ABSTRACT Title of Dissertation: SYSTEMIC AND TRANSGENERATIONAL REGULATION OF GENE EXPRESSION BY SMALL RNAS IN C. ELEGANS Pravrutha Raman, Doctor of Philosophy, 2019 Dissertation directed by: Associate Professor, Antony M. Jose, Department of Cell Biology and Molecular Genetics Development of an organism requires information contained minimally within a single cell. This information is inherited in two forms- the genome sequence and regulatory molecules. Little is understood about the types of the regulatory molecules inherited or the impact of parental experiences on them. However, environmental stimuli can alter gene expression without changing DNA sequence and these changes can be inherited suggesting heritable regulatory molecules are influenced by parental experience. Such changes could require communication of regulatory information between cells within an animal (systemic regulation) and across generations via germ cells (transgenerational regulation). Double- stranded RNA (dsRNA) introduced to an animal can silence a gene of matching sequence within that animal and this silencing can persist in progeny suggesting that RNA has the potential to transfer gene-specific regulatory information. Using RNA silencing in C. elegans, we identify conditions that facilitate systemic and transgenerational regulation of gene expression. Previous work suggested that two forms of dsRNA, short and long, could move between somatic cells to cause systemic silencing. However, we show that the movement of short dsRNA is not an obligatory feature of systemic silencing and that long dsRNA introduced by feeding likely enters every cell to cause silencing. Silencing by dsRNA can also be communicated to the germ cells, however this does not guarantee persistence of silencing in descendants. Even the same target sequence expressed from different genetic contexts shows varying susceptibility to transgenerational silencing. Most tested genes recover from silencing in a few generations suggestive of mechanisms that repair changes induced in ancestors. We characterize a unique gene that is exceptionally susceptible to transgenerational silencing that lasts for >200 generations and find that non-genomic signals mediate its expression pattern in every generation. A forward genetic screen to isolate mutants exhibiting re-activation of gene expression (Rage) after many generations of silencing revealed additional defects indicative of endogenous processes that utilize transgenerational silencing mechanisms. We speculate that homeostatic mechanisms that prevent or preserve induced changes maintain form and function across generations in living systems. SYSTEMIC AND TRANSGENERATIONAL REGULATION OF GENE EXPRESSION BY SMALL RNAS IN C. ELEGANS by Pravrutha Raman Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2019 Advisory Committee: Associate Professor Antony M. Jose, Chair Professor Norma Andrews Associate Professor Kan Cao Professor Eric Haag Professor Leslie Pick ? Copyright by Pravrutha Raman 2019 Dedication This dissertation is dedicated to my father and my best friend for their eternal love of biology. ii Acknowledgements My time in graduate school has been a journey of learning and introspection. Such personal and scholastic growth would have been impossible if it weren?t for the encouragement and guidance from those around me. The impact that Dr. Antony Jose has had as a mentor has been tremendous and I wouldn?t be here today without his training. Antony?s passion for science is infectious and makes working in his lab and in the field a joy. He challenged me to push myself, taught me to be comfortable with the process of learning when I was wrong and engaged constructively even when we had differences in opinion. He has taught me to think critically, ask the right questions and communicate science effectively. I have enjoyed talking to Antony about the philosophy of science and academia over the years and aspire to be a mentor and scientist like him. I would be remiss not to thank the members of the Jose lab, past and present, who tirelessly challenged me to be a better version of myself. I would particularly like to thank those who have made lab a cheerful environment as they inspired me in their own unique ways to do more. Snusha Ravikumar, Julia Marr? and Sindhuja Devanapally have been my support system every step of the way through graduate school. Snusha has been my work-wife, a comforting presence and constant support through all the ups and downs. I am grateful for her not just for her willingness to safeguard my emotional health but also her green smoothie recipes! If it weren?t for Julia I may never had rotated in the Jose lab. Julia has been a relentless cheerleader and has taught me to focus on what brings me joy. Julia and Alex Marr? have been a home away from home through the years. Sindhuja is a force to reckon with in the lab iii and working with her gave me a new perspective in thinking about the science. Her quirky personality has always been a source of amusement. I thank Soriayah Zaghab and Hai Le for the enthusiasm and delight they brought to the lab, Nathan Shugarts for his infectious laid back attitude and Mary Chey for her generosity. I thank Yun Choi for all the food binges and for introducing me to running. I am grateful to the patience of Farida Eteffa and Yixin Lin, mentoring both of whom has been an absolute pleasure. They tolerated my eccentricities and were the optimists I needed to counter my skepticism. I thank Farida for forcing on me her love of ice cream and for enduring my endless demands. Yixin Lin?s hard work and persistence during tough times has been an inspiration. I can only hope they learnt from me half as much as I have learnt from them. My family has always been a source of joy and inspiration and I would not be where I am today without their love and encouragement. I am constantly inspired by my mother?s endless creativity and my father passion for his work. My brother is my best friend and brings much needed tranquility. If it weren?t for my husband I may never have pushed myself to be better or to aim higher. He is the biggest champion of my career and my source of strength. I thank Dr. Leslie Pick, Dr. Zhongchi Liu, Dr. Kan Cao, Dr. Eric Haag and Dr. Norma Andrews for taking time off their precious schedules to be at my meetings and for constructive comments on my work. I thank Dr. Karen Carleton, Dr. Scott Juntti and Dr. Quentin Gaudry for advising me on postdoctoral applications. I am grateful to other professors in the department at UMD and members of the Baltimore worm club for their valuable feedback. iv Table of Contents Dedication ..................................................................................................................... ii Acknowledgements ...................................................................................................... iii List of Tables ............................................................................................................. viii List of Figures .............................................................................................................. ix Chapter 1: General Introduction ................................................................................... 1 1.1 Preface ................................................................................................................. 1 1.2 What is the minimal information required to make an animal? .......................... 1 1.3 Experiences of an animal can be correlated with consequences in descendants 2 1.4 Candidates that can behave as heritable signaling molecules ............................. 4 1.4.1 Microbiome .................................................................................................. 6 1.4.2 Proteins ........................................................................................................ 6 1.4.3 DNA methylation ......................................................................................... 6 1.4.4 Histone modifications .................................................................................. 7 1.4.5 Noncoding RNA .......................................................................................... 8 1.5 RNAs as a signaling molecule ............................................................................ 9 1.5.1 RNA as a therapeutic in disease ................................................................. 10 1.5.2 RNA as a pesticide ..................................................................................... 11 1.5.3 RNA as a defense system ........................................................................... 11 1.6 C. elegans is an ideal model system to study communication via RNA ........... 12 1.6.1 Methods to introduce dsRNA in C. elegans .............................................. 13 1.6.2 RNA silencing in C. elegans ...................................................................... 14 1.6.3 dsRNA as a mobile signal in C. elegans .................................................... 16 Chapter 2: Systemic silencing upon feeding RNAi can be explained by the direct entry of dsRNA into all cells in C. elegans ................................................................ 21 2.1 Preface ............................................................................................................... 21 2.2 Introduction ....................................................................................................... 21 2.3 Materials and methods ...................................................................................... 24 2.3.1 Strains and oligonucleotides ...................................................................... 24 2.3.2 Transgenesis ............................................................................................... 32 2.3.3 Plasmids ..................................................................................................... 41 2.3.4 Genome editing .......................................................................................... 42 2.3.5 Feeding RNAi and scoring associated defects ........................................... 42 2.3.6 Balancing loci ............................................................................................ 45 2.4 Results ............................................................................................................... 47 2.4.1 Expression of a repetitive transgene in a tissue can inhibit RNAi in that tissue ................................................................................................................... 47 2.4.2 Using repetitive transgenes to rescue rde-4 in one somatic tissue can support RNAi within other somatic tissues ........................................................ 54 2.4.3 Rescue of rde-4 in a somatic tissue from repetitive transgenes does not cause detectable silencing in the germline .......................................................... 56 2.4.4 Silencing can occur in rde-4(-) cells when rde-4 mosaic animals ingest dsRNA ................................................................................................................ 57 2.4.5 Spatial patterns of silencing vary with the promoters that drive tissue- specific rescue from repetitive transgenes .......................................................... 60 v 2.4.6 Using repetitive transgenes to express RDE-4 in one tissue and SID-1 in another tissue fails to provide support for the movement of short dsRNAs between cells ....................................................................................................... 64 2.4.7 Single-copy transgenes can restrict RDE-4 activity to specific somatic tissues. ................................................................................................................. 69 2.5 Discussion ......................................................................................................... 70 2.5.1 Challenges and limitations in interpreting experiments that use repetitive transgenes. ........................................................................................................... 71 2.5.2 Efficiency of RNAi could be regulated by expression from repetitive DNA ............................................................................................................................. 72 2.5.3 Silencing by feeding RNAi can be restricted to a tissue with RDE-4 ....... 72 Chapter 3: Epigenetic recovery typically enables rescue from induced transgenerational silencing. ......................................................................................... 75 3.1 Preface ............................................................................................................... 75 3.2 Introduction ....................................................................................................... 75 3.3 Materials and methods ...................................................................................... 78 3.3.1 Short hand used for referring to some transgenes ...................................... 78 3.3.2 Strains and oligonucleotides ...................................................................... 79 3.3.3 Transgenesis ............................................................................................... 87 3.3.4 Genome editing .......................................................................................... 92 3.3.5 Feeding RNAi and scoring associated defects ........................................... 97 3.3.6 Expression of dsRNA ................................................................................ 99 3.3.7 Quantitative RT-PCR ................................................................................. 99 3.3.8 Chromatin Immunoprecipitation-qPCR ................................................... 100 3.3.9 Forward genetic screen to identify mutants with Re Activation of Gene Expression (RAGE) .......................................................................................... 102 3.3.10 Complementation Testing ...................................................................... 103 3.3.11 Microscopy: ........................................................................................... 103 3.3.12 Statistical Analysis: ................................................................................ 104 3.4 Results ............................................................................................................. 105 3.4.1 Transgenerational silencing is uncommon and variable even for the same target sequence .................................................................................................. 105 3.4.2 A resistant gene can recover even after a few generations of silencing. . 110 3.4.3 Transgenerational silencing can be enhanced in somatic cells ................ 111 3.4.4 The dynamics of silencing at an exceptionally susceptible gene ............. 112 3.4.5 T can be silenced simply by mating. ........................................................ 114 3.4.6 Mechanisms of mating-induced silencing ............................................... 116 3.4.7 Non-DNA signals correlated with T can modify gene expression states . 120 3.4.8 Mutants with Re Activation of Gene Expression (Rage) could provide insights into endogenous mechanisms of transgenerational silencing .............. 128 3.5 Discussion ....................................................................................................... 132 3.5.1 Epigenetic recovery prevents the persistence of changes across generations. ........................................................................................................................... 132 3.5.2 Somatic cells can be made susceptible to transgenerational silencing .... 134 3.5.3 Intergenerational versus transgenerational silencing ............................... 135 3.5.4 What makes a gene susceptible to transgenerational silencing? .............. 137 vi 3.5.5 Mutants exhibiting re-activation of gene expression (Rage) can provide insights into the mechanisms of transgenerational silencing. ........................... 139 Chapter 4: General discussion .................................................................................. 142 4.1 Systemic gene regulation by small RNAs ...................................................... 142 4.1.1 How is specificity of silencing mechanisms determined? ....................... 143 4.1.2 small RNAs as a molecular signal for systemic regulation ..................... 145 4.1.3 RNAs as a communication signal between species ................................. 146 4.2 Gene regulation by small RNAs across generations ....................................... 148 4.2.1 small RNAs as a molecular signal for intergenerational regulation ........ 148 4.2.2 Epigenetic recovery prevents the rampant inheritance of silencing ........ 150 4.2.3 small RNAs as a molecular signal for transgenerational regulation ........ 151 Chapter 5: Future directions ...................................................................................... 154 5.1 Preface ............................................................................................................. 154 5.2 Introduction ..................................................................................................... 154 5.3 Materials and methods .................................................................................... 154 5.3.1 Strains ...................................................................................................... 154 5.3.2 P0 & F1 feeding RNAi or F1 only RNAi ................................................ 156 5.3.3 Semiquantitative RT-PCR ....................................................................... 157 5.4 Results ............................................................................................................. 157 5.4.1 The precise recruitment of RNAi factors in a cell is poorly understood . 157 5.4.2 How is spread of silencing across a gene coordinated? ........................... 161 5.4.3 What is the route taken by ingested dsRNA in C. elegans ...................... 163 5.4.4 Identifying endogenous mobile RNAs. .................................................... 166 5.4.5 Germline genes are susceptible to transgenerational silencing ................ 167 5.4.6 Deciphering the cell code of an animal .................................................... 170 Bibliography ............................................................................................................. 173 vii List of Tables Table 2- 1: Strains used. .............................................................................................. 24 Table 2- 2: Oligonucleotides used (5? to 3?, IDT). ..................................................... 28 Table 2- 3: Scoring of gene-specific silencing. .......................................................... 44 Table 3- 1: Strains used. .............................................................................................. 79 Table 3- 2: Oligonucleotides used (5? to 3?, IDT). ..................................................... 84 Table 3- 3: Scoring of gene-specific silencing. .......................................................... 98 Table 5- 1: Strains used. ............................................................................................ 155 Table 5- 2:Oligonucleotides used (5? to 3?, IDT). .................................................... 156 viii List of Figures Fig. 1- 1: Inheritance of changes caused by environmental stimuli could require communication between somatic cells and the germline. ..................................... 5 Fig. 1- 2: A general schematic for the RNA silencing pathway in C. elegans. .......... 15 Fig. 1- 3: Model of C. elegans anatomy. .................................................................... 17 Fig. 2- 1: Two forms of dsRNA can move between cells to cause silencing. ............ 23 Fig. 2- 2: Silencing by feeding RNAi of some genes is reduced in tissues expressing RDE-4 or RDE-1. ................................................................................................ 48 Fig. 2- 3: Expression of any repetitive transgene in a tissue can inhibit silencing by ingested dsRNA. ................................................................................................. 50 Fig. 2- 4: Silencing by feeding RNAi of some genes is reduced in tissues expressing a repetitive transgene of unrelated sequence. ........................................................ 51 Fig. 2- 5: Genes that show robust inhibition of silencing upon expression of an unrelated transgene require the nuclear RNAi pathway for complete silencing. 53 Fig. 2- 6: Tissue-specific rescues of RDE-4 from repetitive transgenes can enable silencing of genes in non-rescued somatic tissues. ............................................. 55 Fig. 2- 7: Tissue-specific rescues of RDE-4 from repetitive transgenes enables silencing of ubiquitous gfp in non-rescued somatic tissues but not germline. ... 57 Fig. 2- 8: rde-4(-) cells can be silenced by ingested RNAi in rde-4 mosaic animals. 59 Fig. 2- 9: Silencing of bli-1 upon feeding RNAi in wild-type animals and in animals with tissue-specific rescue of rde-4 in non-hypodermal cells results in unique patterns of blisters that differ from those in bli-1(-) animals. ............................. 61 Fig. 2- 10: Repetitive transgenes expressing RDE-4 from different promoters enable different spatial patterns of bli-1 silencing in rde-4(-) hypodermis. ................... 63 Fig. 2- 11: Expression of RDE-1 in one somatic tissue can enable silencing of Peft- 3::gfp in other mutant somatic tissues. ............................................................... 65 Fig. 2- 12: Expression of RDE-4 in the somatic tissues of parents does not typically enable feeding RNAi in rde-4(-) progeny. .......................................................... 66 Fig. 2- 13: Misexpression of RDE-4 may be sufficient to explain silencing by short dsRNA upon feeding RNAi. ............................................................................... 69 Fig. 2- 14: Ingested dsRNA probably enters every cell to cause silencing in the soma ............................................................................................................................. 70 Fig. 3- 1: Transgenerational silencing isn?t observed for all genes. ........................... 77 Fig. 3- 2: Transgenerational silencing is uncommon. ............................................... 107 Fig. 3- 3: Transgenerational silencing can be enhanced but animals still recover after a few generations. ............................................................................................. 109 Fig. 3- 4: A susceptible locus can be transgenerationally silenced for 15 generations. ........................................................................................................................... 113 Fig. 3- 5: Mating can induce silencing in progeny. .................................................. 115 Fig. 3- 6: Mating-induced silencing engages the small RNA silencing pathway. .... 119 Fig. 3- 7: The expression of T??? or Tcherry can be altered by heritable DNA- independent silencing signal. ............................................................................ 124 Fig. 3- 8: PRG-1 is maternally required while HRDE-1 may be zygotically required. ........................................................................................................................... 126 ix Fig. 3- 9: Mutants exhibiting Re-activation of gene expression (Rage) displayed additional physical defects. ............................................................................... 129 Fig. 3- 10: Characterization of rage mutants by feeding RNAi. ............................... 131 Fig. 3- 11: Genes expressing the same target can show varying susceptibility to transgenerational silencing. ............................................................................... 137 Fig. 3- 12: RAGE may regulate expression of endogenous genes. ......................... 140 Fig. 5- 1: Loss of the exonuclease eri-1 or transgenic expression of RDE-4 in the . 160 hypodermis can bypass the requirement for NRDE-3 to silence bli-1 in response .. 160 to ingested dsRNA .................................................................................................... 160 Fig. 5- 2: Expression of RDE-4 in the germline can enable silencing in rde-4(-) soma. ........................................................................................................................... 164 x Chapter 1: General Introduction 1.1 Preface Thoughts in this dissertation were influenced by key reviews (cited in the text) in addition to discussions with Antony and members of the Jose lab 1.2 What is the minimal information required to make an animal? Life relies on a cyclical perpetuation with information being inherited from parent to progeny in every generation (reviewed in (1)). In many organisms, it begins minimally as a single cell zygote in each generation and stereotyped development of an organism from this single-cell relies on reproducible gene expression patterns executed with precision. This suggests that all information to make an organism must be contained minimally within this one cell zygote. Before the nature of any heritable molecules was defined, proteins and DNA were thought to be inherited (reviewed in (2)). Prolific work on DNA (3,4) including the theory of the replication of DNA (5-7) and the discovery of mutations (8-10) amongst others, led to emphasis on DNA as the molecule of heredity. However, indications of other molecules that also carry heritable information have persisted for many years, some of the earliest examples being paramutation in maize (11) or cortical inheritance in ciliates (12). Numerous examples of inheritance without changes to DNA are continuing to be uncovered providing further support for additional molecules that transmit information across generations (see section 1.4). Together, the information within a zygote recently described as its cell code (1) consists of two stores of information- the linear genomic sequence (referred to as 1 genetic information in this dissertation) and the arrangement of regulatory molecules (referred to as epigenetic information in this dissertation) that together control gene expression. During life the organism goes through many developmental changes including repeated replication of genetic information, recycling of epigenetic information and replication and division of cellular components. In some organisms there is evidence of epigenetic reprogramming that results in widespread erasure of modifications associated with the genome (13,14). Despite all these changes, animals are able to reproduce to give fairly similar organisms in the next generation. The remarkable reproducibility of gene expression patterns between successive generations argues for the inheritance of a largely similar arrangement of molecules (RNA, DNA, sugars, lipids etc.) in every generation. However, experiences during the lifetime of an animal can result in consequences that are inherited into progeny independent of changed to the DNA (see section 1.3). This ability to tolerate changes in epigenetic information may promote adaptation and evolution. 1.3 Experiences of an animal can be correlated with consequences in descendants The question of whether experiences during the lifetime of an animal can affect the animal and consequently its descendants has been of interest for more than a century. Jean-Baptiste Lamarck in the early 1800s put forth the idea that experiences of an organism during its lifetime are passed on to offspring (15). Years later, Charles Darwin hypothesized that one way to explain how acquired traits are inherited could be through the transfer of packets of information or ?gemmules? from the soma to the germline (16). However, using meticulous multi-generational 2 experiments August Weismann postulated that experiences that affect an organism?s somatic cells do not affect the germplasm and consequently are not inherited into progeny (17). In most model systems during early development, the germline is set aside from the soma (reviewed in (18)), resulting in Weismann?s theory being interpreted as a barrier between the soma and germ cells (germline) that prevented communication of information from the soma to the germline and was referred to as the Weismann barrier. However, experiences of an organism can be correlated with effects in the animal and in descendants without obvious changes to the DNA sequence. This suggests that molecules other than DNA that are inherited can affect gene expression in an animal and its progeny. Recent evidence emerging in many different animals suggests that environmental stimuli may result in physiological changes in an organism and that these changes can be inherited. Longitudinal epidemiological studies in humans have found correlations between environment and physiological changes in descendants. For example, the Swedish ?verkalix study found that quantities of food supply in parents correlated with altered mortality risk in grandchildren (19). However, studies such as these rely on census information, which may be incomplete. In addition, using information like availability of food as a substitute for actual food consumption or nutrition can be misleading as an estimate of diet. Hence, inferences from these studies are of limited scope and do not provide any mechanistic insights. More reliable well-controlled studies have been done using animal model systems that span multiple generations and may give mechanistic insights. In mammals changes in diet or exposure to stimulants have been correlated with 3 physiological changes in the animals and in their progeny. For instance, in mice, low- protein diet in fathers resulted in increase in expression of cholesterol and lipid metabolism genes in offspring, accompanied by up regulation of genes responsible for lipid, steroid, and cholesterol biosynthesis (20). Obesity or changes in diet in male mice were correlated with changes in sperm small RNAs in animals and metabolic disorders in their progeny (21-24). Exposing stressed mice to an olfactory stimulant resulted in gene expression changes in the odor response pathway correlated with anatomical and behavioral defects in the animal and its grandprogeny (25). Such observations suggest that environmental effects may cause changes in the soma and information from the soma may the communicated to the germ cells and subsequently inherited into progeny. This contradicts the Weismann barrier and is indicative of molecules that can act as signals to transmit information between the soma and the germline. However we cannot rule out the possibility that these effects could also be a result of direct changes to the germ cells by environmental stimuli. 1.4 Candidates that can behave as heritable signaling molecules Changes due to exposure to environmental stimuli could be a result of changes to gene expression in cells in the animal. Such changes can result from direct exposure of a cell to the stimulus or the communication of change from the exposed cell to other cells in the animal. The inheritance of changes requires the direct exposure of the germline to the stimulus or communication of information from the soma to the germline (Fig. 1-1). In both cases, molecular signals that can alter gene expression likely carry information between cells (systemic) or across generations (transgenerational). 4 Fig. 1- 1: Inheritance of changes caused by environmental stimuli could require communication between somatic cells and the germline. Somatic cells exposed to environmental stimuli can be affected and display altered phenotypes. This information may be communicated to other somatic cells or the germline. Any inherited effects must be communicated to the germline to be transmitted into the single-cell zygote. In subsequent generations, observed defects may be due to changes in somatic cells. Molecules that carry such information between somatic cells or to the germline would need to be able to cross cell boundaries to enter a cell. In order to be inherited, molecules that are transmitted into progeny through the germline would need to be able to resist epigenetic reprogramming mechanisms in the embryo and persist through cell divisions to retain its effects through development (13,14). In either case such signaling molecules must be able to specifically regulate gene expression changes in its target cells without causing rampant uncontrolled changes to that cell. Below I detail some means that parental effects can be transmitted to progeny in addition to DNA sequence (reviewed in (2,26)). 5 1.4.1 Microbiome Changes to parental microbiome can be inherited into progeny via breast milk, the birth canal or placental transfer in humans (27,28). Wild populations of mice from that were inbred maintained a largely similar gut microbiota for 10 tested generations (29). A low fiber diet resulted in loss of a specific populations of bacteria in mice over generations (30). In flies, changes in temperatures were correlated with changes in microbiota that could be inherited (31). The mechanisms of this transfer or the resulting effects of these changes in the animal remain unclear. However, better tools are required to study the inheritance of microbiota in a controlled manner without risking external transmission. 1.4.2 Proteins Some of the earliest evidence for proteins as heritable molecules (reviewed in(32)) come from the discovery of prions and prion-like proteins (33-35) that have now been discovered in many organisms. These proteins have the ability to self replicate some of their conformations enabling cytoplasmic inheritance across generations. While they seem unaffected by epigenetic reprogramming, most proteins are turned over during development and can be diluted or lost. However, mechanisms that facilitate some proteins? escape from such turnover remain unclear. 1.4.3 DNA methylation Methyl groups added to cytosine or adenine in the DNA sequence of a gene is correlated with altered expression of that gene. Such methylated regions are thought to enhance the formation of heterochromatin and result in transcriptional repression 6 (36). In mammalian systems, environmental changes have been correlated with altered DNA methylation in the animal and in its progeny (37,38). However given that the genome of the germline and early embryo undergoes extensive erasure of marks (epigenetic reprogramming, (13,14)), persistence of effects would suggest the presence of mechanisms that resist such erasure or the presence of alternate carriers of information that enable re-establishment of methylation later in development. 1.4.4 Histone modifications DNA within the nucleus is organized into nucleosomes, which consist of a core of histone proteins with DNA wrapped around them. Amino acids on the histones N terminal tails can be modified by addition of groups including acetyl, phosphoryl and methyl (39-41). More and more such modifications are being discovered and recent estimates (42) suggest that as many as 550 modifications have been detected! These modifications are thought to impact the accessibility of DNA to complexes that regulate processes such as transcription (41). Consistently environmental changes that result in changes in gene expression have been correlated with altered levels of histone modifications in animals and these changes have also been detected in progeny (Highlighted in section 1.3). However, similar to DNA methylation, histone modifications can also undergo reprogramming in the germline and early embryo. While some histone marks can be eliminated, others are required and their loss can result in developmental defects in progeny (43,44) suggesting that there are mechanisms that can differentiate between necessary and dispensable modification. Further studies identifying this machinery can give us insights into the cell codes that are inherited into embryos. 