer. Libraries were sequenced on the Illumina NextSeq500 (SE 75-bp reads) platform. Bioinformatic Analysis For small RNA libraries, sequences were parsed from adapters using FASTQ/A Clipper (options: -Q33 -l 17 -c -n -a TGGAATTCTCGGGTGCCAAGG) and quality filtered using the FASTQ Quality Filter (options: -Q33 -q 27 -p 65) from the FASTX-Toolkit (http://hannonlab.cshl.edu/fastx_toolkit/), mapped to the C. elegans genome WS258 using Bowtie2 v. 2.2.2 (default parameters) (Langmead & Salzberg, 2012), and reads were assigned to genomic features using FeatureCounts (options: -t exon -g gene_id -O --fraction –largestOverlap) which is part of the Subread v. 1.5.1 package (Liao et al., 2013, 2014). Differential expression analysis was done using DESeq2 v. 1.22.2 (Love et al., 2014). To define gene lists from IP experiments, a 2-fold-change cutoff, a DESeq2 adjusted p-value of ≤0.05, and at least 10 RPM in the IP libraries were required to identify genes with significant changes in small RNA levels. Venn Diagrams and Violin plots were generated using the Venndiagram (Chen & Boutros, 2011)and Vioplot R (Adler & Kelly, 2020) packages respectively, and modified in Adobe Illustrator. 112 Materials Supplemental Table 1. Strains used in this study Experimental Models: Organisms/Strains Identifier C. elegans: Strain N2 - wild-type N2 C. elegans: Strain WM200 - alg-4(ok1041) III; alg- 3(tm1155) IV WM300 C. elegans: Strain CB151 - unc-3(e151) X CB151 C. elegans: Strain USC1191 - wago- 10(cmp204[wago-10::2xFLAG]) V USC1191 C. elegans: Strain USC1228 - cmp217[wago- 10Δ3673bp] V USC1228 C. elegans: Strain USC1240 - csr-1a(cmp135) IV;wago-10(cmp217) V USC1240 C. elegans: Strain USC1272 - wago- 10(cmp225[(GFP + loxP + 3xFLAG)::wago-10]) V USC1272 C. elegans: Strain USC1092 - alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg-3]) IV USC1092 C. elegans: Strain USC988 - mut-16(cmp41[mut- 16::mCherry + loxP]) : wago-1(cmp92[(GFP + loxP + 3xFLAG)::wago-1]) I USC988 113 Fig 1 | WAGO-10 is selectively expressed during the fourth larval stage. a, FPKM expression data of wago-10, alg-3, and alg-4 from modENCODE PolyA+ libraries across development (Wormbase.org). The bars represent median value of the libraries used. EE, early embryos; LE, late embryos; YA, young adults. b, qRT-PCR detecting for wago-10, alg-3, and wago-4 expression across development. RNA was isolated from wild-type animals at each larval stage and adult. Relative expression was normalized to Y45F10D.4 and rpl-32, and calculated relative to L1 stage. Statistical significance was calculated for each stage compared to L1. * indicates a p-value ≤ 0.05, ** indicates a p-value ≤ 0.01, and *** indicates a p-value ≤ 0.001. c, Western blot detecting for CSR-1A at each larval stage and gravid adult in 2xFLAG::WAGO-10 strain using α-FLAG antibody. Actin is shown as a loading control. 114 Fig 2 | WAGO-10 is expressed in the spermatogenesis region of the germline. a, Phylogram derived from ClustalW2 alignment of NRDE-3, HRDE-1, and WAGO-10 protein sequences. b, Immunofluorescence staining for FLAG::WAGO-10 in dissected L4 hermaphrodite germline, using α-FLAG. Asterisk indicates distal end. DNA is stained with DAPI, blue. Scale bar, 15µM. c, Immunofluorescent staining of a FLAG::WAGO-10 in dissected L4 hermaphrodite germline, using antibodies against FLAG, and PGL-1. Scale bars represent 5µM. 115 Fig 3 | WAGO-10 is essential for optimizing male sperm fitness. a, Schematic representation of wago-10 deletion allele. cmp217 removes 3973 bp, almost all of wago-10 open reading frame. b, Brood size assay on wago-10 (cmp217) raised at 20ºC and 25ºC (n=10). Two-tail t-tests were performed to determined statistical significance. *** indicates a p-value≤0.001. n.s denotes not significant. c, Male fertility assay performed on wild-type and mutant males crossed to unc-3 hermaphrodite at 25ºC. Unc progeny were scored as hermaphrodite sperm, and wild-type progeny were scored as male sperm. d, An in vitro sperm activation assay was performed on spermatids from wild-type and wago-10 (cmp217) mutants. Animals were raised at 25°C for either one or two generations and each experiment was performed in triplicate. At least 100 spermatids were counted for each replicate for each condition, for a total of at least 400 spermatids per condition. Two-tail T-tests were performed to determine statistical significance. * indicates a p-value ≤ 0.05. n.s denotes not significant. 116 Fig 4 | WAGO-10 binds to 22G-RNAs that do not specifically target spermatogenic genes. a-b, 5’ length and nucleotide distribution in representative input and IP libraries from L4 stage GFP/FLAG::ALG-3 (a) and FLAG::WAGO-10 (b). c, Normalized reads for spermatogenic and oogenic genes at L4 stage from FLAG::WAGO-10 IP compared to input. Spermatogenic and oogenic genes are indicated in yellow and teal respectively. d, Venn diagram indicates overlap of spermatogenic genes and WAGO-10 target genes at L4 stage. Target genes are defined are at least 2-fold enriched in the IP, with at least 10 RPM in IP samples and a DESeq2 adjusted p-value≤0.05. e, Venn diagram indicates overlap of spermatogenic genes and ALG-3 target genes at L4 stage. Target genes are defined are at least 2-fold enriched in the IP, with at least 10 RPM in IP samples and a DESeq2 adjusted p-value≤0.05. 117 Fig 5 | WAGO-10 binds small RNAs that target the same genes as ALG-3. a-b, Normalized reads for ALG-3 and Mutator targets at L4 stage from FLAG::WAGO-10 IP (a) and GFP/FLAG::WAGO-1 IP (b) compared to input. ALG-3 and Mutator targets are indicated in orange and green respectively. c, Venn diagram indicates overlap of ALG-3 and WAGO-10 target genes at L4 stage. Target genes are defined are at least 2-fold enriched in the IP, with at least 10 RPM in IP samples and a DESeq2 adjusted p-value≤0.05. d-e, Enrichment analysis (log2(fold enrichment)) examining the overlap of WAGO-10 (d) and WAGO-1 (e) target genes with known targets of the ALG-3 and mutator small RNA pathways. ALG-3 targets were defined as in (c). *** indicates a p-value≤0.001 118 CHAPTER 4 Localization and Regulation of Argonaute Proteins Dieu An H. Nguyen 1 , Diego Cervantes 1 , Carolyn M. Phillips 1 119 4.1 Introduction RNAi, or RNA interference, employs small RNAs to fight against foreign viral dsRNA, regulate gene expression, chromosome segregation, ensuring fertility. Typically ranged between 18-30 nucleotides, small RNAs are bound by Argonaute proteins to provide sequence specificity in recognizing complementary mRNA transcripts (Czech & Hannon, 2011). There are three major classes of small RNAs, microRNAs (miRNAs), piwi-interacting RNAs (piRNAs), and small-interfering RNAs (siRNAs) (Ghildiyal & Zamore, 2009). Before being loaded into Argonaute proteins, the ribonuclease Dicer cleaves the longer precursor of siRNA small RNAs to the appropriate length, leaving the smaller product with a 5’ monophosphate, a hydroxyl group at the 3’ end, and a subset of which has 2’-O-methylation(Han et al., 2009; Ketting et al., 2001; Montgomery et al., 2012). In C. elegans, the RNA-dependent RNA polymerases (RdRP) amplify a large pool of siRNAs. Unlike products of Dicer, products of RdRP possess three phosphate groups at the 5’ end, though the 3’ hydroxyl group remains unmodified (Pak & Fire, 2007; Sijen et al., 2007). As RNAi is extremely well-conserved, so is the structure of Argonaute proteins, which maintains three functional domains, MID, PIWI, and PAZ, that can be seen across the eukaryotic kingdom (Wilson & Doudna, 2013). The majority of characterized Argonaute proteins in C. elegans are expressed in the germline, where RNAi is characteristically robust. The Argonaute protein, ERGO-1, is expressed strongly in cells undergoing oogenesis, binding to a population of small RNAs that are approximately 26 nucleotides in length with a bias for guanine at the 5’ end (26G-RNAs) to target poorly conserved and recently duplicated genes. The 120 targeting event by ERGO-1 26G-RNA triggers the production of secondary siRNA, typically 22 nucleotides in length also with a guanine at the 5’ end (22G-RNA), by RdRPs (Fischer et al., 2011; Pak & Fire, 2007; Sijen et al., 2007; Vasale et al., 2010). Other 26G-RNA binding Argonaute proteins include the redundant ALG-3 and ALG-4, which target spermatogenesis genes (Conine et al., 2010). ALG-3/4 localize to P granules, which are germ granules that surveil and process RNAs that are situating directly over the nuclear pores in the cytoplasm (Conine et al., 2010). At P granules, Argonaute proteins binding to 26G-RNAs and 22G-RNAs co-exist, targeting complementary mRNA transcripts as they exit the nucleus, regulating them by either direct cleavage the targets by the Argonaute PIWI domain or repressing translation (Jackson & Standart, 2007; Wilson & Doudna, 2013). While many Argonaute proteins congregate at the P granules, how each Argonaute protein identifies its correct siRNA binding substrates remains unclear. Small RNA-associated Argonaute proteins localizing to an RNA processing granule is not unique to C. elegans. In Drosophila, the germline PIWI proteins, Aub and Ago3, localize to a distinct substructure called ‘nuage’, which bears a striking functional similarity to C. elegans P granules (Pek et al., 2012). Both Aub and Ago3 are the main effectors in the ping-pong piRNA amplification cycle, but their functions are non- redundant, and their modes of localizing to nuage distinct. Ago3 is tethered to nuage by the Tudor protein Krimper, whereas Aub localizes to the substructure by binding to piRNA (Webster et al., 2015). Both PIWI proteins contains symmetrically dimethylated arginine (sDMA) at their N-termini RG/RGG motifs, which provides a putative binding 121 platform for Tudor proteins such as Krimper (Kirino et al., 2009; Webster et al., 2015). The sDMA marks at the N-terminus of PIWI proteins are well-documented in other model organisms and are generally thought as a tethering mechanism for Tudor proteins to PIWI proteins (Kirino et al., 2009; Vagin et al., 2009). Of the three PIWI proteins in Drosophila, only Piwi gains entry to the nucleus by binding to primary piRNA (Saito et al., 2010). The C. elegans nuclear Argonaute proteins, HRDE-1 and NRDE-3, share a similar mechanism for translocation in the nucleus, but for many cytoplasmic Argonaute proteins, including the only PIWI protein in C. elegans PRG-1, the mechanism for localization remains unclear (Buckley et al., 2012; Guang et al., 2008). However, PRG-1 expression depends on its piRNA substrates, which are 21 nucleotides in length, with a strong bias for an uracil at the 5’ end (21U) (Albuquerque et al., 2014; Batista et al., 2008). When 21U-biogenesis is compromised, the presumably empty PRG-1 protein drastically degrades, hinting at a stabilization effect small RNA has on its binding cofactor (Albuquerque et al., 2014). In fact, selective degradation of empty AGO is evolutionarily conserved for miRNA- associated Argonaute proteins, observed in Drosophila, plants, mouse cell lines, and mammals (Derrien et al., 2012; Johnston et al., 2010; Martinez & Gregory, 2013; Smibert et al., 2013; Suzuki et al., 2009). Empty, but not miRNA-loaded, Argonaute proteins can be routed to either the ubiquitin proteasome system or macroautophagy, selected by either an E3 ligase or Valosin-containing protein, or VCP, respectively (Kobayashi et al., 2019a, 2019b). Whether empty siRNA-associated Argonaute proteins would be subjected to degradation remains to be investigated. 122 In this study, we sought to understand the mechanisms of localization by siRNA- associated Argonaute proteins and PRG-1, by dissecting the role of small RNA-binding and post-translational modifications of a series of selected Argonaute proteins. We systematically selected Argonaute proteins from distinct small RNA pathways: ERGO-1 and ALG-3 for 26G-RNA binding primary Argonaute proteins, PRG-1 for 21U-RNA binding primary Argonaute protein, and WAGO-1, WAGO-10, and CSR-1 for 22G-RNA binding secondary Argonaute proteins. We demonstrate that, like Drosophila PIWI, many siRNA-associated Argonaute proteins are dimethylated at RG motifs. We also found that PRG-1, ALG-3, and WAGO-1 require small RNA binding to localize to the P granule, much like Drosophila Aub. In fact, when ALG-3 cannot bind to its small RNA partners, it is effectively non-functional, though that does not result in ALG-3 being selectively degraded. Lastly, CSR-1 does not mislocalize from the P granule when it cannot bind to small RNA, but instead causes the granules to have unusual and misshapen phenotypes, resulting in improper germline development, and ultimately sterility. Together, we demonstrate a divergence in localization mechanism of C. elegans siRNA-associated Argonaute proteins from what has been described in other organisms, and that C. elegans Argonaute proteins and their siRNAs do not only participate in silencing harmful genetic transcripts, but also in maintaining a healthy germline environment and consequently fertility. 123 4.2 Creating binding deficient AGO mutants Argonaute proteins bind to small RNA to gain sequence specificity of their target mRNA transcripts. The 5’ phosphate of small RNA is stacked over an invariant tyrosine in the Argonaute MID domain, stabilized by the hydrogen bonds from the side chains by the rest of the conserved motifs in the pocket – tyrosine-lysine-glutamine-lysine (Y-K-Q- K) (Ma et al., 2005). This motif is rigidly conserved across the eukaryotic kingdom (Fig 1a, yellow highlight). Interestingly, sequence alignment of C. elegans worm-specific Argonaute proteins (WAGOs) shows a deviation from this strict sequence, substituting the first of the four conserved residues to histidine (H-K-Q-K) (Fig 1a, blue highlight). Worm-specific Argonaute proteins bind to 22G-RNAs, the amplification products of the RNA-dependent RNA polymerase (RdRP). These 22G-RNAs are triphosphorylated at their 5’ end, rather than monophosphorylated like the primary siRNA made by Dicer and other nucleases through cleavage of longer precursor RNAs (Pak & Fire, 2007; Ruby et al., 2006). The substitution of histidine for tyrosine in the binding motif coupled with the two biochemically distinct groups of small RNAs prompts the question as to whether there is specificity in the binding site of Argonaute proteins for specific small RNA 5’ end modifications. To investigate Argonaute protein localization when they cannot bind to small RNAs, we first selected six representative Argonaute proteins from diverse pathways and labeled each one with either an epitope or fluorescent tag at their endogenous locus using the CRISPR/Cas9 system (Fig 1b). ALG-3, PRG-1, and ERGO-1 represent the primary Argonaute proteins, and WAGO-1, WAGO-10, and CSR-1 the secondary 124 Argonaute proteins. Apart from ERGO-1, all selected proteins localize to the P granules when expressed (Fig 2a, b, and c) (Conine et al., 2010; W. Gu et al., 2009; Nguyen & Phillips, 2021)(and Dissertation Chapter 3, Fig 2). Previous work studying purified MID domain of A. fulgidus Piwi protein, has demonstrated that binding affinity for RNA is most severely reduced when the first two residues of the motifs are replaced with alanines (Ma et al., 2005). Therefore, to generate binding-defective Argonaute proteins, we mutated the first two residues of the conserved binding motifs (Y-K-Q-K or H-K-Q-K) to alanine (A-A-Q-K) using CRISPR/Cas9 (Fig 1b). With the exception of csr-1[HK-to- AA], which is homozygous sterile, all binding-deficient mutant strains are viable and are healthy at permissive temperature. 4.3 Most binding-defective Argonaute proteins are stably expressed We next tested for protein expression in the binding-deficient mutants, to determine whether the mutations affect protein stability. By performing western blot on wild-type ERGO-1, WAGO-1, and WAGO-10, and their binding deficient counterparts, ERGO-1[YK-to-AA], WAGO-1[HK-to-AA] and WAGO-10[HK-to-AA] respectively, we found that the protein expression is comparable between mutants and wild-type, (Fig 1c, e, f). (Elkayam et al., 2012; Tsuboyama et al., 2018). In contrast, prg-1[YK-to-AA] mutants have a significant reduction in protein expression level compared to wild-type (Fig 1d). This result is reminiscent of the reduction in PRG-1 protein levels when piRNAs are unavailable due to the absent of the upstream biogenesis factor, PID- 1(Albuquerque et al., 2014). Because both the binding-defective mutant, prg-1[YK-to- 125 AA] and the genetic mutant, pid-1, share a similar reduction in PRG-1 protein levels, we propose that the decrease in PRG-1[YK-to-AA] expression is a result of PRG-1 instability when it cannot bind to piRNAs. The selective degradation does not seem to apply to siRNA Argonaute proteins, which are stably expressed even when they cannot bind to small RNA. 4.4 Binding-defective Argonaute proteins fail to localize to the P granule P granules sit on the cytoplasmic periphery of the nucleus, covering the nuclear pores and capturing nascent mRNA transcripts. A major site for RNA metabolism, P granules host a massive number of RNA-processing proteins, including multiple Argonaute proteins. We next sought to determine whether small RNA binding is required for the localization of Argonaute proteins to P granules. To this end, we compared the germline expression of PRG-1, WAGO-1, and ALG-3 to their small RNA binding-defective counterparts. As demonstrated previously, WAGO-1, PRG-1, and ALG-3 localize primarily to P granules (Fig 2a, b, and c) (Conine et al., 2010; W. Gu et al., 2009; Nguyen & Phillips, 2021)(and Dissertation Chapter 3, Fig 3.2). In contrast, WAGO-1[HK-to-AA], PRG-1[YK-to-AA], and ALG-3[YK-to-AA] lose association with P granules and are instead diffusely localized to the cytoplasm throughout the gonad (Fig 2a-c). The loss of P-granule association in WAGO-1[HK-to-AA] and ALG-3[YK-to-AA] are unlikely to be an effect of decreased WAGO-1 and ALG-3 protein expression because both wild-type WAGO-1 and WAGO-1[HK-to-AA] and ALG-3 and ALG-3[YK-to- AA] are expressed at similar levels, suggesting that the localization of WAGO-1 and 126 ALG-3 to P granules is small RNA-dependent (Fig 1e, 2d). In contrast, the protein expression of PRG-1[YK-to-AA] is decreased compared to wild-type PRG-1 (Fig 1d), indicating that reduced protein expression could contribute to the reduced P-granule association of PRG-1[YK-to-AA] protein. Thus, WAGO-1 and ALG-3 do not require small RNA binding for protein stability but for localization to P granules, while PRG-1 requires small RNA binding for both protein stability and localization to P granules. 4.5 alg-3[YK-to-AA] mutants phenocopy null alg-3 mutants The fundamental function of an Argonaute protein is to recognize complementary mRNA transcripts and to alter expression, through mRNA cleavage, translational repression, or transcriptional inhibition. Thus, loss of small RNA binding should effectively render Argonaute proteins non-functional. To test this hypothesis, we focused on the spermatogenesis-specific ALG-3, which is only expressed during the fourth larval stage and restricted to the proximal region of the germline. ALG-3 is redundant with its paralog, ALG-4, however only ALG-3 localization has been studied (Conine et al., 2010, 2013). We sought to see if alg-3[YK-to-AA] would phenocopy null alg-3. At 25°C, a null allele of alg-3 alone does not result in sterility, but greatly compromises the brood size (Conine et al., 2010). The fertility of this strain is due to the presence of wild-type ALG-4. Similarly, an alg-4 mutant in the presence of ALG-3 has a reduced brood size but is not sterile (Fig 2e). However, when both alg-3 and alg-4 are disrupted, animals are completely sterile at 25°C (Fig 2e) (Conine et al., 2010). To determine if alg-3[YK-to-AA] mutants would also contribute to temperature sensitive 127 sterility in the presence of the alg-4 mutants, we generated the alg-3[YK-to-AA]; alg-4 strain. Indeed, we found that at elevated temperature, alg-3[YK-to-AA]; alg-4 mutants are completely sterile. Together the data suggests that the loss of small RNA binding does not only renders ALG-3 incapable of localizing to P granules, but also non- functional in maintaining fertility at elevated temperature. 4.6 csr-1[HK-to-AA] mutants are sterile and form misshapen P granules Of the 27 annotated C. elegans Argonaute proteins, only CSR-1 is essential for fertility and development. Similar to a csr-1 null allele or RNAi knockdown of csr-1, csr- 1[HK-to-AA] mutants are sterile and are embryonically lethal (Claycomb et al., 2009; D. L. Updike & Strome, 2009), indicating that the small RNA binding is essential for CSR-1 function. Furthermore, nuclei in the germlines of csr-1[HK-to-AA] mutants lose their uniformity in size and arrangement, at times forming nuclei about two to three times larger than a typical nucleus (Fig 2c, white arrows). Like P granules observed in other csr-1 mutants (Claycomb et al., 2009; Gerson-Gurwitz et al., 2016; D. L. Updike & Strome, 2009), the csr-1[HK-to-AA] mutants have misshapen P granules; though interestingly, some CSR-1[HK-to-AA] still localizes to these unusual P granules that are two to four-fold larger than wild-type P granules (Fig 2c-d). Notably, mutations in ego-1, the RNA-dependent RNA polymerase responsible for synthesizing CSR-1 22G-RNAs, also share a similar phenotype (Vought et al., 2005), suggesting that enlarged P granules are a characteristic phenotype of defects in the CSR-1 pathway. This data demonstrates that small RNA-binding is not absolutely required for CSR-1 association 128 with P granules, which distinguishes it from small RNA binding-defective WAGO-1 and PRG-1. 4.7 The RG motifs in multiple C. elegans Argonaute proteins are dimethylated RG/RGG motifs have been previously demonstrated to be methylation targets of protein arginine methyltransferases (PRMTs) (Kirino et al., 2009; Kirino, Vourekas, Kim, et al., 2010; Vagin et al., 2009). Recently we have shown that the C. elegans Argonaute protein, CSR-1A, is dimethylated and that methylation contributes to CSR-1A small RNA binding specificity. We next sought to determine if other C. elegans Argonaute proteins are also methylated. We examined the protein sequences of all C. elegans Argonaute proteins and determined that ERGO-1, ALG-3, and HRDE-1 sequences contained RG/RGG motifs, therefore we selected these proteins as candidates for arginine dimethylation (DMA) (Fig 4). To this end, we immunoprecipitated ERGO-1, ALG-3, and HRDE-1 and subjected the proteins to mass spectrometry analysis. We identified at least one dimethylation in each of the three Argonaute proteins, and because our mass spectrometry coverage was incomplete and some RG/RGG motifs were not captured, either methylated or unmethylated, the dimethylation of these Argonaute proteins may be even more extensive (Fig 4). Thus, our data indicates that multiple C. elegans Argonaute proteins with RG/RGG motifs are dimethylated. What could be the functions of these methylation marks? In Drosophila, symmetrical DMAs (sDMA) at Aub facilitate Aub interaction with the Tudor protein Krimper. This interaction tethers Aub to the ping-pong site with its partner Ago3, which 129 also associates with Krimper, though not through post-translational modification (Webster et al., 2015). In C. elegans we have demonstrated that the dimethylarginine modification is responsible for CSR-1A small RNA binding specificity, which we hypothesize to be through complexing with Tudor proteins. In both Drosophila and C. elegans, methylation marks are dispensable for AGO localization to their respective substructures, and in C. elegans, the methylation marks on CSR-1A are not required for protein stability (Nguyen & Phillips, 2021). We therefore do not expect the newly identified DMA marks in ERGO-1, ALG-3, and HRDE-1, to be responsible for their localization. Instead, it seems likely that DMAs on AGO are the mechanism that allows for AGO to be sequestered to appropriate places to carry out their functions, possibly by providing a new platform for other proteins, Tudor, or other interacting proteins, to bind and tether to their respective complexes. 4.8 Summary One of our primary interests in this study is to investigate the general principles of C. elegans AGO localization and regulation. To this end, we have systemically generated a series of small RNA binding-defective Argonaute proteins, each belonging to a unique small RNA pathway. We have demonstrated that multiple Argonaute proteins fail to localize correctly in the absence of ther small RNA partners, and that only CSR-1 retains P granule localization, despite a severe disruption in granule morphology. While some Argonaute proteins in other species become destabilized in the absence of their small RNA partners, here we show that only PRG-1 requires small 130 RNA binding for protein stability, while all the other small RNA binding-defective Argonaute proteins tested were stably expressed. One significant caveat in this work is that we have yet to perform northern blots to confirm whether binding-deficient mutants are truly defective in small RNA binding and therefore ‘empty’. Nonetheless, our work demonstrates that mutating two of the four conserved residues in the small RNA binding pocket is sufficient to confer changes in localization and expression levels. 