7 1.4.5 Noncoding RNA Noncoding RNAs can be transcribed from the genome but are not translated and instead are processed to facilitate regulation of gene expression. Noncoding RNA can bind mRNA of matching sequence to facilitate changes in mRNA levels or to direct DNA or histone modifications (45). Changes in gene expression levels by noncoding RNAs have more commonly been associated with decrease in mRNA levels or increase in repressive histone marks, both of which down regulate gene expression. However, noncoding RNAs can also increase expression of genes though little is understood about such regulation. Noncoding RNAs have been detected in many different animals, both within cells and in the extracellular space and distinct classes of RNA can be detected in gametes (45). Some classes of noncoding RNA (such as miRNAs and piRNAs) and their associated protein complexes are highly conserved (46). Environmental stimuli such as changes to diet have been correlated with changes in small RNAs in descendants in C. elegans (47) and altered RNA in sperm have been correlated with metabolic changes in progeny in mice (48,49). These studies (Also highlighted in section 1.3) suggest that exposure to environmental changes can alter RNA populations in the exposed animal and RNA or its derivatives may be inherited. RNAs can direct the deposition of DNA methylation and histone modifications that may lead to changes in gene expression. Evidence in yeast suggests that histone modifications can in turn lead to the production of RNAs (45). This suggests that RNA, DNA methylation and histone modification can alter each other making it difficult to discern which molecules initiate change, facilitate 8 downstream effects or maintain the changes. To prevent such misinterpretations, it is essential to induce change in a specific manner and track subsequent events in the animal and in its progeny. We chose to induce such change using RNA. 1.5 RNAs as a signaling molecule Apart from abundant levels of noncoding RNAs that have been found within animal cells RNAs have been found in extracellular spaces and fluids. These RNAs are excellent candidates for communication signals since they can transmit gene regulatory information in a sequence specific manner between cells and possibly to the next generation. Extracellular RNAs have been found in human blood (50), saliva (51,52), semen (53), placenta (54) and breast milk (55). The composition of RNAs is also found to be changed in disease states (56). However, it remains unclear where these RNAs are made, how they are exported or imported into other cells and what genes these RNAs are changing. The discovery of RNA interference opened up the tantalizing possibility that RNA induced changes can be transmitted between cells and across generations. In C. elegans, it was observed that introducing dsRNA into one cell led to sequence specific silencing of a target gene in that cell, in other cells in the organism as well as in progeny (57). This process of silencing has been detected in many other systems and has provided clues for how RNAs could be transmitted between cells. In plants, when unsilenced scions were grafted onto silenced rootstocks, silencing could be observed in the scions (58). Intercellular connections between cells in plants such as plasmodesmata and phloem can facilitate the transfer of signals like RNA over long distances (59). In animals, fluorescently labeled short interfering RNA (siRNA) or 9 overexpressed miRNAs were able to move between cells through gap junctions (60- 62). Similarly, mRNA could also be transported through intercellular bridges found in germ cells (63). RNAs have been detected in membrane bound vesicles isolated from cell lines, bacteria, parasites and even bodily fluids (reviewed in (64)). The presence of some but not all RNA within these vesicles suggests enrichment of RNA found in the vesicles (65,66). Lastly, miRNAs complexed with their Argonaute protein (Ago2) (50,67) or with high density lipoprotein (HDL) (68) have been found in human plasma. The latter can change between diseases and healthy individuals and retain their function in cultured cells. This suggests that RNA might also be stable when bound by certain proteins and need not be transported via membrane bound structures. Understanding the potential functions of these RNAs could reveal new modes of communication between cells and between generations, help us understand how disease states could be inherited via non-DNA inheritance and potentially reveal new biomarkers that can be used in disease diagnosis. 1.5.1 RNA as a therapeutic in disease dsRNA can be used to silence disease-associated genes in a sequence specific manner making it a promising strategy to treat diseases. However, delivery of dsRNA into the appropriate location (cell/tissue) would require it to cross cell boundaries and to be released into the cytoplasm without being degraded en route. While liquid nanoparticles encapsulating dsRNA have been an effective delivery system in rodents and non-human primates (69-71), efficiency has been very low (72). RNAs encased in fatty nanoparticles tend to collect in the liver and this is being used to reduce the 10 accumulation of misfolded transthyretin (TTR) protein in the liver of patients with hereditary transthyretin amyloidosis caused by mutations in the TTR gene (73,74). Success in clinical trials has led to the approval of the first RNAi based drug to treat hereditary transthyretin amyloidosis. However, further work is required to enable targetted delivery into other tissues. 1.5.2 RNA as a pesticide Killing animals by feeding them double-stranded RNA (dsRNA) that matches an essential gene is a powerful way to control animal pests. For example, expression of long dsRNA in potato plants was recently used to kill Colorado potato beetle that feed on these plants (75). This approach to pest control relies on the ability of many insects and parasitic nematodes to process ingested long dsRNA and use it to silence genes of matching sequence through RNA interference (RNAi) (76-79). However, the mechanisms of gene silencing by ingested dsRNA are not well understood, making it difficult to anticipate resistance mechanisms and therefore design effective dsRNA pesticides. C. elegans, a popular model system can also perform feeding RNAi similar to insects (79). The availability of resources in this model and the ease of genetic manipulation may provide insights into feeding RNAi that are applicable to pests. 1.5.3 RNAi as a defense system There is evidence in multiple animal systems that RNAi can be a defense mechanism mounted against viruses upon infection. But it is still unclear if this requires spread of RNA between cells. Silencing is observed in mammalian cell lines 11 infected with viruses defective in suppressors of RNAi (80-82). In Drosophila anti- viral immunity required the uptake of dsRNA by receptor mediated endocytosis suggesting that the spread of RNA may be required to mount an anti-viral response (83). In C. elegans, RNAi is required for protection against viral infection, however there is contradictory evidence regarding the requirement of spread for this response (84-86). Interestingly, during a viral response in mammals, the ortholog of a C. elegans dsRNA importer, SIDT2, facilitates import of viral dsRNA from endocytic vesicles into the cytosol. More work is required to establish the role for the spread of RNAi in anti-viral responses. 1.6 C. elegans is an ideal model system to study communication via RNA Some of the most compelling evidence for chromatin and small RNAs as molecular signals that regulate gene expression in distant cells or in descendants comes from studies in C. elegans. The removal of histone modifiers temporarily resulted in increased lifespan in descendants (87,88). Growth under heat stress resulted in changes in germline gene expression (89-91) accompanied by changes in small RNAs (90) in descendants. Starvation in parents was correlated with increased lifespan in great-grandprogeny and the generation of small RNAs against nutrition genes for several generations (47,92). However, in all of these studies it is unclear whether the observed effects were caused by the transfer of signals from exposed somatic cells to the germline or if the germline was directly affected. Expressing double-stranded RNA (dsRNA) in neurons resulted in gene-specific silencing in other somatic tissues (93) and in the germline (94). This silencing was completely dependent on a dsRNA specific importer suggesting that dsRNA or its derivatives can 12 move between cells to cause silencing. This suggests that dsRNA are strong candidates for molecular signals that can regulate gene expression systemically and transgenerationally. To study RNAs as candidates for molecular signals we need a system that is well characterized and has the tools to introduce RNA and C. elegans fits the bill for both (95). In addition since C. elegans is a hermaphrodite progeny have nearly genetically identical genomes making interpretations on effects from non-DNA signals more reliable. Rare males in the population can be mated with hermaphrodites and maintained to generate cross progeny making genetic manipulations easy. There are added advantages to using C. elegans as a model to study inheritance of information across generations. The generation time of C. elegans is only 3 days making studies across generations relatively quick. In addition, an impermeable barrier is made surrounding a C. elegans embryo ~14mins after fertilization (96). Parental molecules are unlikely to be able to penetrate this barrier and hence effects observed in an animal are unlikely to be clouded by parental influence after this time frame. Lastly, subsequent goals of the studies elucidated in this dissertation are to understand the cell code of an organism. The reproducible pattern of development of C. elegans from embryo to adulthood and the knowledge of its well-defined lineage will facilitate such studies. 1.6.1 Methods to introduce dsRNA in C. elegans In C. elegans, silencing by dsRNA can be achieved by introducing dsRNA to an animal in many different ways as listed below. Since the discovery of RNAi, many studies have demonstrated the robustness of these techniques and verified gene- 13 specific silencing. Some tissues in C. elegans have been noted to be more resistant to silencing such as the pharynx, neurons (97) and vulval muscles (57). 1. Injection- DsRNA can be introduced by injecting (57) an animal with RNA transcribed in vitro or by injecting readily available fluorescently labeled RNA (98,99). 2. Expression- DsRNA can be expressed in specific tissues from a transgenic array under a tissue-specific promoter (93,94,100-103). 3. Feeding - Bacteria expressing dsRNA from plasmids can be fed to worms and this results in robust silencing (79). 1.6.2 RNA silencing in C. elegans Extensive work from many labs has resulted in a comprehensive understanding of how small RNAs are processed in C. elegans. While more and more genes required for this processing are being discovered, the knowledge currently available can be summarized into a general pathway (Fig. 1-2, reviewed in (104)). Some factors such as Argonaute (Ago) proteins or RNA dependent RNA Polymerases (RdRPs) for instance are common to the processing of small RNA. However, since the worm has multiple Agos and RdRPs, silencing of different genes show differential requirements for these proteins and it is unclear how these factors are recruited (See Discussion). Long dsRNA is processed by an endonuclease DCR-1 (105) and a dsRNA binding protein (106) to generate short dsRNA or 1? siRNA (short interfering RNA). One strand of this 1? siRNA (guide strand) bound by a primary Ago (107) bind an mRNA of matching sequence. In C. elegans, this enables amplification of silencing 14 signals by the production of 2? siRNA that are made by RNA dependent RNA Polymerases (108-110) in perinuclear foci (111-113). 2? siRNA are single-stranded RNA antisense to the mRNA and are usually 22nt in length with a 5? Guanidine (G) bias. While it is unclear what happens to the mRNA after the production of secondary RNAs, more recent work has suggested that an endonuclease RDE-8 cleaves the mRNA prior to the production of secondary and untemplated Uridine (U) are added to the mRNA (114). 2? siRNAs are then bound by secondary Agos, which can act in the cytosol or nucleus. Cytoplasmic Agos and their associated 2? siRNA are thought to result in the destabilization of mRNA though little is known about how this happens (108-110,115). Alternatively, nuclear Agos bound to their siRNA are thought to mediate chromatin modification at the specific gene by binding an actively transcribing mRNA (116-121) and recruiting histone modifiers. Fig. 1- 2: A general schematic for the RNA silencing pathway in C. elegans. While C. elegans has many different non coding RNA, their processing is similar and relies on homologous proteins. 15 While exogenously introduced dsRNA are processed similarly in the soma and the germline, the worm also generates many species of non-coding small RNAs with many widespread roles endogenously (46,104). The most abundant class of small RNAs in the germline are the Piwi-associated RNAs (piRNAs). These highly conserved RNA (122,123) are thought to maintain germline integrity by silencing transposable elements and foreign sequences (124-127). Unlike the endogenously introduced dsRNA, piRNAs can be single-stranded RNA 21 nucleotide RNA with a 5?U bias or can be 26 nucleotide with a G bias (Fig. 1-2, (46)). Once in the cytoplasm, in C. elegans, piRNA bind their own specific 1? Agos (124,125,127-129) and silence genes by engaging a nuclear secondary Ago. 1.6.3 dsRNA as a mobile signal in C. elegans Multiple lines of evidence suggest that dsRNA can move between cells to cause silencing in C. elegans. Firstly, injection of dsRNA into an intestinal cell resulted of silencing in that cell but also in all other somatic cells and in subsequent progeny. However, given the anatomy of the worm, injection could have resulted in spillage of dsRNA into the body cavity providing all cells with direct access to dsRNA from the body cavity (Fig. 1-3). Secondly, the expression of dsRNA in one tissue can result in silencing of matching genes in other somatic tissues (79,93,100,101) and in the germline (94). This silencing is dependent on the dsRNA specific importer SID-1 (130) suggesting that double-stranded species of RNA can move between cells. Lastly, genetic mosaic analysis rescuing proteins required for processing dsRNA suggested that two forms of dsRNA can move between cells- the long dsRNA and 1? siRNA, but not 2? siRNA ((93), also see Section 2.2). However, 16 expression of dsRNA and mosaic analysis used multi-copy transgenes under tissue- specific promoters, which could result in misexpression in unintended tissues in the worm (131). The requirement of SID-1 for silencing by expressed dsRNA suggests that misexpression is unlikely to be responsible for the silencing in distant tissues but this was not tested for experiments with mosaics (discussed in more detail in Chapter 2). Fig. 1- 3: Model of C. elegans anatomy. All tissues in C.elegans are directly in contact with pseudoseulomic fluid in the body cavity. Systemic silencing requires that RNA be exported from the cell where they are made, stabilized in the extracellular space, imported into recipient cells and then be processed in the recipient cells for silencing. Due to their enrichment in extracellular space in many animal models, microRNAs (miRNAs) that are required for development are a candidate for endogenous RNAs that can move between cells. However, a miRNA tested for silencing in distant cells in C. elegans did not show any silencing (132). Very little is known about how dsRNA can be exported from a cell or be stabilized in the extracellular space. Tissue-specific rescues of SID-1 suggested that it was not required for export. While many genetic screens have been done to understand silencing most have only identified genes that are RNAi defective or required for import thus far (130,133-135). It is possible that genes required for these steps may be essential in C. elegans and a screen specifically looking for mutants that 17 are also germline deficient, sterile, embryonic lethal etc. might result in identification of these factors. A genetic screen in C. elegans to find mutants that are Systemic RNAi Defective (Sid) in C. elegans has revealed several genes required for the import of dsRNA (130). SID-1, a very well conserved (130,136,137) transmembrane protein required for import was selective for dsRNA (compared to dsDNA or RNA/DNA hybrids) when expressed in Drosophila S2 cells (138). Using multi-copy transgenic rescues of SID-1, it was found to localize to the membrane of all non-neuronal cells in C. elegans (130), however its mammalian homolog is found to localize to the lysosomal membrane (139). While the loss of sid-1or its homologs has no obvious defects (136,137), recent work in C. elegans is beginning to reveal an endogenous role for systemic silencing using SID-1 in C. elegans ((140,141) See Discussion). SID-2, a transmembrane protein enriched on the apical membrane of the intestine (142) is specifically required for silencing by ingested dsRNA (143) and is likely required for the uptake of dsRNA from the lumen into the intestine. While SID-2 has homologs in other closely related nematodes, differences in the structure (e.g. extracellular domain in C. briggsae) render some species incapable of silencing by ingested dsRNA (144). Expressing the C. elegans SID-2 can however rescue feeding RNAi in other species (142,144) suggesting that while other processing machinery might be intact only the ability to do feeding RNAi may have been lost. SID-5 found in the same screen is an endosome-associated protein and is thought to be required for the uptake of dsRNA from the intestinal lumen and for trafficking across the intestine into the body cavity (145). SID-3, a conserved tyrosine kinase (101) required for the 18 import of dsRNA in C. elegans is homologous to the human cdc-42-associated kinase (146) that facilitates endocytosis. The import of dsRNA has been linked to endocytic vesicles in Drosophila S2 cells (83). Interestingly, the knockdown of orthologs in C. elegans results in loss of silencing (83) and SID-3 was found in intracellular puncta (101). Together, this suggests a possible role for endocytosis in the transport of dsRNA across membranes. Lastly, work on RNAs inherited from parent to progeny in C. elegans has largely been based on RNAs made within the germline of C. elegans. Since the germline is a syncytium RNAs made within the germline need not cross membranes and could be directly deposited into oocytes or sperm in the germline. dsRNA expressed in neurons could silence the germline in a SID-1 dependent manner (94), and silencing could then be inherited for >25 generations in unexposed progeny. However, injection of fluorescently-labeled dsRNA into the extracellular space resulted in the deposition of RNA directly into the oocyte and this deposition required the receptor RME-2 that facilitates receptor mediated endocytosis of yolk proteins from the intestine in C. elegans (98,99). While SID-1 was not required for the deposition of the labeled RNA into oocytes, it was required for silencing in progeny. Together, this suggests that while SID-1 may not be required for inheritance of RNA, it is likely required for the spread of RNA in progeny or for the release of RNA into the cytoplasm. In my dissertation, I aim to understand how information can be communicated between cells and how information can be inherited across generations. By using ingested dsRNA to target specific genes, and using genetic analysis, we show that 19 long dsRNA that is fed to an animal likely enters every cell in the animal to silence genes of matching sequence. However, silencing of genes in the germline and soma rarely persist in descendants. We further characterize silencing of two genes that display distinct susceptibility to transgenerational silencing suggestive of opposing mechanisms that maintain homeostasis. Lastly, using an exemplary gene that is particularly susceptible to silencing we find non-DNA signals that mediate gene expression, identify mechanisms required to maintain silencing at this gene and perform a genetic screen to isolate mutants that exhibit Re Activation of Gene Expression (RAGE). 20 Chapter 2: Systemic silencing upon feeding RNAi can be explained by the direct entry of dsRNA into all cells in C. elegans 2.1 Preface Most of the work presented in this chapter was published with some modifications as: Raman P, Zaghab SM, Traver EC, and Jose AM (2017) The double- stranded RNA binding protein RDE-4 can act cell autonomously during feeding RNAi in C. elegans. Nucleic Acids Research 113(44):12496-501 (147). Soriayah M Zaghab performed some of the experiments to generate the data for Fig. 2-9B, E and F and all of the data for 2-9 D, 2-10A-E, 2-11C. Ed Traver performed some of the experiments to generate the data for Fig. 2-2A, 2-5D, 2-6, 2- 12. All other data were generated by me. Some worm strains were obtained from the Caenorhabditis elegans Genetic stock Center (CGC), the Hunter laboratory (Harvard University), and the Seydoux laboratory (Johns Hopkins University). The Hamza laboratory (University of Maryland) provided bacteria that express dsRNA. Julia Marr? (Jose lab) generated plasmid pJM6 and Yun Choi (Jose lab) generated the plasmid pYC13 used to make strains (See methods for details). 2.2 Introduction Silencing of genes by feeding animals long dsRNA was first demonstrated in the nematode C. elegans (79). Feeding RNAi has since been observed in other animals 21 and has been particularly useful as a pest control ((148), also see section 1.6.2 in intro and section 4.1.3 in Discussion). However, the mechanisms of gene silencing by ingested dsRNA are not well understood, making it difficult to anticipate resistance mechanisms and therefore design effective dsRNA pesticides. Due to the current intractability and lack of sufficient tools in many insect models, C. elegans still remains the best animal model for understanding this process called feeding RNAi. In C. elegans, ingested dsRNA enters the animal through the intestine and can be delivered into the fluid-filled body cavity that surrounds all internal tissues without entry into the cytosol of intestinal cells (Fig. 1-3, (100,145,149)). Entry into the cytosol of any cell requires a dsRNA-selective importer SID-1 (130) ? a conserved protein with homologs in many insects (150). Upon entry into cells, silencing by dsRNA is thought to occur through the canonical RNAi pathway (reviewed in (104), Fig. 2-1A). Long dsRNA is first bound by the dsRNA-binding protein RDE-4, which recruits the endonuclease DCR-1 to generate short dsRNAs (106,151,152). One strand of this short dsRNA duplex is used as a guide by the primary Argonaute RDE- 1 to identify mRNAs of matching sequence (106,107) and to recruit RNA-dependent RNA polymerases (RdRPs) to the mRNA. RdRPs then synthesize numerous secondary small RNAs (108-110) that are used for potent gene silencing within the cytosol by cytosolic Argonautes and/or within the nucleus by nuclear Argonautes (108-110,115,117,118,121,153). To infer whether any derivatives of long dsRNA generated within a cell also move between cells, components of the RNAi pathway were rescued in body-wall muscles using repetitive transgenes and silencing was assayed in other tissues (93) or in the next generation (154). Specifically, when RDE- 22 4 was rescued in body-wall muscles, silencing was detected in the body-wall muscle as well as in the intestine and the skin (93). However, when RDE-1 was rescued in the body-wall muscles, silencing was only detected in the body-wall muscle but not in the intestine or the skin. Similarly, silencing was detectable in rde-4(-) progeny of animals with RDE-4 rescued in the body-wall muscle (154). But, the same was not observed for RDE-1. Together, these tissue-specific rescue experiments suggest that short dsRNAs could be transported from donor cells to initiate gene silencing independent of RDE-4 in recipient cells (Fig. 2-1B). Fig. 2- 1: Two forms of dsRNA can move between cells to cause silencing. (A) Schematic of the route taken by exogenous dsRNA in C. elegans to silence in a cell. dsRNA can enter a cell using the dsRNA-specific importer, SID-1, get processed by DCR-1 and RDE-4 into 1? siRNA. One strand binds a 1? Ago, RDE-1 and finds an mRNA of matching sequence. Further amplification by an RdRP, RRF-1 results in the production of 2? siRNAs that may either cause mRNA destabilization using a cytoplasmic Ago or alterations in histone modifications using a nuclear Ago, both of which result in silencing. (B) Genetic analysis using tissue-specific expression of genes required for RNAi suggests two forms of dsRNA (long dsRNA and 1? siRNA) can move between cells. When expression of RDE-4 was restricted to the muscle, silencing was detected in rde-4(-) somatic cells. The same was not seen when RDE-1 was rescued in the muscle. 23 In this chapter, we extend previous observations using additional promoters, report an inhibitory effect of repetitive transgenes, and discover conditions for cell autonomous silencing in animals with tissue-specific rescue of rde-4. While expression of rde-4(+) in intestine, hypodermis, or neurons using a repetitive transgene can enable silencing in unrescued somatic tissues, silencing can be inhibited within tissues that express a repetitive transgene. Single-copy transgenes that express rde-4(+) in body-wall muscles or hypodermis, however, enable silencing selectively in the rescued tissue but not in other tissues. These results suggest that silencing by the movement of short dsRNA between cells is not an obligatory feature of feeding RNAi in C. elegans. We speculate that similar control of dsRNA movement could modulate tissue-specific silencing by feeding RNAi in other invertebrates. 2.3 Materials and methods 2.3.1 Strains and oligonucleotides All strains were cultured on Nematode Growth Medium (NGM) plates seeded with 100?l OP50 at 20?C and mutant combinations were generated using standard methods (155). Strains used in this chapter are listed in Table 2-1 and oligonucleotides used in this chapter are listed in Table 2-2. Table 2- 1: Strains used. Strain name Genotype 24 N2 wild type AMJ8 juIs73[Punc-25::gfp] III AMJ57 sid-1(qt9) V; jamEx12[Pmyo-3::sid-1(+)::unc-54 3? UTR & pHC183] (strain generated in (100)after multiple passages) AMJ58 rde-1(ne219) I; jamEx1[Pmyo-3::rde-1(+)::rde-1 3'UTR & Pmyo- 3::DsRed2::unc-54 3?UTR] AMJ66 rde-4(ne301); jamEx4[Pmyo-3::rde-4(+)::rde-4 3?UTR & pHC183] (strain generated in (93) after multiple passages). AMJ151 rde-4(ne301); mIs11[Pmyo-2::gfp::unc-54 3?UTR & gut::gfp::unc-54 3?UTR & pes-10::gfp::unc-54 3?UTR] jamIs3[Pmyo-2:DsRed::unc-54 3?UTR] jamIs4[Pmyo-3::rde-4(+)::rde-4 3?UTR & pHC183)] II (Outcrossed to AMJ8) AMJ162 rde-4(ne301); jamEx24[Pnas-9::rde-4(+)::rde-4 3?UTR & Pnas- 9::gfp::unc-54 3?UTR] AMJ188 rde-4(ne301); jamEx3[Prgef-1::rde-4(+)::rde-4 3?UTR & Prgef- 1::gfp::unc-54 3?UTR] AMJ190 rde-4(ne301) III; oxSi221[Peft-3p::gfp + cb-unc-119(+)] II AMJ210 jamEx24[Pnas-9::rde-4(+)::rde-4 3?UTR & Pnas-9::gfp::unc-54 3?UTR] AMJ212 rde-4(ne301) III; jamEx47[Pwrt-2::rde-4(+)::rde-4 3?UTR & Pwrt- 2::gfp::unc-54 3?UTR] AMJ217 rde-4(ne301) III; jamEx52[Punc-54::rde-4(+)::rde-4 3?UTR & Punc- 25 54::gfp::unc-54 3?UTR] AMJ220 rde-4(ne301) III; jamEx55[Psid-2::rde-4(+)::rde-4 3?UTR & Psid- 2::gfp::unc-54 3?UTR] AMJ229 rde-4(ne301) III; oxSi221 II; jamEx4 AMJ237 rde-4(ne301) III; jamEx65[Pmyo-3::rde-4(+)::rde-4 3?UTR & Pmyo- 3::gfp::unc-54 3?UTR] AMJ238 rde-4(ne301) III; jamEx66[Pmyo-3::rde-4(+)::rde-4 3?UTR & Pmyo- 3::gfp::unc-54 3?UTR] AMJ239 rde-4(ne301) III; jamEx67[Pmyo-3::rde-4(+)::rde-4 3?UTR & Pmyo- 3::gfp::unc-54 3?UTR] AMJ268 rde-4(ne301) III; sid-1(qt9) V; jamEx3 AMJ269 rde-4(ne301) III; sid-1(qt9) V AMJ290 rde-4(ne301) III; jamEx89[Punc-119c::rde-4(+)::rde-4 3?UTR & Punc-119c::gfp::unc-54 3?UTR)] AMJ303 jamEx77[Pnas-9::gfp::unc-54 3?UTR] AMJ311 rde-4(ne301) III; sid-1(qt9) V; jamEx12 AMJ314 oxSi221 II; unc-119(ed3) III (?); rde-1(ne219) V AMJ331 oxSi221 II; unc-119(ed3) III (?); rde-1(ne219) V; jamEx1 AMJ343 rde-4(ne301) III; sid-1(qt9) V; jamEx3 jamEx12 AMJ383 eri-1(mg366) IV; jamEx77 AMJ385 nrde-3(tm1116) X; jamEx77 AMJ488 ergo-1(tm1860) V; jamEx77 26 AMJ565 jamSi6 II; rde-4(ne301) III; unc-119(ed3) III (?) AMJ749 bli-1(jam14) II AMJ783 jamEx194[Pmyo-3::DsRed::unc-54 3?UTR] ? line 1 AMJ784 jamEx195[Pmyo-3::DsRed::unc-54 3?UTR] ? line 2 AMJ785 jamEx196[Pmyo-3::DsRed::unc-54 3?UTR] ? line 3 AMJ788 rde-1(ne219) V; jamEx199[Psid-2::rde-1(+)::rde-1 3?UTR & Psid- 2::gfp::unc-54 3?UTR] AMJ793 jamEx203[Prgef-1::bli-1-dsRNA & Prgef-1::gfp::unc-54 3?UTR] AMJ804 rde-4(ne301) III; K08F4.2(K08F4.2::gfp) IV AMJ805 oxSi221 II; unc-119(ed3) III(?); jamEx196 AMJ806 rde-4(ne301) III; nrIs20 IV [Pur-5::sur-5::gfp::sur-5 3?UTR]; jamEx204 [Psur-5::rde-4(+)::rde-4 3?UTR & Psur-5::DsRed] AMJ821 nrde-3(tm1116) X; jamEx203 AMJ822 sid-1(qt9) V; jamEx203 AMJ824 rde-4(ne301) III; K08F4.2(K08F4.2::gfp) IV; jamEx4 AMJ912 jamSi28[Pmyo-3::rde-4(+)::rde-4 3?UTR] II; rde-4(ne301) III EG4322 ttTi5605 II; unc-119(ed3) III EG6070 oxSi221 II; unc-119(ed3) III GR1373 eri-1(mg366) IV HC195 nrIs20 IV [Pur-5::sur-5::gfp::sur-5 3?UTR] HC196 sid-1(qt9) V 27 HC737 rde-4(ne301) III; nrIs20 IV HC780 rrf-1(ok589) I JH3197 K08F4.2(K08F4.2::gfp) IV WM27 rde-1(ne219) V WM49 rde-4(ne301) III WM156 nrde-3(tm1116) X WM158 ergo-1(tm1860) V YY160 nrde-1(gg88) III YY186 nrde-2(gg91) II YY453 nrde-4(gg129) IV Table 2- 2: Oligonucleotides used (5? to 3?, IDT). Name Sequence P1 gttccattcatcggcatgag P2 ttcatcattctgttcaattcttcatggatttaaccaaactaacg P3 cgttagtttggttaaatccatgaagaattgaacagaatgatgaa P4 cactgcagagaatgagtgtg P5 gtagaggtcagaggcatag P6 ttcatcattctgttcaattcttcatgagtaaaggagaagaacttttc P7 gaaaagttcttctcctttactcatgaagaattgaacagaatgatgaa P8 cggtcataaactgaaacgtaac P9 tcacggaactgacttcttg P10 acctgccaattgtttctcggatggatttaaccaaactaacg 28 P11 cgttagtttggttaaatccatccgagaaacaattggcaggt P12 ctgacatgcaaatggtgtg P13 ctgaaacgtaacatatgataagg P14 acctgccaattgtttctcggatgagtaaaggagaagaacttttc P15 gaaaagttcttctcctttactcatccgagaaacaattggcaggt P16 aattggatggatggcttctg P17 caaaaccctgatattttcaggaaatggatttaaccaaactaacg P18 cgttagtttggttaaatccatttcctgaaaatatcagggttttg P19 ctgcctattgggactcaacg P20 caaaaccctgatattttcaggaaatgagtaaaggagaagaacttttc P21 gaaaagttcttctcctttactcatttcctgaaaatatcagggttttg P22 ttgttgccgccaatttgc P23 ccatttgaaagaagcgagaaatcatggatttaaccaaactaacg P24 cgttagtttggttaaatccatgatttctcgcttctttcaaatgg P25 ggtatgagagagtggcagag P26 ccatttgaaagaagcgagaaatcatgagtaaaggagaagaacttttc P27 gaaaagttcttctcctttactcatgatttctcgcttctttcaaatgg P28 aggctgcaacaaagatcagg P29 tctctccgtacgtgtactctatcactgccggc P30 cgaccactagatccatctagaaatggatttaaccaaactaacg P31 cgttagtttggttaaatccatttctagatggatctagtggtcg P32 agagagctcgaggtagaggtcagaggcatag 29 P33 ggtcggctataataagttcttg P34 cggtcataaactgaaacgtaac P35 cgttagtttggttaaatccatctgaaaaaaaaaagagttttc P36 gcttcttctttggagcagtcatctgaaaaaaaaaagagttttc P37 gaaaactcttttttttttcagatgactgctccaaagaagaagc P38 gaaaactcttttttttttcagatggatttaaccaaactaacg P39 gtacaaatgacacagagccg P40 cttctcgcctctaagaaattc P41 aaatgacacagagccgacg P42 ctgaaacgtaacatatgataagg P43 cggtcataaactgaaacgtaac P44 tttctcggatgtcctcgaattttcccga P45 gctcaactggagaagtgtga P46 gcaggtgatttcacgacttc P47 gaaaagttcttctcctttactcatgaagaattgaacagaatgatg P48 tgctatgaggcacaggtaac P49 cgttagtttggttaaatccattgaacagaatgatgaattgcg P50 cgcaattcatcattctgttcaatggatttaaccaaactaacg P51 tctctccctgcaggatttcaaatgtggaataaacctgt P52 gagagaactagtgtagaggtcagaggcatag P53 ccattgatctttgcccactc P54 gctttcttgcttgttgcctg 30 P55 atttaggtgacactatagattactcctggctgacctgttttagagctagaaatagcaag P56 tggcaccgagtcggtgc P57 atttaggtgacactatagctaccataggcaccacgaggttttagagctagaaatagcaag P58 cgctcgtctctcaaattcag P59 agactagaagttttttttagttagatgaggttagatcacactac P60 gtagtgtgatctaacctcatctaactaaaaaaaacttctagtct P61 cccatctttccagatatacc P62 atatagactgcggtgactgg P63 tcccttcatgtcttcgcaac P64 cacttgaacttcaatacggcaagatgagaatgactggaaaccgtaccgcatgcggtgcctatggtagcgga gcttcacatggcttcagaccaacagccta P65 tcgggaaaattcgaggacatttcctgaaaatatcagggtttt P66 aaaaccctgatattttcaggaaatgtcctcgaattttcccga P67 cgataatctcgtgacactcg P68 gaggaggtgcacggaaaccgtcgtcgtcgtcgatgc P69 gcatcgacgacgacgacggtttccgtgcacctcctc P70 tttcttccaggtagtccagg P71 cccccttatatcagcacac P72 gttgaccatcttgtccgttc P73 cctggactacctggaagaaacgtcgtcgtcgtcgatgc P74 gcatcgacgacgacgacgtttcttccaggtagtccagg P75 atgcccagaactatccaagg 31 P76 gtttccgtgcacctcctc P77 tatagtcctgtcgggtttcg P78 cgttagtttggttaaatccatcatttctagatggatctagtg P79 cactagatccatctagaaatgatggatttaaccaaactaacg P80 tctctccctgcaggtgaccatgattacgccaagc P81 gagagacctgcagggtagaggtcagaggcatag P82 ataaggagttccacgcccag P83 ctagtgagtcgtattataagtg P84 tgaagacgacgagccacttg P85 caaggaagaactctgtacgg P86 aacgttagtttggttaaatccattctgaaaacaaaatgtaaagttca P87 tgaactttacattttgttttcagaatggatttaaccaaactaacgtt P88 gaaattgaagacgcaacaaaaac P89 cttcttctttggagcagtcattctgaaaacaaaatgtaaagttca P90 tgaactttacattttgttttcagaatgactgctccaaagaagaag P91 tctcaaggatcttaccgctg P92 acgcatctgtgcggtatttc 2.3.2 Transgenesis To make strains,N2 gDNAwas used (unless otherwise specified) as a template to amplify promoter or gene regions. To amplify gfp to be used (unless otherwise specified) as a coinjection marker, a plasmid containing gfp sequence was used as a template. All PCRs were performed with Phusion Polymerase (New England 32 Biolabs?NEB), unless otherwise specified, according to the manufacturer?s recommendations. The final fusion products were purified using PCR Purification Kit (QIAquick, Qiagen). To express rde-4(+) in the body-wall muscle under the myo-3 promoter: The wild-type rde-4 gene was expressed under the control of the myo-3 promoter from extrachromosomal arrays (AMJ66 (93), AMJ237, AMJ238, and AMJ239) or from an integrated array (AMJ151)). Expression from extrachromosomal arrays: To make Pmyo-3::rde-4(+)::rde-4 3?UTR, the myo-3 promoter (Pmyo-3) was amplified with primers P28 and P31, and rde-4(+) was amplified with primers P30 and P4. The two PCR products were used as templates to generate the Pmyo-3::rde-4(+) fusion product with primers P29 and P32. To make Pmyo-3::gfp::unc-54 3?UTR, gDNA from a strain with a transgene that expresses Pmyo-3::gfp::unc-54 3?UTR (HC150 (ccIs4251 [pSAK2 (Pmyo-3::nlsGFP- LacZ) & pSAK4 (Pmyo-3::mitoGFP), & dpy-20 subclone)] I; qtIs3 [pBMW14(Pmyo- 2::GFP--unc-22--PFG)] III; mIs11 [Pmyo-2::gfp, gut::gfp, pes-10::gfp] IV sid(qt25)) was used as template to directly amplify the fusion product with the primers P33 and P34 using Long-Template Expand Polymerase (Roche). WM49 animals were microinjected with a 1:1 mixture (10 ng/?l) of Pmyo-3::rde-4(+)::rde-4 3?UTR and Pmyo-3::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate three independent transgenic lines (AMJ237, AMJ238, and AMJ239). Expression from an integrated array: A strain with two spontaneous integration events that generated jamIs3 and jamIs4 was designated as AMJ151 (rde- 4(ne301) III; mIs11 jamIs3 jamIs4 IV). Microinjection of pHC448 at 38 ng/?l in 10 33 mM Tris (pH 8.5) into rde-4(ne301) III; mIs11 generated jamIs3. Subsequent microinjection of a mix of Pmyo-3::rde-4 and pHC183 (as described earlier in (93)) generated jamIs4. The resultant strain was then outcrossed by mating with AMJ8 (juIs73) to generate juIs73/rde-4(ne301) heterozygotes and picking their self progeny that lack juIs73. To express rde-4(+) in the body-wall muscle under the unc-54 promoter: To make Punc-54::rde-4(+)::rde-4 3?UTR, the unc-54 promoter (Punc-54) was amplified with primers P22 and P24, and rde-4(+) and rde-4 3?UTR was amplified with primers P23 and P4. The two PCR products were used as templates to generate the Punc-54::rde-4(+)::rde-4 3?UTR fusion product with primers P25 and P5. To make Punc-54::gfp::unc-54 3?UTR, Punc-54 was amplified using primers P22 and P27 and gfp::unc-54 3?UTR was amplified from pPD95.75 using primers P26 and P8. The two PCR products were used as templates and Punc-54::gfp::unc-54 3?UTR fusion product was generated using the primers P25 and P13. WM49 animals were microinjected with a 1:1 mixture (10 ng/?l) of Punc-54::rde-4(+)::rde-4 3?UTR and Punc-54::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ217. To express rde-4(+) in the hypodermis under the nas-9 promoter: To make Pnas-9::rde-4(+)::rde-4 3?UTR, the nas-9 promoter (Pnas-9) was amplified using the primers P1 and P3, and rde-4(+) and rde-4 3?UTR was amplified using the primers P2 and P4. The two PCR products were used as templates to generate the Pnas-9::rde-4(+)::rde-4 3?UTR fusion product with primers P40 and P5. To make Pnas-9::gfp::unc-54 3?UTR, Pnas-9 was amplified with primers P1 and P7, 34 and gfp::unc-54 3?UTR was amplified from pPD95.75 using the primers P6 and P8. WM49 animals were microinjected with a 2:1:1 mixture of Pnas-9::rde-4(+)::rde-4 3?UTR (10 ng/?l), Pnas-9 (with gfp overlap) (5 ng/?l) and gfp (with Pnas-9 overlap) (5 ng/?l) in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ162. To make strain AMJ210, AMJ162 was crossed with AMJ8 males. F2 cross progeny that were homozygous for juIs73 (which is linked to rde-4(+)) and contained the jamEx24[Pnas-9::rde-4(+)::rde-4 3?UTR & Pnas-9::gfp::unc-54 3?UTR] transgene were passaged for one generation to ensure homozygosity of juIs73 and then crossed with N2 males. A representative F2 progeny of this cross that lacked juIs73 (i.e. was homozygous for rde-4(+)) but contained the jamEx24[Pnas-9::rde- 4(+)::rde-4 3?UTR & Pnas-9::gfp::unc-54 3?UTR] transgene was designated as AMJ210. To express rde-4(+) in the hypodermis under the wrt-2 promoter: To make Pwrt-2::rde-4(+)::rde-4 3?UTR, the wrt-2 promoter (Pwrt-2) was amplified using the primers P9 and P11, and rde-4(+)::rde-4 3?UTR was amplified using the primers P10 and P4. The two PCR products were used as templates to generate the Pwrt-2::rde-4(+)::rde-4 3?UTR fusion product with primers P12 and P5. To make Pwrt-2::gfp::unc-54 3?UTR, Pwrt-2 was amplified using primers P9 and P15, and gfp::unc-54 3?UTR was amplified from pPD95.75 using primers P14 and P8. The two PCR products were used as templates to generate Pwrt-2::gfp fusion product with primers P12 and P13. WM49 animals were microinjected with a 1:1 mixture (10 ng/?l) of Pwrt-2::rde-4(+)::rde-4 3?UTR and Pwrt-2::gfp::unc-54 35 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ212. This strain grew slowly for the first ~4 generations, but became comparable to other strains in later generations. To express rde-4(+) in the intestine under the sid-2 promoter: To make Psid-2::rde-4(+)::rde-4 3?UTR, the sid-2 promoter (Psid-2) was amplified using the primers P16 and P18, and rde-4(+) along with rde-4 3?UTR was amplified using the primers P17 and P4. The two PCR products were used as templates and the Psid-2::rde-4(+)::rde-4 3?UTR fusion product was generated using the primers P19 and P5. To make Psid-2::gfp::unc-54 3?UTR, Psid-2 was amplified using the primers P16 and P21, and gfp::unc-54 3?UTR was amplified from pPD95.75 using the primers P20 and P8. The two PCR products were used as templates and the Psid-2::gfp::unc-54 3?UTR fusion product was generated using the primers P19 and P13. WM49 animals were microinjected with a 1:1 mixture (10 ng/?l) of Psid-2::rde- 4(+)::rde-4 3?UTR and Psid-2::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ220. To express rde-4(+) in the neurons under the rgef-1 promoter: rde-4(ne301); qtEx136; jamEx3[Prgef-1::rde-4(+)::rde-4 3?UTR & Prgef- 1::gfp::unc-54 3?UTR] animals (described in (93)) were crossed with WM49 and rde- 4(ne301); jamEx3[Prgef-1::rde-4(+)::rde-4 3?UTR & Prgef-1::gfp::unc-54 3?UTR] progeny were isolated and designated as AMJ188. To express rde-4(+) in the neurons under the unc-119 promoter: To make Punc-119::rde-4(+)::rde-4 3?UTR, the unc-119 promoter (Punc119) was amplified using the primers P39 and P35, and rde-4(+)::rde-4 3?UTR was 36 amplified using the primers P38 and P4. The two PCR products were used as templates to generate the Punc-119::rde-4(+)::rde-4 3?UTR fusion product with primers P41 and P5. To make Punc-119::gfp::unc-54 3?UTR, Punc119 was amplified using primers P39 and P36, and gfp::unc-54 3?UTR was amplified from pBH34.21 using the primers P37 and P42. The two PCR products were used as templates and the Punc-119::gfp::unc-54 3?UTR fusion product was generated using the primers P41 and P43. WM49 animals were microinjected with a 1:1 mixture (10 ng/?l) of Punc- 119::rde-4(+)::rde-4 3?UTR and Punc-119::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ290. To express rde-1(+) in the body-wall muscles under the myo-3 promoter: As described in (93). To express rde-1(+) in the intestine under the sid-2 promoter: To make Psid-2::rde-1(+)::rde-1 3?UTR, the sid-2 promoter (Psid-2) was amplified using the primers P16 and P65, and rde-1(+)::rde-1 3?UTR was amplified using the primers P66 and P45. The two PCR products were used as templates to generate the Psid-2::rde-1(+)::rde-1 3?UTR fusion product with primers P19 and P46. Coinjection marker Psid-2::gfp::unc-54 3?UTR, was made as described for AMJ220. WM27 animals were microinjected with a 1:1 mixture (10 ng/?l) of Psid- 2::rde-1(+)::rde-1 3?UTR and Psid-2::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ788. To express rde-1(+) in the hypodermis under the wrt-2 promoter: 37 To make Pwrt-2::rde-1(+)::rde-1 3?UTR, the wrt-2 promoter (Pwrt-2) was amplified using the primers P9 and P43, and rde-1(+)::rde-1 3?UTR was amplified using the primers P44 and P46. The two PCR products were used as templates to generate the Pwrt-2::rde-1(+)::rde-1 3?UTR fusion product with primers P12 and P46. Pwrt-2::gfp::unc-54 3?UTR was made as described for AMJ212. WM27 animals were microinjected with a 1:1 mixture (10 ng/?l) of Pwrt-2::rde-1(+)::rde-1 3?UTR and Pwrt-2::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. Three representative transgenic lines were designated as AMJ631, AMJ632, and AMJ633. To express rde-1(+) in neurons under the rgef-1 promoter: Made as described in (93). To express gfp in the hypodermis under the nas-9 promoter: To make Pnas-9::gfp::unc-54 3?UTR, the nas-9 promoter (Pnas-9) was amplified with primers P1 and P6, and gfp::unc-54 3?UTR was amplified from pPD95.75 using primers P47 and P8. The two PCR products were used as templates and Pnas-9::gfp::unc-54 3?UTR fusion product was generated using the primers P40 and P13. N2 animals were microinjected with a Pnas-9::gfp::unc-54 3?UTR in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ303. To express DsRed in the body-wall muscle under the myo-3 promoter: N2 animals were microinjected with pHC183 (Pmyo-3::DsRed::unc-54 3?UTR, made as described in (93)) in 10 mM Tris (pH 8.5) to generate 3 transgenic lines designated as AMJ783, AMJ784 and AMJ785. 38 To express sid-1(+) in the body-wall muscles under the myo-3 promoter: As described in (100). To express rde-4(+) in the hypodermis under the nas-9 promoter from a single-copy transgene: EG4322 animals were microinjected with a mixture of pJM6 (22.5ng/?l) and the coinjection markers pCFJ601 (50ng/?l), pMA122 (10 ng/?l), pGH8 (10 ng/?l), pCFJ90 (2.5 ng/?l), and pCFJ104 (5 ng/?l) (plasmids described in (156)) to generate a transgenic line as described earlier (156). This isolated line was crossed into AMJ8 males and the resulting rde-4(+)/juIs73 male progeny were crossed to WM49, and homozygozed for the single-copy insertion and rde-4(-) to generate AMJ565. The integration of Pnas-9::rde-4(+)::rde-4 3?UTR in AMJ565 was verified by genotyping for the presence of Pnas-9::rde-4(+) using primers P53 and P54. To express rde-4(+) in the body-wall muscle under the myo-3 promoter from a single-copy transgene: EG4322 animals were microinjected with a mixture of pPR1 (22.5ng/?l) and the coinjection markers pCFJ601 (50ng/?l), pMA122 (10 ng/?l), pGH8 (10 ng/?l), pCFJ90 (2.5 ng/?l), and pCFJ104 (5 ng/?l) (plasmids described in (156)) to generate a transgenic line as described earlier (156). This isolated line was crossed into AMJ8 males and the resulting rde-4(+)/juIs73 male progeny were crossed to WM49, and homozygozed for the single-copy insertion and rde-4(-) to generate AMJ912. The integration of Pmyo-3::rde-4(+)::rde-4 3?UTR in AMJ912 was verified by genotyping for the presence of an insertion at the Mos site using primers P82-P84. To express bli-1-dsRNA in the neurons under the rgef-1 promoter: 39 To make Prgef-1::bli-1-dsRNA sense strand, the rgef-1 promoter (Prgef-1) was amplified with primers P67 and P68 and a 1kb region in exon 3 of bli-1 was amplified using primers P69 and P70. The two PCR products were used as templates and Prgef-1::bli-1-dsRNA sense fusion product was generated using the primers P71 and P72. To make Prgef-1::bli-1-dsRNA antisense strand, the rgef-1 promoter (Prgef- 1) was amplified with primers P67 and P73 and bli-1 was amplified using primers P74 and P75. The two PCR products were used as templates and Prgef-1::bli-1- dsRNA antisense fusion product was generated using the primers P71 and P76. N2 animals were microinjected with a 1:1:1 ratio of sense Prgef-1::bli-1-dsRNA, antisense Prgef-1::bli-1-dsRNA and Prgef-1::gfp::unc-54 3?UTR (as described in (93)) in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ793. To express rde-4(+) in all somatic nuclei under the sur-5 promoter: To make Psur-5::rde-4(+)::rde-4 3?UTR, the sur-5 promoter (Psur-5) was amplified using the primers P85 and P86, and rde-4(+) along with rde-4 3?UTR was amplified using the primers P87 and P4. The two PCR products were used as templates and the Psur-5::rde-4(+)::rde-4 3?UTR fusion product was generated using the primers P88 and P5. To make Psur-5::DsRed, Psur-5 was amplified using the primers P85 and P89, and nuclear localized DsRed was amplified from pGC306 using the primers P90 and P91. The two PCR products were used as template and the Psur- 5::DsRed fusion product was generated using the primers P88 and P92. HC737 animals were microinjected with a 1:1 mixture (10 ng/?l) of Psur-5::rde-4(+)::rde-4 40 3?UTR and Psur-5::DsRed in 10 mM Tris (pH 8.5) to generate transgenic lines. A representative transgenic line was designated as AMJ806. 2.3.3 Plasmids The plasmid pJM6 (made by Julia Marr?, Jose lab) was used to make Si[Pnas- 9::rde-4(+)::rde-4 3_UTR]. The nas-9 promoter (Pnas-9) was amplified using primers P48 and P49 and rde-4(+)::rde-4 3 UTR was amplified using primers P50 and P4. The two PCR products were used as templates to generate the Pnas-9::rde- 4(+)::rde-4 3? UTR fusion product with primers P51 and P52. This fused product was purified and cloned into pCFJ151 using the SbfI and SpeI restriction enzymes (NEB) to generate pJM6. The plasmid pPR1 was used to make Si[Pmyo-3::rde-4(+)::rde-4 3? UTR]. The myo-3 promoter (Pmyo-3) was amplified using primers P77 and P78 and rde- 4(+)::rde-4 3? UTR was amplified using primers P79 and P4. The two PCR products were used as templates to generate the Pmyo-3::rde-4(+)::rde-4 3? UTR fusion product with primers P80 and P81. This fused product was purified and cloned into pCFJ151 using the SbfI restriction enzymes (NEB) to generate pPR1. The plasmid pYC13 (made by Yun Choi, Jose lab) is a derivative of pUC::unc-119 sgRNA with a different sgRNA (gift from John Calarco, Addgene plasmid #46169). All other plasmids were as described earlier (pHC448 (93), pPD95.75 (gift from Andrew Fire, Addgene plasmid #1494), pBH34.21 (157), pCFJ151, pCFJ601, pMA122, pGH8, pCFJ90, pCFJ104 (all from (156), pL4440 (79), pHC183 (100)and pGC306 (a gift from Jane Hubbard, Addgene plasmid #19658)). 41 2.3.4 Genome editing To generate bli-1 null mutants, Cas9-based genome editing employing a co- conversion strategy was used (158). To prepare guide RNAs, the scaffold DNA sequence was amplified from pYC13 using primers P55 and P56 for bli-1, and primers P57 and P56 for the co-conversion marker dpy-10. The amplified DNA templates were purified (PCR Purification Kit, Qiagen), transcribed (SP6 RNA polymerase, NEB), and tested in vitro for cutting efficiency (Cas9,NEB). For injection into animals, homology template for repair (repair template) was amplified from N2 gDNA using Phusion polymerase and gene specific primers. P58 and P59 were used to amplify a region immediately upstream of the 5? region of bli-1 and P60 and P61 were used to amplify a region immediately downstream of the 3? region of bli-1 using Phusion Polymerase (NEB). The two PCR products were used as templates to generate the repair template with primers P62 and P63 using Phusion Polymerase (NEB) and the fused product was purified (PCR Purification Kit, Qiagen). Homology template for dpy-10 was a single-stranded DNA oligo (P64). Wild-type animals were injected with 3.5 pmol/?l of bli-1 guide RNA, 2.4pmol/?l of dpy-10 guide RNA, 0.06 pmol/?l of bli-1 homology repair template, 0.6 pmol/?l of dpy-10 homology repair template and 1.6 pmol/?l of Cas-9 protein (PNA Bio Inc.). 2.3.5 Feeding RNAi and scoring associated defects Feeding RNAi: RNAi experiments were performed at 20?C on NGM plates supplemented with 1 mM IPTG (Omega Bio-Tek) and 25g/ml Carbenicillin (MP Biochemicals) (RNAi plates). 42 One generation or F1-only feeding RNAi. A single L4 or young adult (1 day older than L4) animal (P0) was placed on an RNAi plate seeded with 5?l of OP50 Escherichia coli and allowed to lay eggs. After 1 day, when most of the OP50 E. coli was eaten, the P0 animal was removed, leaving the F1 progeny. 100?l of an overnight culture of RNAi food (E. coli which express dsRNA against a target gene) was added to the plate. Two or three days later, the F1 animals were scored for gene silencing by measuring gene specific defects (Table 2-3). All RNAi E. coli clones were from the Ahringer library (159) and generously supplied by Iqbal Hamza (University of Maryland), with the exception of unc-54 RNAi, which was made by inserting a fragment of unc-54 DNA into pL4440 and transforming HT115(DE3) E. coli cells with the resultant plasmid. Control RNAi by feeding E. coli containing the empty dsRNA-expression vector (pL4440), which does not produce dsRNA against any gene, was done in parallel with all RNAi assays. Fluorescence intensity of L4 animals fed dsRNA against gfp was measured using ImageJ (National Institute of Health? NIH). Animals with an intensity of >6000 (a.u.) in a fixed area within the gut immediately posterior to the pharynx were considered not silenced. Based on these criteria, 91.7% of wild-type animals and a 100% of animals expressing DsRed in the muscle (Ex[Pmyo-3::DsRed]) showed silencing. In these silenced animals intensity of gfp was measured from below the pharynx to the end of the vulva and the background intensity for the same area was measured for each animal. The intensity of gfp after background subtraction was plotted for each worm (Fig. 2-4E). Two generations or P0 & F1 Feeding RNAi. The experiments in all Figures (except Fig. 2-2B, 2-6, 2-7, 2-8, 2-11A&B, 2-12) were performed by feeding both the 43 P0 and F1 generations, as described earlier (100). Control RNAi was done in parallel with all RNAi assays. Three or four days after P0 animals were subjected to RNAi, the F1 animals were scored for gene silencing by measuring gene-specific defects (Table 2-3). In general, no difference in gene silencing was observed between F1- only feeding RNAi and P0 & F1 feeding RNAi for rde-4 mutants with tissue-specific rescue (e.g. see ?F1 RNAi? and ?P0 & F1 RNAi? Fig. 2-11). However, to test for inheritance of RDE-4 in Fig. 2-12, only F1 RNAi was used to avoid any inheritance of dsRNA that could occur when P0 & F1 RNAi is used (98). For each RNAi experiment testing rde-4 function, feeding of N2 and WM49 was performed alongside as controls. Table 2- 3: Scoring of gene-specific silencing. Gene Expression Defect scored upon RNAi unc-22 body-wall muscle L4 or young adults twitch within 2 minutes in response to 3mM levamisole (Sigma Aldrich). unc-54 body-wall muscle Inability to move backwards upon touching the head and lack of sinusoidal movement. act-5 intestine Failure to develop to the 4th larval stage in 3 days. dpy-7 hypodermis Short, fat L4- or adult-staged animals bli-1 hypodermis Presence of fluid-filled blisters on adults pos-1 germline Animals that laid >90% unhatched eggs par-1 germline Animals that laid >66% unhatched eggs par-2 germline Animals that laid >80% unhatched eggs 44 For bli-1 defects upon RNAi as well as upon Cas9-based genome editing, the pattern of blister formation was scored. Each animal was partitioned into eight roughly equal sections (a to h) as shown in Fig. 2-9F with the vulva being the mid point of the animal. Sections with >50% of their length covered by a blister were marked black and sections with a discontinuous blister were marked grey. Animals that did not follow the anterior more than posterior and dorsal more than ventral susceptibility pattern (a > b > c > d > e > f > g > h) were culled as variants for each genotype and the relative aggregate blister formation in each section among worms with altered susceptibility were computed using a score of black = 1.0 and grey = 0.5 for each section of every worm. The computed values for each section in all worms of a strain were summed and normalized to the value of the highest section for that strain. To compare multiple strains, these values for each strain were multiplied by the fraction of worms that showed a blister in that strain. Using these measures of normalized relative aggregate blister formation among animals with variant susceptibility, we generated heat maps (160), where black indicates highest frequency of blisters and white indicates the lowest frequency of blisters among the sections of all strains that are being compared. 2.3.6 Balancing loci Integrated transgenes expressing gfp were used to enable identification of mutant chromosomes in progeny of heterozygous animals. Animals were scored as homozygous mutants if they lacked both copies of the transgene. The rde-4(ne301) allele on Chr III was balanced by juIs73 . About 99% (153/ 155) of the progeny of rde-4(ne301)/juIs73 that lacked fluorescence were found to be homozygous rde- 45 4(ne301) animals either by Sanger sequencing (n = 96) or by resistance to pos-1 RNAi (n = 59). 46 2.4 Results 2.4.1 Expression of a repetitive transgene in a tissue can inhibit RNAi in that tissue Until recently, studies examining the function of C. elegans genes have relied on the use of repetitive transgenes often coupled with tissue-specific promoters (e.g. (161-163)) and/or RNAi (e.g.(164-166)). For example, rescue of rde-4 or rde-1 using the myo-3 promoter to drive expression in body-wall muscles from an extrachromosomal repetitive transgene (Ex[Pmyo-3::rde(+)]) was employed to examine the systemic response to RNAi (93,154). While silencing in rde-1(-) animals with Ex[Pmyo-3::rde-1(+)] was observed only in body-wall muscles, silencing in rde-4(-) animals with Ex[Pmyo-3::rde-4(+)] was observed in both body-wall muscles and other tissues. We found that in both cases, silencing of one gene within body-wall muscle cells was substantially reduced compared to that in wild-type animals in response to feeding RNAi (compare silencing of unc-22 to unc-54 in Fig. 2-2A, left. Also see Table 2-3). Similar reduction in silencing was observed in hypodermal cells upon hypodermal rescue of rde genes (compare dpy-7 silencing to bli-1 silencing in Fig. 2-2A, right. Also see Table 2-3). When a ubiquitously expressed target gene was targeted for silencing in rde-4(-); Ex[Pmyo-3::rde-4(+)] animals, decreased silencing was observed only within body-wall muscles despite robust silencing in other tissues (gfp silencing in Fig. 2-2B). 47 Fig. 2- 2: Silencing by feeding RNAi of some genes is reduced in tissues expressing RDE-4 or RDE-1. (A) Silencing by feeding RNAi of some endogenous genes is reduced in tissues expressing rde-4(+) or rde-1(+) from a repetitive transgene. (Left) Wild-type animals, mutant animals (rde-1(-), top or rde-4(-), bottom) or mutant animals with tissue-specific rescues in the body-wall muscles (Ex[Pmyo-3::rde(+)]) were fed dsRNA against unc-22 or unc-54 and the fractions of animals that showed silencing (fraction silenced) were determined. (Right) Wild-type animals, mutant animals (rde- 1(-), top or rde-4(-), bottom) or mutant animals with tissue specific rescue in the hypodermis (Ex[Pwrt-2::rde(+)]) were fed dsRNA against hypodermal genes (dpy-7 or bli-1,green) and the fractions of animals that showed silencing (fraction silenced) were determined. Asterisks indicates p<0.01 (compared to wild-type animals). Error bars indicate 95% confidence intervals (CI), n>22 animals. (B) Silencing of gfp in body-wall muscles that express RDE-4 from a repetitive transgene is reduced despite potent silencing in other rde-4(-) somatic tissues. Representative images of animals with gfp expression (black) in all somatic cells (Peft-3::gfp) in a wild-type background (left) or rde-4(-) background with rde-4(+) expressed in body-wall muscles (Ex[Pmyo-3::rde-4(+)) (right) that were fed bacteria that express dsRNA against gfp (gfp RNAi) or control RNAi are shown. Tissues that show reduced silencing (pharynx, muscles) are labeled, insets are brightfield images, and scale bar = 50 ?m. Also see methods for details on making of the inverted grey-scale images presented in all figures. The extent of gene silencing varied based on the specific promoter used (e.g. silencing of dpy-7 when wrt-2 promoter was used (Fig. 2-2A, right) was greater than when the nas-9 promoter was used (Fig. 2-3A)). Thus, rescuing rde-4 or rde-1 using 48 tissue-specific promoters and repetitive transgenes does not reliably restore tissue- restricted silencing of all genes. To test if these cases of reduced silencing could be explained by insufficient levels of rde expression, we overexpressed RDE-4 in the hypodermis of wild-type animals (Ex[Pnas-9::rde-4(+)]) and examined silencing. Feeding RNAi of neither bli-1 nor dpy-7 resulted in detectable silencing (Fig. 2-3B, left), suggesting that the observed lack of silencing is not because there was insufficient rde-4(+) expression. An alternative possibility is that such lack of silencing for both tested genes could reflect co-suppression (167) of rde-4 because of rde-4(+) expression from a repetitive transgene. Co-suppression is a phenomena that was first observed in plants, where the introduction of a transgenic copy of a gene resulted in the silencing of the transgenic and endogenous copy of that gene. However, this hypothesis cannot explain the differential susceptibility of bli-1 and dpy-7 that was observed when rde-4(+) was expressed under the wrt-2 promoter. Alternatively, expression of any repetitive transgene could inhibit RNAi and the extent of inhibition could vary based on the promoter used and on the target gene being tested. To test these possibilities, we expressed gfp from a repetitive transgene in the hypodermis (Ex[Pnas-9::gfp]) and examined silencing of bli-1 and of dpy-7 by feeding RNAi. Surprisingly, no silencing was detected when gfp was expressed (Fig. 2-3B, right) and using a single-copy transgene to express rde-4(+) in the hypodermis (Si[Pnas-9::rde-4(+)]) enabled potent silencing of both dpy-7 and bli-1 by feeding RNAi (Fig. 2-3C). Furthermore, this silencing could be inhibited by additionally expressing Ex[Pnas-9::gfp] (Fig. 2- 3C). Together, our results suggest that inhibition of silencing is the result of 49 expression from a repetitive transgene and not because of rde-4 expression or co- suppression. Fig. 2- 3: Expression of any repetitive transgene in a tissue can inhibit silencing by ingested dsRNA. (A) Silencing of bli-1 and dpy-7 is inhibited by the expression of rde-4 from a repetitive transgene using a different promoter in the hypodermis. rde-4(-) animals that express RDE-4 in the hypodermis (Ex[Pnas-9::rde-4(+)]) from extrachromosomal repetitive DNA (array) were fed dsRNA against hypodermal genes (dpy-7 or bli-1, green). The fractions of animals either with or without the arrays that showed silencing (fraction silenced) were determined. (B) Silencing of a gene by ingested dsRNA within a tissue can be inhibited by the expression of a repetitive transgene of unrelated sequence within that tissue. (Left) Wild-type animals that express RDE-4 in the hypodermis (Ex[Pnas-9::rde-4(+)]) from extrachromosomal repetitive DNA (array) were fed dsRNA against hypodermal genes (dpy-7 or bli-1, green). (Right) Wild-type animals that express GFP in the hypodermis (Ex[Pnas- 9::gfp]) from extrachromosomal repetitive DNA (array) were fed dsRNA against hypodermal genes (dpy-7 or bli-1, green). The fractions of animals either with or without the arrays that showed silencing (fraction silenced) were determined. (C) Expression of RDE-4 from a single-copy transgene within a tissue does not inhibit feeding RNAi in that tissue. rde-4(-) animals that express RDE-4 in the hypodermis from a single-copy transgene (Si[Pnas-9::rde-4(+)]) or that additionally express gfp in the hypodermis (Ex[Pnas-9::gfp]) from a repetitive transgene (array) were fed dsRNA against dpy-7 or bli-1 and were analyzed as in (B). Asterisks indicate p<0.01 (compared to animals without Ex[Pnas-9::gfp]). Error bars indicate 95% confidence intervals (CI), n>22 animals. Consistently inhibition of silencing of unc-54, unc-22, and gfp was also observed in body-wall muscles when we expressed DsRed from repetitive transgenes in body-wall muscles (Fig. 2-4). Together, our results suggest that expression from 50 repetitive transgenes in a tissue can interfere with silencing by ingested dsRNA within that tissue. Fig. 2- 4: Silencing by feeding RNAi of some genes is reduced in tissues expressing a repetitive transgene of unrelated sequence. (A) Silencing of unc-22 and unc-54 can be inhibited by the expression of a repetitive transgene of unrelated sequence in body-wall muscles. Three independent lines of wild-type animals expressing DsRed in body-wall muscles (Ex[Pmyo-3::DsRed]) from extrachromosomal repetitive transgenes (array) were fed dsRNA against body- wall muscle genes (unc-22 or unc-54, magenta). The fractions of animals either with or without the arrays that showed silencing (fraction silenced) were determined (n=50). Error bars indicate 95% CI and asterisks indicate p<0.01 (compared to animals without array). (B-D) Representative images of animals with gfp expression in all somatic cells (GFP) that were fed gfp RNAi (B) or animals that in addition express DsRed in the body-wall muscle (Pmyo-3::DsRed) that were fed control RNAi (C) or gfp RNAi (D) are shown. Expression of GFP or DsRed is shown in black except in merged images in (C) and (D) that show overlap of GFP and DsRed expression (DsRed = magenta and GFP = green). Insets are brightfield images and scale bar = 50 ?m. (E) Quantification of gfp silencing in animals with or without DsRed expression in body-wall muscles. In animals expressing DsRed in body-wall 51 muscles (Ex[Pmyo-3::DsRed]), gfp fluorescence was brighter in body-wall muscles than in other tissues. Red lines indicate median GFP fluorescence. Asterisks indicates p<0.01 (compared to wild-type animals). One explanation of these results could be that repetitive transgenes produce dsRNAs (168,169) that compete with ingested dsRNA for engaging the gene silencing machinery within a cell. Such competition between pathways is the reason silencing by feeding RNAi can be enhanced in animals that lack genes required solely for the processing of endogenous dsRNA (170-172). Consistent with this hypothesis, we found that loss of the RdRP rrf-3 or the endonuclease eri-1, factors required for endogenous dsRNA production but not loss of lin-15 that enhances RNAi as a consequence of soma to germline transformation (173) overcame the inhibition of feeding RNAi (Fig. 2-5A). However, loss of downstream Argonaute proteins (the primary Argonaute ergo-1 or the secondary Argonaute nrde-3) did not overcome inhibition. Because RRF-3 and ERI-1 are not required for the production of dsRNA from repetitive transgenes (174), these results suggest that silencing by feeding RNAi in the presence of expression from a repetitive transgene could be enabled by loss of dsRNA production at endogenous genes (175). While these results are consistent with competition between ingested dsRNA and dsRNA from repetitive transgenes, other underlying mechanisms are also possible (see Discussion). Consistent with a role for the nuclear RNAi pathway in silencing bli-1, we found that bli-1 silencing by feeding RNAi also depended on components of the feeding RNAi pathway including factors that act downstream of NRDE-3 (Fig. 2-5B, right) (116,118). This dependence on the nuclear RNAi pathway was also observed when bli-1 was targeted for silencing by dsRNA expressed in neurons (Fig. 2-5C), suggesting that irrespective of the source of 52 the dsRNA, NRDE-3 is required for silencing bli-1. Certain genes appear to be more susceptible to inhibition by expression from repetitive transgenes. For example, unc- 54 was more susceptible than unc-22 (Fig. 2-2A, left) and bli-1 was more susceptible than dpy-7 (Fig 2-2, right). While the basis for these differences is unclear, we found silencing of bli-1 and unc-54 but not of unc-22 showed a dependence on the secondary Argonaute NRDE-3 (Fig. 2-5D). Additional experiments are needed to establish mechanistic links, if any, between nrde-3-dependence and inhibition of RNAi by expression from repetitive transgenes. Fig. 2- 5: Genes that show robust inhibition of silencing upon expression of an unrelated transgene require the nuclear RNAi pathway for complete silencing. (A) Inhibition of bli-1 silencing by expression of any repetitive DNA can be relieved by the loss of some components of the endogenous RNAi pathway. (Left) Loss of rrf- 3 or eri-1 but not of lin-15, ergo-1 or nrde-3 relieves inhibition of bli-1 silencing caused by hypodermal expression of repetitive DNA. Extent of silencing (fraction silenced) in response to bli-1 feeding RNAi of animals with or without an extrachromosomal array that expresses gfp in the hypodermis (Ex[Pnas-9::gfp]) in a wild-type, rrf-3(-), eri-1(-), ergo-1(-) or nrde-3(-) background were determined (n>28 animals). Error bars indicate 95% CI and asterisks indicate p<0.01 (compared to wild- type animals with or without the array, respectively). (Right) Schematic of endogenous RNAi. Endogenous RNA recruits the RNA dependent RNA polymerase RRF-3, the exonuclease ERI-1, the endonuclease DCR-1, the primary Argonaute ERGO-1, and the secondary Argonautes (e.g. NRDE-3) to cause silencing. (B) Genes that show robust inhibition of silencing upon expression of a repetitive transgene appear to also require the nuclear Argonaute NRDE-3 for complete silencing. Wild- type animals and nrde-3(-) animals were fed dsRNA against unc-22, unc-54, or bli-1 53 and the fractions of animals that showed silencing (fraction silenced) were determined. Error bars indicate 95% CI and asterisks indicate p<0.01 (compared to wild-type animals). (C) Silencing of bli-1, by ingested dsRNA requires multiple components of the nuclear RNAi pathway. Wild-type, nrde-3(-), nrde-2(-), nrde-1(-), or nrde-4(-) animals were fed dsRNA against bli-1 and the percentage of gravid adult animals that showed silencing (% Bli) was determined (n = number of gravid adults scored for silencing). (D) Silencing of bli-1 by neuronal dsRNA requires the nuclear Argonaute nrde-3. Silencing of bli-1 by dsRNA against bli-1 expressed under a neuronal promoter (Ex[Prgef-1::bli-1-dsRNA]) in a wild-type, sid-1(-), or nrde-3(-) background were measured as in (C). Taken together, our analyses suggest that expression of any repetitive transgene in a tissue can inhibit silencing of some genes within that tissue and using a single-copy transgene can avoid this problem. 2.4.2 Using repetitive transgenes to rescue rde-4 in one somatic tissue can support RNAi within other somatic tissues When repetitive transgenes were used to express the dsRNA-binding protein RDE-4 in body-wall muscles of rde-4(-) animals, silencing of genes that function in other somatic tissues was observed upon feeding RNAi (93). In contrast, when repetitive transgenes were used to express the Argonaute protein RDE-1 in body-wall muscles, silencing was restricted to body-wall muscles in rde-1(-) animals (93,176). These results were used to infer the possible intercellular movement of short dsRNAs generated downstream of RDE-4 but upstream of RDE-1 (Fig. 2-1B). To test if this observation extended to other tissues (intestine, hypodermis, and neurons), we similarly used repetitive transgenes to perform tissue-specific rescues of rde-1 and rde-4 and assessed silencing of genes expressed in somatic tissues (Fig. 2-6). In all cases, silencing by feeding RNAi was observed only in one RNAi-sensitive somatic tissue when RDE-1 was expressed in that somatic tissue (Fig. 2-6, top). In contrast, 54 silencing was observed in all tested somatic tissues when RDE-4 was expressed in any one somatic tissue (Fig. 2-6, bottom). This silencing in rde-4(-) soma was also observed when a ubiquitously expressed gene (Pgtbp-1::gtbp-1::gfp) was targeted for silencing in rde-4(-); Ex[Pmyo-3::rde-4(+)] animals or in Ex[Prgef-1::rde-4(+)] animals (Fig. 2-7). Fig. 2- 6: Tissue-specific rescues of RDE-4 from repetitive transgenes can enable silencing of genes in non-rescued somatic tissues. Wild-type animals (dark grey), mutant animals (rde-1(-) or rde-4(-), white), and mutant animals with tissue-specific rescues (colors within worms) of rde-1 or rde-4 were fed dsRNA against genes expressed in somatic tissues (the body-wall muscles (unc-22, magenta), intestine (act-5, blue), hypodermis (dpy-7, green)), or in the germline (pos-1, par-1, or par-2, grey) and the fractions of animals that showed silencing (fraction silenced) were determined. Wild-type genes (rde-1 or rde-4) were expressed in the body-wall muscles (Ex[Pmyo-3::rde(+)], magenta), in the intestine (Ex[Psid-2::rde(+)], blue), in the hypodermis (Ex[Pwrt-2::rde(+)], green), or in neurons (Ex[Prgef-1::rde(+)], orange). Error bars indicate 95% CI and n>24 animals. Asterisks indicate p<0.01 (compared to wild-type animals). 