4.9 Methods C. elegans strains Strains were maintained at 20°C on NGM plates seeded with OP50 E. coli according to standard conditions unless otherwise stated(Brenner, 1974). All strains used in this project are listed in Supplementary Table 1. Plasmid and strain construction Plasmid-based CRISPR: All fluorescent and epitope tags were integrated at the endogenous loci by CRISPR genome editing(Dickinson et al., 2013, 2015). For all CRISPR insertions of fluorescent tags, we generated homologous repair templates using the primers listed in Supplementary Table 2. gfp::3xFLAG::ergo-1 were assembled into pDD282 (Addgene #66823) by isothermal assembly according to published protocols(Dickinson et al., 2015; Gibson et al., 2009). To protect the repair template from cleavage, we introduced silent mutations at the site of guide RNA targeting by incorporating these mutations into one of the homology arm primers or, if necessary, by 131 performing site-directed mutagenesis(Dickinson et al., 2013). All guide RNA plasmids were generated by ligating oligos containing the guide RNA sequence into BsaI-digested pRB1017 (Addgene #59936)(Arribere et al., 2014). Guide RNA sequences are provided in Supplementary Table 2. GFP CRISPR injection mixes included 25-50 ng/μl repair template, 50 ng/μl guide RNA plasmid, 50 ng/μl eft-3p::cas9-SV40_NLS::tbb-2 3’UTR (Addgene #46168), 2.5–10 ng/μl mCherry co-injection markers. The gfp::3xFLAG::ergo- 1 construct was injected into the wild-type strain. For wago-1[HK->AA] mutation, the injection mix included 50 ng/μl wago-1 AA repair oligo, 25 ng/μl, wago-1 guide RNA plasmids, 20 ng/µl rol-6 repair template, 25 ng/µl rol-6 guide RNA (pJA42, Addgene #59930), and 50 ng/µl eft-3p::Cas9 (pJW1259, Addgene #61251). Mixture was injected into USC988 (cmp41[mut-16::mCherry + loxP]: cmp92[(GFP + loxP + 3xFLAG) :: wago- 1]) animals. F1 animals with the Rol phenotype were isolated and genotyped by PCR to identify animals with the correct repair, which would yield a restriction site not present under wild-type condition(Arribere et al., 2014). All repair template sequences are provided in Supplementary Data 2. Protein-based CRISPR: For prg-1 [YK->AA], alg-3[YK->AA], ergo-1[YK->AA], csr-1[HK- >AA], and wago-10 [HK->AA] mutations, we used an oligo repair template and RNA guide (Supplementary Table 2). All injection mixes included 0.25 μg/μl Cas9 protein (IDT), 100 ng/μl tracrRNA (IDT), 14 ng/μl dpy-10 crRNA, 42 ng/μl gene-specific crRNA, and 110 ng/μl of each repair template. The prg-1 [YK->AA] injection mix was injected into USC1232 (prg-1(cmp220[(mKate2 +loxP + 3xMyc)::prg-1) I. The alg-3 [YK->AA] was 132 injected into USC1092 (alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg-3]) IV. The ergo-1[YK- >AA] injection mix was injected into USC1046 ergo-1(cmp94[(GFP + loxP +3xFLAG)::ergo-1]) V. The csr-1[HK->AA] injection mix was injected into USC1137 csr- 1(cmp173[(GFP + loxP + 3xFLAG)::csr-1]) IV. The wago-10 [HK->AA] injection mix was injected into USC1191 wago-10(cmp204[wago-10::2xFLAG]) V. Following injection, F1 animals with the Rol phenotype were isolated and genotyped by PCR to identify animals with the mutations of interest(Dokshin et al., 2018; Paix et al., 2015). The csr-1 [HK-to- AA] animals are sterile and needs to be maintained with a balancer. Live imaging Live imaging of C. elegans was performed in M9 buffer containing sodium azide to prevent movement. For immunofluorescence, C. elegans were dissected in egg buffer containing 0.1% Tween-20 and fixed in 1% formaldehyde in egg buffer as described(Phillips et al., 2009). Animals were dissected at the L4 (48 hours post hatching) or young adult stage (52 hours post hatching). Imaging was performed on a DeltaVision Elite microscope (GE Healthcare) using a 60x N.A. 1.42 oil-immersion objective. When data stacks were collected, three-dimensional images are presented as maximum intensity projections. Images were pseudocolored using the SoftWoRx package or Adobe Photoshop. 133 Western blots C. elegans were synchronized at 20°C by bleaching gravid adult animals and maintaining starved L1 larvae for at least 24 hours before plating on OP50. For sample collection, animals were harvested after 68 hours (gravid adults) on OP50 for PRG-1, ERGO-1, and WAGO-1 samples. Animals were harvest 48 hours (L4) on OP50 for ALG-3 and WAGO- 10 samples. Approximately 400 L4s or 200 gravid adults were loaded per lane. Proteins were resolved on 4-12% Bis-Tris polyacrylamide gels (Thermo Fisher), transferred to nitrocellulose membranes (Thermo Fisher), and probed with mouse anti-Myc 1:1,000 (ThermoFisher 13-2500), mouse anti-FLAG 1:1,000 (Sigma, F1804), or mouse anti-actin 1:10,000 (Abcam ab3280). Secondary HRP antibodies were purchased from Thermo Fisher. Brood size analysis Wild-type and mutant C. elegans strains were maintained at 20°C prior to temperature- shift experiments. Animals were either maintained at 20°C or shifted to 25°C as L4 larvae and ten of their progeny were picked to individual plates for 25°C brood size analysis. To score the complete brood, each animal was moved to a fresh plate every day until egg- laying was complete. After allowing the progeny 2-3 days to develop, the total number of animals on each plate was counted. 134 Immunoprecipitation/Mass Spectrometry Analysis For mass spectrometry experiments to identify post-translational modifications, starting with ~1.25 million synchronized USC1092 (alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg- 3]) IV L4 stage animals (~48 hours at 20°C after L1 arrest), and ~1 million each of synchronized USC1046 ergo-1(cmp94[(GFP + loxP +3xFLAG)::ergo-1]) V, USC1137 csr- 1(cmp173[(GFP + loxP + 3xFLAG)::csr-1]) IV, and WM285 [flag::wago-9/hrde-1, cb-unc- 119(+)] II; unc-119(ed3) III adult animals (~68 hours at 20°C after L1 arrest) were collected in IP Buffer (50 mM Tris-Cl pH 7.4, 100 mM KCl, 2.5 mM MgCl2, 0.1% Igapal CA-630, 0.5 mM PMSF, cOmplete Protease Inhibitor Cocktail (Roche 04693159001), and RNaseOUT Ribonuclease Inhibitor (Thermo Fisher 10777019)), frozen in liquid nitrogen, and homogenized using a mortar and pestle. Wild-type animals were prepped alongside as a negative control (~500, 000 synchronized adult, and ~750, 000 synchronized L4 animals). After further dilution into IP buffer (1:10 packed worms:buffer), insoluble particulate was removed by centrifugation and 10% of sample was taken as “input.” The remaining lysate was used for the immunoprecipitation. Immunoprecipitation was performed at 4°C for 1 hour with pre-conjugated anti-FLAG affinity matrix (Sigma Aldrich A22220), then washed at least 3 times in immunoprecipitation buffer. A fraction of each sample was analyzed by western blot to confirm efficacy of immunoprecipitation. After immunoprecipitation, a fraction of each sample was analyzed by western blot to confirm efficacy of immunoprecipitation. 2x sample buffer was added to the remainder of each sample, followed by gel electrophoresis (4-12% Bis-Tris polyacrylamide gels, Thermo Fisher) and overnight colloidal Coomassie staining. Bands containing 135 immunoprecipitated protein were excised from gel and cut into approximately 1 mm 3 pieces. Gel pieces were then subjected to a modified in-gel chymotrypsin digestion procedure(Shevchenko et al., 1996). Gel pieces were washed and dehydrated with acetonitrile for 10 min. followed by removal of acetonitrile. Pieces were then completely dried in a speed-vac. Rehydration of the gel pieces was with 50 mM ammonium bicarbonate solution containing 12.5 ng/µl modified sequencing-grade chymotrypsin (Promega) at 4ºC. After 45 min., the excess chymotrypsin solution was removed and replaced with 50 mM ammonium bicarbonate solution to just cover the gel pieces. Samples were then incubated at room temperature overnight. Peptides were later extracted by removing the ammonium bicarbonate solution, followed by one wash with a solution containing 50% acetonitrile and 1% formic acid. The extracts were dried in a speed-vac (~1 hr) and stored at 4ºC until analysis. On the day of analysis, the samples were reconstituted in 5 - 10 µl of HPLC solvent A (2.5% acetonitrile, 0.1% formic acid). A nano-scale reverse-phase HPLC capillary column was created by packing 2.6 µm C18 spherical silica beads into a fused silica capillary (100 µm inner diameter x ~30 cm length) with a flame-drawn tip(Peng & Gygi, 2001). After equilibrating the column each sample was loaded via a Famos auto sampler (LC Packings) onto the column. A gradient was formed and peptides were eluted with increasing concentrations of solvent B (97.5% acetonitrile, 0.1% formic acid). As peptides eluted, they were subjected to electrospray ionization and then entered into an LTQ Orbitrap Velos Pro ion-trap mass spectrometer (Thermo Fisher). Peptides were detected, isolated, and fragmented by collision-induced dissociation to produce a tandem 136 mass spectrum of specific fragment ions for each peptide. Peptide sequences (and hence protein identity) were determined by matching protein databases with the acquired fragmentation pattern by the software program, Sequest (Thermo Fisher)(Eng et al., 1994). All databases include a reversed version of all the sequences and the data was filtered to between a one and two percent peptide false discovery rate. 137 Materials Supplemental Table 1. Strains used in this study Experimental Models: Organisms/Strains C. elegans: Strain N2 - wild-type N2 C. elegans: Strain WM200 - alg-4(ok1041) III; alg-3(tm1155) IV WM300 C. elegans: Strain USC1092 - alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg-3]) USC1092 C. elegans: Strain USC1276 - alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg-3], cmp230[YK-->AA]) IV USC1276 C. elegans: Strain USC1342 - alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg-3]); alg-4(ok1041) USC1342 C. elegans: Strain USC1341 - alg-3(cmp155[(GFP + loxP + 3xFLAG)::alg-3], cmp230[YK-->AA]) IV; alg-4(ok1041) USC1341 C. elegans: Strain USC1232 - prg-1(cmp220[(mKate2 + loxP + 3xMyc)::prg-1]) I USC1232 C. elegans: Strain USC1277- prg-1(cmp220[(mKate2 + loxP + 3xMyc)::prg-1]), cmp231[YK-->AA]) I USC1277 C. elegans: Strain 1191 - wago-10(cmp204[wago-10::2xFLAG]) V C. elegans: Strain USC1287 - wago-10(cmp204[wago-10::2xFLAG]), cmp240[HK to AA] V USC1287 C. elegans: Strain1046 - ergo-1(cmp94[(GFP + loxP + 3xFLAG)::ergo-1]) V USC1046 C. elegans: Strain USC1281 - ergo-1(cmp94[(GFP + loxP + 3xFLAG)::ergo- 1]), cmp234[YK-->AA] USC1281 C. elegans: Strain USC988 - mut-16(cmp41[mut-16::mCherry + loxP]) : wago- 1(cmp92[(GFP + loxP + 3xFLAG)::wago-1] I USC988 C. elegans: Strain USC1196 - mut-16(cmp41[mut-16::mCherry + loxP]) : wago-1(cmp92[(GFP + loxP + 3xFLAG)::wago-1], cmp207[HK-->AA]) I USC988 C. elegans: Strain USC1137 - csr-1(cmp173[(GFP + loxP + 3xFLAG)::csr-1]) IV USC1137 138 C. elegans: Strain1295 - csr-1(cmp173[(GFP + loxP + 3xFLAG)::csr-1]), cmp244[HK to AA])/nT1 IV USC1295 C. elegans: WM285 - [flag::wago-9/hrde-1, cb-unc-119(+)] II; unc-119(ed3) III WM285 Supplemental Table 2. Oligonucleotide sequences used in this study Primers for plasmid-based CRISPR 3xFLAG::GFP::ergo- 1 guide RNA TGAGCTATAACAACGGCGG 5' homology arm - F CGTATGGACCCAAAATGGGC 5' homology arm - R TTTTGTTAGTCTGGAAGTGAGAAG 3' homology arm - F ATGAGCTACAATAATGGAGGCGGCGGCGGTGG 3' homology arm - R TCCGCCAAATCTGCAATTTGC wago-1[HK->AA] 5' guide RNA TAAGGAAAACTTACTGTGG 3' guide RNA TCAGAGCAGTACAAAGCCT repair template GTAAGTGCAAGTATGTCTTGATGATCACTGATGACGCAAT CGTTCACCTCGCGAAGCAATATGCAGCCTTGGAACAGAG AACAATGATGATCGTCCAGGATATGAAAATTTCCAAA Primers for protein-based CRISPR prg-1[YK->AA] guide RNA GAATGGGGCACTCGACACAT repair template GGAGGACATTCACATGCTCGTCGTAATGCTCGCTGACGA CAACAAAACTCGAGCCGACAGTCTCGCGAAGTTTCTATG TGTCGAGTGCCCCATTCCCAACCAATGCGTGAACTTGCG TACG alg-3[YK->AA] 5' guide RNA ACATCTACAGTAAGATTCGC 3' guide RNA TTTTATTTCAGTGACTGTGA repair template CAGTTTGTGATTGCATCATTGTAGTTCTCCAGTCCAAAAA CTCGGACATCGCCATGACTGTGGCGGAACAATCTGACAT TGTTCACGGAATTATGTCCCAATGCGTGCTCATG ergo-1[YK->AA] 5' guide RNA GCGAGAGCCACTGTGTATTC 3' guide RNA GCAAGATACTTGAGAACATC 139 repair template CGAGACGGCGAAGTCATCGTTCCGATTGTTTTTGCCGTT TTCCAGGCGCGTGCCACCGTCTACTCTGGCAACAATAAT GAGTATAATGATGCTAATGTTCTCGCGTATCTTGCCGATA ACAAGTATGGAATCCACACTCAAGGAATTCTCGAG csr-1[HK-->AA] guide RNA TCATACGCTCTTCGAATTTG repair template CGTCTTTCACTCCGTTCGTCTTGTTTATTTCCGATGACGT TCCTAACATTGCTGAGTGTCTCGCATTCGAAGAGCGTAT GAGTGACATTCCAACGCAGCACGTACTTCTCAAAAA wago-10[HK->AA] guide RNA TTCAAGACATCGTGAACATC repair template CTGCTTGCAAACCTGAATAGTCTGCAAGCCGACGGAAGC CTCATAATACgctAgcACATCGgcAACATCGGGCTTCTTCTC TTTGGCAATTGCAACCAGAATATTCACTCC 140 Fig 1 | Modifying Argonaute proteins binding pockets to render binding-deficient mutants. a, Sequence alignment of conserved 5’-phosphate binding residues across various species of Argonaute proteins. Prefix Af, Archaeoglobus fulgidus; Aa, Aquifex aeolicus; Hs, Homo Sapiens; Ce, Caenorhabditis elegans. Asterisk indicates conserved Y(or H)-K-Q-K. b, Schematic representation of fluorescent or epitope tagging of C. elegans Argonaute proteins. Asterisks indicate the first two conserved residues of the binding site that were mutated to Alanine. Note in CSR-1, the mutation affects both isoforms. c-f, Western blot to detect wild-type GFP::FLAG::ERGO-1 and GFP::FLAG::ERGO-1[YK-to-AA] (c), wild- type GFP::FLAG::WAGO-1 and GFP::FLAG::WAGO-1[HK-to-AA] (e), and wild-type FLAG::WAGO-10 and FLAG::WAGO-10[HK-to-AA] strains (f) using a-FLAG antibody, and wild-type mKate::Myc::PRG-1 and mKate::Myc::PRG-1[YK-to-AA] strains (d) using a-Myc antibody. Actin is shown as a loading control. 