55 Together, these results are consistent with the interpretation that short dsRNAs are transported between somatic cells when long dsRNA is processed in one somatic tissue by RDE-4. Alternatively, it is formally possible that the RDE-4 protein or mRNA is transported between somatic tissues. 2.4.3 Rescue of rde-4 in a somatic tissue from repetitive transgenes does not cause detectable silencing in the germline If short dsRNAs are exported from a somatic tissue into the pseudocoelomic fluid, then they could be imported and used for silencing in all tissues, including the germline. Therefore, we tested if expression of RDE-4 in a somatic tissue (body-wall muscles, intestine, hypodermis or neurons) using repetitive transgenes was sufficient to enable silencing of genes expressed in the germline (pos-1, par-1 or par-2) (See Table 2-3 for defects scored). In contrast to the soma, no silencing was detectable within the germline in any case (Fig. 2-6, grey bars). This lack of silencing in the germline was also observed when a ubiquitously expressed gene (Pgtbp-1::gtbp- 1::gfp) was targeted for silencing in rde-4(-); Ex[Pmyo-3::rde-4(+)] animals and rde- 4(-); Ex[Prgef-1::rde-4(+)] animals (Fig. 2-7). These results suggest that the germline does not accumulate sufficient levels of short dsRNAs for gene silencing. Alternatively, all tested repetitive transgenes that express rde-4(+) could be broadly expressed within somatic tissues but not within the germline, which is known to have powerful mechanisms that silence repetitive transgenes (177). Such misexpression could be sufficient to explain the gene silencing observed for rde-4 rescues but not for 56 rde-1 rescues despite both rescues using the same promoter if low levels of RDE-4 but not RDE-1 are sufficient for function. Fig. 2- 7: Tissue-specific rescues of RDE-4 from repetitive transgenes enables silencing of ubiquitous gfp in non-rescued somatic tissues but not germline. Representative images of animals with gfp expression (black) in all somatic and germline cells (Pgtbp- 1::gtbp-1::gfp) in a wild-type background, rde-4(-) background, or rde-4(-) background with rde- 4(+) expressed in body-wall muscles (Ex[Pmyo-3::rde-4(+)) or in the neurons (Ex[Prgef-1::rde- 4(+)) that were fed control RNAi or gfp RNAi are shown. In all cases, 50 L4-staged animals were analyzed and the majority phenotype is shown. Insets are brightfield images and scale bar = 50?m. 2.4.4 Silencing can occur in rde-4(-) cells when rde-4 mosaic animals ingest dsRNA Effects of short dsRNAs derived from ingested dsRNA can also be evaluated using mosaic analysis (178). In this approach, mosaic animals that result from the loss of rescuing DNA during mitosis are examined. Perdurance of protein or mRNA after 57 the loss of rescuing DNA in ancestral cells can complicate the interpretation of effects in descendent cells, which is a different caveat compared to that for tissue-specific rescue. Therefore, to complement our analysis using tissue-specific rescue, we examined silencing by ingested dsRNA in rde-4(-) mosaic animals (Fig. 2-8A). Specifically, we co-expressed rde-4(+) and DsRed under the control of the sur-5 promoter, which drives expression in all somatic cells, and examined silencing of gfp upon feeding RNAi of an integrated Psur-5::sur-5::gfp transgene in mosaic animals. This results in animals with an extrachromosomal array where cells that have the DNA for rde-4(+) expression are marked with DsRed expression. In all 19 mosaic animals examined, silencing was observed in rde-4(+) as well as in rde-4(-) cells (Fig. 2-8B-D). Consistent with inhibition of silencing in tissues with multi copy arrays, we observed reduced silencing in cells that showed expression of rde-4(+) than in cells that lacked expression of rde-4(+) (e.g. the single binucleated intestinal cell in Fig. 2-8D). Taken together with the results from tissue-specific rescue of rde- 4, these results are consistent with the hypothesis that some derivatives of ingested dsRNA generated in somatic tissues with rde-4(+) expression may be transported between cells. However, given that maternal rde-4 (mRNA or protein) can perdure in mutant progeny up to adulthood (98), it is likely that the observations here are due to the perdurance of rde-4 mRNA or RDE-4 protein from precursor cells. 58 Fig. 2- 8: rde-4(-) cells can be silenced by ingested RNAi in rde-4 mosaic animals. (A) Schematic of rde-4 mosaic analysis. Animals with mitotic loss of rescuing DNA resulting in some mutant cells (rde-4(-)) and some wild-type cells (rde-4(+) with co- expressed DsRed) were fed dsRNA targeting a gene expressed in both mutant cells and wild-type cells and the resultant silencing (feeding RNAi - ?) was examined. (B to D) Representative images of animals with gfp expression in all somatic nuclei (Is[Psur-5::sur-5::gfp]) in wild-type animals (B), rde-4(-) animals (C) or rde-4 mosaic animals with rde-4 rescue marked with DsRed expression (Ex[Psur-5::rde- 4(+)&DsRed]) (D) that were fed control dsRNA (control RNAi, left) or dsRNA against gfp (gfp RNAi, right) are shown (n > 9 in (B) and n > 10 in (C); and n>12 in (D)). Merged images in (D) show overlap of gfp and DsRed expression (red channel = magenta; green channel = green; and merge = black). Insets are brightfield images and scale bar = 50?m. 59 2.4.5 Spatial patterns of silencing vary with the promoters that drive tissue- specific rescue from repetitive transgenes If transport of short dsRNAs, rather than misexpression in somatic tissues, is the reason for the observed silencing, then silencing could be more common in cells that are near the source of dsRNA. However, when an animal is only scored as silenced versus not silenced in response to feeding RNAi, such qualitative differences between animals are overlooked. Examination of such differences requires a target gene whose silencing in subsets of cells can be discerned in each animal. We found that null mutants of the hypodermal gene bli-1 result in a fluid-filled sac (blister) along the entire worm (Fig. 2-9 A&B, (155)), and that blisters that form upon feeding RNAi in wild-type animals had a different pattern (Figure 2-9 C-F). Specifically, upon bli-1 feeding RNAi, anterior sections of the worm tended to be more susceptible to silencing when compared to posterior sections (Figure 2-9 C-E, and see methods section 2.3.5), resulting in a stereotyped pattern of relative susceptibility to blister formation (Figure 2-9F). This bias in the tendency to form blisters likely reflects the graded uptake of dsRNA from the anterior to the posterior in the intestine upon feeding RNAi. These characteristics of blister formation as a result of bli-1 silencing enable examination of qualitative differences, if any, between silencing in wild-type animals and in animals with tissue-specific rde-4 rescue. 60 Fig. 2- 9: Silencing of bli-1 upon feeding RNAi in wild-type animals and in animals with tissue-specific rescue of rde-4 in non-hypodermal cells results in unique patterns of blisters that differ from those in bli-1(-) animals. (A) Design of Cas9-based genome editing to generate a bli-1 null mutant. The bli-1 gene (exons, blue boxes; introns, grey lines) was targeted by a single guide RNA (sgRNA) that cuts within the gene (orange *) and was repaired with a double- stranded DNA template (green flanking the gene). (B) A null mutation in bli-1 results in blisters that cover the entire body. A representative image of a bli-1 null mutant animal generated by Cas9-based genome editing. Scale bar = 50 ?m. (C) Feeding RNAi of bli-1 in wild-type animals results in blisters that cover part of the body. A representative image of a wild-type animal fed dsRNA against bli-1 (bli-1 RNAi). Scale bar = 50 ?m. (D) A representative animal that illustrates scoring of bli-1 silencing in each section of the worm. The animal was divided into 8 sections (a through h, see F) and each section was scored for presence of a full blister (black), partial blister (grey), or no blister (white) as indicated in the inset. Scale bar = 50 ?m. (E) Wild-type animals display a stereotyped pattern of susceptibility to bli-1 feeding RNAi. Progeny of wild-type animals targeted by Cas9-based genome editing, bli-1 null mutant animals, and wild-type animals exposed to feeding RNAi were scored for blister patterns as described in D (n>45 gravid adult animals). Unlike sections in the progeny of animals that were injected with Cas9/sgRNA or in bli-1 null mutants, sections in wild-type animals that were subject to bli-1 feeding RNAi showed a stereotyped frequency of blister formation (a > b > ? > h). (F) Susceptibility to bli-1 feeding RNAi decreases from anterior to posterior hypodermis in wild-type animals. (top) Schematic of hypodermal sections (a through h) scored for blister formation. (bottom) Consensus relative frequency of blister formation in each hypodermal section of wild-type animals upon bli-1 feeding RNAi. The frequency ranged from 1.0 (black, a) to 0.06 (~white, h). 61 To systematically analyze such differences, we culled animals with a pattern of blister formation that differed from a consensus blister susceptibility pattern observed in most wild-type animals (Fig. 2-10 A&B and see methods section 2.3.5). We found that unlike in wild-type animals, in animals with tissue-specific rescue of rde-4 from repetitive transgenes, patterns of blisters that differed from the reference pattern were common (Fig. 2-10). Furthermore, the pattern of variant blister susceptibility differed depending on the promoter used for tissue-specific rescue (rgef-1 or unc-119 for neurons and myo-3 or unc-54 for body-wall muscles) (Fig. 2- 10 C-E). Intriguingly, animals with different promoters that drive expression in the same tissue showed similar patterns of silencing. Additionally, silencing in rde-4(-) cells was not robustly inhibited by the presence of a repetitive array (Fig. 2-10F). While these results could provide a case for the transport of short dsRNAs to nearby cells from the tissue where long dsRNA is processed by RDE-4, similar results would also be obtained if misexpression from each tissue-specific promoter occurred in subsets of hypodermal cells. 62 Fig. 2- 10: Repetitive transgenes expressing RDE-4 from different promoters enable different spatial patterns of bli-1 silencing in rde-4(-) hypodermis. (A) RDE-4 expression in intestinal cells near the tail correlates with blister formation in rde-4(-) hypodermis near the tail upon bli-1 RNAi. (B and C) Blister scoring method to detect variations in the pattern of blister formation upon bli-1 feeding RNAi. (B) The pattern of blister formation in rde-4(-) animals with rde-4(+) expressed in the intestine (Psid-2) was examined (step I) and animals following the stereotyped order of susceptibility to bli-1 RNAi (a > b > ? > h) were removed (step II). The remaining animals, which show variant susceptibility to bli-1 feeding RNAi, were aggregated (step III and IV) and collapsed into a heat map (step V) to assess frequency of variant blisters in each section. Grey bounding box (step II-step IV) indicates the total number of worms that showed blister formation in each strain (n = 50 gravid adult animals) (C) The pattern of bli-1 silencing in rde-4(-) animals with RDE-4 expressed from neuronal promoters (Prgef-1 & Punc-119) or body-wall muscle promoters (Pmyo-3 & Punc-54) were examined and variant blisters were isolated as in (B) (n = 50 adult animals). (D) The pattern of blisters that result from silencing of bli-1 in rde-4(-) hypodermis are characteristic of the tissue that expresses RDE-4. Patterns of blister formation in response to ingested bli-1 RNAi were examined in wild-type animals or in rde-4(-) animals that express RDE-4 under 63 neuronal promoters (Prgef-1 or Punc-119, orange) or under body wall muscle promoters (Pmyo-3 or Punc-54, magenta). For each strain, sections with full (black) or partial (grey) blister formation in animals that showed a variation in the order of susceptibility (i.e. were not a > b > ? > h) were plotted. Grey bounding box and n are as in (B). (E) The patterns of blisters that result from silencing of bli-1 in rde-4(-) hypodermis are different from the consensus blister pattern. Aggregate patterns of blister formation among animals that deviate from the consensus susceptibility order (consensus bli-1 RNAi susceptibility in (Fig. 2-9F)) for each strain (variant susceptibility, % variants) are shown. All strains being compared were normalized together (black, section with highest frequency of blisters in all strains; white, section with the lowest frequency of blisters in all strains). Schematic of worms indicate locations of variant blisters (thick black shading) on worms with rde-4(+) expressed in neurons (orange) or in body-wall muscles (magenta). (F) Silencing by short dsRNA is not inhibited by the expression of a repetitive transgene. Wild-type animals or rde-4(-) animals with rde-4 rescues in the muscle (rde-4(-); Is(Pmyo-3::rde-4)) with or without a repetitive transgene expressed in the skin (Pnas-9::gfp) were fed dsRNA against skin genes, dpy-7 or bli-1. Error bars indicate 95% CI and asterisks indicate p<0.01 (compared to wild-type animals). 2.4.6 Using repetitive transgenes to express RDE-4 in one tissue and SID-1 in another tissue fails to provide support for the movement of short dsRNAs between cells Consideration of the following observations on repetitive transgenes leaves open the possibility that misexpression of rde-4(+) in somatic tissues explains the silencing in all somatic tissues despite the use of tissue-specific promoters. First, the formation of a repetitive transgene can generate rearrangements that result in novel promoter elements (169,179,180) that could lead to misexpression despite the use of well-characterized tissue-specific promoters. In support of this possibility, animals expressing gfp in all somatic tissues (Peft-3::gfp) showed some silencing in non- muscle cells even in rde-1(-); Ex[Pmyo-3::rde-1(+)] animals (Fig. 2-11). Second, repetitive transgenes can be selectively silenced within the germline (177,181), potentially explaining the observed lack of silencing in the germline of animals with 64 tissue-specific rde-4 rescue (Fig. 2-6 and Fig. 2-7). Third, because repetitive transgenes could be expressed at low levels within the germline (e.g. heat shock promoter (131), and because feeding RNAi in all somatic tissues was observed in rde- 4(-) progeny of heterozygous parents (98) we hypothesize that inherited RDE-4 from misexpression in the germline of parents could be responsible for feeding RNAi in progeny. While such silencing enabled by inherited RDE-4 was not detectable in most cases, it was detectable when unc-22 silencing was examined in rde-4(-) progeny of animals with rde-4 rescued using the myo-3 promoter (Fig. 2-12). Fig. 2- 11: Expression of RDE-1 in one somatic tissue can enable silencing of Peft-3::gfp in other mutant somatic tissues. Representative images of animals with gfp expression (black) in all somatic cells (Peft-3::gfp) in a wild-type background (A and C), rde-1(-) background (A), or rde- 1(-) background with rde-1(+) expressed in body-wall muscles (Ex[Pmyo-3::rde- 65 1(+)) (B and C). Animals were fed control RNAi or gfp RNAi for one generation (A & B) or for two generations (C). Insets are brightfield images and scale bar = 50 ?m. Fig. 2- 12: Expression of RDE-4 in the somatic tissues of parents does not typically enable feeding RNAi in rde-4(-) progeny. (A-D) Progeny of wild-type animals, rde-4(-) animals, or rde-4(-) animals expressing RDE-4 in the muscle (Pmyo-3 or Punc-54), hypodermis (Pnas-9), or intestine (Psid- 2) were fed dsRNA (F1-only feeding RNAi) against unc-22, unc-54, act-5, or bli-1 66 and the fractions of animals that showed silencing (fraction silenced) were determined. Error bars indicate 95% CI and asterisks indicate p<0.01 (animals without array compared to rde-4(-) animals). (E) Parental expression of RDE-4 from an integrated Pmyo-3::rde-4(+) transgene also enables feeding RNAi of unc-22 in rde-4(-) progeny. The rde-4(+) progeny (Is[Pmyo-3::rde-4(+)]/?) and rde-4(-) progeny (?/?) of parent animals expressing RDE-4 from an integrated repetitive array in the muscle (Is[Pmyo-3::rde-4(+)]/?)) were subjected to feeding RNAi (F1 RNAi) of unc-22 and the fractions of animals that showed silencing were determined (fraction silenced). 100% of wild-type and 0% of rde-4(-) control animals showed silencing of unc-22. Error bars indicate 95% CI and n>15 animals. 67 We attempted to obtain additional evidence for the movement of short dsRNAs derived from ingested dsRNA. Tissues that express high levels of SID-1 act as sinks for dsRNA (149) and the entry of both long dsRNA and short dsRNAs generated upon processing by RDE-4 are expected to require SID-1. Therefore, if no silencing occurs when processing by RDE-4 is restricted to one tissue but import through SID-1 (130,138,182)is restricted to another tissue, it would provide strong support for the transport of short dsRNAs processed from ingested dsRNAs between tissues. We generated sid-1(-); rde-4(-) animals in which rde-4 was rescued using the neuronal promoter Prgef-1 and sid-1 was rescued using the body-wall muscle promoter Pmyo-3 (Fig. 2-13A). We observed silencing in these animals upon feeding RNAi of the body-wall muscle gene unc-54 (Fig. 2-13B). Consistent with the restriction of SID-1-dependent dsRNA entry into body-wall muscle cells, we did not detect any silencing of the hypodermal gene bli-1 upon feeding RNAi. However, these results leave open several possibilities. If misexpression occurred, then these results are consistent with either rde-4(+) misexpression in muscle cells or sid-1(+) misexpression in neurons but not the hypodermis. Misexpression of rde-4(+) implies that the direct entry of long dsRNA into each cell is sufficient to explain the observed silencing. Selective misexpression of sid-1(+) implies that entry of long dsRNA into neurons and subsequent production of short dsRNAs in neurons followed by their entry into the body-wall muscles caused silencing. Finally, even if there was no misexpression, the interpretation of results from this experiment could be complicated by unexpected interactions between the two repetitive transgenes used. 68 Fig. 2- 13: Misexpression of RDE-4 may be sufficient to explain silencing by short dsRNA upon feeding RNAi. (A) Expected outcomes of test to distinguish movement of short dsRNAs from other possibilities (e.g. misexpression of RDE-4) in animals with tissue-specific rescue of RDE-4 from a repetitive transgene. (Left) If RDE-4 is expressed only in neurons from the rgef-1 promoter (Prgef-1::rde-4(+)) in a rde-4(-); sid-1(-); Ex[Pmyo-3::sid-1(+)] background no silencing is expected in muscles, which can import dsRNA (have SID- 1) but not process dsRNA (lack RDE-4). (Right) If RDE-4 is expressed in neurons and in muscles from the rgef-1 promoter (Prgef-1::rde-4(+)) in a rde-4(-); sid-1(-); Ex[Pmyo-3::sid-1(+)] background silencing can occur in muscles, which can import dsRNA (have SID-1) and process dsRNA (have RDE-4). See text for additional possibilities. (B) RDE-4 is likely misexpressed in muscles when expressed under the neuronal rgef-1 promoter from a repetitive transgene. Wild-type animals, mutant animals (rde-4(-), sid-1(-), or rde-4(-);sid-1(-)) or mutant animals with rde-4(+) and/or sid-1(+) expressed from repetitive transgenes (as schematized in (A)) were fed dsRNA against a gene expressed in the body-wall muscle (unc-54) or in the hypodermis (bli-1) and the fractions of animals that showed silencing (fraction Unc or fraction Bli) were determined. No silencing was observed in rde-4(-) or in sid-1(-) animals. Error bars indicate 95% CI, n>24 animals and asterisks indicate p<0.01 (compared to wild-type animals). 2.4.7 Single-copy transgenes can restrict RDE-4 activity to specific somatic tissues. Because the formation of a repetitive transgene can generate rearrangements (169,179,180) that complicate interpretations, we used single-copy transgenes to re- examine if the expression of RDE-4 in one tissue could enable silencing in other somatic tissues. We expressed RDE-4 from a single copy transgene using the myo-3 promoter (Si[Pmyo-3::rde-4(+)]) or the nas-9 promoter (Si[Pnas-9::rde-4(+)]) in rde-4(-) animals. In both cases, silencing was restricted to the intended tissue with 69 RDE-4 expression (Fig. 2-14A). These results suggest that upon expressing RDE-4 using a single copy transgene short dsRNAs made in one tissue are not sufficient to cause silencing in another tissue. Furthermore, these results also do not support the possibility that RDE-4 protein or mRNA moves between cells, suggesting that RDE-4 can provide restricted function within the cells where it is made (Fig. 2-14B). Thus the apparent intercellular transport of short dsRNAs seen in experiments using repetitive transgenes could be simply because of high levels of expression achieved using such transgenes or may have required some features of repetitive transgenes. Fig. 2- 14: Ingested dsRNA probably enters every cell to cause silencing in the soma (A) Expression of RDE-4 from a single-copy transgene reveals a requirement for RDE-4 within a tissue for silencing by ingested dsRNA in that tissue. Wild-type animals or rde-4(?) animals that express RDE-4 from a single-copy transgene in the body-wall muscle (Si[Pmyo-3::rde-4(+)]) or in the hypodermis (Si[Pnas-9::rde- 4(+)]) were fed dsRNA against unc-22 (magenta), act-5 (blue) or bli-1 (green). No silencing was observed in rde-4(-) or in sid-1(-) animals. Error bars indicate 95% CI, n>24 animals and asterisks indicate p<0.01 (compared to wild-type animals). (B) Model: Silencing by entry of ingested long dsRNA into each tissue could account for tissue-restricted silencing in animals with single-copy rescue of RDE-4. 2.5 Discussion We have shown that expression from repetitive transgenes within a tissue can selectively inhibit feeding RNAi in that tissue. While repetitive transgenes enabled 70 silencing in by RDE-4 in unrescued tissues, we have generated single-copy transgenes that can restrict RDE-4 activity to specific tissues. 2.5.1 Challenges and limitations in interpreting experiments that use repetitive transgenes. Partial characterization of the molecular structure of repetitive transgenes in C. elegans by Southern blotting (179) or Illumina sequencing (169) reveal that they can be made of >100 copies with at least some copies including complex rearrangements. These rearrangements can create unique sequences in promoter regions and within coding regions. When promoter or 3? UTR sequences are changed, it can result in the expression of a gene in unintended tissues even when well- characterized promoters were used to make the repetitive transgene. When coding sequences are changed, it can result in the production of new variants of a protein with different biochemical properties that can result in unpredictable functions. Repeats are associated with heterochromatin factors and histone modifications, some of which can also occur upon feeding RNAi at the target gene (e.g. H3K9me3 (183)). Recent reports suggest that loss of H3K9me3 can cause RNA:DNA hybrid-associated instability of the repeats (174). This possibility could alter inferences about the repetitive transgene-dependent inhibition observed for feeding RNAi of some genes. For example, the bli-1 silencing observed in eri-1(-) mutants despite expression from repetitive transgenes could be associated with changes in DNA and/or chromatin. Furthermore, when two repetitive transgenes are used, chromatin factors could be titrated from one transgene by the other, resulting in changed levels of expression. 71 2.5.2 Efficiency of RNAi could be regulated by expression from repetitive DNA Our discovery that expression from repetitive DNA within a tissue can interfere with silencing by feeding RNAi within that tissue could impact studies that use RNAi to infer the function of a gene. For example, RNAi of a gene in strains that express fluorescent reporters within a tissue from a repetitive transgene could be specifically inhibited in that tissue. While many RNAi screens have successfully identified novel components of biological processes using strains that express repetitive transgenes (e.g. (119,174,183), it may be possible to identify additional components if our results are also taken into account in the design of future screens. The efficiency of feeding RNAi differs between tissues and is a concern for the application of feeding RNAi to combat animal pests (78). For example, in C. elegans, genes expressed in neurons are relatively refractory to silencing by feeding RNAi (noted in (184)). One reason for such reduced silencing could be that neurons have high levels of expression from endogenous repetitive DNA. Consistent with this possibility, both silencing in tissues with expression from repetitive DNA and silencing in neurons are enhanced upon loss of the exonuclease ERI-1 (164) or the RdRP RRF-3 (97). Similarly, tissue-specific expression from endogenous repetitive DNA could explain differential sensitivity to RNAi among insect tissues. 2.5.3 Silencing by feeding RNAi can be restricted to a tissue with RDE-4 The ability of dsRNA expressed in one cell to cause SID-1-dependent silencing in other cells (100,130) revealed that dsRNA or its derivatives can be exported from cells, and be imported into cells, in C. elegans. Observations with 72 repetitive transgenes presented here and in previous studies (93,154) are consistent with a ?transit? model for feeding RNAi where dsRNA first enters the cytosol of a cell, is subsequently processed within the cytosol of that cell, and finally exported for silencing in distant cells. However, results from strains with single-copy rescues of rde-4 that we have generated suggest that in some cases cell-autonomous processing of long dsRNA by RDE-4 could be sufficient to account for silencing upon feeding RNAi. Specifically, because dsRNA can be transported across intestinal cells without entry into the cytosol (100,145,149) and reach the pseudocoelomic fluid that bathes all C. elegans tissues, the direct entry of dsRNA into all cells that show silencing and subsequent processing by RDE-4 in each cell could be sufficient to explain the systemic response to feeding RNAi in these strains. Due to the lack of better methods, previous studies have similarly used repetitive transgenes to rescue genes in a tissue-specific manner and have made inferences based on their activity within that cell and in other cells. With improvements in techniques, these studies may need to be reevaluated to eliminate effects due to unintended misexpression of repetitive transgenes. For instance, to understand transport of dsRNA in C. elegans, using rescues from repetitive transgenes, it was inferred that dsRNA can move from the lumen of the intestine and across the intestine without entering the cytoplasm of the intestine to distant tissues using transcytosis. This was done using repetitive transgenes to rescue the dsRNA specific importer SID-1 in non-intestine tissues, however, the misexpression of SID-1 in the intestine could have led to entry of dsRNA into the intestine and then into distant tissues. 73 Taken together with recent studies, our results suggest that several characteristics of feeding RNAi in many insects and parasitic nematodes (78,185,186) could be similar to those in C. elegans. First, a long dsRNA binding protein, Staufen was found to be essential for RNAi in Coleopteran insects (187). Lepd-SL1 cells from the coleopteran insect, L. decemlineata that displayed a resistance to RNAi, were found to have low levels of Staufen. Second, long dsRNA (>60 bp) is preferentially ingested (64) and realization of this preference was crucial for developing plastid expression as an effective strategy to deliver long dsRNA into crop pests (75). Third, dsRNA can be detected in intestinal cells and in internal tissues upon feeding RNAi (188). Next, with the exception of dipteran insects, most invertebrates have homologs of the dsRNA importer SID-1 (150). Finally, silencing initiated by feeding RNAi can persist for multiple generations (189). These similarities suggest that insights gleaned using the tractable animal model C. elegans are likely to be applicable to many invertebrates, including agronomically important insect and nematode pests. 74 Chapter 3: Epigenetic recovery typically enables rescue from induced transgenerational silencing. 3.1 Preface The work in this chapter was done with two undergrads that I mentored during graduate school- Farida Eteffa and Yixin Lin and some of the work was done by a fellow lab member, Sindhuja Devanapally. Farida Eteffa and I generated the data in Figs. 3-2B, 3-3E&H, 3-9 and 3-10A. Farida Eteffa generated the remaining data for Fig. 3-3. Yixin Lin generated data in Fig. 3-6C. Sindhuja Devanapally (Jose lab) generated all data in Fig. 3-5 and some data in Fig. 3-6B&C, Fig. 3-8A (See Figure legend for details). All other data was generated by me. Some worm strains were obtained from the Caenorhabditis elegans Genetic stock Center, the Seydoux laboratory (Johns Hopkins University), the Cohen-Fix laboratory (National Institute of Health) and the Fire lab (Stanford University). Dr. Henrik Bringmann (Max Planck Institute) provided us with information regarding strains with GPR-1 overexpression (Fig. 3-7). The Hamza laboratory (University of Maryland) provided bacteria that express dsRNA against gfp. 3.2 Introduction Increasing evidence in many organisms including C. elegans suggests that environmental stimuli can have consequences that can persist in descendants (Also see General Introduction). These consequences are likely due to changes in gene 75 expression. For effects to persist across generations, changes must be transmitted to progeny through the germ cells (or the germline). This suggests that environmental stimuli are either affecting gene expression in the germline directly or information is being communicated from the soma to the germline and consequently to progeny (Fig. 1-2). However, in these studies complex environmental stimuli such as alterations to diet or endocrine disruption (See introduction) may be generating a cascade of effects making it hard to tease apart direct effects from indirect effects. Manipulating the expression of a single gene can help us clearly decipher mechanisms that perpetuate patterns of gene expressions across generations. One way to achieve targeted manipulation of gene expression is to use dsRNA to silence a gene of matching sequence and observe the persistence of these changes across generations (57). Work from many labs have shown that different genes can be successfully targeted for silencing by small RNAs (Fig. 3-1) and silencing can persist across generations for some targets. While silencing of a target in the germline by injection or ingestion of dsRNA can be due to the direct uptake of dsRNA from the body cavity into the germline, silencing of the germline by expression of dsRNA in a somatic cell (neurons) suggests that dsRNA can move from soma to the germline (94). The persistence of silencing of genes expressed in the soma is rare (Fig. 3-1) and even for germline genes, silencing isn?t always inherited in descendants. Furthermore, fluorescently labeled dsRNA injected into the body cavity can be directly packaged into oocytes suggesting that at least in the first generation after exposure, dsRNA can be directly deposited into progeny (98,99). Therefore, true transgenerational silencing 76 must persist for more than a few generations and this likely engages mechanisms that actively preserve these changes in every generation. Mechanisms that enable such silencing across generations could also inform how organisms spatially and temporally regulate gene expression patterns in an animal, enabling expression of genes in some tissues at some times but not others. Fig. 3- 1: Transgenerational silencing is not observed for all genes. Summary of studies that have analyzed inheritance of silencing for 1 or more generations. Here we show that the same sequence in different genetic contexts can show differential susceptibility to persistence of silencing across generations. This suggests that for the same sequence, mechanisms that promote and prevent persistence of changes may be engaged. Enhancing silencing was not sufficient to change the susceptibility of genes to silencing. Using a paradigmatic gene that could be silenced 77 transgenerationally by expressed dsRNA (94), by feeding RNAi (this thesis) and simply by mating (190), we studied the mechanisms that maintain silencing for >200 generations at this gene. Persistence of silencing was not correlated with tested repressive histone marks and required the secondary Argonaute HRDE-1 indicative of 2? small RNAs being required for silencing in every generations. We use genetic manipulation to show that diffusible parental signals independent of DNA can alter gene expression states in progeny. Lastly, we conducted a genetic screen on animals that were silenced for >200 generations and isolated mutants that exhibited Re Activation of Gene Expression (Rage). These mutants additionally displayed physical defects that could reveal endogenous processes that rely on transgenerational silencing in C. elegans. 3.3 Materials and methods 3.3.1 Short hand used for referring to some transgenes For ease, we use short hand notations to refer to some genes in this thesis as below (Also see Fig. 3-5). 1. T refers to the single-copy recombinant gene, Pmex-5::mCherry::H2B::tbb- 2:3'UTR::gpd-2 operon::GFP::H2B::cye-1 3'UTR, which expresses mCherry and GFP in the germline under the mex-5 promoter. Both proteins are detected within the nucleus since they are tagged to the H2B coding sequence. This gene is likely an operon so mCherry and gfp are probably transcribed as a single pre-mRNA. Lastly, T specifically refers to this gene when it is expressed in the animal and not silenced. 78 2. iT refers to the above transgene once it has been silenced by mating-induced silencing for >150 generations, where i stands for inactive. 3. T??? is the truncated form T (Pmex-5::mCherry::cye-1 3'UTR) that was generated using Cas-9 mediated genome editing to delete sequences from T (some flanking regions were deleted as well). This transgene expresses mCherry in the cytoplasm of the germline. 4. Tcherry refers to a recombinant gene that was re-created as a new strain to express the same sequence as T??? (Pmex-5::mCherry::cye-1 3'UTR) and also expresses mCherry in the cytoplasm of the germline. 5. iTcherry refers to the above transgene once it has been silenced by mating-induced silencing for >5 generations, where i stands for inactive. 6. gpr-1 oe refers to the expression of gpr-1 from a single-copy transgene in the germline (Pmex-5::gpr-1::smu-1 3?UTR) in addition to the endogenous gpr-1. 3.3.2 Strains and oligonucleotides All strains were cultured on Nematode Growth Medium (NGM) plates seeded with 100?l OP50 at 20?C and mutant combinations were generated using standard methods (155). Strains used in this chapter are listed in Table 3-1 and oligonucleotides used in this chapter are listed in Table 3-2. All oligonucleotides were ordered from Integrated DNA Technologies (IDT) uncles otherwise mentioned. Table 3- 1: Strains used. Strain Name Genotype AMJ174 oxSi487[Pmex-5::mCherry::H2B::tbb-2:3'UTR::gpd-2 operon::GFP::H2B::cye-1 3'UTR + unc-119(+)] dpy-2(jam29) II; 79 unc-119(ed3)? III; lin-2(jam30) X AMJ471 jamEx140 [Prgef-1::gfp-dsRNA:: unc-54 3?UTR & Pmyo- 2::DsRed::unc-54 3?UTR] AMJ506 prg-1(tm872) I; oxSi487 II; unc-119 (ed3)? III AMJ552 oxSi487 dpy-2(jam33) II; unc-119(ed3)? III AMJ577 hrde-1(tm1200) III [4x] AMJ587 mut-2(jam9) I. AMJ675 oxSi487 II; unc-119(ed3)? hrde-1(tm1200) III AMJ685 K08F4.2::gfp [Pgtbp-1::gtbp-1::gfp] IV; jamEx140 AMJ819 K08F4.2::gfp eri-1(mg366) IV AMJ842 K08F4.2::gfp eri-1(mg366) IV; jamEx140 AMJ844 oxSi487 dpy-2(e8) II; unc-119(ed3)? III AMJ928 jamSi27[Pmex-5::mCherry::cye-1 3'UTR] II AMJ930 dpy-10(jam68) II AMJ975 sur-5(jam79[sur-5::gfp]) X AMJ1021 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(- ) (mutant l in Fig. 3-10) AMJ1023 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant g in Fig. 3-10) AMJ1025 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant m in Fig. 3-10) AMJ1031 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) 80 (mutant n in Fig. 3-10) AMJ1032 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant o in Fig. 3-10) AMJ1034 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant a in Fig. 3-10) AMJ1035 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant h in Fig. 3-10) AMJ1036 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant d in Fig. 3-10) AMJ1038 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant b in Fig. 3-10) AMJ1042 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant i in Fig. 3-10) AMJ1043 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant c in Fig. 3-10) AMJ1044 dpy-2(jam38) II; hrde-1(tm1200) unc-93(jam48) III AMJ1091 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant j in Fig. 3-10) AMJ1094 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant f in Fig. 3-10) AMJ1097 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) (mutant e in Fig. 3-10) AMJ1098 oxSi487 dpy-2(jam29) II; unc-119(ed3)? III; lin-2(jam30) X; rage(-) 81 (mutant k in Fig. 3-10) AMJ1116 oxSi487 dpy-2(e8) II; unc-119(ed3)? III; met-2(n4256) III AMJ1117 oxSi487 dpy-2(e8) II; unc-119(ed3)? III; met-2(n4256) III AMJ1118 oxSi487 dpy-2(e8) II; unc-119(ed3)? III; met-2(n4256) III AMJ1126 mut-16(pk710) I; oxSi487 dpy-2(e8) II; unc-119(ed3)? III AMJ1127 mut-16(pk710) I; oxSi487 dpy-2(e8) II; unc-119(ed3)? II AMJ1128 mut-16(pk710) I; oxSi487 dpy-2(e8) II; unc-119(ed3)? III AMJ1135 mut-2(jam9) I; oxSi487 dpy-2(e8) II; unc-119(ed3)? III AMJ1136 mut-2(jam9) I; oxSi487 dpy-2(e8) II; unc-119(ed3)? III AMJ1137 met-2(n4256) III ; K08F4.2::gfp IV AMJ1138 met-2(n4256) III ; K08F4.2::gfp IV AMJ1139 met-2(n4256) III ; K08F4.2::gfp IV AMJ1142 oxSi487 dpy-2(e8) II; unc-119(ed3)? III; pgl-1(ct131) IV him- 3(e1147) IV? AMJ1143 oxSi487 dpy-2(e8) II; unc-119(ed3)? III; pgl-1(ct131) IV him- 3(e1147) IV? AMJ1157 oxSi487 dpy-2(jam33) II; unc-119(ed3)? III; rde-8(jam75) IV AMJ1158 oxSi487 dpy-10(jam82) II dpy-2(jam33) II; unc-119(ed3)? III; rde- 8(jam76) IV AMJ1170 jamSi37 II [Pmex-5::mCherry::cye-1 3'UTR + unc-119(+)]; unc- 119(ed3) III AMJ1174 dpy-10(jam106) jamSi37 [Pmex-5::mCherry::cye-1 3'UTR] II; unc- 82 119(ed3) III AMJ1176 jamSi27 II; K08F4.2::gfp IV AMJ1186 jamSi37 II; unc-119(ed3)? III AMJ1206 set-32(jam46) I; oxSi487 dpy-2(e8) II; unc-119(ed3)? III AMJ1207 oxSi487 dpy-2(e8) II heri-1(jam47) II; unc-119(ed3)? III AMJ1236 jamSi37 II; unc-119(ed3?) III; K08F4.2::gfp IV AMJ1238 dpy-10(jam106) jamSi37 II AMJ1240 dpy-10(jam106) jamSi37 II; ccTi1594 [mex-5p::GFP::gpr-1::smu-1 3'UTR + Cbr-unc-119(+), III: 680195] unc-119(ed3?) III AMJ1267 dpy-10(jam106) jamSi37 II; ccTi1594 unc-119(ed3?) III AMJ1268 dpy-10(jam106) jamSi37 II; ccTi1594 unc-119(ed3?) III EG6787 oxSi487 II; unc-119 (ed3) III GR1373 eri-1(mg366) IV JH3197 K08F4.2::gfp IV JH3270 pgl-1::gfp C-term MT13293 met-2(n4256) III NL1810 mut-16(pk710) I OCF62 jfSi1[Psun-1::GFP + cb-unc-119(+)] II; ltIs38[(pAA1) pie- 1::GFP::PH(PLC1delta1) + unc-119(+)] PD1594 ccTi1594 unc-119(ed3) III SS2 pgl-1(ct131) him-3(e1147) IV. TX189 unc-199(ed3) III; teIs1 [(pRL475) oma-1p::oma-1::GFP + 83 (pDPMM016) unc-119(+)] Table 3- 2: Oligonucleotides used (5? to 3?, IDT). Primer Sequence P1 GGAACATATGGGGCATTCG P2 CAGACCTCACGATATGTGGAAA P3 CACAAATCCGAACTATCTGAC P4 CTTGGTGTCGTAAACTTTCTG P5 ATCTGAGCAGCATTCATCTTC P6 CATTTGTGCATTTCCTTCCA P7 ATGCTTGTGAAATCCGGGTA P8 TCGGGAGACAGCATCATTTG P9 TCCTATAACATTGACGCGCAC P10 TCGATGACGACAAGAAGCTC P11 AAAGACTTCCGCTCTGACAC P12 GGGATACAGTCTTCAAGTAAC P13 AAACGCACTGTTCTTCGGATC P14 TTCTGGATACTCCTCGGATG P15 ATTTAGGTGACACTATAGGATTACTCATAATGACATGGTTTT AGAGCTAGAAATAGCAAG P16 AAAAGCACCGACTCGGT P17 ATTTAGGTGACACTATAGCGTTGGTGATGGTGATGAGGTTTT AGAGCTAGAAATAGCAAG 84 P18 ATTTAGGTGACACTATAGCTACCATAGGCACCACGAGGTTTT AGAGCTAGAAATAGCAAG P19 ATCTGATTATTATATTTCAGATTACTCATAATTAATGTATTCA ATTTGTTAATATATTTC P20 CACTTGAACTTCAATACGGCAAGATGAGAATGACTGGAAAC CGTACCGCATGCGGTGCCTATGGTAGCGGAGCTTCACATGGC TTCAGACCAACAGCCTA P21 TATTTGATCCTAAGCGACGTG P22 ACCAACTTGATGACACTACTG P23 ATACAGTAAGGCGATTGTGAG P24 ATTTAGGTGACACTATAGTGCTTCGATAGATCTCGAGGTTTT AGAGCTAGAAATAGCAAG P25 ATTTAGGTGACACTATAGTTCAGCTTACAATGGACTAGTTTT AGAGCTAGAAATAGCAAG P26 TTAATTCTTAACAAAAAACTGTTTCCGCTCCTACGGATACAA CTACATGAAAAATCATCT P27 TGGATTCTCCTGCTCTGAAG P28 GAGGTTCGCAGATGTTCTTG P29 ATTTAGGTGACACTATAGAGTAGTTACTGATGAGCTGGTTTT AGAGCTAGAAATAGCAAG P30 ATTTAGGTGACACTATAGTCGAGCTGTAGGCTCTTGGGTTTT AGAGCTAGAAATAGCAAG 85 P31 GCUACCAUAGGCACCACGAGGUUUUAGAGCUAUGCU P32 GAGAGATTCAAAAGAACAAAAAAGCCGCAGAGAGCCTACA GCTCGATCTGTAGAGTGTTT P33 TCATCGACCATGTCAGGTAC P34 CTTCTTCGGCAAACGGATTC P35 CAATATCCACAACACGCTCG P36 ATTTAGGTGACACTATAGAAATGCTCAGAGATGCTCGGTTTT AGAGCTAGAAATAGCAAG P37 CTGCCGATTCACCACAATTTC P38 ATCCATCTTCTCCAGGCATAC P39 GGACCACGTGGAGTTCCAGGACATCCAGGTTTTCCAGGTGAC CCAGGAGAGTATGGAATT P40 CGTCTCTTGATATTCCTTGC P41 CAAGCGAATGGAAGTGGTCCT P42 TCACATACACATCTTCTGCACC P43 TTGGTAGAAGCTGCATCACTTT P44 GAGATTCAAGGTCCACATGGAGG P45 ATGGAAGTGGTCCTCCCTTGG P46 TCTTCGGCGCTAATCTTTTC P47 CACGAGTTCGAGATCGAG P48 GGAAGCTGAAAATTTAAATAATCAG P49 TTCTGTCAGTGGAGAGGG 86 P50 GTGTTGGCTGAAAATTTAAATAAT 3.3.3 Transgenesis All PCRs for genotyping during strain making were performed with home- made Taq polymerase. To make Pgtbp-1::gtbp-1::gfp eri-1(-) animals To generate Pgtbp-1::gtbp-1::gfp eri-1(-) animals, Pgtbp-1::gtbp-1::gfp males were mated with eri-1(-) hermaphrodites. In the F1 generation, cross progeny hermaphrodites were isolated based on the presence of green fluorescence in all cells. F2 progeny with green fluorescence were singled out and were genotyped a day later for the eri-1(-) mutation by PCR and gel electrophoresis using primers P1 and P2. Homozygosity of Pgtbp-1::gtbp-1::gfp was determined based on the presence of green fluorescence in F3 progeny. Plates with all F3 progeny showing green fluorescence was indicative of the F2 parent being homozygous for Pgtbp-1::gtbp- 1::gfp. F3 progeny from F2 parents that were Pgtbp-1::gtbp-1::gfp eri-1(-) were singled out and their genotypes were reconfirmed using the same approaches described above. One line was isolated for analysis and designated as AMJ819. To make met-2(-); Pgtbp-1::gtbp-1::gfp animals To generate met-2(-); Pgtbp-1::gtbp-1::gfp animals, Pgtbp-1::gtbp-1::gfp males were mated with met-2(-) hermaphrodites. In the F1 generation, cross progeny hermaphrodites were isolated based on the presence of green fluorescence in all cells. F2 progeny with green fluorescence were singled out and were genotyped a day later for the met-2(-) mutation by PCR and gel electrophoresis using primers P3, P4, and 87 P5. Homozygosity of Pgtbp-1::gtbp-1::gfp was determined based on the presence of green fluorescence in F3 progeny. Plates with all F3 progeny showing green fluorescence was indicative of the F2 parent being homozygous for Pgtbp-1::gtbp- 1::gfp. F3 progeny from F2 parents that were Pgtbp-1::gtbp-1::gfp eri-1(-) were singled out and their genotypes were reconfirmed using the same approaches described above. Three lines were isolated from different F1 parents for analysis and designated as AMJ1137, AMJ1138, and AMJ1139. To make Pgtbp-1::gtbp-1::gfp; T??? animals To generate Pgtbp-1::gtbp-1::gfp; T??? animals, Pgtbp-1::gtbp-1::gfp males were mated with jamSi27 (T???) hermaphrodites. In the F1 generation, cross progeny hermaphrodites were isolated based on the presence of green fluorescence in all cells. F2 progeny with green fluorescence and red fluorescence in the germline were singled out. Homozygosity of both transgenes was determined based on the presence of green and red fluorescence in F3 progeny. Plates with all F3 progeny showing green fluorescence were indicative of the F2 parent being homozygous. F3 progeny were singled out and their genotypes were reconfirmed using the same approaches described above and imaging a representative population (>25 animals). Animals used from this cross were designated as AMJ1176. To make Pgtbp-1::gtbp-1::gfp; Tcherry animals To generate Pgtbp-1::gtbp-1::gfp; Tcherry animals, Pgtbp-1::gtbp-1::gfp males were mated with jamSi37 (Tcherry) hermaphrodites. In the F1 generation, cross progeny hermaphrodites were isolated based on the presence of green fluorescence in all cells. F2 progeny with green fluorescence and red fluorescence in the germline 88 were singled out. Homozygosity of both transgenes was determined based on the presence of green and red fluorescence in F3 progeny. Plates with all F3 progeny showing green fluorescence were indicative of the F2 parent being homozygous. F3 progeny were singled out and their genotypes were reconfirmed using the same approaches described above and imaging a representative population (>25 animals). Animals used from this cross were designated as AMJ1236. To make dpy-10(-) Tcherry; gpr-1 oe animals To generate dpy Tcherry; gpr-1 oe animals, ccTi1594 (gpr-1 oe) males were mated with dpy-10(-) jamSi37 (dpy-10(-) Tcherry- AMJ1174) hermaphrodites. In the F1 generation, cross progeny hermaphrodites were isolated based on their Rol phenotype, which is characteristic of dpy-10(+/-) animals. F2 progeny that were Dpy were singled out, allowed to have progeny for 24-48h and then imaged for the presence of green fluorescence (from ccTi1594) and red fluorescence (jamSi37) in the germline. Homozygosity of both transgenes was determined based on the presence of green and red fluorescence in F3 progeny by imaging. Animals used from this cross were designated as AMJ1240. To make dpy iTcherry; gpr-1 oe animals To generate dpy iTcherry; gpr-1 oe animals, ccTi1594 (gpr-1 oe) males were mated with dpy-10(-) jamSi37 (dpy-10(-) iTcherry- AMJ1238) hermaphrodites. In the F1 generation, cross progeny hermaphrodites were isolated based on their Rol phenotype, which is characteristic of dpy-10(+/-) animals. F2 progeny that were Dpy were singled out, allowed to have progeny for 24-48h and then imaged for the presence of green fluorescence (from ccTi1594) and genotyped for the presence of 89 Tcherry since this would be off. Homozygosity was determined based on the presence of green fluorescence by imaging and genotyping for iTcherry in F3 progeny. Animals were imaged to ensure iTcherry was OFF and two lines were frozen as AMJ1268 and AMJ1267. To make mut-2(-); T dpy-2(-) To generate mut-2(-); T dpy-2(-) animals, mut-2(-) males were mated with iT dpy-2 (AMJ844) hermaphrodites. In the F1 generation, cross progeny hermaphrodites that would be nonDpy were isolated since dpy-2(+/-) animals look wild-type. Homozygosity of T was determined based on animals being Dpy since the mutation in dpy-2 is closely linked to T in the genome. F2 progeny that were Dpy were singled out and F3 progeny were pooled and genotyped for the mut-2(-) mutation by PCR and gel electrophoresis using primers P6 and P7. Plates with all Dpy F4 progeny were indicative of the F2 parent being homozygous for T dpy-2(-). F4 progeny from F3 parents that were mut-2(-); T dpy-2(-) were singled out and their genotypes were reconfirmed using the same approaches described above. Three lines were isolated from different F1 parents for analysis and designated as AMJ1135 and AMJ1136. Animals were scored after the F4 generation. To make T dpy-2(-); pgl-1(-) To generate T dpy-2(-); pgl-1(-) animals, pgl-1(-) males were mated with iT dpy-2 hermaphrodites. In the F1 generation, cross progeny hermaphrodites that would be nonDpy were isolated since dpy-2(+/-) animals look wild-type. Homozygosity of T was determined based on animals being Dpy since the mutation in dpy-2 is closely linked to T in the genome. F3 progeny were pooled and genotyped a day later for the 90 pgl-1(-) mutation by PCR and gel electrophoresis using primers P8 and P9. Plates with all Dpy F3 progeny were indicative of parents being homozygous for T dpy-2(-). F4 progeny from F3 parents that were T dpy-2(-); pgl-1(-) were singled out and their genotypes were reconfirmed using the same approaches described above. More worms were picked and genotyped with newly designed primers, P10 & P11. Two lines were isolated from different F2 parents for analysis and designated as AMJ1142, and AMJ1143. Animals were scored in after the F5 generation. To make mut-16(-); T dpy-2(-) To generate mut-16(-); T dpy-2(-) animals, mut-16(-) males were mated with iT dpy-2 hermaphrodites. In the F1 generation, cross progeny hermaphrodites that would be nonDpy were isolated since dpy-2(+/-) animals look wild-type. Homozygosity of T was determined based on animals being Dpy since the mutation in dpy-2 is closely linked to T in the genome. F2 progeny that were Dpy were singled out and were genotyped a day later for the mut-16(-) mutation by PCR and sequencing using primers P12-P14 (used for sequencing). Plates with all Dpy F3 progeny were indicative of the F2 parent being homozygous for T dpy-2(-). F3 progeny from F2 parents that were mut-16; T dpy-2(-) were singled out and their genotypes were reconfirmed using the same approaches described above. Three lines were isolated from different F1 parents for analysis and designated as AMJ1126, AMJ1127, and AMJ1128. Animals were scored after the F4 generation. To make met-2(-); T dpy-2(-) To generate met-2(-); T dpy-2(-) animals, met-2(-) males were mated with iT dpy-2 hermaphrodites. In the F1 generation, cross progeny hermaphrodites that would 91 be nonDpy were isolated since dpy-2(+/-) animals look wild-type. Homozygosity of T was determined based on animals being Dpy since the mutation in dpy-2 is closely linked to T in the genome. F2 progeny that were Dpy were singled out and were genotyped a day later for the met-2(-) mutation by PCR and gel electrophoresis using primers P3-P5. Plates with all Dpy F3 progeny were indicative of the F2 parent being homozygous for T dpy-2(-). F3 progeny from F2 parents that were met-2(-); T dpy-2(- ) were singled out and their genotypes were reconfirmed using the same approaches described above. Three lines were isolated from different F1 parents for analysis and designated as AMJ1116 AMJ1117, and AMJ1118. Animals were scored in F4, F6 and F8 generation from all 3 lines. 3.3.4 Genome editing All PCRs to generate sgDNA used to transcribe sgRNA for genome editing were performed with Phusion Polymerase (New England Biolabs?NEB), unless otherwise specified, according to the manufacturer?s recommendations. The final fusion products were purified using PCR Purification Kit (QIAquick, Qiagen). All PCRs for genotyping during strain making were performed with home-made Taq polymerase. Making mutants to test requirement of genes for the maintenance of transgenerational silencing To generate mutants, we performed genome editing using Cas-9. Mutant alleles to be made were designed such that dsDNA breaks were introduced at two far away locations. Homology templates were designed such that they contained homologous sequences on either side of the cuts to enable repair such that sequences 92 for essential domains in the protein were removed. To prepare guide RNAs, the scaffold DNA sequence was amplified from pPDD162. To make rde-8(-); T To generate rde-8 mutants, Cas9-based genome editing employing a co- conversion strategy was used (158). To prepare guide RNAs, the scaffold DNA sequence was amplified using primers P15 and P16 to target the first site (sgRNA 1) and P17 and P16 to target the second site (sgRNA 2) in rde-8. The scaffold DNA sequence was amplified using primers P18 and P16 to target dpy-10. The amplified DNA templates were purified (PCR Purification Kit, Qiagen) and transcribed (SP6 RNA polymerase, NEB). Homology templates used for repair of rde-8 (P19) and dpy- 10 (P20) were single-stranded oligos. iT (AMJ552) animals were injected with 8.1 pmol/?l of rde-8 sgRNA 1, 10.9 pmol/?l of rde-8 sgRNA 2, 6.9 pmol/?l of dpy-10 sgRNA, 13.5 pmol/?l of rde-8 homology repair template, 6.5 pmol/?l of dpy-10 homology repair template and 0.6 pmol/?l of Cas-9 protein (PNA Bio Inc). P0 animals were injected and F1 progeny that either showed Rol or Dpy were singled out suggestive of editing to the dpy-10 gene. Descendants of singled out F1 progeny were genotyped later for rde-8 mutations using primers P21-P23. Progeny that showed the presence of edits in rde-8 were isolated and were genotyped a day later. Progeny that showed homozygous edits in rde-8 upon genotyping were isolated and re-genotyped for rde-8(-)reconfirmed using the same approaches described above. Since the genotype of dpy-10 was not of consequence to our experiments, we did not select for a specific genotype of dpy-10. Two lines were isolated from 93 different F1 parents for analysis and designated as AMJ1157 and AMJ1158. Animals were scored after generations F5. To make set-32(-); T To generate set-32 mutants, Cas9-based genome editing employing a co- conversion strategy was used (137). To prepare guide RNAs, the scaffold DNA sequence was amplified using primers P24 and P16 to target the first site (sgRNA 1) and P25 and P16 to target the second site (sgRNA 2) in set-32. The scaffold DNA sequence was amplified using primers P18 and P16 to target dpy-10. The amplified DNA templates were purified (PCR Purification Kit, Qiagen) and transcribed (SP6 RNA polymerase, NEB). Homology templates used for repair of set-32 (P26) and dpy-10 (P20) were single-stranded oligos. iT animals were injected with 3.9 pmol/?l of set-32 sgRNA 1, 3.9 pmol/?l of set-32 sgRNA 2, 2.8pmol/?l of dpy-10 sgRNA, 7.50 pmol/?l of set-32 homology repair template, 7.50 pmol/?l of dpy-10 homology repair template and 0.6 pmol/?l of Cas-9 protein (PNA Bio Inc.). 12 P0 animals were injected and 98 F1 progeny that either showed Rol or Dpy were singled out suggestive of editing to the dpy-10 gene. F1s were allowed to lay progeny for two days and then genotyped for set-32 using primers P27 and P28. F2 progeny of F1 animals that showed the presence of edits in set-32 were isolated and were genotyped a day later. F3 progeny from F2 animals that showed homozygous edits in set-32 upon genotyping were isolated and re-genotyped for set-32 using the same approaches described above. Since the genotype of dpy-10 was not of consequence to our experiments, we did not choose for any specific genotype of dpy- 94 10. 1 line was isolated for analysis and designated as AMJ1206. Animals were scored in the F4 generation. To make heri-1(-); T To generate heri-1 mutants, Cas9-based genome editing employing a co- conversion strategy was used (137). To prepare guide RNAs, the scaffold DNA sequence was amplified using primers P29 and P16 to target the first site (sgRNA 1) and P30 and P16 to target the second site (sgRNA 2) in heri-1. The amplified DNA templates were purified (PCR Purification Kit, Qiagen) and transcribed (SP6 RNA polymerase, NEB). crRNA (P31) and tracr RNA (IDT) was used to edit the co- conversion marker dpy-10. Homology templates used for repair of heri-1 (P32) and dpy-10 (P20) were single-stranded oligos. iT animals were injected with 3.7 pmol/?l of heri-1 sgRNA 1, 3.7 pmol/?l of heri-1 sgRNA 2, 2.3 pmol/?l of dpy-10 crRNA, 2.7 pmol/?l tracrRNA, 7.5 pmol/?l of heri-1 homology repair template, 7.5 pmol/?l of dpy-10 homology repair template and 0.6 pmol/?l of Cas-9 protein (PNA Bio Inc.). 12 P0 animals were injected and 15 F1 progeny that either showed Rol or Dpy were singled out suggestive of co-editing to the dpy-10 gene. F1s were allowed to lay progeny for two days and then genotyped for heri-1 using primers P33-P35. F2 progeny of F1 animals that showed the presence of edits in heri-1 were isolated and were genotyped two days later. F3 progeny from F2 animals that showed homozygous edits in heri-1 upon genotyping were isolated and re-genotyped for heri- 1. Since the genotype of dpy-10 was not of consequence to our experiments, we did not choose for any specific genotype of dpy-10. One line was isolated for analysis and designated as AMJ1207. Animals were scored at the F2 and F3 generations. 95 Making a worm suitable for a genetic screen We wanted to conduct a forward genetic screen to isolate mutants that show a Re-activation of gene expression (Rage) after being silenced for >200 generations. Previous genetic analysis on these animals that were transgenerationally silenced has suggested that parental deposition of proteins required for maintenance of transgenerational silencing can enable persistence of silencing in mutant progeny (190). In order to capture such mutants in our screen as well, we decided to screen both F2 and F3 progeny of P0 mutagenized animals. However, in prior screens that I have performed, I have frequently observed that mutagenized animals are unable to continue to have progeny beyond two generations post mutagenesis due to starvation likely caused by overcrowding on the plate. Such starvation can alter gene expression and might hinder our ability to isolate mutants with restored expression of mCherry from T. We customized the strain used for our screen by generating a loss of function mutation in the lin-2 gene, which prevents egg laying causing a ?bag of worms? phenotype and restricting the brood size of each animal. This reduction in brood size would prevent overcrowding and enable us to isolate both F2 and F3 mutant progeny. To generate lin-2 mutants, Cas9-based genome editing employing a co-conversion strategy was used (137). To prepare guide RNAs, the scaffold DNA sequence was amplified using primers P36 and P16 to target lin-2 and primers P18 and P16 to target a marker gene dpy-2 by the co-conversion strategy. The amplified DNA templates were purified (PCR Purification Kit, Qiagen), transcribed (SP6 RNA polymerase, NEB) and tested in vitro with amplified dpy-2 template (P37 and P38) for cutting 96 efficiency (Cas-9, NEB). As the strain to be screened was in a dpy-2(-) background, a 60bp single-stranded DNA oligo was used as the homology template (P39) to repair the dpy-2 mutation to make it wild type and silent mutations were introduced into this template to prevent re-cutting by sgRNA after repair. Animals that had been silenced for over 200 generations were injected with 1.5 pmol/?l of Cas-9 protein, 7.4 pmol/?l of dpy-2 guide RNA, 6.2 pmol/?l of lin- 2 guide RNA, 0.6 pmol/?l of dpy-2 homology repair template. Injections were performed by Dr. Antony Jose. No homology template was used to generate the mutation in lin-2, as non-homologous end joining would likely generate a loss of function mutations. NonDpy F2 animals displaying the bag of worms phenotype were isolated and genotyped. Sequencing of isolated mutants revealed a 122bp insertion containing portions of the dpy-2 homologous repair template. Two lines were isolated and designated as AMJ174 and AMJ175. Sequencing of AMJ174 revealed repair of the dpy-2 mutation by the homologous repair template. In AMJ175, dpy-2 was probably repaired by non-homologous end joining indicated by the lack of silent mutations that were in the homologous repair template. AMJ174 was used to perform the genetic screen to isolate rage mutants. 3.3.5 Feeding RNAi and scoring associated defects Feeding RNAi: RNAi experiments were performed at 20?C on NGM plates supplemented with 1 mM IPTG (Omega Bio-Tek) and 25g/ml Carbenicillin (MP Biochemicals) (RNAi plates). Single-generation (P0 Feeding RNAi) 97 This feed was performed as in (98) and was used in all Figures with feeding RNAi except Fig. 3-10. Two generations (P0 & F1 Feeding RNAi) This feed was performed in the same way as in Chapter 2 (See 2.3.5). This feed was used in Fig. 3-10. Multigenerational feeds (i.e. P0-F2 Feeding RNAi) For this feed, P0 animals, their progeny (F1), and their grandprogeny (F2) were subjected to feeding RNAi. F1 and F2 animals were scored at larval stage 4 (L4) to assess potency of the RNAi food and then siblings were transferred to a new plate with RNAi food to prevent starvation. Similar to the P0 Feeding RNAi protocol (See Fig. 3-3C), F2 adults (24 hours post L4) were washed four times with M9 buffer to remove residual dsRNA and transferred to a plate with OP50. F3 progeny were then scored for inherited silencing effects. This feed was used in Fig. 3-3. Feeding RNAi to Identify Distinct Classes of Mutants To determine whether rage mutants contained mutations in different genes, two generation RNAi feeds against three different target genes was performed in the same way as in Chapter 2 (See Table 2-3 for expected defect and method of scoring). Three different genes with easily distinguishable defects upon silencing were selected based on their known expression pattern (soma vs. germline) and the RNAi components (nuclear vs. cytoplasmic) they have been previously observed to utilize as summarized in Table 3-3. Table 3- 3: Scoring of gene-specific silencing. dsRNA Food Location RNAi pathway 98 unc-22 Soma (Muscle) Cytoplasmic (147) bli-1 Soma (Skin) Nuclear (126) gfp Germline Nuclear (94) based on data from expressed dsRNA. 3.3.6 Expression of dsRNA To study inherited silencing, we expressed dsRNA from an extrachromosomal array that is mitotically unstable. Animals that express the array will have both progeny that inherit the array and those that don?t. We used an array expressing dsRNA in neurons from Ex[Prgef-1::gfp-dsRNA] (92) and evaluated progeny that don?t have the array to measure inherited silencing since parents were exposed to dsRNA from the array but progeny are not. This assay was used in Fig. 3-3. 3.3.7 Quantitative RT-PCR Total RNA was isolated using TRIzol (Fisher Scientific) from 50-100?l pellets of mixed-stage animals. 3 biological replicates were isolated by pelleting animals from 3 different plates of the same strain. RNA was extracted by chloroform extractions, precipitated using isopropanol, washed with Ethanol and resuspended in 20-30?l of nuclease-free water. 2?l of resuspended RNA was set aside to run on a gel and the remaining was DNase treated using homemade DNase buffer (100mM Tris- HCl, pH 8, 5mM CaCl2, 25mM MgCl2), and incubated with 0.25?l DNase I (NEB, 2units/?l) at 37?C for 60mins followed by heat inactivation and 75?C for 10mins. Pre- and Post- DNase treated RNA were run on a 1% agarose gel to check for the 99 presence of rRNA bands. RNA concentration was measured and equal amounts (500ng-1000ng) of RNA were converted to cDNA using SuperScript III Reverse Transcriptase (Invitrogen) according to the manufacturer?s recommendations. For cDNA conversion 3-5 technical replicates were done for each biological replicate of each sample. RT primer P40 was used for R11A8.1 and P41 for mCherry to convert mRNA. qRT-PCR was done on cDNA using LightCycler 480 SYBR Green I Mastermix (Roche) guidelines according to the manufacturer?s recommendations. Primers P42 and P43 were used for analysis of R11A8.1 and P44 and P45 were used for analysis of mCherry. Fold change was calculated using 2^-Ct method and samples were normalized to total RNA. 3.3.8 Chromatin Immunoprecipitation-qPCR This protocol was adapted from (116).300-500?l frozen mixed-stage worm pellets were used for each chromatin immunoprecipitation experiment. 3 biological replicates were done for every strain and worms from each sample were split into 100?l pellets. Frozen pellets were crushed by grinding with mortar and pestle. Crushed pellets were resuspended in 1 ml buffer A (15 mM Hepes-Na, pH 7.5, 60 mM KCl, 15 mM NaCl, 0.15 mM beta-mercaptoethonal, 0.15 mM spermine, 0.15 mM spermidine, 0.34M sucrose, 1XHALT protease and phosphatase inhibitor cocktail). To crosslink, formaldehyde was added to a final concentration of 2%, and incubated at room temperature for 15mins. The formaldehyde was quenched by adding 0.1ml 1M Tris HCl (pH 8). The lysate was spun at 15,000xg for 1min at 4?C. The resulting pellets were washed twice with ice-cold buffer A by centrifuging between washes. The pellets were resuspended in 0.3 ml buffer A with 2 mM CaCl2. 100 Micrococcal nuclease (Roche) was added to a final concentration of 0.3 U/?l and incubated for 5 minutes incubation at 37?C (invert the tube several times per minute). EGTA to a final concentration of 20mM was added to stop the digestion reaction. Centrifuge at 15,000xg for 1 min at 4?C. Wash the resulting pellets with 300?l of ice- cold RIPA buffer (1XPBS, 1% NP40, 0.5% Sodium Deoxycholate, 0.1% SDS, 1XHALT protease and phosphatase inhibitor and 2 mM EGTA). Centrifuge at 15,000xg for 1 min at 4?C. Resuspend the pellet after washes in 0.8 ml ice-cold RIPA buffer, and solubilized by shearing (191). Samples were kept on ice at all times except during shearing. All sheared lysate for each biological replicate were pooled and split equally to precipitate for all marks being measured. Centrifuge sheared lysates at 15,000xg for 2 min. 80?l of the supernatant was set-aside at -20?C for " input" libraries and the remaining supernatant was used for IP. Antibodies were chosen based on their efficiency in C. elegans (192). Added 2 ?g of anti-H3 antibody (Abcam, ab1791) or 3 ?g of anti-H3K9me1 antibody (Abcam, ab8896) or 3 ?g of anti-H3K9me2 antibody (Abcam, ab1220) or 2 ?g of anti-H3K9me3 antibody (Abcam, ab8898) and agitated gently at 4?C overnight. Added 50 ?l of protein A Dynabeads (10% slurry in 1x PBS buffer) added and shook for 2hr at 4?C. The beads were then washed four times (four minutes/wash) with ice-cold 600 ?l LiCl washing buffer (100mM Tris HCl, pH 8, 500mM LiCl, 1%NP-40, 1% Sodium deoxycholate) using a magnetic stand (DynaMag-2 Magnet, Thermo Scientific) to discard supernatant after every wash. Beads and input were incubated with 450 ?l worm lysis buffer (0.1M Tris HCl, pH 8, 100 mM NaCl, 1% SDS) containing 200 ?g/ml protease K at 65?C for 4 hours with agitation every 30 minutes to elute the 101 immunoprecipitated nucleosome and reverse crosslinks. Organic extraction and DNA precipitation was done to isolate DNA. DNA obtained was measured by qPCR using LightCycler 480 SYBR Green I Mastermix (Roche) according to the manufacturer?s recommendations. Primers P46 and P43 were used for analysis of R11A8.1, P47 and P48 were used for analysis of mCherry and P49 and P50 were used for analysis of gfp. Fold change was calculated using 2^-??Ct method and samples were normalized to co-immunoprecipitated control gene, R11A8.1 (Fig. 3-6). 3.3.9 Forward genetic screen to identify mutants with Re Activation of Gene Expression (RAGE) This screen was performed on a strain mutated for lin-2 with T silenced for >200generations (AMJ174, refer to section 3.3.4). 10-20 35mm plates of near-starved animals (P0) of all life stages were mutagenized using 1mM N-Ethyl-N-Nitrosourea (ENU, Toronto Research Chemicals) for 4-6 hours. Mutagenized animals were washed four times with wash buffer (0.01% Triton X-100 in M9) and 2-3 adult animals were placed on NG plates seeded with OP50. Over the next 3 weeks, F1, F2, and F3 progeny were screened to isolate mutants exhibiting re-activation of gene expression (Rage), as indicated by presence of mCherry fluorescence. Animals that displayed Rage were singled out (up to 7 animals from each P0 plate) and followed up to confirm persistence of Rage defect. If the defect persisted in descendants, the mutants were frozen. Mutants that were sterile or had arrested embryos (Fig. 3-9) were either discarded or frozen in 5?l of M9 buffer. This pilot screen was performed twice to generate a total of 42 mutants. 102 3.3.10 Complementation Testing Complementation testing was performed on the mutants by mating mutant (T; rage(-)) males with hrde-1(-) hermaphrodite animals (See Fig 3-10). To distinguish cross progeny from self-progeny the hrde-1(-) hermaphrodite animals contained a dpy-2 mutation so that self-progeny would be Dpy phenotype while cross progeny would not. In addition, male progeny are likely cross progeny and can be confidently scored for fluorescence. Therefore F1 males and nonDpy hermaphrodites were scored for silencing of T. Animals scored were categorized as exhibiting bright, dim, or not detectable expression (190). 3.3.11 Microscopy: Animals to be imaged were paralyzed in 5?L of 3mM levamisole for up to 5 minutes, mounted on slides with a 2% agarose pad, and imaged using an AZ100 Nikon microscope and the Photometrics Cool SNAP HQ2 camera at a fixed magnification under non-saturating conditions. Images being compared were inverted on Adobe Photoshop such that gfp or mCherry expression is visible in black and then adjusted identically (levels adjustment) for display. In experiments with feeding RNAi exposure was set for control RNAi fed animals for each strain and the gfp RNAi fed counterparts were imaged at the same exposure. Control and experimental animals were all imaged at non-saturating conditions either at a fixed exposure as determined previously (190) or by setting exposure to their respective controls. In Fig. 3-7, for crosses with GPR-1 overexpression, cross progeny contained GFP from either sur-5::gfp or gtbp-1::gfp in addition to GFP expressed from Pmex- 103 5::gfp::gpr-1. Since the GFP expressed from Pmex-5::gfp::gpr-1 was very dim compared to sur-5::gfp and gtbp-1::gfp exposure was set at non-saturating conditions for the brighter GFP (sur-5::gfp or gtbp-1::gfp). Previous reports have suggested that the pharynx, neurons (94), and vulval muscles (57) are resistant to silencing by dsRNA and hence were not included in our scoring. In some cases, animals expressing bright fluorescence were scored by eye at a fixed magnification of 5X using the Olympus MVX10 fluorescent microscope and only representative images were taken (Fig. 3-3). Representative mutants with sterility or arrested embryos were imaged under non-saturating conditions but we were not able to image age matched controls since it was difficult to track the age of these mutants. 