141 Fig 2 | Binding-deficient mutants fail to localize to P granules. a-c, Live imaging of GFP::FLAG::WAGO-1 and GFP::FLAG::WAGO-1[HK-to-AA] (a) and mKate::myc::PRG-1 and mKate::myc-1::PRG-1[YK-to-AA] (b) in gravid adult germline, and GFP::FLAG::ALG-3 and GFP::FLAG::ALG-3[YK-to-AA] (c) in L4 germline. d, Western blot to detect wild-type GFP::FLAG::ALG-3 and GFP::FLAG::ALG-3[YK-to-AA] strains, using a-FLAG antibody. Actin is shown as a loading control. e, Brood size assay on GFP::FLAG::ALG-3, GFP::FLAG::ALG-3; alg-4, and GFP::FLAG::ALG-3[YK-to- AA]; alg-4 raised at 25C (n=10). Two-tail t-tests were performed to determined statistical significance. 142 Fig 3 | csr-1[HK-to-AA] mutants are sterile and form misshapen P granules. a, Live imaging of wild-type GFP::FLAG::CSR-1 and GFP::FLAG::CSR-1[HK-to-AA] in gravid adult germline. Scale bars, 25µM. b, Quantification of GFP::CSR-1 granule size in GFP::FLAG::CSR-1 and GFP::FLAG::CSR-1[HK-to- AA]. Granule diameter was defined as the distance between two furthest points within one focus on a given z-plane using the measuring software of the SoftWoRx package. At least 50 nuclei were scored per genotype. Fig 4 | Argonaute proteins are dimethylated at the RG motifs. Graphical display of dimethylation detected on the primary Argonaute proteins, ERGO-1 and ALG-3 and the secondary Argonaute protein HRDE-1 by mass spectrometry following IP. 143 Appendix A: Knockdown of CSR-1 results enlarged germline nuclei Dieu An H. Nguyen, Jenny Zhao, Diala Alhousari AID::CSR-1 is completely knockdown when auxin is added Null csr-1 mutants result in sterility and embryonic lethality. This makes the mutant difficult to work with on a larger scale. To bypass this problem, we employed the Auxin-inducible degron (AID) system to conditionally knockdown CSR-1 that has been endogenously labeled with a degron tag (L. Zhang et al., 2015). The strain was provided by Jessica Kirshner from the John Kim Lab. We first asked if auxin can completely knockdown CSR-1 expression. To do this, we resuspended auxin in ethanol, as described, in a series of dilution, starting with 0mM, 2.5mM, 5mM, and 10mM. We added either 0mM, 2.5mM, 5mM, or 10mM of dissolved auxin to our degron-tagged CSR-1 (AID::CSR-1). We found that 10mM was required to induce complete sterility of every AID::CSR-1 animals on the plates after 24 hours, a phenotype observed in null csr-1 mutant. To investigate if adding auxin degrades all CSR-1 from the system, we performed western blot on whole-cell lysates of AID::CSR-1 animals that had been treated with either 0mM or 10mM of auxin. We collected the animals at various timepoints to determine how long it would take for auxin to induce CSR-1 degradation. We found that after four to eight hours, CSR-1 expression was decreased compared to 144 wild-type, and CSR-1 expression was completely gone after 24 hours (Appendix A, a). Thus we confirmed that adding auxin to AID::CSR-1 is sufficient to knockdown CSR-1. Next, we asked what happens to the germline after CSR-1 is knockdown by auxin. To do this, we treated the AID::CSR-1 animals with either 0mM or 10mM auxin for 24 hours and immunostained the dissected germline. We found that pachytene nuclei in the germlines that had been treated with 0mM appeared to be uniform, whose condensed chromosomes are decorated by the synaptonemal complex proteins HTP-3 and SYP-1 (Appendix A, b). In contrast, nuclei in germlines of AID::CSR-1 animals that had been treated by 10mM of auxin lack this uniformity, with some appeared to be two to three times larger than their neighboring nuclei (Appendix A, c). There are still wild- type pachytene nuclei stained by HTP-3 and SYP-1 in these germlines, but the enlarged nuclei lose staining for these pachytene hallmarks. DNA in these large nuclei is not even condensed, suggesting that they might be stuck in the interphase of mitosis. These animals are sterile. Despite the fact CSR-1 has already been implicated in multiple facets of germline regulations, these experiments shed light on a possible new role of CSR-1 in cell division regulation. 145 Appendix A | Absence of CSR-1 results in enlarged mitotic nuclei in the germline a, Western blot detecting for CSR-1 in the endogenously tagged AID::CSR-1; psun-1::tir- 1::Ruby animals that had been treated with either 0mM or 10mM auxin, using a-CSR-1. Actin is used as loading control. b-c, Immunofluorescent staining of AID::CSR-1; psun-1::tir-1::Ruby of germlines that had been treated with either 0mM (b) or 10mM (c) auxin after 24 hours, using antibodies against HTP-3 and SYP-1 and DAPI to stain DNA. Arrows indicate germline distal tip. Scale bars represent 25µM. 146 References Adams-Cioaba, M. A. & Min, J. (2009). Structure and function of histone methylation binding proteins. Biochemistry and Cell Biology, 87(1), 93–105. https://doi.org/10.1139/o08-129 Albuquerque, B. F. M. de, Luteijn, M. J., Rodrigues, R. J. C., Bergeijk, P. van, Waaijers, S., Kaaij, L. J. T., Klein, H., Boxem, M. & Ketting, R. F. (2014). PID-1 is a novel factor that operates during 21U-RNA biogenesis in Caenorhabditis elegans. Genes & Development, 28(7), 683–688. https://doi.org/10.1101/gad.238220.114 Amiri, A., Keiper, B. D., Kawasaki, I., Fan, Y., Kohara, Y., Rhoads, R. E. & Strome, S. (2001). An isoform of eIF4E is a component of germ granules and is required for spermatogenesis in C. elegans. Development, 128(20), 3899–3912. Aoki, K., Moriguchi, H., Yoshioka, T., Okawa, K. & Tabara, H. (2007). In vitro analyses of the production and activity of secondary small interfering RNAs in C. elegans. The EMBO Journal, 26(24), 5007–5019. https://doi.org/10.1038/sj.emboj.7601910 Arribere, J. A., Bell, R. T., Fu, B. X. H., Artiles, K. L., Hartman, P. S. & Fire, A. Z. (2014). Efficient marker-free recovery of custom genetic modifications with CRISPR/Cas9 in Caenorhabditis elegans. Genetics, 198(3), 837–846. https://doi.org/10.1534/genetics.114.169730 Ashe, A., Sapetschnig, A., Weick, E.-M., Mitchell, J., Bagijn, M. P., Cording, A. C., Doebley, A.-L., Goldstein, L. D., Lehrbach, N. J., Pen, J. L., Pintacuda, G., Sakaguchi, A., Sarkies, P., Ahmed, S. & Miska, E. A. (2012). piRNAs can trigger a multigenerational epigenetic memory in the germline of C. elegans. Cell, 150(1), 88– 99. https://doi.org/10.1016/j.cell.2012.06.018 Avgousti, D. C., Palani, S., Sherman, Y. & Grishok, A. (2012). CSR-1 RNAi pathway positively regulates histone expression in C. elegans. The EMBO Journal, 31(19), 3821–3832. https://doi.org/10.1038/emboj.2012.216 Bagijn, M. P., Goldstein, L. D., Sapetschnig, A., Weick, E.-M., Bouasker, S., Lehrbach, N. J., Simard, M. J. & Miska, E. A. (2012). Function, targets, and evolution of Caenorhabditis elegans piRNAs. Science, 337(6094), 574–578. https://doi.org/10.1126/science.1220952 Barberán-Soler, S., Fontrodona, L., Ribó, A., Lamm, A. T., Iannone, C., Cerón, J., Lehner, B. & Valcárcel, J. (2014). Co-option of the piRNA Pathway for Germline- Specific Alternative Splicing of C. elegans TOR. Cell Reports, 8(6), 1609–1616. https://doi.org/10.1016/j.celrep.2014.08.016 147 Barucci, G., Cornes, E., Singh, M., Li, B., Ugolini, M., Samolygo, A., Didier, C., Dingli, F., Loew, D., Quarato, P. & Cecere, G. (2020). Small-RNA-mediated transgenerational silencing of histone genes impairs fertility in piRNA mutants. Nature Cell Biology, 22(2), 235–245. https://doi.org/10.1038/s41556-020-0462-7 Batista, P. J., Ruby, J. G., Claycomb, J. M., Chiang, R., Fahlgren, N., Kasschau, K. D., Chaves, D. A., Gu, W., Vasale, J. J., Duan, S., Conte, D., Luo, S., Schroth, G. P., Carrington, J. C., Bartel, D. P. & Mello, C. C. (2008). PRG-1 and 21U-RNAs Interact to Form the piRNA Complex Required for Fertility in C. elegans. Molecular Cell, 31(1), 67–78. https://doi.org/10.1016/j.molcel.2008.06.002 Bezler, A., Braukmann, F., West, S. M., Duplan, A., Conconi, R., Schütz, F., Gönczy, P., Piano, F., Gunsalus, K., Miska, E. A. & Keller, L. (2019). Tissue- and sex-specific small RNAomes reveal sex differences in response to the environment. PLoS Genetics, 15(2), e1007905. https://doi.org/10.1371/journal.pgen.1007905 Bolger, A. M., Lohse, M. & Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics, 30(15), 2114–2120. https://doi.org/10.1093/bioinformatics/btu170 Bouwman, P., Aly, A., Escandell, J. M., Pieterse, M., Bartkova, J., Gulden, H. van der, Hiddingh, S., Thanasoula, M., Kulkarni, A., Yang, Q., Haffty, B. G., Tommiska, J., Blomqvist, C., Drapkin, R., Adams, D. J., Nevanlinna, H., Bartek, J., Tarsounas, M., Ganesan, S. & Jonkers, J. (2010). 53BP1 loss rescues BRCA1 deficiency and is associated with triple-negative and BRCA-mutated breast cancers. Nature Structural & Molecular Biology, 17(6), 688–695. https://doi.org/10.1038/nsmb.1831 Brangwynne, C. P., Eckmann, C. R., Courson, D. S., Rybarska, A., Hoege, C., Gharakhani, J., Jülicher, F. & Hyman, A. A. (2009). Germline P granules are liquid droplets that localize by controlled dissolution/condensation. Science, 324(5935), 1729–1732. https://doi.org/10.1126/science.1172046 Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics, 77(1), 71–94. Buckley, B. A., Burkhart, K. B., Gu, S. G., Spracklin, G., Kershner, A., Fritz, H., Kimble, J., Fire, A. & Kennedy, S. (2012). A nuclear Argonaute promotes multigenerational epigenetic inheritance and germline immortality. Nature, 489(7416), 447–451. https://doi.org/10.1038/nature11352 Burkhart, K. B., Guang, S., Buckley, B. A., Wong, L., Bochner, A. F. & Kennedy, S. (2011). A pre-mRNA-associating factor links endogenous siRNAs to chromatin regulation. PLoS Genetics, 7(8), e1002249. https://doi.org/10.1371/journal.pgen.1002249 148 Carmell, M. A., Xuan, Z., Zhang, M. Q. & Hannon, G. J. (2002). The Argonaute family: tentacles that reach into RNAi, developmental control, stem cell maintenance, and tumorigenesis. Genes & Development, 16(21), 2733–2742. https://doi.org/10.1101/gad.1026102 Cecere, G., Hoersch, S., O’Keeffe, S., Sachidanandam, R. & Grishok, A. (2014). Global effects of the CSR-1 RNA interference pathway on the transcriptional landscape. Nature Structural & Molecular Biology, 21(4), 358–365. https://doi.org/10.1038/nsmb.2801 Chen, H. & Boutros, P. C. (2011). VennDiagram: a package for the generation of highly- customizable Venn and Euler diagrams in R. BMC Bioinformatics, 12(1), 35. https://doi.org/10.1186/1471-2105-12-35 Claycomb, J. M., Batista, P. J., Pang, K. M., Gu, W., Vasale, J. J., Wolfswinkel, J. C. van, Chaves, D. A., Shirayama, M., Mitani, S., Ketting, R. F., Conte, D. & Mello, C. C. (2009). The Argonaute CSR-1 and its 22G-RNA cofactors are required for holocentric chromosome segregation. Cell, 139(1), 123–134. https://doi.org/10.1016/j.cell.2009.09.014 Conine, C. C., Batista, P. J., Gu, W., Claycomb, J. M., Chaves, D. A., Shirayama, M. & Mello, C. C. (2010). Argonautes ALG-3 and ALG-4 are required for spermatogenesis- specific 26G-RNAs and thermotolerant sperm in Caenorhabditis elegans. Proceedings of the National Academy of Sciences of the United States of America, 107(8), 3588–3593. https://doi.org/10.1073/pnas.0911685107 Conine, C. C., Moresco, J. J., Gu, W., Shirayama, M., Conte, D., Yates, J. R. & Mello, C. C. (2013). Argonautes promote male fertility and provide a paternal memory of germline gene expression in C. elegans. Cell, 155(7), 1532–1544. https://doi.org/10.1016/j.cell.2013.11.032 Courchaine, E. M., Barentine, A. E. S., Straube, K., Lee, D.