3.3.12 Statistical Analysis: Error bars indicate 95% confidence intervals (p < 0.01) for single proportions calculated using Wilson's estimates with a continuity correction. Significance of differences between two strains or conditions was determined using pooled Wilson's estimates. Significance for graphs with bright, dim and off expression were measured using a chi-square test. Significance for ChIP experiments in Fig. 3-6 was measured using Students t-test. 104 3.4 Results 3.4.1 Transgenerational silencing is uncommon and variable even for the same target sequence Many studies in C. elegans have used different sources of dsRNA to target many different genes and assess silencing in the animal that was exposed to the RNA and/or persistence of silencing across generations as summarized in Fig. 3-1. Based on these studies, we observe that silencing of genes in the soma is rarely inherited for more than one generation and that transgenerational silencing is extremely variable. However the high variability between studies could be due to differences in source of dsRNA, biased or unbiased propagation of silenced animals, differences in target genes, tissue where target is expressed, mode of propagation, difference in researcher etc. A more rigorous and well controlled analysis is necessary to make any definitive conclusions about the prevalence of transgenerational silencing. We decided to use feeding RNAi against the same target sequence expressed in the germline in different gene contexts and assess silencing across generations. We fed dsRNA against gfp to animals expressing gfp from a single copy transgene in different gene contexts i.e., different promoters and UTRs and located at different positions in the genome but with the same sequence of gfp (Fig. 3-2). Five different strains were used- animals expressing pgl-1::gfp or Ppie-1::gfp::pH with sun-1::gfp or gtbp-1::gfp or oma-1::gfp or Pmex-5::mCherry::gfp (Since the sequence for sun- 1::gfp has not been verified yet, this was not scored for silencing). Single generation feeding RNAi was used (See methods section 3.3.5) and P0 animals that were fed 105 dsRNA for 24hours were assessed for silencing. The presence of silencing in this animal would indicate if the RNAi food was effective and would enable us to confidently score descendants for transgenerational silencing. Animals were categorized based on the intensity of fluorescence as bright, dim or off (representative animals in Fig. 3-2). Thereafter, animals were propagated blindly in every generation and stage matched animals were scored for expression of gfp from generation F1 to F5. Fluorescently labeled dsRNA injected outside the germline in C. elegans can be directly deposited into the oocytes and subsequently inherited into progeny (98,99). While it is unclear if dsRNA fed to an animal can also be deposited into progeny, silencing detected in the F1 progeny of P0 parents exposed to dsRNA could be due to deposition of fed dsRNA. Hence, true transgenerational silencing is likely only detectable from the F2 generation. All five targets showed a persistence of silencing in the F1 generation but only gtbp-1::gfp and Pmex-5::mCherry::gfp showed silencing in the F2. By F5 four out of the five targets had completely recovered to bright gfp expression, however one target, Pmex-5::mCherry::gfp (hereafter, referred to as T) showed persistence of silencing at least up to the F5. This suggests that transgenerational silencing is infrequent and a majority of tested genes could recover from silencing induced even within the germline Furthermore, the same target sequence (gfp) can show variability in transgenerational silencing. Our analysis gives us the unique advantage of being able to compare genes that have the same target sequence to analyze how susceptibility to transgenerational silencing is 106 determined. We chose gtbp-1:;gfp and T to study differences in susceptibility to transgenerational silencing. Fig. 3- 2: Transgenerational silencing is uncommon. (A) Schematic for feeding RNAi to assess transgenerational silencing. P0 animals were fed dsRNA for 24hours and then assessed for silencing. Descendants for up to 5 generations (F1-F5) that were not exposed to dsRNA were assessed for inherited silencing. Also see methods section 3.3.5. (B) The same sequence can show variability in transgenerational silencing upon feeding RNAi. Five target genes expressing gfp (green) were exposed to control RNAi or dsRNA against gfp (gfp RNAi). The target genes were low (Ppie-1::gfp::pH) or single copy (Pmex- 5::mCherry::gfp) transgenes or endogenous gene tags (gtbp-1::gfp, pgl-1::gfp, oma- 1::gfp). The fraction of P0 animals and F1-F5 descendants (fract. progeny) were analyzed for silencing (Silencing) and categorized as bright (dark green), dim (light 107 green) or off (grey) based on intensity of fluorescence. The number of generations after the P0 that showed silencing (Gens silenced) is noted. Representative animals for each gene are shown with a colored dot to indicate the level of expression seen in bright, dim or off animals. Scale bar = 50 ?m and n indicates the number of animals scored. The data for gtbp-1;:gfp is the same as in Fig. 3-3E. T was also susceptible to transgenerational silencing when dsRNA was expressed in ancestors (190). The silencing persisted for >25 generations after loss of expressed dsRNA. We tested if gtbp-1::gfp was susceptible to transgenerational silencing upon parental exposure to dsRNA expressed in neurons. dsRNA against gfp was expressed from an extrachromosomal array, Prgef-1::gfp-dsRNA. Due to the mitotically unstable nature of extrachromosomal arrays, some progeny inherit the array while others do not (Fig. 3-3A). Those that do not inherit the array have no source of dsRNA and any silencing detected is likely inherited from parents that had the array. While expression of Prgef-1::gfp-dsRNA enabled silencing of gfp in animals with the dsRNA array, little silencing was detected in siblings that had lost the array (Fig. 3-3B, left). This suggests that similar to the results from feeding RNAi, this gene is less susceptible to transgenerational silencing upon exposure to dsRNA. Thus far our data suggests that differences in the persistence of silencing seen by others or us (Fig. 3-1) are unlikely to be due to the differences in source of dsRNA or the differences in the sequence being targeted for silencing. 108 Fig. 3- 3: Transgenerational silencing can be enhanced but animals still recover after a few generations. (A, B) Exposure to expressed dsRNA does not enable detectable transgenerational silencing of gtbp-1::gfp. (A) Animals expressing dsRNA against gfp (Prgef-1::gfp- dsRNA, orange) from a mitotically unstable array have progeny with the array (w/ array) and without the array (w/o array). (B) Animals w/ or w/o array expressing ubiquitous gfp (gtbp-1::gfp) in a wild type or eri-1(-) background were scored for expression of gfp in the germline as in Fig. 3-2B. (C, D) Increasing ancestral exposure to dsRNA can enhance inherited silencing of the germline. (C) Ancestral exposure to dsRNA can be increased by feeding animals dsRNA (Feeding RNAi, orange) for multiple generations. Progeny were propagated such that some were continued to be fed dsRNA for multiple generation (one, two, or three, C) while siblings were scored for gfp expression as in Fig. 3-2B. Also see methods section 3.3.5. (D) Animals expressing gtbp-1::gfp in a wild type or eri-1(-) background were fed dsRNA against gfp for multiple generations and scored for expression of gfp in the germline as in B. (E) Loss of methyl-transferase MET-2, can enhance transgenerational silencing of gtbp-1::gfp. Animals expressing ubiquitous gtbp-1::gfp in a wild type or a met-2(-) background (3 isolated lines) were fed dsRNA against gfp for a single generation and scored for expression of gfp in the germline as in B (Same 109 data as in Fig. 3-2B). (F-H) Expression of gtbp-1::gfp in the soma could be analyzed for silencing across generations upon exposure to expressed dsRNA (F), prolonged ancestral exposure to dsRNA by feeding (G) and upon loss of met-2(-) (H). (F) Animals from B scored for expression of gfp in the soma. (G) Animals in D scored for expression of gfp in soma. (H) Animals in E scored for expression of gfp in soma. n indicates the number of animals scored. 3.4.2 A resistant gene can recover even after a few generations of silencing. Using gtbp-1::gfp as a representative gene that shows resistance to transgenerational silencing, we tested whether enhancing silencing can prolong the maintenance of silencing. To enhance silencing, we first tested if ancestral exposure to dsRNA for multiple generations could enhance silencing in progeny. Animals were fed gfp-dsRNA for 2 or 3 generations (see methods section 3.3.5, Fig. 3-3 C) and then assessed for silencing. We did not see a significant increase in persistence of silencing even after prolonged exposure to dsRNA in ancestors (Fig. 3-3 D, left). Next, we tested if mutating the exonuclease eri-1, which is thought to degrade siRNAs in the endogenous RNAi pathway could enhance transgenerational silencing (185). The loss of eri-1 is thought to impede the endogenous RNAi pathway, making shared factors more available for silencing by exogenous RNA such as fed dsRNA. The loss of eri-1 coupled with expressed dsRNA (Fig. 3-3B right) or multi- generational feeding RNAi (Fig. 3-3D right) in parents did not significantly enhance silencing across generations. Lastly, we tested if the loss of the methyl transferase met-2, which is thought to be required for mono- and di- methylation of the 9th lysine of the histone 3 (H3K9me1 and H3K9me2, (193)) can enhance transgenerational silencing. Previous work has shown that loss of methyl transferases required for mono-, di- and tri- 110 methylation of H3K9 (H3K9me1, H3K9me2 and H3K9me3) can lead to >30 generations of silencing of some germline genes (194). Similarly, we observed that in the absence of MET-2, silencing of gtbp-1::gfp persisted for at least 4 generations which is more than that observed in a wild-type background (Fig. 3-3E). However, in all 3 lines tested, expression of gfp was recovered at least by the 7th generation. Together, our results suggest that while prolonged exposure to dsRNA in ancestors or loss of eri-1 were not able to enhance transgenerational silencing, loss of met-2 led to an increase in the persistence of silencing. Nevertheless, gtbp-1::gfp was still able to recover from induced silencing even after several generations. Together these data are indicative of competing mechanisms that may promote or prevent persistence of silencing across generations. The differences in silencing observed could be due to specific machinery associated with a gene that dictate the persistence of silencing or recovery from silencing (See discussion). 3.4.3 Transgenerational silencing can be enhanced in somatic cells While most somatic tissue (except pharynx, neurons (94) and vulval muscles (57)) are susceptible to silencing when exposed to dsRNA, transgenerational silencing of the soma is rare (Fig. 3-1). gtbp-1::gfp is expressed ubiquitously in the worm and hence was useful to study both somatic and germline silencing. Expression of dsRNA in parental neurons did not result in detectable silencing of the soma in progeny (Fig. 3F) in a wild type background or in an eri-1(-) background. Upon increased ancestral exposure to dsRNA by feeding for multiple generations, we saw some silencing in unexposed progeny soma but silencing did not 111 persist beyond one generation (Fig. 3-3G, left). While, the loss of eri-1 dramatically enhanced silencing of unexposed F1 progeny, silencing did not persist beyond the F1 (Fig. 3-3G, right). Silencing in the F1 could be indicative of dsRNA deposited from the parent (98,99) into the first cell zygote where soma and germline have not been partitioned. This signal might be sufficient to silence genes in the soma, however the persistence of silencing beyond the soma might require mechanisms that actively transmit information into the embryo (See Discussion). However, the loss of met-2 enabled persistence of silencing for up to 5 generations in some lines (Fig. 3-3H) but expression was recovered by F7 similar to the germline. Together, these results suggest that consistent with other previous reports we also observe that the soma is less susceptible to transgenerational silencing than the germline. Enhancement observed due to the loss of eri-1 or met-2 suggests that these proteins may directly or indirectly contribute to the resistance of silencing seen in the soma. 3.4.4 The dynamics of silencing at an exceptionally susceptible gene The transgene T (Pmex-5::mCherry::gfp) showed exceptional susceptibility to transgenerational silencing (Fig. 3-1) when compared to other genes that also had gfp and were targeted for silencing by the same RNAi food at the same time. We fed and analyzed silencing of T in two independent experiments (both shown in fig. 3-4B). In both cases, silencing was detected from P0-F2 generations (P0 not analyzed in one attempt). We propagated F1 progeny from 3 different P0 animals in each repeat and observed that while all F2s were silenced, there was variation in maintenance of 112 silencing across generations. In both cases, recovery of silencing could be observed from the F3, but by F8 at least, we observed one line that had completely recovered silencing. Two other lines showed persistence of silencing (for at least 15 generations in one case), with one line showing more robust silencing than the other (Fig. 3-4B). Fig. 3- 4: A susceptible gene can be transgenerationally silenced for 15 generations. (A) Structure of Pmex- 5::mCherry::gfp (T). mCherry and gfp are expressed from the mex-5 promoter under the tbb-2 and cye-1 3?UTR respectively. Each fluorophore is tagged with the same sequence of h2b enabling nuclear expression. (B) Feeding dsRNA against gfp in T can also result in silencing of mCherry in T. Animals expressing Pmex-5::mCherry::gfp (T) were fed gfp-dsRNA (gfp RNAi) for a single generation and descendants were scored for expression of gfp and mCherry as in Fig. 3-2B and presented as a pie chart. Animals were categorized as bright (dark magenta or green), dim (light magenta or green), or silenced (grey) for gfp (shades of green) or mCherry (shades of magenta). All generations shown were scored for silencing of gfp by imaging, except for the F2 generation in one feed (left), which was scored by eye. n indicates the number of animals scored. T is a single-copy transgene that consists of a bicistronic operon that expresses mCherry and gfp in the germline, presumably, as one transcript before splicing (Fig. 3-4A). We used this gene to understand silencing of an operon, by measuring expression of mCherry and gfp in the same animals when gfp was targeted for silencing. While mCherry was not detectably silenced in the initial generations upon gfp RNAi, by the F3 generation mCherry begins to show silencing and this silencing 113 correlated with gfp expression in different lines. In lines where gfp showed a recovery from silencing, mCherry also showed a recovery from silencing and in lines where gfp showed persistence of silencing, mCherry also showed a persistence of silencing. Together, our data suggests that even for this exceptionally susceptible transgene, transgenerational silencing is not always a given. Interestingly, the recovery of gfp expression in all lines in F3 and the initiation of mCherry silencing in the same generation suggests that there could be differences in the mechanism of silencing in the early vs. late generations (See discussion). 3.4.5 T can be silenced simply by mating. We chose to study T as an exemplary gene to understand how silencing can be perpetuated across generations and what makes a gene susceptible to transgenerational silencing. Work from our lab has found that this same transgene could be silenced by mating. Paternal inheritance of T but not maternal inheritance of T resulted in loss of expression of both GFP and mCherry from T in a majority of progeny ((190), Fig. 3-5). Other tested genes that contained gfp or mCherry sequences were not susceptible to such ?mating-induced silencing? (168). The susceptibility of T but not other genes containing similar sequences (like gfp or mCherry) could be due to the bicistronic nature of T or the gene context of T including but not restricted to specific sequences of the 3?UTR or histone tag sequences in T. To identify sequences that might determine susceptibility of T to silencing, Cas-9 mediated genome editing was used to edit T to the minimal sequence necessary for expression of a fluorescent marker. This resulted in a transgene T??? 114 expressing mCherry in the germline from the mex-5 promoter and the cye-1 3?UTR ((168) Fig. 3-5, done by Sindhuja Devanapally). This minimal sequence was also susceptible to mating-induced silencing suggesting that the sequences that were eliminated from T were not required for susceptibility to silencing. To eliminate any effects observed due to background mutations, the minimal version of T was reinserted into the genome at the same location as T. This transgene Tcherry also continued to be susceptible to mating-induced silencing (Fig. 3-5, done by Sindhuja Devanapally ). Fig. 3- 5: Mating can induce silencing in progeny. (A) Schematic of transgenes used to study mating-induced silencing. Structure of three transgenes, T (same as Fig. 3-4A), T??? (Pmex-5::mCherry::tbb-2 created by editing T) and Tcherry (Pmex-5::mCherry::tbb-2, a newly made transgene) (B) Hermaphrodites (left) or males (right) that carry transgenes T, T??? or Tcherry were mated with wild-type (+) males or hermaphrodites, respectively, and expression of gfp and mCherry fluorescence was scored in cross progeny as in Fig. 3-4B. n indicates the number of animals scored and orange indicates the presence of a marker (physical defect) near the transgene that enables easy assessment of cross progeny. Sindhuja Devanapally (Jose lab) generated data in this figure. If a silenced F1 was chosen and then propagated in an unbiased manner, silencing was inherited for >200 generations (168). This was different from silencing of T by feeding RNAi where silenced F1 animals still showed differences in inheritance of silencing in descendants (Fig. 3-4B). In addition, mCherry similar to 115 gfp was also silenced in the F1 by mating and this silencing persisted for >200 generations. To avoid the variability of silencing observed by feeding RNAi, we decided to use mating-induced silencing as a reliable way to initiate and maintain transgenerational silencing of T in further studies. 3.4.6 Mechanisms of mating-induced silencing To identify the mechanisms that resulted in the silencing of T by mating, we began by testing the requirement of genes that have been implicated in silencing by small RNAs (Fig 3-6A). Silencing by small RNAs from outside the germline would likely require the dsRNA importer SID-1 (130). Alternatively, silencing within the C. elegans germline can be initiated by piRNAs (46), a highly conserved class of small RNAs. piRNAs encompass 21U RNA (21nt long with a 5? Uridine bias) (124,125,127,128) and 26G RNA (26nt long with a 5? G) (195-197). Within a cell, small RNAs are first processed by primary Agos such as RDE-1 for dsRNA (88,89) or PRG-1 for 21U RNA (128) or ERGO-1 for 26G RNA (198) and can find an mRNA of matching sequence. Recent work has suggested that in the germline, mRNA is cleaved by the endoribonuclease RDE-8 (114)and results in 3? uridylated fragments of mRNA. It remains unclear what protein is required for the uridylation of the mRNA but one possibility is the uridyl transferase MUT-2 that is essential for silencing (199). The uridylated fragments of mRNA can then act as a template for secondary siRNA production by RdRPs (127) such as RRF-1 or EGO-1. The processing of small RNAs in the cytoplasm are thought to occur in perinuclear foci including the P granules (200), Z granules (112,113) and mut granules (111). The loss 116 of some proteins (such as PGL-1, ZNFX-1 and MUT-16) prevents the formation of these foci and can result in defective RNAi. Secondary siRNAs bound to secondary nuclear Argonautes such as NRDE-3 (116-118,120) or HRDE-1 (121) are thought to enter the nucleus and induce changes to histone modifications that are subsequently thought to lead to changes in gene expression. While SID-1, RDE-1, RRF-1, and NRDE-3 were not required for initiation of mating-induced silencing (190), the primary Argonaute PRG-1 ((168) and Fig. 3-6B) and the secondary Argonaute HRDE-1 (Fig. 3-6B) were required for initiation of mating-induced silencing. The requirement of ERGO-1, MUT-2, RDE-8, EGO-1 and MUT-16 are yet to be tested. Maintenance of silencing in subsequent generations (tested after 150 generations of silencing) required the uridyl transferase MUT-2, the mutator foci associated protein MUT-16 (Fig. 3-6C) and secondary Argonaute HRDE-1 ((168) and Fig. 3-6C). Maintenance did not require SID-1, RDE-1, RRF-1, NRDE-3 (168) or RDE-8 (Fig. 3-6C). The requirement of HRDE-1 for silencing has been correlated with an increase in repressive histone marks such as H3K9me3 (116) and H3K27me3 (120). The establishment of H3K9me3 is likely a step-wise process that starts with the establishment of mono- and di- methylation of H3K9 (193). We tested the requirement of two methyl transferases- met-2 that is required for the establishment of H3K9 mono- and di- methylation (H3K9me1 and H3K9me2) and set-32 (193,201- 203) that is thought to be required for the establishment of H3K9 tri-methylation (H3K9me3) along with another methyl transferase set-25. Neither met-2 nor set-32 117 was required for the maintenance of silencing at T. A recently identified chromodomain protein, HERI-1 (204) that is thought to associate with SET-32 was also not required for maintenance of silencing (Fig. 3-6C). To identify changes in mRNA and histone marks, we used an endogenous HRDE-1 target gene, R11A8.1, as a control to control for variations between samples and experiments as this gene should be unaffected upon mating-induced silencing. R11A8.1 was previously shown to be regulated by HRDE-1 (121) and consistent with this, we saw a modest increase in mRNA levels of R11A8.1 (Fig. 3-6D, left) and a 6 fold decrease in H3K9me3 modification (Fig. 3-6D, right) in hrde-1(-). As expected R11A8.1 mRNA levels (Fig. 3-6E, left) remained unchanged in animals with T silenced for >200 generations (iT). Consistent with the silencing being mediated by small RNAs we detected a 47 fold decrease in mRNA in iT animals (Fig. 3-6E, right). We compared H3K9me1, me2 and me3 (and H3 as a control) between animals expressing T or in animals with iT. Consistent with the lack of requirement of met-2 and set-25, we did not detect any significant changes to the H3K9 methylation upon silencing (Fig. 3-6F). 118 Fig. 3- 6: Mating-induced silencing engages the small RNA silencing pathway. (A) Schematic of a small RNA pathway in the C. elegans germline. (B) Mating- induced silencing requires the Argonautes PRG-1 and HRDE-1 for initiation. Silencing was initiated as in Fig. 3-5B in a wild type background or in different mutant (prg-1(-) or hrde-1(-)) backgrounds Expression of mCherry and gfp was scored in cross progeny as in Fig. 3-4. Sindhuja Devanapally (Jose lab) generated data in this figure for wild type and prg-1(-). Asterisk indicates P < 0.01 (?2 test) compared to wild type. (C) iT hermaphrodites that had remained silenced for many generations were mated with mutant males and heterozygous cross progeny were allowed to give homozygous mutant F2 progeny and animals from the F3 generation onwards were scored (Also see Methods). Expression of mCherry and gfp was scored as in Fig. 3-4. Sindhuja Devanapally (Jose lab) generated data in this figure for prg- 1(-) and hrde-1(-). Use of prg-1(-/+) males owing to the poor mating by prg-1(-) males in B and C is indicated (?).Asterisk indicates P < 0.01 (?2 test) compared to wild type. (D) R11A8.1 mRNA and H3K9me3 levels are affected upon loss of HRDE-1. Changes to mRNA and H3K9me3 levels of R11A8.1 (blue) between wild type and hrde-1(-) animals were measured. mRNA levels were measured by qRT- PCR (left) and H3K9me3 levels were measured by ChIP-qPCR (right). Results were normalized to wild type. (E) mCherry mRNA levels are decreased upon by mating- induced silencing. Changes to mRNA levels of R11A8.1 (blue) and mCherry (magenta) between T, iT and wild-type animals were measured as in D. No mRNA was detected in wild-type animals. (F) H3K9 methylation is unaffected upon by mating-induced silencing. Changes to H3, H3K9me1, H3K9me2 and H3K9me3 levels were measured at mCherry (magenta) and gfp (green) between T and iT as in D. Results were normalized to R11A8.1 measured from each samples respective IP 119 and then to T. No significant difference was detected between T and iT using a student?s t-test. Each dot in D-F is one biological replicate with 5 technical replicates and three biological replicates were tested in each experiment. Together, our data suggests that small RNAs in the germline mediate the initiation and maintenance of mating-induced silencing. While the initiation requires both PRG-1 and HRDE-1, maintenance does not require PRG-1 but still requires HRDE-1, MUT-2 and MUT-16. Interestingly, despite the requirement of HRDE-1, silencing of T was not correlated with changes in tested repressive H3K9 methylation. This would suggest that either other chromatin marks such as H3K27 methylation may be altered or that changes to chromatin are not required to maintain silencing of T. 3.4.7 Non-DNA signals correlated with T can modify gene expression states To analyze DNA-independent signals (such as small RNAs or proteins) that can mediate changes to gene expression, we used a tool developed in C.elegans that prevents the paternal and maternal nuclei from fusing in embryos (205,206). The maternal inheritance of a G protein regulator, GPR-1, when it is overexpressed maternally, increases forces that pull on spindle poles converting single bipolar mitotic spindle into two monopolar spindles in the nucleus and prevents the maternal and paternal nuclei from fusing. This allows the contents of the paternal nucleus to be inherited into cells of the P lineage and the contents of the maternal nucleus to be inherited into the AB lineage. By way of such non-mendelian segregation in most cross progeny, paternal DNA is inherited into all germline cells and select somatic cells (such as the intestine and body wall muscles) and maternal DNA is only 120 inherited into the somatic cells (207) (Fig. 3-7 A, middle). A smaller fraction of progeny either have maternal DNA in the germline and some soma and paternal DNA in most somatic cells (Fig. 3-7 A, right) or undergo mendelian segregation with paternal and maternal DNA in all cells (Fig. 3-7 A, left). Initial reports of this tool did not result in a high frequency of non-mendelian cross progeny (205). However, new strains made with gpr-1 overexpression have shown much higher segregation of non-mendelian progeny (206). gpr-1 was expressed in the germline from a Pmex-5::gfp::gpr-1::smu-1 3?UTR transgene (referred to as gpr-1 oe for gpr-1 overexpression) under the mex-5 promoter and tagged with gfp. To analyze the robustness of this tool in our hands, we tested two different single-copy fluorescent genes as markers to identify the segregation of paternal and maternal DNA (Fig. 3-7A)- one expressed nuclear-localized GFP primarily in all somatic cells, sur-5::gfp and another expressed cytoplasmic GFP in all tissues, gtbp-1::gfp. When gpr-1 oe hermaphrodites were crossed with males carrying the fluorescent marker genes, >95% of cross progeny showed non-mendelian segregation with paternal DNA in germline and some somatic cells (based on presence of GFP in the intestinal cells for sur-5::gfp and in intestinal and germline for gtbp-1::gfp, Fig. 3-7A, middle) and showed segregation of maternal DNA into other somatic cells (based on absence of GFP in the canal cells for sur-5::gfp and absence of GFP in the pharyngeal cells and neurons for gtbp-1::gfp, Fig. 3-7A, middle). A much smaller population of cross progeny (<5%) showed the opposite segregation (with maternal DNA in the germline and intestine and paternal DNA in the canal cells, pharyngeal cells and neurons, Fig. 3-7A, right) or mendelian segregation 121 (uniform expression in all cells, Fig. 3-7A left). We decided to use gtbp-1::gfp as the marker to identify non-mendelian cross progeny as the differences in fluorescence were more easily visible. To analyze effects of parental signals on T in the germline, we had to ensure that T (and the accompanying marker gene, gtbp-1::gfp) was always inherited from the male since the majority of non-mendelian cross progeny would inherit paternal T into the germline. Since paternal inheritance of gpr-1 oe did not seem to show the same segregation effects (data not shown) and males of gpr-1 oe strains were difficult to obtain and maintain we did not do reciprocal crosses as controls. Lastly, since the gpr-1 oe transgene also expressed gfp, we decided to use the minimal version of T i.e. T??? or Tcherry (Fig. 3-5) for these analyses to prevent any misinterpretation of data. Using this tool, we identified three conditions that affect the stability of gene expression from T??? or Tcherry. First, we observed that mating-induced silencing did not require the parental nuclei to mingle (Fig. 3-7B, 3rd line). Specifically, non- mendelian progeny that had paternally inherited T??? in the germline and probably had no maternal DNA in the germline cells exhibited mating-induced silencing of T???. Second, the presence of a maternal copy of Tcherry prevents silencing of a paternally inherited copy of Tcherry (Fig. 3-7C, 6th line). Specifically, non-mendelian progeny that had maternally inherited Tcherry in most somatic cells but paternally inherited Tcherry in the germline did not show any mating-induced silencing of Tcherry. Consistent with this, Tcherry/+ hermaphrodites could prevent mating- induced silencing of paternally inherited Tcherry (Fig 3-7C, 5th line). Lastly, the presence of a maternal copy of Tcherry that has been silenced for >5 generations 122 (iTcherry) enables the silencing of a na?ve paternally inherited copy of Tcherry (Fig. 3-7D, 7th line). Specifically, non-mendelian progeny that had maternally inherited a silenced copy of Tcherry in most somatic cells but paternally inherited an expressed copy of Tcherry in the germline displayed robust silencing of the paternally inherited copy of Tcherry in the germline. In this case, mating-induced silencing contributes to the observed silencing as well since Tcherry is inherited paternally. 123 Fig. 3- 7: The expression of T??? or Tcherry can be altered by heritable DNA- independent silencing signal. (A) gpr-1 overexpression results in non-mendelian segregation of parental DNA. sur- 5::gfp or gtbp-1::gfp males were mated with hermaphrodites expressing gpr-1 from a mex-5 promoter in the germline in addition to endogenous gpr-1 (gpr-1 oe, blue). Expression was scored in cross progeny as in Fig. 3-4B. Representative images show the anterior half of the worm with gfp expression in black. Brackets/arrow indicate cells with DNA from each parent and scale bar- 50?m. Schematics depict outcome of each cross. Paternally present marker (gtbp-1::gfp or sur-5::gfp, blue) is inherited through the sperm (cloud shape) and the oocyte (circle) has maternally inherited 124 overexpressed GPR-1 (gpr-1 oe) or no overexpressed gpr-1. S and O label DNA inherited through sperm and oocyte, respectively. Chromosome with the transgene (colored boxes) or without the transgene (black line) is as indicated. (B) Mating- induced silencing does not require parental nuclei to fuse. Animals expressing T???; gtbp-1::gfp were mated with either wild-type (+) or gpr-1 oe animals. Expression in the germline (F1 germline genotype indicated) was scored in cross progeny (fract F1) as in Fig. 3-4B. Schematics depicts outcome of gpr-1 oe cross. Asterisk indicates P < 0.01 (?2 test) compared to mating- induced silencing of T???. (C) Maternal presence of Tcherry can protect progeny from mating-induced silencing. Self progeny of animals expressing Tcherry or Tcherry;gtbp-1::gfp or cross progeny of males expressing Tcherry; gtbp-1::gfp or gtbp-1:gfp mated with hermaphrodites that were either wild type or expressed Tcherry (homozygous or hemizygous) or Tcherry; gpr-1 oe were analyzed. Expression in the germline (F1 germline genotype indicated) was scored (fract F1) in self progeny and male and hermaphrodite cross progeny as in Fig. 3-4B. Schematics depicts outcome of non-mendelian progeny of gpr-1 oe cross. (D) Maternal presence of a silenced Tcherry can enable silencing of a na?ve copy of Tcherry. Animals that stayed silenced after mating-induced silencing of Tcherry for >5 generations were designated as iTcherry. Cross progeny of males expressing Tcherry; gtbp-1::gfp or iTcherry mated with hermaphrodites that were either wild type or expressed iTcherry, Tcherry, gpr-1 oe or Tcherry; gpr-1 oe were analyzed. Expression in the germline (F1 germline genotype indicated) was scored (fract F1) in cross progeny as in Fig. 3-4B. Schematics depicts outcome of non-mendelian progeny of gpr-1 oe cross. Asterisk indicates P < 0.01 (?2 test) compared to mating- induced silencing of Tcherry in non-mendelian progeny. Orange represents chromosomal marker with a recessive mutation on the same chromosome as T used to help identify cross progeny. Since parental factors should only be present within the one cell embryo at the same time and segregated immediately after in the non-mendelian progeny generated upon GPR-1 overexpression, our results would suggest that diffusible signals that are unlikely to be DNA or chromatin are inherited from parents into progeny. These signals are likely present in the one-cell embryo and transmitted into subsequent cells. However, we cannot yet rule out the possibility that signals can be communicated between tissues later in development. Together, our data suggests that DNA- independent signals can be inherited across generations and these signals can determine the expression patterns of a gene in the animal. 125 Our work has suggested that factors required for RNAi including PRG-1 and HRDE-1 ((190), Fig. 3-6 B&C) initiate and/or maintain mating-induced silencing. Maternal absence of PRG-1 was sufficient to prevent the initiation of mating-induced silencing (Fig. 3-8A, top, Sindhuja Devanapally). However, the parental absence of HRDE-1 from either parent did not prevent mating induced silencing in progeny. The absence of HRDE-1 in both parents and consequently the zygote prevented mating- induced silencing (Fig. 3-8A, bottom). Together, these data suggests that to initiate mating-induced silencing, PRG-1 and presumably its associated piRNAs are maternally inherited (Fig. 3-8B). However zygotically expressed HRDE-1 or HRDE- 1 inherited from either parent is required for mating-induced silencing (Fig. 3-8B), though further analysis is required to distinguish these possibilities. On the other hand, for maintenance of silencing across generations, PRG-1 is not required, but parentally deposited HRDE-1 and presumably its associated 2? RNAs are sufficient to perpetuate silencing. Fig. 3- 8: PRG-1 is maternally required while HRDE-1 may be zygotically required. Top, PRG-1 is maternally required to initiate mating-induced silencing. Males expressing T or T; prg-1(+/-) were mated with either wild-type (+) or prg-1(-) hermaphrodites. Bottom, Zygotic presence of HRDE-1 can enable mating-induced 126 silencing. Males expressing T or hrde-1(-);T were mated with either wild type (+) or +;hrde-1(-) hermaphrodites. Expression in the germline was scored in cross progeny (fract F1) as in Fig. 3-4B. Asterisk indicates P < 0.01 (?2 test) compared to indicated cross. (B) Model for initiation and maintenance of mating-induced silencing. 127 3.4.8 Mutants with Re Activation of Gene Expression (Rage) could provide insights into endogenous mechanisms of transgenerational silencing Testing the requirements of transgenerational silencing of T revealed that most of the genes tested were not required for maintenance of silencing. While we haven?t tested all known genes that have been implicated in RNA silencing, a forward genetic screen could identify new factors that mediate transgenerational silencing. Animals that had been silenced for more than 200 generations were mutagenized (see methods section 3.3.9) and 42 mutants exhibiting Re-activation of gene expression or Rage were isolated from screening 72,000 haploid genomes (Fig. 3-9). The mutants broadly fell into 3 categories with immediately obvious endogenous defects ? 