-R., Bewersdorf, J. & Neugebauer, K. M. (2021). DMA-tudor interaction modules control the specificity of in vivo condensates. Cell. https://doi.org/10.1016/j.cell.2021.05.008 Crittenden, S. L., Bernstein, D. S., Bachorik, J. L., Thompson, B. E., Gallegos, M., Petcherski, A. G., Moulder, G., Barstead, R., Wickens, M. & Kimble, J. (2002). A conserved RNA-binding protein controls germline stem cells in Caenorhabditis elegans. Nature, 417(6889), 660–663. https://doi.org/10.1038/nature754 Czech, B. & Hannon, G. J. (2011). Small RNA sorting: matchmaking for Argonautes. Nature Reviews Genetics, 12(1), 19–31. https://doi.org/10.1038/nrg2916 de Albuquerque, B. F. M., Placentino, M. & Ketting, R. F. (2015). Maternal piRNAs Are Essential for Germline Development following De Novo Establishment of Endo- 149 siRNAs in Caenorhabditis elegans. Developmental Cell, 34(4), 448–456. https://doi.org/10.1016/j.devcel.2015.07.010 Derrien, B., Baumberger, N., Schepetilnikov, M., Viotti, C., Cillia, J. D., Ziegler-Graff, V., Isono, E., Schumacher, K. & Genschik, P. (2012). Degradation of the antiviral component ARGONAUTE1 by the autophagy pathway. Proceedings of the National Academy of Sciences, 109(39), 15942–15946. https://doi.org/10.1073/pnas.1209487109 Dhar, S., Vemulapalli, V., Patananan, A. N., Huang, G. L., Lorenzo, A. D., Richard, S., Comb, M. J., Guo, A., Clarke, S. G. & Bedford, M. T. (2013). Loss of the major Type I arginine methyltransferase PRMT1 causes substrate scavenging by other PRMTs. Scientific Reports, 3(1), 1311. https://doi.org/10.1038/srep01311 Dickinson, D. J., Pani, A. M., Heppert, J. K., Higgins, C. D. & Goldstein, B. (2015). Streamlined Genome Engineering with a Self-Excising Drug Selection Cassette. Genetics, 200(4), 1035–1049. https://doi.org/10.1534/genetics.115.178335 Dickinson, D. J., Ward, J. D., Reiner, D. J. & Goldstein, B. (2013). Engineering the Caenorhabditis elegans genome using Cas9-triggered homologous recombination. Nature Methods, 10(10), 1028–1034. https://doi.org/10.1038/nmeth.2641 Dokshin, G. A., Ghanta, K. S., Piscopo, K. M. & Mello, C. C. (2018). Robust Genome Editing with Short Single-Stranded and Long, Partially Single-Stranded DNA Donors in Caenorhabditis elegans. Genetics, 210(3), 781–787. https://doi.org/10.1534/genetics.118.301532 Dosztányi, Z., Csizmók, V., Tompa, P. & Simon, I. (2005). The Pairwise Energy Content Estimated from Amino Acid Composition Discriminates between Folded and Intrinsically Unstructured Proteins. Journal of Molecular Biology, 347(4), 827–839. https://doi.org/10.1016/j.jmb.2005.01.071 Duchaine, T. F., Wohlschlegel, J. A., Kennedy, S., Bei, Y., Conte, D., Pang, K., Brownell, D. R., Harding, S., Mitani, S., Ruvkun, G., Yates, J. R. & Mello, C. C. (2006). Functional proteomics reveals the biochemical niche of C. elegans DCR-1 in multiple small-RNA-mediated pathways. Cell, 124(2), 343–354. https://doi.org/10.1016/j.cell.2005.11.036 Edgar, L. G., Wolf, N. & Wood, W. B. (1994). Early transcription in Caenorhabditis elegans embryos. Development (Cambridge, England), 120(2), 443–451. Elbaum-Garfinkle, S., Kim, Y., Szczepaniak, K., Chen, C. C.-H., Eckmann, C. R., Myong, S. & Brangwynne, C. P. (2015). The disordered P granule protein LAF-1 drives phase separation into droplets with tunable viscosity and dynamics. 150 Proceedings of the National Academy of Sciences of the United States of America, 112(23), 7189–7194. https://doi.org/10.1073/pnas.1504822112 Eng, J. K., McCormack, A. L. & Yates, J. R. (1994). An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. Journal of the American Society for Mass Spectrometry, 5(11), 976–989. https://doi.org/10.1016/1044-0305(94)80016-2 Fassnacht, C., Tocchini, C., Kumari, P., Gaidatzis, D., Stadler, M. B. & Ciosk, R. (2018). The CSR-1 endogenous RNAi pathway ensures accurate transcriptional reprogramming during the oocyte-to-embryo transition in Caenorhabditis elegans. PLoS Genetics, 14(3), e1007252. https://doi.org/10.1371/journal.pgen.1007252 Feric, M., Vaidya, N., Harmon, T. S., Mitrea, D. M., Zhu, L., Richardson, T. M., Kriwacki, R. W., Pappu, R. V. & Brangwynne, C. P. (2016). Coexisting Liquid Phases Underlie Nucleolar Subcompartments. Cell, 165(7), 1686–1697. https://doi.org/10.1016/j.cell.2016.04.047 Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E. & Mello, C. C. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 391(6669), 806–811. https://doi.org/10.1038/35888 Fischer, S. E. J., Montgomery, T. A., Zhang, C., Fahlgren, N., Breen, P. C., Hwang, A., Sullivan, C. M., Carrington, J. C. & Ruvkun, G. (2011). The ERI-6/7 helicase acts at the first stage of an siRNA amplification pathway that targets recent gene duplications. PLoS Genetics, 7(11), e1002369. https://doi.org/10.1371/journal.pgen.1002369 Friedland, A. E., Tzur, Y. B., Esvelt, K. M., Colaiácovo, M. P., Church, G. M. & Calarco, J. A. (2013). Heritable genome editing in C. elegans via a CRISPR-Cas9 system. Nature Methods, 10(8), 741–743. https://doi.org/10.1038/nmeth.2532 Friend, K., Campbell, Z. T., Cooke, A., Kroll-Conner, P., Wickens, M. P. & Kimble, J. (2012). A conserved PUF-Ago-eEF1A complex attenuates translation elongation. Nature Structural & Molecular Biology, 19(2), 176–183. https://doi.org/10.1038/nsmb.2214 Frøkjær-Jensen, C., Davis, M. W., Hopkins, C. E., Newman, B. J., Thummel, J. M., Olesen, S.-P., Grunnet, M. & Jorgensen, E. M. (2008). Single-copy insertion of transgenes in Caenorhabditis elegans. Nature Genetics, 40(11), 1375–1383. https://doi.org/10.1038/ng.248 Gent, J. I., Lamm, A. T., Pavelec, D. M., Maniar, J. M., Parameswaran, P., Tao, L., Kennedy, S. & Fire, A. Z. (2010). Distinct Phases of siRNA Synthesis in an 151 Endogenous RNAi Pathway in C. elegans Soma. Molecular Cell, 37(5), 679–689. https://doi.org/10.1016/j.molcel.2010.01.012 Gerson-Gurwitz, A., Wang, S., Sathe, S., Green, R., Yeo, G. W., Oegema, K. & Desai, A. (2016). A Small RNA-Catalytic Argonaute Pathway Tunes Germline Transcript Levels to Ensure Embryonic Divisions. Cell, 165(2), 396–409. https://doi.org/10.1016/j.cell.2016.02.040 Ghildiyal, M. & Zamore, P. D. (2009). Small silencing RNAs: an expanding universe. Nature Reviews Genetics, 10(2), 94–108. https://doi.org/10.1038/nrg2504 Gibson, D. G., Young, L., Chuang, R.-Y., Venter, J. C., Hutchison, C. A. & Smith, H. O. (2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature Methods, 6(5), 343–345. https://doi.org/10.1038/nmeth.1318 Goh, W.-S. S., Seah, J. W. E., Harrison, E. J., Chen, C., Hammell, C. M. & Hannon, G. J. (2014). A genome-wide RNAi screen identifies factors required for distinct stages of C. elegans piRNA biogenesis. Genes & Development, 28(7), 797–807. https://doi.org/10.1101/gad.235622.113 Gu, S. G., Pak, J., Guang, S., Maniar, J. M., Kennedy, S. & Fire, A. (2012). Amplification of siRNA in Caenorhabditis elegans generates a transgenerational sequence-targeted histone H3 lysine 9 methylation footprint. Nature Genetics, 44(2), 157–164. https://doi.org/10.1038/ng.1039 Gu, W., Shirayama, M., Conte, D., Vasale, J., Batista, P. J., Claycomb, J. M., Moresco, J. J., Youngman, E. M., Keys, J., Stoltz, M. J., Chen, C.-C. G., Chaves, D. A., Duan, S., Kasschau, K. D., Fahlgren, N., Yates, J. R., Mitani, S., Carrington, J. C. & Mello, C. C. (2009). Distinct argonaute-mediated 22G-RNA pathways direct genome surveillance in the C. elegans germline. Molecular Cell, 36(2), 231–244. https://doi.org/10.1016/j.molcel.2009.09.020 Guang, S., Bochner, A. F., Pavelec, D. M., Burkhart, K. B., Harding, S., Lachowiec, J. & Kennedy, S. (2008). An Argonaute transports siRNAs from the cytoplasm to the nucleus. Science, 321(5888), 537–541. https://doi.org/10.1126/science.1157647 Guo, S. & Kemphues, K. J. (1995). par-1, a gene required for establishing polarity in C. elegans embryos, encodes a putative Ser/Thr kinase that is asymmetrically distributed. Cell, 81(4), 611–620. https://doi.org/10.1016/0092-8674(95)90082-9 Han, T., Manoharan, A. P., Harkins, T. T., Bouffard, P., Fitzpatrick, C., Chu, D. S., Thierry-Mieg, D., Thierry-Mieg, J. & Kim, J. K. (2009). 26G endo-siRNAs regulate spermatogenic and zygotic gene expression in Caenorhabditis elegans. Proceedings of the National Academy of Sciences of the United States of America, 106(44), 18674–18679. https://doi.org/10.1073/pnas.0906378106 152 Huberdeau, M. Q., Zeitler, D. M., Hauptmann, J., Bruckmann, A., Fressigné, L., Danner, J., Piquet, S., Strieder, N., Engelmann, J. C., Jannot, G., Deutzmann, R., Simard, M. J. & Meister, G. (2017). Phosphorylation of Argonaute proteins affects mRNA binding and is essential for microRNA-guided gene silencing in vivo. The EMBO Journal, 36(14), 2088–2106. https://doi.org/10.15252/embj.201696386 Hulsen, T., Vlieg, J. de & Alkema, W. (2008). BioVenn – a web application for the comparison and visualization of biological lists using area-proportional Venn diagrams. BMC Genomics, 9(1), 488. https://doi.org/10.1186/1471-2164-9-488 Hutvagner, G. & Simard, M. J. (2008). Argonaute proteins: key players in RNA silencing. Nature Reviews Molecular Cell Biology, 9(1), 22–32. https://doi.org/10.1038/nrm2321 Ishidate, T., Ozturk, A. R., Durning, D. J., Sharma, R., Shen, E., Chen, H., Seth, M., Shirayama, M. & Mello, C. C. (2018). ZNFX-1 Functions within Perinuclear Nuage to Balance Epigenetic Signals. Molecular Cell, 70(4), 639-649.e6. https://doi.org/10.1016/j.molcel.2018.04.009 Izumi, N. & Tomari, Y. (2014). Diversity of the piRNA pathway for nonself silencing: worm-specific piRNA biogenesis factors. Genes & Development, 28(7), 665–671. https://doi.org/10.1101/gad.241323.114 Jackson, R. J. & Standart, N. (2007). How Do MicroRNAs Regulate Gene Expression? Sci. STKE, 2007(367), re1–re1. https://doi.org/10.1126/stke.3672007re1 Jin, J., Xie, X., Chen, C., Park, J. G., Stark, C., James, D. A., Olhovsky, M., Linding, R., Mao, Y. & Pawson, T. (2009). Eukaryotic Protein Domains as Functional Units of Cellular Evolution. Science Signaling, 2(98), ra76–ra76. https://doi.org/10.1126/scisignal.2000546 Johnston, M., Geoffroy, M.-C., Sobala, A., Hay, R. & Hutvagner, G. (2010). HSP90 Protein Stabilizes Unloaded Argonaute Complexes and Microscopic P-bodies in Human Cells. Molecular Biology of the Cell, 21(9), 1462–1469. https://doi.org/10.1091/mbc.e09-10-0885 Kelly, W. G., Schaner, C. E., Dernburg, A. F., Lee, M.-H., Kim, S. K., Villeneuve, A. M. & Reinke, V. (2002). X-chromosome silencing in the germline of C. elegans. Development, 129(2). Ketting, R. F., Fischer, S. E., Bernstein, E., Sijen, T., Hannon, G. J. & Plasterk, R. H. (2001). Dicer functions in RNA interference and in synthesis of small RNA involved in developmental timing in C. elegans. Genes & Development, 15(20), 2654–2659. https://doi.org/10.1101/gad.927801 153 Kim, D., Langmead, B. & Salzberg, S. L. (2015). HISAT: a fast spliced aligner with low memory requirements. Nature Methods, 12(4), 357–360. https://doi.org/10.1038/nmeth.3317 Kim, J., Daniel, J., Espejo, A., Lake, A., Krishna, M., Xia, L., Zhang, Y. & Bedford, M. T. (2006). Tudor, MBT and chromo domains gauge the degree of lysine methylation. EMBO Reports, 7(4), 397–403. https://doi.org/10.1038/sj.embor.7400625 Kirino, Y., Kim, N., Planell-Saguer, M. de, Khandros, E., Chiorean, S., Klein, P. S., Rigoutsos, I., Jongens, T. A. & Mourelatos, Z. (2009). Arginine methylation of Piwi proteins catalysed by dPRMT5 is required for Ago3 and Aub stability. Nature Cell Biology, 11(5), 652–658. https://doi.org/10.1038/ncb1872 Kirino, Y., Vourekas, A., Kim, N., Alves, F. de L., Rappsilber, J., Klein, P. S., Jongens, T. A. & Mourelatos, Z. (2010). Arginine Methylation of Vasa Protein Is Conserved across Phyla*. Journal of Biological Chemistry, 285(11), 8148–8154. https://doi.org/10.1074/jbc.m109.089821 Kirino, Y., Vourekas, A., Sayed, N., Alves, F. de L., Thomson, T., Lasko, P., Rappsilber, J., Jongens, T. A. & Mourelatos, Z. (2010). Arginine methylation of Aubergine mediates Tudor binding and germ plasm localization. RNA, 16(1), 70–78. https://doi.org/10.1261/rna.1869710 Klass, M., Wolf, N. & Hirsh, D. (1976). Development of the male reproductive system and sexual transformation in the nematode Caenorhabditis elegans. Developmental Biology, 52(1), 1–18. https://doi.org/10.1016/0012-1606(76)90002-6 Knutson, A. K., Egelhofer, T., Rechtsteiner, A. & Strome, S. (2017). Germ Granules Prevent Accumulation of Somatic Transcripts in the Adult Caenorhabditis elegans Germline. Genetics, 206(1), 163–178. https://doi.org/10.1534/genetics.116.198549 Kobayashi, H., Shoji, K., Kiyokawa, K., Negishi, L. & Tomari, Y. (2019a). Iruka Eliminates Dysfunctional Argonaute by Selective Ubiquitination of Its Empty State. Molecular Cell, 73(1), 119-129.e5. https://doi.org/10.1016/j.molcel.2018.10.033 Kobayashi, H., Shoji, K., Kiyokawa, K., Negishi, L. & Tomari, Y. (2019b). VCP Machinery Mediates Autophagic Degradation of Empty Argonaute. Cell Reports, 28(5), 1144-1153.e4. https://doi.org/10.1016/j.celrep.2019.07.003 LaMunyon, C. W. & Ward, S. (1998). Larger sperm outcompete smaller sperm in the nematode Caenorhabditis elegans. Proceedings. Biological Sciences, 265(1409), 1997–2002. https://doi.org/10.1098/rspb.1998.0531 Langmead, B. & Salzberg, S. L. (2012). Fast gapped-read alignment with Bowtie 2. Nature Methods, 9(4), 357–359. https://doi.org/10.1038/nmeth.1923 154 Larson, A. G., Elnatan, D., Keenen, M. M., Trnka, M. J., Johnston, J. B., Burlingame, A. L., Agard, D. A., Redding, S. & Narlikar, G. J. (2017). Liquid droplet formation by HP1α suggests a role for phase separation in heterochromatin. Nature, 547(7662), 236–240. https://doi.org/10.1038/nature22822 Lee, H.