13 mutants had arrested embryos and hence were not viable for further analysis (Fig. 3-9B top), 14 mutants had ectopic expression of mCherry and were sterile (if there was a germline in these animals it was not well developed and/or easily visible, Fig. 3-9B bottom) and 15 of the remaining mutants were viable with some other associated defects. 11 of the 15 viable mutants displayed enlarged nuclei in the germline (Fig. 3-9C), 1 had ectopic expression of mCherry in the head and tail (Fig. 3-9D, left), another mutant had a shrunken germline (Fig. 3-9D, right) and one other had an abnormal intestine (Fig. 3-9E). 128 Fig. 3- 9: Mutants exhibiting Re-activation of gene expression (Rage) displayed additional physical defects. (A) Schematic of forward genetic screen performed to isolate rage mutants. (B-E) 42 mutants were isolated from the RAGE screen. (B) 27/42 were mutants that had arrested embryos (top, red brackets) or ectopic expression of mCherry (bottom, red brackets). (C) 11/15 non-sterile mutants had enlarged nuclei (left, red arrows) in the germline. (D&E) 3/15 displayed additional defects, such as ectopic expression in the head and tail (D left, red arrows), a small germline (D right, red bracket), as well as a morphological defect in the intestine (E, red brackets). Scale bar= 50 ?m. We further characterized the isolated mutants by feeding dsRNA against 3 different target genes- unc-22 expressed in the muscle, bli-1 expressed in the skin and gfp (from T) expressed in the germline. These genes were chosen as target since they could be used to distinguish between mutants based on their site of action and mechanism of RNAi (See table 3-3 in methods). Specifically, unc-22 and bli-1 are expressed in the soma and gfp is expressed in the germline. Additionally, unc-22 silencing upon feeding RNAi is likely cytoplasmic while bli-1 silencing is likely nuclear based on their requirement for the nuclear Ago NRDE-3 (147). Silencing of gfp in T by expressed dsRNA required the nuclear Ago HRDE-1 (94) but its requirements for silencing by feeding RNAi haven?t yet been tested. Upon feeding RNAi (two generation feed, see methods section 3.3.5), the 15 viable mutants showed differences in susceptibility to silencing of the three target genes (Fig 3-10A) and could be categorized into 5 different classes. Class 1 mutants showed silencing of all 129 three targets upon feeding RNAi, similar to hrde-1(-) animals, and mutants in this class could include alleles of hrde-1. Class 2 mutants showed silencing of unc-22 and bli-1 but not gfp. Genes defective in these mutants could be germline specific thus susceptible to silencing in the soma. The class 3 mutant showed silencing of gfp and unc-22 but not bli-1. Class 4 mutants showed silencing of unc-22 but not gfp or bli-1. Genes defective in these mutants could be nuclear factors not required for cytoplasmic silencing of unc-22. Lastly, class 5 mutants did not show silencing of gfp, bli-1 or unc-22. Genes defective in these mutants are RNAi defective and likely required for all feeding RNAi. We tested three of the 15 viable mutants for defects in initiation of silencing but upon crossing these mutants with wild-type hermaphrodites we were able to initiate silencing (Fig 3-10B). This suggests that maternal presence of RAGE is not required for initiation of mating-induced silencing and that rage alleles in these three mutants were recessive. We performed complementation analysis on three mutants, two of which behaved similar to hrde-1 upon feeding RNAi (Class 1). All three mutants complemented hrde-1 suggesting that these three mutants were unlikely to be alleles of hrde-1 (Fig 3-10C). Our initial attempts to generate males or mate other mutants weren?t successful so we were unable to perform more rigorous complementation tests with all mutants. 130 Fig. 3- 10: Characterization of rage mutants by feeding RNAi. (A) Rage mutants could be categorized into 5 distinct classes upon feeding RNAi. RAGE mutants were fed control dsRNA or dsRNA against gfp (green), bli-1, and unc-22 for two-generations and progeny were scored for silencing (fraction silenced). Only fraction of animals that showed silencing were plotted (fraction silenced). Dim (light green) and no (grey) expression of gfp was plotted together as silenced. Magenta dot next to mutant names indicates presence of enlarged nuclei. Error bars indicate 95% CI. (B) Tested rage mutants could initiate transgenerational silencing. Three rage mutants or hrde-1 mutants expressing T were mated with wild-type animals and expression of mCherry was scored as in Fig. 3-4B. (C) Complementation tests suggest three mutants are unlikely to be alleles of hrde-1. Three rage mutants or hrde-1 mutants expressing T were mated with hrde-1 mutant animals and expression of mCherry was scored as in Fig. 3-4B. 131 Together, our results suggest that the Rage screen has likely given us at least 5 different types of mutant animals (based on feeding RNAi). In addition, while the associated endogenous defects could be due to background mutations the high frequency of these defects amongst isolated mutants suggest that it is likely related to the rage mutation. These defects could lead us to understand endogenous phenomena that are affected by mutations in genes that are required for the maintenance of silencing across generations. 3.5 Discussion We demonstrate that not all genes display the same level of susceptibility to transgenerational silencing. While transgenerational silencing of a resistant gene could be enhanced, descendants eventually recover from induced silencing. We characterized a gene that is exceptionally susceptible to transgenerational silencing and identified non-DNA signals that can alter gene expression states. Lastly, we have isolated 42 mutants that display Re activation of gene expression (Rage) that can provide insights into mechanisms that enable the persistence of silencing for >200 generations. 3.5.1 Epigenetic recovery prevents the persistence of changes across generations. Many previous studies have observed that not all genes are equally susceptible to transgenerational silencing (Fig. 3-1). However, these studies had many variables including source of dsRNA and differences in target sequences, possibly clouding inferences. Using a consistent method of inducing gene silencing, one generation feeding RNAi, to target the same sequence expressed from different genes, we 132 observe differences in the persistence of silencing. Based on the rate of recovery from silencing, genes may be categorized as resistant, intermediate or susceptible (Fig. 3- 11). Resistant genes are those that are unaffected by silencing in parents. Since dsRNA can be directly deposited from parents into progeny (98,99), even genes that show persistence of silencing for only one generation would be characterized as resistant. Intermediate genes are those that undergo changes for several generations but eventually return to their original state. Lastly, susceptible genes are those that maintain the changed state for many generations (or indefinitely). Of the genes we tested, most genes were resistant or intermediate with the exception of one gene. The ability of the majority of tested genes to return to their original state of expression is evidence of mechanisms that actively enable repair from an induced change. Alternatively, this recovery could be due to dilution of signals deposited from exposed ancestors over time. Together, these mechanisms enable epigenetic recovery (See General Discussion). We observe that enhancing RNA silencing by increasing ancestral exposure to dsRNA or by removing the exonuclease, ERI-1 was not sufficient to dramatically enhance transgenerational silencing. This is in contradiction to another study that observed enhanced transgenerational silencing upon loss of eri-1 (208). ERI-1 is an exonuclease that acts in the endogenous RNAi pathway and loss of ERI-1 results in increased availability of shared RNAi factors and consequently enhanced silencing. Changes in histone marks have been correlated with expression states of genes. MET- 2, a conserved methyl transferase, that is thought to be required for the repression of gene expression has been correlated with silencing (193). Unexpectedly, more recent 133 work has suggested that the loss of MET-2 results in enhanced transgenerational silencing (194) which would suggest that MET-2 is required to maintain expression. We also observed that loss of MET-2 enhanced transgenerational silencing. However, this silencing does not persist indefinitely and similar to intermediate genes, expression recovered by F7. One explanation for this observation could be that removal of MET-2 causes loss of repressive marks at this gene and consequently increased transcription. An increase in mRNA may then lead to an increased response to silencing upon RNAi, potentially because more template is available for amplification i.e., the production of secondary small RNAs. Future experiments will be able to test this hypothesis. Occasionally genes can remain silenced for many generations (such as T) and this suggests that opposing mechanisms in the cell can maintain an induced change. These two different states indicate that opposing mechanisms that prevent or promote the persistence of change can together maintain transgenerational homeostasis (1). 3.5.2 Somatic cells can be made susceptible to transgenerational silencing Interestingly, we and others have observed that the soma is usually resistant to transgenerational silencing (Fig 3-1 and Fig. 3-3). One case of transgenerational silencing in the soma (up to 6 generations) has been observed when expression from the same gene was silenced in the germline (up to 13 generations) (209). Even in this case, silencing in the soma was only detectable when the germline was also silenced. Signals that silence genes in the germline must be inherited into the one cell embryo, hence, these signals can be available for silencing in all cells that originate from this one cell bottleneck. However, the same gene expressed in the soma and germline can 134 show differences in susceptibility to inherited silencing (Fig. 3-3). These differences could be due to sensitivity to dosage of inherited signals or nature of the signal and the requirement of different machinery in soma and germline to process inherited signals. An alternate hypothesis could be that silencing signals are being asymmetrically partitioned into the germline from the one cell zygote. Together, these are indicative of mechanisms that resist transgenerational silencing in the soma. We observed that loss of met-2 could enhance the inheritance of transgenerational silencing of the soma. It remains to be determined if this is due to the presence of more silencing signals in the embryo resulting in detectable silencing of the soma or because MET-2 actively resists silencing in the soma. 3.5.3 Intergenerational versus transgenerational silencing Most tested genes are susceptible to silencing upon exposure to dsRNA; however, a few genes are able to maintain this induced silencing indefinitely. Our work suggests that silencing in earlier generations (intergenerational) may be different from silencing in later generations (transgenerational). We observe an obvious switch in the expression of GFP from T in the 3rd generation after silencing where most animals show a recovery from silencing. However, descendants from siblings behave differently with some recovering completely and others succumbing to transgenerational silencing. Interestingly, at around the same generation, we also begin to detect silencing of mCherry expressed from T. Together these are indicative of a possible change in mechanisms after a few generations of silencing. Amplification mechanisms proposed in C. elegans suggest that 2? siRNA can be made 5? to the target sequence. Since T is an operon, mCherry 135 and gfp are encoded in a single RNA transcript and it is possible that the silencing of mCherry we observe is due to amplification of 2? siRNA from this transcript. However, our data would suggest that silencing due to this amplification is only visible in later generations indicative of an increased spread of silencing across generations. Alternatively it is possible that silencing of mCherry in later generations is due to silencing at the pre-mRNA level in later generations. Previous work on silencing of operons in C. elegans has shown that not all operons show silencing of both genes (117,210-212). In one case (210), targeting the same gene that is part of 2 neighboring operons showed silencing of a gene 3? to it and not of the gene 5? to it. These studies provide further evidence that there may be differences in the silencing of genes even when the same sequence is targeted. We hypothesize that the silencing observed from P0-F2 in T may be similar to the silencing observed at resistant or intermediate genes. However, from the F3, different silencing mechanisms that facilitate escape from epigenetic recovery may enable persistence of silencing indefinitely (Fig. 3-11). Furthermore, our results suggest that the silencing dynamics at a gene and the persistence of silencing is largely gene dependent and so caution must be exercised when generalizing results. 136 Fig. 3- 11: Genes expressing the same target can show varying susceptibility to transgenerational silencing. While all genes are silenced in the P0 upon exposure to dsRNA, the persistence of silencing in progeny varies between genes. Based on this, genes can be categorized as resistant, intermediate or susceptible. Mechanisms that maintain silencing in the first few generations (blue hexagon) may be different from those in later generations (blue star, magenta small RNA) 3.5.4 What makes a gene susceptible to transgenerational silencing? While experiments that enhance silencing have given us clues about machinery that might enable recovery, we still do not have an understanding of what dictates susceptibility or resistance to transgenerational silencing. Several observations that we have made might guide us in understanding determinants of susceptibility: 137 First, we see that the same sequence (gfp) can show varying susceptibility to transgenerational silencing. This suggests that the target sequence may not dictate susceptibility to silencing. This is contrary to previous arguments that suggest that foreign sequences such as gfp are more susceptible to silencing within the germline (124,126). Second, the context of gfp in every target gene was different suggesting that gene contexts might dictate susceptibility. This can include differences in surrounding sequences such as promoter sequences or UTR sequences. But this may also include specific locations in the genome. Third, different sources of dsRNA did not result in a dramatic difference in susceptibility to RNAi. A susceptible gene was silenced transgenerationally for many generations by expressed dsRNA (94), by feeding RNAi and by mating. Alternatively, a resistant gene could not be silenced transgenerationally by expressed dsRNA or by feeding RNAi. Lastly, differences or similarities in the machinery recruited to silence a gene might determine if a gene remains silenced indefinitely. The Ago PRG-1 was not required for silencing of T by feeding RNAi (Farida Eteffa, data not shown) but was required for silencing of T by mating. However, both resulted in long-term silencing of T. The Ago HRDE-1 was required for inherited silencing of both gtbp-1::gfp and T (Farida Eteffa, data not shown) and yet, while T remained silenced indefinitely, gtbp- 1::gfp expression bounced back. These preliminary results suggest that PRG-1 and HRDE-1 might merely be recruited to silence and may not determine susceptibility to silencing. 138 Taken together, we hypothesize that each gene has a set of machinery that are recruited to regulate gene expression, the cell code at the locus, and the differences in cell codes can determine downstream processing machinery. Therefore, changing the genetic context of the same target sequence can change its mechanism of silencing due to the different cell code in its new genetic context. 3.5.5 Mutants exhibiting re-activation of gene expression (Rage) can provide insights into the mechanisms of transgenerational silencing. A forward genetic screen conducted on animals silenced for > 200 generations resulted in the isolation of 42 mutants that exhibit Rage. These mutants exhibited a variety of physical defects that may give us a window into the processes that engage transgenerational silencing in an animal. A majority of the mutants were sterile or showed embryonically arrest defects suggesting that essential genes are probably required for maintenance of transgenerational silencing. While these additional defects could be due to background mutations resulting from mutagenesis, the large number of mutants that exhibit sterility or embryonic arrest with Rage at the same time suggests that the same gene is likely causing both effects. The embryonic arrest defects suggest that genes within the RNAi pathway that regulate developmental timing in C. elegans may be mutated (213). All isolated sterile mutants exhibited ectopic expression suggesting a model in which RAGE may regulate gene expression of endogenous genes (Fig. 3- 12) expressed in the soma and germline. For instance, imagine an endogenous gene that is essential for fertility whose expression is silenced in the soma but not in the germline by RAGE (Fig. 3-12, left). Similarly, expression of T in the germline may 139 require RAGE to silence this gene in the soma. When we trigger silencing, RAGE is now recruited to T in the germline to silence (Fig. 3-12, middle). Upon mutagenesis, loss of RAGE disrupts all silencing in the animal, this includes silencing of T in the soma and germline resulting in ectopic expression of T and silencing of the endogenous gene that is crucial for fertility resulting in sterility (Fig. 3-12, right). Fig. 3- 12: RAGE may regulate expression of endogenous genes. A hypothetical endogenous gene (blue) required for fertility is silenced by RAGE in the soma but not in the germline. RAGE is also required to silence T (magenta and green) in the soma. Upon initiation of transgenerational silencing, RAGE is recruited to T to silence it in the germline as well. Loss of RAGE due to mutagenesis disrupts silencing of T in the soma and the germline, resulting in ectopic expression, but also disrupts silencing of the endogenous gene required for fertility, resulting in sterility. Of the remaining viable mutants, a large majority showed enlarged nuclei in the germline, reminiscent of apoptotic nuclei or of endoreduplication in the nucleus. Interestingly, mutants of mut-2 and mut-16 that showed expression of T after silencing for >200 generations also showed similar enlarged nuclei (Yixin Lin, data not shown). Animals lacking heterochromatin proteins that repress transcription in C. elegans also exhibit defects such as increased germline apoptosis and loss of fertility (214). This suggests that the screen is giving mutants that are required for the maintenance of silencing and further work will identify alleles of known genes and new genes in related processes. Initial characterization of viable mutants upon feeding RNAi suggests the isolation of at least 5 different genes based on the varying 140 behavior of the mutants (Fig. 3-10). Identification and characterization of isolated mutants can provide insights into the mechanisms underlying transgenerational silencing. 141 Chapter 4: General discussion Parental experiences have been correlated with molecular changes in the animal and in descendants. However, due to the widespread effects of parental experiences, it has been a challenge to identify what molecules facilitate spread of changes between cells in an animal and across generations. Using targeted silencing of a single gene by dsRNA in C. elegans, we demonstrate that 1. dsRNA fed to animals likely enters every cell in the animal to enable silencing, 2. transgenerational silencing is rare and susceptibility to silencing is not determined by the sequence being targeted for silencing, and 3. genes that are required for transgenerational silencing are likely also required for essential processes. 4.1 Systemic gene regulation by small RNAs In the fight for survival, organisms have developed many ways to defend themselves and promote their fitness. Amongst other mechanisms (215-217), small RNAs are primarily thought to be one way for organisms to defend themselves from foreign genetic elements (46,218-221). However, other essential functions have been identified as well. For example, in C. elegans, in the absence of telomerases, small RNAs can promote telomeric silencing using Agos, WAGO-1 and HRDE-1, to perhaps change the heterochromatin at telomeres and promote telomere stability. In Drosophila (140), small RNA pathways are essential for the maintenance of telomere stability (222,223) and facilitate heterochromatin-mediated silencing. These pathways are conserved across evolution and are predicted to be fast evolving (224,225). Ago 142 proteins and their sRNAs are central to silencing pathways and have been discovered in all clades of life (46). This strong conservation suggests a functional importance that has led to the preservation of this complex. 4.1.1 How is specificity of silencing mechanisms determined? Silencing of every gene does not always depend on the same factors. For example, we find that silencing of bli-1 unlike unc-22 and unc-54 is completely reliant on the nuclear Argonaute NRDE-3. Many species have multiple Argonaute proteins that bind specific RNA (221). Some prokaryotic Agos have also been shown to bind short single-stranded DNA and regulate gene expression (226,227). C. elegans has 27 different Agos and yet our data suggest that only one Ago is capable of silencing bli-1 upon feeding RNAi! Agos have been characterized based on the presence of 4 key domains- the PIWI domain that interacts with the small RNA and the mRNA and may have catalytic activity (228,229), the MID domain that coordinates the 5? end of the small RNA, the less conserved PAZ domain that coordinates the 3? end of the small RNA (230) and the least conserved N- domain that likely facilitates RNA duplex unwinding during loading of the guide RNA (231) and prevents extended duplex formation during target recognition (232). Phylogenetic analysis has revealed that Agos have acquired insertions that might enable specific protein-protein interactions but further work is required to identify the function of these newly acquired segments (221). In many systems, different non-coding RNAs have dedicated Agos and sometimes RdRPs but it isn?t clear how this specificity is bestowed. 143 One way that RNAi machinery can be distinguished for specific functions is based on the type of RNA that can interact with the protein. Animals and plants have RNA that differ both in length and end-modifications (233) and these difference may facilitate interactions with specific proteins. For example, Non-coding RNAs can have varied structures and these structures might specifically only bind certain proteins. In. C. elegans, classes of sRNAs other than miRNAs (234,235) such as 21U RNAs, 26G RNA and 22G RNAs have been discovered. While 21U RNAs are thought to be made by RNA Polymerase II (236,237), 26G and 22G RNAs are products of different RdRPs (108,110,238,239). In addition to the differences in their biogenesis, each of these binds specific Ago proteins. While 21U RNA bind PRG-1, 26G RNA can bind ALG-3/4 or ERGO-1 and sub populations of 22G RNA can bind different Worm AGOs (WAGOs). Such regulation exist in other plants and animals where RNAs of a specific size or with specific modifications can bind specific Argonautes, however, it remains unclear if and how these differences are detected by Ago proteins. In the case of 22Gs in C. elegans, the same size RNA with the same modification is able to bind different Argonautes! Structural analysis of the MID domain of prokaryotic Agos suggests that this domain is most likely to bind 5? monophosphorylated RNAs with the strongest affinity (240). However, even amongst those analyzed some bind other 5? modifications and some others are unable to bind nucleic acids. This suggests that while the structure of an Ago might lend some specificity, additional features are required to facilitate precise binding. Another way that specificity can be achieved is by temporal or spatial regulation of expression of proteins and their RNAs. In C. elegans, 26G RNA that 144 bind ALG-3/4 are specific to the male germline or the spermatogenic germline in hermaphrodites (129,241,242). On the other hand, 26G RNAs that bind ERGO-1 are unique to the oogenic germline and to embryos (129,198,239) In this case, the expression of these RNAs in certain tissues may confer specificity. Genes required for the formation of perinuclear foci have been associated with RNAi defects (111- 113,200) and some Agos have been found to localize to such perinuclear foci. Restricting the presence of Agos and their RNAs to specific regions within the cell, such as specific foci, may also promote specificity. This would suggest that both proteins and RNA are able to localize to these foci and more work is required to understand how such localization can be achieved. 4.1.2 small RNAs as a molecular signal for systemic regulation RNAs are an excellent candidate for molecules that can carry information between cells since they can regulate gene expression in a precise sequence specific manner. There is increasing evidence for the movement of RNA between cells in many organisms. RNA have been detected in bodily fluids and vesicles (243) in animals and these vesicles could be transported between cells in culture (244). In mice, the loss of the endonuclease Dicer in adipose tissues led to a decrease in miRNAs in exosomes and correlated with a decrease in Fgf-21 (Fibroblast Growth Factor-21) mRNA in hepatic cells (245). tRNA fragments abundant in mature sperm in mice are thought to be imported from vesicles from surrounding somatic cells (48,49) and can regulate gene expression both in sperm and in embryos. In addition, these RNA could be altered by changes to dietary intake. In Drosophila, the dArc mRNA bound to the Arc1 protein is transferred from motorneurons to muscles via 145 vesicles (246). In plants, RNA can be transmitted through plasmodesmata (247-249) and phloem (249) without crossing cell boundaries. A large diversity of RNA can be found in extracellular vesicles (243,250) but the functional relevance of these RNAs and the mechanisms of the export and import of these vesicles remain unclear. Targeted experiments in C. elegans address whether dsRNA expressed in one tissue can move to other distant tissues. RNAs introduced by injection (57), ingestion (79) or by expression in somatic cells (100) have led to silencing in distant cells. The requirement of a dsRNA specific importer, SID-1 for silencing suggests that the molecules transmitting information can be RNA and double-stranded in nature. The endogenous role of SID-1 has remained elusive since its discovery 17 years ago (130). However, recent work in C. elegans is beginning to uncover the role of systemic RNAi. Systemic silencing via SID-1 can suppress germline immortality defects associated with loss of the Ago PRG-1 upon activation of a stress resistance transcription factor (251). SID-1 has also been implicated in the maintenance of germline stem cell arrest during quiescence (141), in regulating gene expression upon dauer (an arrested developmental stage upon overcrowding or starvation C. elegans) formation (252) and in maternal transmission of dauer signals during infection (253). From these studies, it is seems likely that the role of systemic RNA may be more apparent under conditions of stress. 4.1.3 RNAs as a communication signal between species RNAs can be used as communication signals between organisms. Small RNAs can be used by pathogens to promote infections in their hosts. For example, the grey mold fungus, Botrytis cinerea, produces small RNAs that can downregulate 146 immunity genes in Arabidopsis. Extracellular vesicles containing Ago and distinct siRNAs are released by the gastrointestinal nematode Heligmosomoides bakeri (254). Similar small RNA mediated pathogenesis has also been detected for infections caused by the Verticillium dahlia in multiple plants (255), by Escherichia coli in C. elegans (256) and by Trypanosoma cruzi in mammalian cells (257). RNAs from hosts such as plants can also be transported into interacting microbes or pests. Fungi, Verticillium dahlia contained miRNAs from cotton plants that they infected (258), and couple of these miRNAs could downregulate fungal virulence genes. miRNAs from mice and humans are thought to regulate mRNAs in gut bacteria (259) and intestinal miRNA depleted mice show uncontrolled growth of gut microbiota and colitis correlated with impaired regulation of bacterial mRNAs. Many invertebrates including insects (like honeybees, ticks and mites) and nematodes (like C. elegans), can also ingest dsRNA (260). RNA derived from Wolbachia bacteria has been detected in Aedes aegypti, Drosophila melanogaster, and Drosophila simulans (261). RNAs from tomato and corn can be found in Colorado potato beetles and western corn rootworms (188). In addition, spraying dsRNA on plants inhibited fungal infection (255,262). Such findings, have led to the investigation of dsRNA as a pesticide (263). Our work on feeding RNAi in C. elegans (Chapter 2) suggests that long ingested dsRNA can enter all cells to cause silencing. Consistent with this, both insects and fungi can take up long dsRNA with a preference for longer RNA in some cases (262,264). In addition, we see that expression of repetitive DNA can inhibit silencing in a cell and this can be a potential cause for resistance of silencing in some tissues. Understanding the most effective 147 ways to silence are likely to be useful in understanding the best target genes in insects to use dsRNA as an effective pesticide and to anticipate resistance mechanisms. Together, clues from different organisms suggest that it is possible that experiences of an animal such as infection or harsh environmental conditions can either lead to the transfer of small RNAs or changes to endogenous small RNAs in an animal. These RNAs could be used as mobile signals to communicate information to other cells and even to the next generation to confer protection from harsh environmental conditions in progeny. 4.2 Gene regulation by small RNAs across generations From studies in many organisms it has become evident that information can be transferred into descendants by non-genomic molecules. Furthermore, changes in the environment that affect organisms can be transmitted to descendants via these molecules. This inheritance tends to span from just a few generations to an indefinite number of generations. However, little in known about why such variability is observed or the mechanisms that perpetuate inheritance of these molecules. In this dissertation, we investigated small RNAs as a carrier of information across generations. 4.2.1 small RNAs as a molecular signal for intergenerational regulation While many studies in mammals have shown stimuli such as changes in dietary intake (265,266) or olfactory stimulation during stress (25) in parents being correlated with changes in progeny and even grandprogeny, in mammals, parental exposure to any stimulus could result in in utero exposure of progeny and the germ 148 cells of the progeny (and consequently grandprogeny) to the initial stimulus. So any effects in mammals can only be considered transgenerational if maternal exposure results in defects in the F3 animals or if paternal exposure to a stimulus results in defects in progeny. However, studies in C. elegans, where the progeny are less likely to be exposed in utero, have shown that exposure to starvation in parents can be correlated with increased small RNAs against genes regulating metabolism in descendants. We and others (Fig. 3-1) have seen that inheritance of silencing upon targeted silencing of a gene is vastly variable with most somatic genes rarely showing inheritance for more than one generation. The observation that RNA injected into the extracellular space in C. elegans hermaphrodites could be directly deposited into oocytes (98,99) indicates that the silencing detected in immediate progeny could simply reflect silencing due to the trigger introduced in parents and may not specific transgenerational gene silencing mechanisms. This suggests that even in other organisms where silencing triggered in parents is detectable in progeny, deposition of the original trigger could explain inherited effects without the need to engage transgenerational mechanisms (26). Evaluation of persistence of effects for many generations may be necessary to distinguish between such intergenerational effects (seen only for a few generation) and truly transgenerational effects (seen for many generations). Currently, these observations suggest that the effects observed for a few generations could simply be a result of dilution of the initial trigger. Such transient inheritance may be one way for organisms to facilitate immediate adaptation to harsh environmental conditions without inducing permanent changes. 149 4.2.2 Epigenetic recovery prevents the rampant inheritance of silencing We see that dsRNA against the same target gene (gfp) in different genetic contexts showed variability in the persistence of silencing across generations. Most tested genes recovered from silencing within a few generations (intergenerational silencing) suggesting that transgenerational silencing is uncommon. Even the one gene that was transgenerationally silenced, showed differences in transgenerational silencing with some progeny recovering from silencing. Recovery from a change has been detected in other studies too. While temperature changes that result in gene expression changes persist for 14 generations, recovery of gene expression can be detected even after a few generations (90). A well-known recovery mechanism that can actively prevent the inheritance of changes is epigenetic reprogramming (13,14) that results in the erasure of histone and DNA modification in the germline and in the early embryo. Effects that persist for a few generations may be able to resist this erasure by engaging silencing machinery or by transmitting molecules (such as RNA) that remain unaffected by this machinery. Epigenetic recovery could also be due to dilution of the trigger molecule resulting in the eventual loss of silencing or mechanisms that actively repair changes induced in parents. Multiple lines of evidence from our work suggest that there may be repair of silencing induced in parents. Firstly, the loss of methyl transferases, met-2 ((172) and this dissertation), set-25 and set-32 (194) can prolong transgenerational silencing, suggesting that met-2 may be a factor that resists transgenerational silencing or makes a gene more resistant to silencing by changing the chromatin landscape of that gene 150 or changing the chromatin landscape throughout the genome making RNAi factors more available for silencing at a targeted gene (194). Second, a recent screen to identify mutants that show prolonged inherited silencing (204) was able to isolate genes that prevent or reduce transgenerational silencing One gene from this screen, heri-1 is thought to recruit set-32 to prevent silencing (204). However, in contradiction to their roles in preventing transgenerational silencing, these methyl transferases have also been associated with persistence of intergenerational and transgenerational silencing (89,119,126,201,267-269). This suggests that the same methyl transferases can be recruited to both prevent and prolong transgenerational silencing. In C. elegans, some populations of 22G RNAs bind specific WAGOs (Worm specific AGOs) and downregulate expression (115,117,124,153) while other 22G RNA bind an Argonaute CSR-1 and facilitate expression of genes (270). Similar to CSR-1, ALG-3/4 can promote gene expression but only at higher temperatures (242). Given this ability of Argonautes, RNAs and methyl transferases to perform opposing functions, it is conceivable that the same machinery is being recruited to facilitate silencing and epigenetic recovery by regulating gene expression. Other molecules that are yet to be identified may decide if a gene goes through recovery from silencing or persistence of silencing. Repair mechanisms would prevent the rampant spread of changes in every generation that may be detrimental to organisms and in this way form and function can be maintained in every generation. 