-C., Gu, W., Shirayama, M., Youngman, E., Conte, D. & Mello, C. C. (2012). C. elegans piRNAs mediate the genome-wide surveillance of germline transcripts. Cell, 150(1), 78–87. https://doi.org/10.1016/j.cell.2012.06.016 Liao, Y., Smyth, G. K. & Shi, W. (2013). The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic Acids Research, 41(10), e108– e108. https://doi.org/10.1093/nar/gkt214 Liao, Y., Smyth, G. K. & Shi, W. (2014). featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics, 30(7), 923–930. https://doi.org/10.1093/bioinformatics/btt656 Liu, H., Wang, J.-Y. S., Huang, Y., Li, Z., Gong, W., Lehmann, R. & Xu, R.-M. (2010). Structural basis for methylarginine-dependent recognition of Aubergine by Tudor. Genes & Development, 24(17), 1876–1881. https://doi.org/10.1101/gad.1956010 Liu, K., Chen, C., Guo, Y., Lam, R., Bian, C., Xu, C., Zhao, D. Y., Jin, J., MacKenzie, F., Pawson, T. & Min, J. (2010). Structural basis for recognition of arginine methylated Piwi proteins by the extended Tudor domain. Proceedings of the National Academy of Sciences of the United States of America, 107(43), 18398–18403. https://doi.org/10.1073/pnas.1013106107 Love, M. I., Huber, W. & Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550. https://doi.org/10.1186/s13059-014-0550-8 Ma, J.-B., Ye, K. & Patel, D. J. (2004). Structural basis for overhang-specific small interfering RNA recognition by the PAZ domain. Nature, 429(6989), 318–322. https://doi.org/10.1038/nature02519 Ma, J.-B., Yuan, Y.-R., Meister, G., Pei, Y., Tuschl, T. & Patel, D. J. (2005). Structural basis for 5’-end-specific recognition of guide RNA by the A. fulgidus Piwi protein. Nature, 434(7033), 666–670. https://doi.org/10.1038/nature03514 Manage, K. I., Rogers, A. K., Wallis, D. C., Uebel, C. J., Anderson, D. C., Nguyen, D. A. H., Arca, K., Brown, K. C., Rodrigues, R. J. C., Albuquerque, B. F. M. de, Ketting, R. F., Montgomery, T. A. & Phillips, C. M. (2020). A Tudor domain protein, SIMR-1, promotes siRNA production at piRNA-targeted mRNAs in C. elegans. ELife, 9, e56731. https://doi.org/10.7554/elife.56731 155 Maniar, J. M. & Fire, A. Z. (2011). EGO-1, a C. elegans RdRP, modulates gene expression via production of mRNA-templated short antisense RNAs. Current Biology : CB, 21(6), 449–459. https://doi.org/10.1016/j.cub.2011.02.019 Mao, H., Zhu, C., Zong, D., Weng, C., Yang, X., Huang, H., Liu, D., Feng, X. & Guang, S. (2015). The Nrde Pathway Mediates Small-RNA-Directed Histone H3 Lysine 27 Trimethylation in Caenorhabditis elegans. Current Biology : CB, 25(18), 2398–2403. https://doi.org/10.1016/j.cub.2015.07.051 Marnik, E. A., Almeida, M. V., Cipriani, P. G., Chung, G., Caspani, E., Karaulanov, E., Butter, F., Sharp, C. S., Zinno, J., Gan, H. H., Piano, F., Ketting, R., Gunsalus, K. & Updike, D. L. (2021). The Caenorhabditis elegans TDRD5/7-like protein, LOTR-1, interacts with the helicase ZNFX-1 to balance epigenetic signals in the germline. BioRxiv. https://doi.org/10.1101/2021.06.18.448978 Marnik, E. A., Fuqua, J. H., Sharp, C. S., Rochester, J. D., Xu, E. L., Holbrook, S. E. & Updike, D. L. (2019). Germline Maintenance Through the Multifaceted Activities of GLH/Vasa in Caenorhabditis elegans P Granules. Genetics, 213(3), 923–939. https://doi.org/10.1534/genetics.119.302670 Martinez, N. J. & Gregory, R. I. (2013). Argonaute2 expression is post-transcriptionally coupled to microRNA abundance. RNA, 19(5), 605–612. https://doi.org/10.1261/rna.036434.112 McCarter, J., Bartlett, B., Dang, T. & Schedl, T. (1999). On the Control of Oocyte Meiotic Maturation and Ovulation inCaenorhabditis elegans. Developmental Biology, 205(1), 111–128. https://doi.org/10.1006/dbio.1998.9109 Meister, G., Landthaler, M., Patkaniowska, A., Dorsett, Y., Teng, G. & Tuschl, T. (2004). Human Argonaute2 Mediates RNA Cleavage Targeted by miRNAs and siRNAs. Molecular Cell, 15(2), 185–197. https://doi.org/10.1016/j.molcel.2004.07.007 Mészáros, B., Erdős, G. & Dosztányi, Z. (2018). IUPred2A: context-dependent prediction of protein disorder as a function of redox state and protein binding. Nucleic Acids Research, 46(W1), gky384-. https://doi.org/10.1093/nar/gky384 Molliex, A., Temirov, J., Lee, J., Coughlin, M., Kanagaraj, A. P., Kim, H. J., Mittag, T. & Taylor, J. P. (2015). Phase separation by low complexity domains promotes stress granule assembly and drives pathological fibrillization. Cell, 163(1), 123–133. https://doi.org/10.1016/j.cell.2015.09.015 Montgomery, T. A., Rim, Y.-S., Zhang, C., Dowen, R. H., Phillips, C. M., Fischer, S. E. J. & Ruvkun, G. (2012). PIWI associated siRNAs and piRNAs specifically require the Caenorhabditis elegans HEN1 ortholog henn-1. PLoS Genetics, 8(4), e1002616. https://doi.org/10.1371/journal.pgen.1002616 156 Nguyen, D. A. H. & Phillips, C. M. (2021). Arginine methylation promotes siRNA-binding specificity for a spermatogenesis-specific isoform of the Argonaute protein CSR-1. Nature Communications, 12(1), 4212. https://doi.org/10.1038/s41467-021-24526-6 Nishida, K. M., Okada, T. N., Kawamura, T., Mituyama, T., Kawamura, Y., Inagaki, S., Huang, H., Chen, D., Kodama, T., Siomi, H. & Siomi, M. C. (2009). Functional involvement of Tudor and dPRMT5 in the piRNA processing pathway in Drosophila germlines. The EMBO Journal, 28(24), 3820–3831. https://doi.org/10.1038/emboj.2009.365 Ortiz, M. A., Noble, D., Sorokin, E. P. & Kimble, J. (2014). A new dataset of spermatogenic vs. oogenic transcriptomes in the nematode Caenorhabditis elegans. G3 (Bethesda, Md.), 4(9), 1765–1772. https://doi.org/10.1534/g3.114.012351 Paix, A., Folkmann, A., Rasoloson, D. & Seydoux, G. (2015). High Efficiency, Homology-Directed Genome Editing in Caenorhabditis elegans Using CRISPR-Cas9 Ribonucleoprotein Complexes. Genetics, 201(1), 47–54. https://doi.org/10.1534/genetics.115.179382 Pak, J. & Fire, A. (2007). Distinct populations of primary and secondary effectors during RNAi in C. elegans. Science, 315(5809), 241–244. https://doi.org/10.1126/science.1132839 Panier, S. & Boulton, S. J. (2014). Double-strand break repair: 53BP1 comes into focus. Nature Reviews Molecular Cell Biology, 15(1), 7–18. https://doi.org/10.1038/nrm3719 Pavelec, D. M., Lachowiec, J., Duchaine, T. F., Smith, H. E. & Kennedy, S. (2009). Requirement for the ERI/DICER complex in endogenous RNA interference and sperm development in Caenorhabditis elegans. Genetics, 183(4), 1283–1295. https://doi.org/10.1534/genetics.109.108134 Pek, J. W., Patil, V. S. & Kai, T. (2012). piRNA pathway and the potential processing site, the nuage, in the Drosophila germline. Development, Growth & Differentiation, 54(1), 66–77. https://doi.org/10.1111/j.1440-169x.2011.01316.x Peng, J. & Gygi, S. P. (2001). Proteomics: the move to mixtures. Journal of Mass Spectrometry, 36(10), 1083–1091. https://doi.org/10.1002/jms.229 Phillips, C. M., Brown, K. C., Montgomery, B. E., Ruvkun, G. & Montgomery, T. A. (2015). piRNAs and piRNA-Dependent siRNAs Protect Conserved and Essential C. elegans Genes from Misrouting into the RNAi Pathway. Developmental Cell, 34(4), 457–465. https://doi.org/10.1016/j.devcel.2015.07.009 157 Phillips, C. M., McDonald, K. L. & Dernburg, A. F. (2009). Meiosis, Volume 2, Cytological Methods. Methods in Molecular Biology, 558, 171–195. https://doi.org/10.1007/978-1-60761-103-5_11 Phillips, C. M., Montgomery, B. E., Breen, P. C., Roovers, E. F., Rim, Y.-S., Ohsumi, T. K., Newman, M. A., Wolfswinkel, J. C. van, Ketting, R. F., Ruvkun, G. & Montgomery, T. A. (2014). MUT-14 and SMUT-1 DEAD box RNA helicases have overlapping roles in germline RNAi and endogenous siRNA formation. Current Biology : CB, 24(8), 839–844. https://doi.org/10.1016/j.cub.2014.02.060 Phillips, C. M., Montgomery, T. A., Breen, P. C. & Ruvkun, G. (2012). MUT-16 promotes formation of perinuclear mutator foci required for RNA silencing in the C. elegans germline. Genes & Development, 26(13), 1433–1444. https://doi.org/10.1101/gad.193904.112 Pitt, J. N., Schisa, J. A. & Priess, J. R. (2000). P Granules in the Germ Cells of Caenorhabditis elegans Adults Are Associated with Clusters of Nuclear Pores and Contain RNA. Developmental Biology, 219(2), 315–333. https://doi.org/10.1006/dbio.2000.9607 Placentino, M., Domingues, A. M. J., Schreier, J., Dietz, S., Hellmann, S., Albuquerque, B. F., Butter, F. & Ketting, R. F. (2021). Intrinsically disordered protein PID‐2 modulates Z granules and is required for heritable piRNA‐induced silencing in the Caenorhabditis elegans embryo. The EMBO Journal, 40(3), e105280. https://doi.org/10.15252/embj.2020105280 Quarato, P., Singh, M., Cornes, E., Li, B., Bourdon, L., Mueller, F., Didier, C. & Cecere, G. (2021). Germline inherited small RNAs facilitate the clearance of untranslated maternal mRNAs in C. elegans embryos. Nature Communications, 12(1), 1441–14. https://doi.org/10.1038/s41467-021-21691-6 Reinke, V., Gil, I. S., Ward, S. & Kazmer, K. (2004). Genome-wide germline-enriched and sex-biased expression profiles in Caenorhabditis elegans. Development, 131(2), 311–323. https://doi.org/10.1242/dev.00914 Reuter, M., Chuma, S., Tanaka, T., Franz, T., Stark, A. & Pillai, R. S. (2009). Loss of the Mili-interacting Tudor domain–containing protein-1 activates transposons and alters the Mili-associated small RNA profile. Nature Structural & Molecular Biology, 16(6), 639–646. https://doi.org/10.1038/nsmb.1615 Riddle, D. L., Blumenthal, T., Meyer, B. J. & Priess, J. R. (1997). C. elegans II. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press. Rocheleau, C. E., Cullison, K., Huang, K., Bernstein, Y., Spilker, A. C. & Sundaram, M. V. (2008). The Caenorhabditis elegans ekl (Enhancer of ksr-1 Lethality) Genes 158 Include Putative Components of a Germline Small RNA Pathway. Genetics, 178(3), 1431–1443. https://doi.org/10.1534/genetics.107.084608 Rodrigues, R. J. C., Domingues, A. M. de J., Hellmann, S., Dietz, S., Albuquerque, B. F. M. de, Renz, C., Ulrich, H. D., Sarkies, P., Butter, F. & Ketting, R. F. (2019). PETISCO is a novel protein complex required for 21U RNA biogenesis and embryonic viability. Genes & Development, 33(13–14), 857–870. https://doi.org/10.1101/gad.322446.118 Ruby, J. G., Jan, C., Player, C., Axtell, M. J., Lee, W., Nusbaum, C., Ge, H. & Bartel, D. P. (2006). Large-scale sequencing reveals 21U-RNAs and additional microRNAs and endogenous siRNAs in C. elegans. Cell, 127(6), 1193–1207. https://doi.org/10.1016/j.cell.2006.10.040 Saha, S., Weber, C. A., Nousch, M., Adame-Arana, O., Hoege, C., Hein, M. Y., Osborne-Nishimura, E., Mahamid, J., Jahnel, M., Jawerth, L., Pozniakovski, A., Eckmann, C. R., Jülicher, F. & Hyman, A. A. (2016). Polar Positioning of Phase- Separated Liquid Compartments in Cells Regulated by an mRNA Competition Mechanism. Cell, 166(6), 1572-1584.e16. https://doi.org/10.1016/j.cell.2016.08.006 Saito, K., Ishizu, H., Komai, M., Kotani, H., Kawamura, Y., Nishida, K. M., Siomi, H. & Siomi, M. C. (2010). Roles for the Yb body components Armitage and Yb in primary piRNA biogenesis in Drosophila. Genes & Development, 24(22), 2493–2498. https://doi.org/10.1101/gad.1989510 Sasaki, T., Shiohama, A., Minoshima, S. & Shimizu, N. (2003). Identification of eight members of the Argonaute family in the human genome☆. Genomics, 82(3), 323– 330. https://doi.org/10.1016/s0888-7543(03)00129-0 Schisa, J. A., Pitt, J. N. & Priess, J. R. (2001). Analysis of RNA associated with P granules in germ cells of C. elegans adults. Development, 128(8), 1287–1298. https://doi.org/10.1242/dev.128.8.1287 Schreier, J., Dietz, S., Domingues, A. M. de J., Seistrup, A.-S., Nguyen, D. A. H., Gleason, E. J., Ling, H., L’Hernault, S. W., Phillips, C. M., Butter, F. & Ketting, R. F. (2020). A membrane-associated condensate drives paternal epigenetic inheritance in C. elegans. BioRxiv, 2020.12.10.417311. https://doi.org/10.1101/2020.12.10.417311 Seth, M., Shirayama, M., Gu, W., Ishidate, T., Conte, D. & Mello, C. C. (2013). The C. elegans CSR-1 argonaute pathway counteracts epigenetic silencing to promote germline gene expression. Developmental Cell, 27(6), 656–663. https://doi.org/10.1016/j.devcel.2013.11.014 Seydoux, G & Fire, A. (1994). Soma-germline asymmetry in the distributions of embryonic RNAs in Caenorhabditis elegans. Development, 120(10), 2823–2834. 159 Seydoux, Geraldine, Mello, C. C., Pettitt, J., Wood, W. B., Priess, J. R. & Fire, A. (1996). Repression of gene expression in the embryonic germ lineage of C. elegans. Nature, 382(6593), 713–716. https://doi.org/10.1038/382713a0 Shakes, D. C., Wu, J., Sadler, P. L., LaPrade, K., Moore, L. L., Noritake, A. & Chu, D. S. (2009). Spermatogenesis-specific features of the meiotic program in Caenorhabditis elegans. PLoS Genetics, 5(8), e1000611. https://doi.org/10.1371/journal.pgen.1000611 She, X., Xu, X., Fedotov, A., Kelly, W. G. & Maine, E. M. (2009). Regulation of heterochromatin assembly on unpaired chromosomes during Caenorhabditis elegans meiosis by components of a small RNA-mediated pathway. PLoS Genetics, 5(8), e1000624. https://doi.org/10.1371/journal.pgen.1000624 Sheth, U., Pitt, J., Dennis, S. & Priess, J. R. (2010). Perinuclear P granules are the principal sites of mRNA export in adult C. elegans germ cells. Development, 137(8), 1305–1314. https://doi.org/10.1242/dev.044255 Shevchenko, A., Wilm, M., Vorm, O. & Mann, M. (1996). Mass Spectrometric Sequencing of Proteins from Silver-Stained Polyacrylamide Gels. Analytical Chemistry, 68(5), 850–858. https://doi.org/10.1021/ac950914h Shirayama, M., Seth, M., Lee, H.-C., Gu, W., Ishidate, T., Jr., D. C. & Mello, C. C. (2012). piRNAs Initiate an Epigenetic Memory of Nonself RNA in the C. elegans Germline. Cell, 150(1), 65–77. https://doi.org/10.1016/j.cell.2012.06.015 Shukla, A., Yan, J., Pagano, D. J., Dodson, A. E., Fei, Y., Gorham, J., Seidman, J. G., Wickens, M. & Kennedy, S. (2020). poly(UG)-tailed RNAs in genome protection and epigenetic inheritance. Nature, 582(7811), 283–288. https://doi.org/10.1038/s41586- 020-2323-8 Sievers, F., Wilm, A., Dineen, D., Gibson, T. J., Karplus, K., Li, W., Lopez, R., McWilliam, H., Remmert, M., Söding, J., Thompson, J. D. & Higgins, D. G. (2011). Fast, scalable generation of high‐quality protein multiple sequence alignments using Clustal Omega. Molecular Systems Biology, 7(1), 539. https://doi.org/10.1038/msb.2011.75 Sijen, T., Fleenor, J., Simmer, F., Thijssen, K. L., Parrish, S., Timmons, L., Plasterk, R. H. A. & Fire, A. (2001). On the Role of RNA Amplification in dsRNA-Triggered Gene Silencing. Cell, 107(4), 465–476. https://doi.org/10.1016/s0092-8674(01)00576-1 Sijen, T., Steiner, F. A., Thijssen, K. L. & Plasterk, R. H. A. (2007). Secondary siRNAs result from unprimed RNA synthesis and form a distinct class. Science, 315(5809), 244–247. https://doi.org/10.1126/science.1136699 160 Singh, M., Cornes, E., Li, B., Quarato, P., Bourdon, L., Dingli, F., Loew, D., Proccacia, S. & Cecere, G. (2021). Translation and codon usage regulate Argonaute slicer activity to trigger small RNA biogenesis. Nature Communications, 12(1), 3492. https://doi.org/10.1038/s41467-021-23615-w Smardon, A., Spoerke, J. M., Stacey, S. C., Klein, M. E., Mackin, N. & Maine, E. M. (2000). EGO-1 is related to RNA-directed RNA polymerase and functions in germ- line development and RNA interference in C. elegans. Current Biology, 10(4), 169– 178. https://doi.org/10.1016/s0960-9822(00)00323-7 Smibert, P., Yang, J.-S., Azzam, G., Liu, J.-L. & Lai, E. C. (2013). Homeostatic control of Argonaute stability by microRNA availability. Nature Structural & Molecular Biology, 20(7), 789–795. https://doi.org/10.1038/nsmb.2606 Smith, H. E. (2014). Nematode sperm motility. WormBook, 1–15. https://doi.org/10.1895/wormbook.1.68.2 Smith, J., Calidas, D., Schmidt, H., Lu, T., Rasoloson, D. & Seydoux, G. (2016). Spatial patterning of P granules by RNA-induced phase separation of the intrinsically- disordered protein MEG-3. ELife, 5, 803. https://doi.org/10.7554/elife.21337 Spike, C. A., Bader, J., Reinke, V. & Strome, S. (2008). DEPS-1 promotes P-granule assembly and RNA interference in C. elegans germ cells. Development, 135(5), 983– 993. https://doi.org/10.1242/dev.015552 Strom, A. R., Emelyanov, A. V., Mir, M., Fyodorov, D. V., Darzacq, X. & Karpen, G. H. (2017). Phase separation drives heterochromatin domain formation. Nature, 547(7662), 241–245. https://doi.org/10.1038/nature22989 Strome, S. & Wood, W. B. (1983). Generation of asymmetry and segregation of germ- line granules in early C. elegans embryos. Cell, 35(1), 15–25. https://doi.org/10.1016/0092-8674(83)90203-9 Suen, K. M., Braukmann, F., Butler, R., Bensaddek, D., Akay, A., Lin, C.-C., Milonaitytė, D., Doshi, N., Sapetschnig, A., Lamond, A., Ladbury, J. E. & Miska, E. A. (2020). DEPS-1 is required for piRNA-dependent silencing and PIWI condensate organisation in Caenorhabditis elegans. Nature Communications, 11(1), 4242. https://doi.org/10.1038/s41467-020-18089-1 Suzuki, Y., Minami, M., Suzuki, M., Abe, K., Zenno, S., Tsujimoto, M., Matsumoto, K. & Minami, Y. (2009). The Hsp90 Inhibitor Geldanamycin Abrogates Colocalization of eIF4E and eIF4E-Transporter into Stress Granules and Association of eIF4E with eIF4G*. Journal of Biological Chemistry, 284(51), 35597–35604. https://doi.org/10.1074/jbc.m109.036285 161 Tang, W., Tu, S., Lee, H.-C., Weng, Z. & Mello, C. C. (2016). The RNase PARN-1 Trims piRNA 3’ Ends to Promote Transcriptome Surveillance in C. elegans. Cell, 164(5), 974–984. https://doi.org/10.1016/j.cell.2016.02.008 Teixeira, F. K., Okuniewska, M., Malone, C. D., Coux, R.-X., Rio, D. C. & Lehmann, R. (2017). piRNA-mediated regulation of transposon alternative splicing in the soma and germ line. Nature, 552(7684), 268–272. https://doi.org/10.1038/nature25018 Thivierge, C., Makil, N., Flamand, M., Vasale, J. J., Mello, C. C., Wohlschlegel, J., Conte, D. & Duchaine, T. F. (2011). Tudor domain ERI-5 tethers an RNA-dependent RNA polymerase to DCR-1 to potentiate endo-RNAi. Nature Structural & Molecular Biology, 19(1), 90–97. https://doi.org/10.1038/nsmb.2186 Tsai, H.-Y., Chen, C.-C. G., Conte, D., Moresco, J. J., Chaves, D. A., Mitani, S., Yates, J. R., Tsai, M.-D. & Mello, C. C. (2015). A ribonuclease coordinates siRNA amplification and mRNA cleavage during RNAi. Cell, 160(3), 407–419. https://doi.org/10.1016/j.cell.2015.01.010 Uebel, C. J., Anderson, D. C., Mandarino, L. M., Manage, K. I., Aynaszyan, S. & Phillips, C. M. (2018). Distinct regions of the intrinsically disordered protein MUT-16 mediate assembly of a small RNA amplification complex and promote phase separation of Mutator foci. PLoS Genetics, 14(7), e1007542. https://doi.org/10.1371/journal.pgen.1007542 Updike, D. L., Hachey, S. J., Kreher, J. & Strome, S. (2011). P granules extend the nuclear pore complex environment in the C. elegans germ line. Journal of Cell Biology, 192(6), 939–948. https://doi.org/10.1083/jcb.201010104 Updike, D. L. & Strome, S. (2009). A genomewide RNAi screen for genes that affect the stability, distribution and function of P granules in Caenorhabditis elegans. Genetics, 183(4), 1397–1419. https://doi.org/10.1534/genetics.109.110171 Updike, D. & Strome, S. (2010). P granule assembly and function in Caenorhabditis elegans germ cells. Journal of Andrology, 31(1), 53–60. https://doi.org/10.2164/jandrol.109.008292 Vagin, V. V., Wohlschlegel, J., Qu, J., Jonsson, Z., Huang, X., Chuma, S., Girard, A., Sachidanandam, R., Hannon, G. J. & Aravin, A. A. (2009). Proteomic analysis of murine Piwi proteins reveals a role for arginine methylation in specifying interaction with Tudor family members. Genes & Development, 23(15), 1749–1762. https://doi.org/10.1101/gad.1814809 Vasale, J. J., Gu, W., Thivierge, C., Batista, P. J., Claycomb, J. M., Youngman, E. M., Duchaine, T. F., Mello, C. C. & Conte, D. (2010). Sequential rounds of RNA- dependent RNA transcription drive endogenous small-RNA biogenesis in the ERGO- 162 1/Argonaute pathway. Proceedings of the National Academy of Sciences of the United States of America, 107(8), 3582–3587. https://doi.org/10.1073/pnas.0911908107 Vought, V. E., Ohmachi, M., Lee, M.-H. & Maine, E. M. (2005). EGO-1, a putative RNA- directed RNA polymerase, promotes germline proliferation in parallel with GLP- 1/notch signaling and regulates the spatial organization of nuclear pore complexes and germline P granules in Caenorhabditis elegans. Genetics, 170(3), 1121–1132. https://doi.org/10.1534/genetics.105.042135 Wan, G., Bajaj, L., Fields, B., Dodson, A. E., Pagano, D., Fei, Y. & Kennedy, S. (2021). ZSP-1 is a Z granule surface protein required for Z granule fluidity and germline immortality in Caenorhabditis elegans. The EMBO Journal, 40(3), e105612. https://doi.org/10.15252/embj.2020105612 Wan, G., Fields, B. D., Spracklin, G., Shukla, A., Phillips, C. M. & Kennedy, S. (2018). Spatiotemporal regulation of liquid-like condensates in epigenetic inheritance. Nature, 557(7707), 679–683. https://doi.org/10.1038/s41586-018-0132-0 Wang, J., Saxe, J. P., Tanaka, T., Chuma, S. & Lin, H. (2009). Mili Interacts with Tudor Domain-Containing Protein 1 in Regulating Spermatogenesis. Current Biology, 19(8), 640–644. https://doi.org/10.1016/j.cub.2009.02.061 Ward, J. D. (2015). Rapid and Precise Engineering of the Caenorhabditis elegans Genome with Lethal Mutation Co-Conversion and Inactivation of NHEJ Repair. Genetics, 199(2), 363–377. https://doi.org/10.1534/genetics.114.172361 Ward, S & Carrel, J. S. (1979). Fertilization and sperm competition in the nematode Caenorhabditis elegans. Developmental Biology, 73(2), 304–321. https://doi.org/10.1016/0012-1606(79)90069-1 Ward, Samuel, Hogan, E. & Nelson, G. A. (1983). The initiation of spermiogenesis in the nematode Caenorhabditis elegans. Developmental Biology, 98(1), 70–79. https://doi.org/10.1016/0012-1606(83)90336-6 Webster, A., Li, S., Hur, J. K., Wachsmuth, M., Bois, J. S., Perkins, E. M., Patel, D. J. & Aravin, A. A. (2015). Aub and Ago3 Are Recruited to Nuage through Two Mechanisms to Form a Ping-Pong Complex Assembled by Krimper. Molecular Cell, 59(4), 564–575. https://doi.org/10.1016/j.molcel.2015.07.017 Wedeles, C. J., Wu, M. Z. & Claycomb, J. M. (2013). Protection of germline gene expression by the C. elegans Argonaute CSR-1. Developmental Cell, 27(6), 664– 671. https://doi.org/10.1016/j.devcel.2013.11.016 163 Wei, M.-T., Elbaum-Garfinkle, S., Holehouse, A. S., Chen, C. C.-H., Feric, M., Arnold, C. B., Priestley, R. D., Pappu, R. V. & Brangwynne, C. P. (2017). Phase behaviour of disordered proteins underlying low density and high permeability of liquid organelles. Nature Chemistry, 9(11), 1118–1125. https://doi.org/10.1038/nchem.2803 Wilson, R. C. & Doudna, J. A. (2013). Molecular mechanisms of RNA interference. Annual Review of Biophysics, 42, 217–239. https://doi.org/10.1146/annurev-biophys- 083012-130404 Xu, F., Feng, X., Chen, X., Weng, C., Yan, Q., Xu, T., Hong, M. & Guang, S. (2018). A Cytoplasmic Argonaute Protein Promotes the Inheritance of RNAi. Cell Reports, 23(8), 2482–2494. https://doi.org/10.1016/j.celrep.2018.04.072 Yen, C.-A., Ruter, D. L., Turner, C. D., Pang, S. & Curran, S. P. (2020). Loss of flavin adenine dinucleotide (FAD) impairs sperm function and male reproductive advantage in C. elegans. ELife, 9, e52899. https://doi.org/10.7554/elife.52899 Yigit, E., Batista, P. J., Bei, Y., Pang, K. M., Chen, C.-C. G., Tolia, N. H., Joshua-Tor, L., Mitani, S., Simard, M. J. & Mello, C. C. (2006). Analysis of the C. elegans Argonaute family reveals that distinct Argonautes act sequentially during RNAi. Cell, 127(4), 747–757. https://doi.org/10.1016/j.cell.2006.09.033 Youngman, E. M. & Claycomb, J. M. (2014). From early lessons to new frontiers: the worm as a treasure trove of small RNA biology. Frontiers in Genetics, 5, 416. https://doi.org/10.3389/fgene.2014.00416 Zhang, C., Montgomery, T. A., Gabel, H. W., Fischer, S. E. J., Phillips, C. M., Fahlgren, N., Sullivan, C. M., Carrington, J. C. & Ruvkun, G. (2011). mut-16 and other mutator class genes modulate 22G and 26G siRNA pathways in Caenorhabditis elegans. Proceedings of the National Academy of Sciences, 108(4), 1201–1208. https://doi.org/10.1073/pnas.1018695108 Zhang, L., Ward, J. D., Cheng, Z. & Dernburg, A. F. (2015). The auxin-inducible degradation (AID) system enables versatile conditional protein depletion in C. elegans. Development, 142(24), 4374–4384. https://doi.org/10.1242/dev.129635