4.2.3 small RNAs as a molecular signal for transgenerational regulation Despite recovery mechanisms, there have been multiple instances of transgenerational inheritance (silencing >1 generation), suggesting that genes can 151 evade epigenetic repair. For instance, the phenomenon of paramutation has been reported in plants (271-273), mice (273) and flies (274,275) where a silenced allele can convert expression of a homologous allele. Some transgenes introduced into the C. elegans germline can be silenced permanently by recruiting piRNAs (124,126,127,276-278). But the most well-known example is silencing of transposons or repetitive elements that has been seen in many organisms. While RNA silencing is thought to protect from selfish genetic elements, not all genes introduced into the genome get targeted for silencing. In C. elegans, based on their recruitment to silence transgenic sequences introduced in the germline, piRNAs have been proposed as molecules that defend the genome from changes to maintain continuity of the germline across generations (46,124,126). However, not all transgenic sequences are silenced suggesting that if piRNAs are in fact a way to silence ?foreign? genes, it is unclear how a gene is deemed as being foreign. Similarly, little is understood about how transposable elements or repetitive sequences that get integrated into the genome are targeted for transgenerational silencing. The Pmex-5::mCherry::gfp (T) gene provides us with the unique opportunity to understand how silencing can be initiated and how silencing in earlier generations can be different from later generations. Given the differential susceptibility to transgenerational silencing between siblings for T, we can begin to identify differences, if any, that allow a sequence to stay silenced or to get repaired. Our data also suggest that it is likely that the genetic context of a sequence could dictate its susceptibility to transgenerational silencing. Genetic context could include the 152 regulatory sequences used to express the gene such as the UTRs or the promoter or could include the location of the gene in the genome etc. Different regulatory sequences probably require different arrangements of molecules that regulate expression of the gene. Differences in these regulatory molecules could result in distinct susceptibilities to transgenerational silencing. In C. elegans, the prevalence of periodic An/Tn rich clusters (PATCs) have been correlated with gene expression (279). However, PATCs seem to be absent outside the Caenorhabditis species and even in C. elegans, some sequences with PATCs are susceptible to silencing. This suggests that the lack of PATCs alone are insufficient to explain susceptibility to transgenerational silencing though they may contribute to it. Further work will be required to find the determinants of susceptibility to transgenerational silencing. However, the prevalence of sequence features that allow the persistence of change likely contributes to variation and evolution of an organism. The presence of mechanisms that perpetuate acquired changes that improve fitness is beneficial to an organism and might be a reason why transgenerational silencing is coopted. One of the main functions of mechanisms that enable persistence of change may be to ensure the repression of foreign elements such as repetitive DNA in every generation (90). However, for some organisms such as asexual species that may have a reduced chance of genetic variation in every generation, epigenetic variation may be a source of phenotypic diversity (280). In Arabidopsis, the high rate of spontaneous ?epimutations? at methylated sites is thought to have resulted in diversity among its natural populations (281). This suggests that the persistence of changes may aid adaptation and drive evolutionary novelty. 153 Chapter 5: Future directions 5.1 Preface All data in this chapter was generated by Pravrutha Raman and a part of Fig 5- 2A was published in (98). Some worm strains were made by Soriayah Zaghab and Julia Marr? in the Jose lab. 5.2 Introduction In this dissertation we investigated the spread of silencing in an animal and the perpetuation of silencing across generations. In this chapter, directions that will further our understanding of these processes will be discussed. While all directions are based on work delineated in previous chapters, some preliminary observations in support of proposed future directions will be described here. 5.3 Materials and methods 5.3.1 Strains All strains were cultured on Nematode Growth Medium (NGM) plates seeded with 100?l OP50 at 20?C and mutant combinations were generated using standard methods (155). All strains used are listed in Table 5-1 or in Chapter 2 and all oligonucleotides used are listed in Table 5-2 or in Chapter 2. All worm strains were made by standard genetic crosses. 154 Table 5- 1: Strains used. Strain Name Genotype N2 wild type AMJ142 rde-1(ne219) V; rde-4(ne301) III AMJ183 rde-4(ne301); nrde-3(tm1116) AMJ190 rde-4(ne301) III; oxSi221 [Peft-3p::gfp + cb-unc-119(+)] II AMJ285 jamSi1 II [[Pmex-5::rde-4(+)]]; rde-4(ne301) III AMJ314 oxSi221 II; unc-119(ed3) III (?); rde-1(ne219) V AMJ315 rde-4(ne301) III; mIs11[Pmyo-2::gfp::unc-54 3?UTR & gut::gfp::unc-54 3?UTR& pes-10::gfp::unc-54 3?UTR] jamIs3 jamIs3[Pmyo-2:DsRed::unc-54 3?UTR] jamIs4[Pmyo-3::rde- 4(+)::rde-4 3?UTR & pHC183)] IV; nrde-3(tm1116) X AMJ345 jamSi2 [Pmex-5::rde-1(+)] II; rde-1(ne219) V- Used in (98) AMJ494 rde-4(ne301) III; nrde-3(tm1116) X; jamEx3 [Prgef-1::rde- 4(+)::rde-4 3?UTR & Prgef-1::gfp::unc-54 3?UTR] AMJ495 rde-4(ne301) III; nrde-3(tm1116) X; jamEx55[Psid-2::rde- 4(+)::rde-4 3?UTR & Psid-2::gfp::unc-54 3?UTR] AMJ564 rde-4(ne301) III; eri-1(mg366) IV AMJ565 jamSi6 [Pnas-9::rde-4(+)::rde-4 3?UTR] II; unc-119(ed3) III rde- 4(ne301) III AMJ611 jamSi6 [Pnas-9::rde-4(+)::rde-4 3?UTR] II; unc-119(ed3) III rde- 4(ne301) III; nrde-3(tm1116) V- Used in (98) AMJ612 rde-4(ne301) III; eri-1(mg366) IV; nrde-3(tm1116) V 155 AMJ618 rde-4(ne301) III; rde-1(ne219) V; jamEx3 AMJ616 rrf-1(ok589) I; rde-4(ne301) III [2x outcrossed] AMJ617 rrf-1(ok589) I; rde-4(ne301) III; jamEx3 WM27 rde-1(ne219) V WM49 rde-4(ne301) III WM156 nrde-3(tm1116) X Table 5- 2:Oligonucleotides used (5? to 3?, IDT). Primer Sequence P1 CAGTGTGCTTGTAAATCGGC P2 TGCTCTTCGGCAGTTGCTTC P3 GCAAAGAATCTTGCAGCATGG P4 GAACACACCCAGACTGAAGA P5 GACGAGCAAATGCTCAACG P6 TCGTCTTCGGCAGTTGCTTC 5.3.2 P0 & F1 feeding RNAi or F1 only RNAi These methods were performed as described in chapters 2. P0&F1 RNAi was performed for experiments in Fig. 5-1 and F1 only RNAi was performed for experiments in Fig. 5-2. Silencing of target genes were scored as described in Chapters 2. bli-1 scoring was done as described in Chapter 2. 156 5.3.3 Semiquantitative RT-PCR RNA from each strain was isolated from 50 L4-staged animals as described earlier (92). Primer, P1 was used to reverse transcribe the sense strand of rde-4 and P2 was used to reverse transcribe the sense strand of tbb-2. The resulting cDNA was used as a template for PCR (30 cycles for both rde-4 and tbb-2) using Taq polymerase and gene-specific primers (P3 and P4 for rde-4 and P5 and P6 for tbb-2). Intensity of the bands were quantified as described previously (92). The level of rde- 4(+) mRNA in wild-type was set as 1.0 and that in other strains with the ne301 mutation were reported relative to that of wild-type after subtracting the level of rde- 4(ne301) mRNA in WM49 (0.3 in Fig. 5-1D). 5.4 Results We observe differences in the mechanisms used for both silencing at a specific gene and within a cell and little is known about how specificity of silencing via a particular RNAi pathway is established for specific genes and how spread of silencing is regulated across a gene. Both of these directions are discussed in this section. 5.4.1 The precise recruitment of RNAi factors in a cell is poorly understood C. elegans has 27 Agos (282) and yet some Argonautes are completely required for certain functions, suggesting that they cannot substitute function for each other. Two instances of this in this dissertation are- the nuclear Argonaute, NRDE-3 is completely required for the silencing of a skin gene, bli-1 and HRDE-1 is completely required for the initiation and inheritance of mating induced silencing. 157 The requirement of NRDE-3 remained unchanged whether we used feeding RNAi or expressed dsRNA to silence bli-1, suggesting that the method of introducing dsRNA does not affect this requirement. However, we found 3 cases where this requirement could be overcome. First, upon removal of the exonuclease ERI-1 (185) we see robust silencing of bli-1 (Fig5-1A). This silencing suggests that in an eri-1(-) background the Ago recruited to silence bli-1 is altered. Alternatively, it is possible that this change in recruitment is only seen upon loss of NRDE-3. NRDE-3 is thought to be the secondary Ago in the endogenous RNAi pathway and loss of ERI-1 results in the mislocalization of NRDE-3 to the cytoplasm (116) and presumably no nrde-3- mediated endogenous silencing in the nucleus. By tagging NRDE-3 with a fluorophore, we can test if NRDE-3 localizes to the nucleus in eri-1(-) animals upon feeding dsRNA against bli-1. The localization of NRDE-3 to the nucleus might suggest that as long as NRDE-3 is present, silencing of bli-1 is facilitated by NRDE-3 and an alternate Ago is only recruited upon loss of both nrde-3 and eri-1. The alternate possibility is that NRDE-3 doesn?t localize to the nucleus and yet silencing of bli-1 is observed. This could indicate that an alternate Ago is being used or that NRDE-3 is able to function in the cytoplasm to mediate silencing of bli-1. By immunoprecipitating NRDE-3 and looking for bli-1 2?siRNA, we can test if NRDE-3 still binds small RNA targeting bli-1. Second, NRDE-3 independent silencing of bli-1 was observed when RDE-4 was rescued both from a multi copy array and a single copy (Fig 5-1B). Silencing in the former case was weak and was observed upon expression of RDE-4 all tested 158 tissues, not just the skin. However, silencing in the latter case was stronger and was only observed when RDE-4 was expressed in the skin (compare Fig. 5-1 to Fig. 2- 14). Since endogenous RDE-4 has been primarily detected in the germline (283), in both cases it is possible that overexpression of RDE-4 in the skin (via misexpression from multi copy transgenes or intended expression from single copy transgene) results in NRDE-3 independent silencing. While we detected increased expression from multi-copy transgenes, we didn?t detect increased expression from single-copy transgenes (Fig5-1C). However, because RT-PCR was performed on whole worms it is possible that expression in the hypodermis was increased but not detectable. In all these cases, the pattern of bli-1 silencing was altered from controls that had NRDE-3 (Fig. 5-1D). A large part of the hypodermis in C. elegans is a syncytium (hyp7) covering the ventral part of the worm (284) and so the patterns of bli-1 silencing observed when NRDE-3 is present could be indicative of silencing in specific nuclei within the syncytium since we often see animals with blisters encasing only parts of the syncytium. However, upon NRDE-3-independent silencing, we mostly see blisters along the entire length of the worm and this might be suggestive of cytoplasmic silencing. There does seem to be a preference for blistering on the ventral side in all analyzed cases that we don?t completely understand yet. By tagging bli-1 with a nuclear localized fluorophore and then testing for silencing of bli-1 by feeding RNAi, we might be able to see if each nucleus silences as a separate unit or if multiple nuclei can silence using the same factors available in their shared cytoplasm. 159 Fig. 5- 1: Loss of the exonuclease eri-1 or transgenic expression of RDE-4 in the hypodermis can bypass the requirement for NRDE-3 to silence bli-1 in response to ingested dsRNA (A,B) Loss of the exonuclease eri-1 or transgenic expression of RDE-4 in the hypodermis can bypass the requirement for NRDE-3 to silence bli-1 in response to ingested dsRNA. (A) wild-type animals, eri-1(-) animals or eri-1(-) animals that in addition lack rde-4 or nrde-3 were fed dsRNA against bli-1 and analyzed for silencing. (B) Animals that lack rde-4 or double mutant animals that lack rde-4 and either rde-1, rrf-1 or nrde-3 with or without transgenic expression of RDE-4 in the skin (Pnas-9, green), neurons (Prgef-1, orange), intestine (Psid-1, blue) or in body- wall muscles (Pmyo-3, magenta) were fed dsRNA against bli-1 and analyzed for silencing. (C) Animals expressing rde-4(+) from repetitive transgenes express higher levels of rde-4(+) mRNA compared to animals expressing rde-4(+) from a single- copy transgene. Semiquantitative RT-PCR of rde-4 mRNA and tbb-2 mRNA (control) in wild-type animals, rde-4(-) animals, or rde-4(-) animals expressing rde- 4(+) using a repetitive transgene (Ex(nas-9), Ex(myo-3), or Ex(rgef-1)) or a single- copy transgene (Si(nas-9)). Levels of rde-4(+) mRNA relative to that in wild-type animals are indicated for each strain. Asterisk indicates band corresponding to genomic DNA. (D) Patterns of blisters upon silencing bli-1 in nrde-3(-) animals differs from that in wild-type animals but are similar across multiple tissues that express rde-4(+). Aggregate patterns of variation in blister formation in eri-1(-) animals, nrde-3(-); eri-1(-) animals, rde-4(-) or nrde-3(-); rde-4(-) animals that express rde-4(+) in the hypodermis (Pnas-9, green), neurons (Prgef-1, orange), intestine (Psid-2, blue), or body-wall muscles (Pmyo-3, magenta) upon bli-1 RNAi were determined as in Fig. 2-10. The percentage of animals with blister patterns that 160 deviated from wild type (% variants) and the total number of gravid adult animals analyzed (n) is indicated for each strain. Argonautes aren?t the only proteins that exhibit specificity. C. elegans has 4 RdRPs, each of which could be completely required for silencing in most cases but be expendable in some conditions (103,285). A bigger challenge remains in understanding how RNAi factors like Argonautes or RdRPs get their specificity and bind only certain small RNA. Differences in specificity could be due to differences in localization to distinct tissues or differences in localization within a cell. To understand this a large undertaking to tag these proteins by Cas9-mediated genome editing might be necessary. The discovery of multiple membraneless granules such as the P,Z,M granules (111-113,200) that are required for RNAi suggest that proteins and small RNAs could be finding each other by localizing to different parts of the cell but how small RNAs are sorted into these specific granules remains to be studied. It would also be informative to analyze localization and silencing across development since it is unclear if and how RNAi occurs in earlier developmental stages. 5.4.2 How is spread of silencing across a gene coordinated? C. elegans has amplification mechanisms that are required for robust silencing of a gene. These amplification mechanisms include the production of secondary small RNA by an RdRP on an mRNA (109,212). Previous work has suggested that feeding dsRNA matching a sequence resulted in silencing of sequences 5? to the targeted sequence. This in addition with the detection of secondary RNA made largely against sequences 5? to the trigger RNA but not 3? of the trigger suggested that secondary RNA are made mainly upstream of the primary trigger. 161 Consistent with this we observe silencing of mCherry in T, which is upstream of gfp and encoded in the same operon (Fig. 3-4) upon gfp RNAi. However, contradictory to what was observed before we don?t see silencing of mCherry for the first few generations after silencing. This suggests that the silencing we observe may be utilizing different machinery in latter generations that now enable silencing of the entire operon or that amplification of silencing in earlier stages is not sufficient to detect silencing of mCherry in earlier generations. Sequencing small RNAs from animals in different generations post-feeding RNAi can inform if secondary RNAs are made against mCherry in earlier or later generations. An alternate explanation might be differences observed due to the differences in histone marks. Inheritance of silencing upon feeding RNAi is dependent on the nuclear Argonaute HRDE-1 (Farida Eteffa, data not shown, (121)), which is thought to facilitate the deposition of repressive marks, H3K9me3 or H3K27me3 (116-118,120). ChIP-seq to determine the spread of repressive marks if any across T upon feeding RNAi can be used to determine if spread of repressive histone marks into mCherry is correlated with silencing of mCherry. While spread of chromatin marks has been detected (119), this spread seems to be controlled by unknown mechanisms. Though the deposition of repressive modifications is not necessary for the persistence of silencing across generations (this dissertation and (89,194,201,267,268)), it is possible that RNAs direct methylation upon feeding RNAi. Lastly, targeting silencing of mCherry and gfp by switching the position of sequences within the operon and targeting each sequence for silencing is likely to provide more insights about the dynamics of silencing at this gene. 162 5.4.3 What is the route taken by ingested dsRNA in C. elegans We show that the use of multi-copy transgenes to rescue RDE-4 likely resulted in misexpression of RDE-4 in other somatic tissues and was hence misinterpreted as movement of processed dsRNA. Single-copy transgenes used to rescue RDE-4 in the skin and muscle was only sufficient to rescue silencing within the tissue of RDE-4 expression and not in other somatic tissues. This suggests that long dsRNA likely enters every cell to cause silencing. However, little is known about the localization of endogenous RDE-4 in the worm. Whole worm in situ hybridization (283) suggests that RDE-4 is primarily expressed in the germline. RDE-4 in the maternal germline expressed from a single copy transgene resulted in silencing within progeny that lack RDE-4 (98). However, the same was not true for RDE-1. To test for the spread of silencing when RDE-4 is expressed in the germline and prevent maternal deposition of protein, we introduced RDE-4 through the male and tested silencing in progeny (Fig. 5-2A). Progeny were fed dsRNA against a germline gene (pos-1) and genes expressed in the soma (muscle gene unc-22, hypodermal gene dpy-7 and intestinal gene act-5). Surprisingly, silencing against all tested targets was detectable. This silencing could be due to paternal deposition of RDE-4 in embryos. However, rde-4(-) progeny of males with RDE-4 in the germline did not exhibit any silencing (Fig. 5-2B). This suggests that dsRNA that is fed is being processed in the germline by RDE-4 and then can silence genes in the soma. We cannot however rule out the possibility that RDE-4 is being misexpressed from this single-copy transgenes in the soma. When we rescued RDE-1 in the germline under the same promoter from a single-copy transgene and introduced 163 it from the male parent, we only observed silencing of pos-1 in the germline (98)and act-5 in the intestine but not of unc-22 in muscles (Fig. 6-1B). It is possible that the Pmex-5 promoter is misexpressed in the intestine since no silencing is detectable in the muscle. These results together suggest that maybe in some cases RDE-4 processed dsRNA can move between cells to cause silencing. Fig. 5- 2: Expression of RDE-4 in the germline can enable silencing in rde-4(-) soma. (A) Paternally inherited rde-4 expressed in the germline (grey, Pmex-5) can enable silencing in the soma. rde-4(-) or rde-1(-) males with transgenic expression of RDE-4 or RDE-1 in the germline (Pmex-5) were mated with rde-4(-) or rde-1(-) hermaphrodites respectively (left). Cross progeny were fed dsRNA against unc-22 (magenta, body-wall muscles), dpy-7 (green, hypodermis), act-5 (blue, intestine) or pos-1 (grey, germline) and analyzed for silencing (right). (B) rde-4(-) or rde-1(-) hemizygous males with transgenic expression of RDE-4 or RDE-1 in the germline (Pmex-5) were mated with rde-4(-) or rde-1(-) hermaphrodites respectively (left). Cross progeny with or without rde-4 or rde-1 expression were fed dsRNA against unc-22 and bli-1 and analyzed for silencing. Orange + represents the presence of a 164 fluorescent marker used to distinguish genotypes of progeny. Wild-type animals and rde-4(-) or rde-1(-) animals were fed dsRNA against all target genes as well. Silencing by dsRNA requires the ingested dsRNA specific importer SID-2 as well as SID-1 (130,142). It is thought that SID-2 is required for the entry of dsRNA from the lumen of the gut into gut cells. However, the rescue of SID-1 in the muscle using a multi-copy transgene enabled silencing in the muscle by ingested RNAi (100). This suggested that dsRNA could move across the gut without entering its cytoplasm possibly in vesicles to reach muscle cells. An alternate possibility is that SID-1, which was expressed from a multi-copy transgene, was misexpressed in the gut and hence dsRNA was in fact able to enter the gut and subsequently cause silencing in the muscle. Retesting the requirement of SID-1 in the gut to silence genes in other tissues will inform us about the path that dsRNA takes from the lumen into the worm. If RDE-4 is primarily expressed in the germline (as detected by in situ hybridization (283)) it is possible that ingested RNAs can enter the germline get processed by RDE-4 and then move to other somatic cells to silence genes. Based on this, tagging RDE-4 to identify its localization would be a first step to understand how dsRNA can silence in the soma at all. Next, small RNA sequencing of animals with RDE-4 rescued in the germline and with or without SID-1 can be used to identify species of RNA made from fed dsRNA that might enable spread of silencing. By first looking at the species of RNA made from ingested dsRNA, we can identify specific sizes or modifications that might distinguish mobile RNAs from all RNAs. Identification of these specific species can help us look for similar RNAs in wild-type animals to identify endogenous RNAs that move. Previous studies have used rde-4(-) 165 animals to identify endogenous dsRNA (286) but it is yet to be determined if these species can spread between cells. Lastly, viral dsRNA has been labeled to visualize trafficking in live cells (287). This technique fuses one half of a fluorescent protein (N or C terminal moiety) to a dsRNA binding protein and the other half to dsRNA such that binding results in fluorescence. While tagging this dsRNA in bacteria to be ingested might require more optimization, it might plausible to inject tagged dsRNA to detect when and where dsRNA binds the endogenously tagged RDE-4. While animals can be silenced by expressing, feeding or injecting dsRNA, it is unclear what species of RNA are made by bacteria that express dsRNA or by multi- copy transgenes that have been used to express dsRNA. When different sizes of in vitro transcribed dsRNA was injected into worms, more potent silencing was observed for longer RNA (138). However, it is formally possible that multiple sizes of RNA can silence but only some can move between cells. Identifying what sizes are made from bacteria used for feeding RNAi and from multi-copy transgenes and then injecting fluorescently labeled dsRNA of different sizes could determine what sizes of RNA can move between cells. 5.4.4 Identifying endogenous mobile RNAs. Proteins required specifically for the import of dsRNA such as SID-1, SID-2 and SID-3 have been identified in C. elegans(101,130,142,182). SID-1 and SID-3 are highly conserved while SID-2 is only conserved in some species of nematodes but has been lost in others. However, the rescue of C. elegans SID-2 can enable feeding RNAi in species that have lost SID-2 suggesting that the other components required for silencing have been preserved. However, the loss of these transport specific 166 proteins does not result in any obvious defects in animals. RNAseq to identify small RNAs and mRNAs that are changed between wild-type animals and sid-1(-), sid-2(-) or sid-3(-) animals may help identify the endogenous role of these proteins. Alternatively, it is possible that these proteins are only required under conditions of stress. SID-1 was seen to localize to cell membranes when expressed from a multi copy array (130). However, another multi copy array expressing SID-1 in the muscle showed localization of SID-1 to vesicular structures near the nucleus (Julia Marr?, data not shown) and not at the cell membrane. The mammalian homolog of SID-1, SIDT2, also localizes to intracellular vesicles (139). However, these structures may be a caveat of expressing SID-1 in the soma since SID-1 is mostly detectable in the germline (283,288,289). An endogenous tag of SID-1 is essential to understand its localization and function in a cell. If SID-1 is indeed found in vesicles, isolating SID- 1 specific vesicles either by immunoprecipitating SID-1 or by ultracentrifugation to isolate vesicles of the appropriate size and then sorting out SID-1 enriched vesicles can help us identify endogenous species of dsRNA that may require SID-1 to enter the cytosol. 5.4.5 Germline genes are susceptible to transgenerational silencing Our analyses of silencing of the same sequence in different genes suggest that transgenerational silencing is uncommon and can vary even for the same target sequence. Some methods of enhancing RNAi, such as loss of met-2 led to increased transgenerational silencing but yet did not establish stable silencing at the target gene. Together, this suggests that while most tested genes can be silenced upon exposure to 167 dsRNA, mechanisms that repair induced changes may prevent the maintenance of silencing in descendants. Our work gives us the unique opportunity to compare two genes with the same target sequence that have different susceptibility to transgenerational silencing. Using Cas-9 based genome editing, we can swap out regulatory sequences that enable expression from these genes or locations of the entire sequence within the genome and analyze silencing in different gene contexts. We can also place the genes next to each other or under the control of the same regulatory sequences and see if this changes susceptibility to silencing. Such genome editing experiments will guide us in understanding what makes some sequences susceptible to silencing. Future work may require us to identify the molecules that mediate these decisions. The Rage screen we performed is likely to result in the identification of genes required for persistence of silencing, however, a large majority of the mutants isolated from our screen showed other defects. Performing a screen that is biased towards isolating mutants that showed sterility and embryonic arrest might lead to the isolation of genes that use transgenerational silencing for essential processes in C. elegans. The screen done in this thesis did not explicitly differentiate between F2 and F3 mutant progeny. If these inviable mutants were F2 progeny, a clonal screen isolating many F1 animals can be done to isolate heterozygous siblings and identify mutations in homozygous progeny. However, maternal deposition of RAGE into F2 progeny could rescue silencing in F2s making this screen challenging. Alternative strategies might be to use an RNAi screen to knockdown genes across the C. elegans genome and look for Rage. Our analysis has suggested that RNAi factors used to 168 facilitate feeding RNAi are not necessary for maintaining silencing of T across generations so obstruction of maintenance of silencing at T by feeding RNAi is less likely. Given the availability of balancer chromosomes in C. elegans (290,291) now, maintenance and future studies of inviable animals is C. elegans should be easier. Single worm gDNA sequencing has been successfully performed recently (292), and a last resort, one that we are currently exploring is to sequence the genome of single animals. Using this, we can sequence every inviable mutant isolated from the screen and use in silico complementation analysis to identify genes that may be required for silencing. The only gene that was exceptionally susceptible to transgenerational silencing was also susceptible to silencing simply upon mating. It is unclear if this is a feature that is unique to T or if genes that are susceptible to transgenerational silencing can also be susceptible to mating-induced silencing. Nevertheless, the observation of mating-induced silencing suggests that similar endogenous mechanisms of epigenetic conflict might exist in C. elegans. Of all the genes that were tested for the initiation of mating-induced silencing, so far only PRG-1 and HRDE-1 are seen to be required. Based on their requirement for silencing of other genes that don?t exhibit mating induced silencing ((121,124) and Farida Eteffa, data not shown), it is unlikely that they dictate susceptibility to mating-induced silencing. A genetic screen can be done to identify genes required for mating induced silencing. However, such a screen would be extremely tedious and would require making males of mutants to mate and assess silencing. Additionally, mating-induced silencing is not observed in all progeny making analysis of lack of silencing even more difficult. 169 Performing an RNAi screen to knockdown genes across the genome in parents, might allow us to find animals that are mutant for mating-induced silencing. By mating males carrying T to wild type hermaphrodites without T and feeding then dsRNA we can look for progeny that show complete expression of T. Plates with male cross progeny (indicative of cross progeny) showing complete expression of T will likely be from parents with genes required for mating-induced silencing knocked down. 5.4.6 Deciphering the cell code of an animal The information contained within minimally a single cell can drive the development of an organism. Understanding what regulatory molecules are contained in this single cell is essential to our understanding of how animals develop and evolve. Our data clarifies that persistence of changes across generations is rare and can vary even for the same sequence. This suggests that the regulatory machinery at each sequence may be key to understanding what information is inherited or used in every generation. By immunoprecipitating the gene of interest such as T vs. gtbp- 1::gfp followed by Mass spectrometry we can identify what molecules regulate that expression state of these genes, before and after inducing silencing. It remains unclear how environmental experiences can be translated into changes in gene expression within an animal or in its progeny. Some candidates to regulate such change include histone marks, DNA methylation and small RNA. Given the differences in detection of modifications upon silencing and the interdependent regulation of these molecules (See introduction), it has been difficult to tease apart the affect that individual changes have in an animal. While our studies suggest that RNA can be communication signals that facilitate maintenance of 170 change, it remains unclear if chromatin or DNA modification can do the same. Using recently described dCas-9 based approaches (293) where a histone-methylase can be fused to dCas-9 and targeted to a specific sequence, we can systematically alter the histone and DNA modifications on a gene to assess the effects of these modifications independent of RNA. In addition, examining is the 3D arrangement of regulatory molecules and interacting chromatin can illuminate the effect that 3D arrangements can have on gene expression (294). Given the well-defined lineage of the C. elegans embryo from one cell through to adulthood and its stereotyped pattern of development in every generation it gives us the unique opportunity to ask what molecules are present in the one cell zygote and if these molecules are consistent between animals. While this is a very broad questions and it isn?t even clear if we know all the molecules present within a cell. We can begin by analyzing the movement and localization of known proteins from immediately after fertilization of the oocyte through till the zygote stage using live imaging techniques (49,295,296). Coupled with tools like gpr-1 overexpression used in this thesis, such techniques can help us distinguish maternal and paternal factors that are diffusible and required for the development of an animal. Recent advances in isolation and analysis of transcriptomes, epigenomes at the single cell level (297,298)combined with the ability to sort embryos will serve as a good starting point to isolate molecules present in the single cell embryo. Using the same techniques, we can then ask how the cell code or components of the cell code change upon exposure to environmental stimuli and how that affects the development of an organism and cell codes of progeny. 171 Identifying cell codes is the first step to understanding how a living system is built and preserved for generations and is at the heart of understanding how life has evolved. 172 Bibliography 1. Jose, A.M. (2018) Replicating and Cycling Stores of Information Perpetuate Life. Bioessays, 40, e1700161. 2. Tikhodeyev, O.N. (2018) The mechanisms of epigenetic inheritance: how diverse are they? Biol Rev Camb Philos Soc, 93, 1987-2005. 3. Avery, O.T., MacLeod, C.M. and McCarty, M. (1944) Studies on the chemical nature of the substance inducing transformation of pseumococcal types., 79, 137?158. 4. Hershey, A.D. and Chase, M. (1952) Independent Functions of Viral Protein and Nucleic Acid in Growth of Bacteriophage. The Journal of General Physiology, 36, 39-56. 5. Watson, J.D. and Crick, F.H.C. (1953) A structure of deoxyribonucleic acid., 171, 964?967. 6. Portin, P. (2014) The birth and development of the DNA theory of inheritance: sixty years since the discovery of the structure of DNA. J Genet, 93, 293-302. 7. Hershey, A.D. (1970) Genes and hereditary characteristics. 226, 697-700. 8. Ingram, V.M. (1957) Genemutations in human haemoglobin: the chemical difference between normal and sickle cell haemoglobin. 180, 326?328. 9. Sarabhai, A.S