Abstract (if available)

Abstract Small RNAs are at the core of the extremely well-conserved gene silencing mechanism called RNA interference, or RNAi, and exist in great abundance to regulate many cellular processes, such as gene expression, chromosome segregation, transposon suppression, and viral defense. They typically range from 18–30 nucleotides and couple with a protein co-factor called Argonaute to downregulate complementary mRNAs. C. elegans has ~27 Argonaute proteins, working concertedly in different pathways to ensure proper gene expression and genome integrity. Deregulation of these Argonaute effectors severely compromises the organism’s fitness, as it allows for de-repression of harmful genetic elements or inappropriate chromosome segregation, etc. While some Argonaute proteins have been extensively studied, the general requirements for Argonaute protein localization remain elusive. Furthermore, the sorting mechanism by which Argonaute proteins employ to correctly identify their correct binding partner has not been investigated. This thesis set out to address these gaps in the field, by examining a series of Argonaute proteins and characterizing their localizations, small RNA binding capacity, and effector functions in the C. elegans germline.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

SIMR-1 facilitates robust silencing of piRNA target loci in the C. elegans germline

PDF

The organization of small RNA pathways within C. elegans germ granules: mutator foci formation, regulation, and interaction

PDF

Genetic basis of diet-dependent responses across the lifespan in Caenorhabditis elegans

PDF

Dissecting novel roles for MAFR-1 in reproduction and metabolic homeostasis

PDF

Swimming exercise reduces native ⍺-synuclein protein species in a transgenic C. elegans model of Parkinson’s disease and confers downstream protective effects independent of food restriction…

PDF

Genetics and the environment: evaluating the role of noncoding RNA in autism spectrum disorder

PDF

C. elegans topoisomerase II regulates chromatin architecture and DNA damage for germline genome activation

PDF

Elucidating the mechanism behind the negative outcomes associated with hyperactive SKN-1

PDF

Identification of factors that underly genome-wide transcriptional repression during the development of the Caenorhabditis elegans germline.

PDF

Defining novel molecular mechanisms to restore balance of a stress response homeostat across spatial and temporal boundaries

PDF

Discovery of mature microRNA sequences within the protein- coding regions of global HIV-1 genomes: Predictions of novel mechanisms for viral infection and pathogenicity

PDF

Phenotypic and multi-omic characterization of novel C. elegans models of Alzheimer's disease

PDF

The role of mitochondria in the male reproductive capacity of Caenorhabditis elegans

PDF

Characterization of genetic and physiological responses to environmental stress in Caenorhabditis elegans across the lifespan

PDF

Placenta growth factor-miRNAs-lncRNAs axis in the regulation of ET-1 gene involved in pulmonary hypertension in sickle cell disease

PDF

Characterizing the physiological roles and regulatory mechanisms of Maf1

PDF

The role of miRNA and its regulation in pulmonary hypertension in sickle cell disease

PDF

Dps contributes to typical growth, survival, and genome organization in E. coli

PDF

Mechanisms for regulating the expression of the TATA-binding protein

PDF

The role of microRNAs in cancer

Asset Metadata

Creator Nguyen, Dieu An Hoang (author)

Core Title Regulation of Caenorhabditis elegans small RNA pathways: an examination of Argonaute protein RNA binding and post-translational modifications in C. elegans germline

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Molecular Biology

Degree Conferral Date 2021-12

Publication Date 12/02/2021

Defense Date 09/22/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag C. elegans,gene regulation,germ granules,germline,OAI-PMH Harvest,RNAi,siRNA,small RNAs

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Phillips, Carolyn Marie (committee chair), Benayoun, Berenice (committee member), Curran, Sean (committee member), Forsburg, Susan (committee member), Michael, Matthew (committee member)

Creator Email annednguyen@gmail.com,dieuanng@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC17789613

Unique identifier UC17789613

Legacy Identifier etd-NguyenDieu-10268

Document Type Dissertation

Format application/pdf (imt)

Rights Nguyen, Dieu An Hoang

Type texts

Source 20211206-wayne-usctheses-batch-901-nissen (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu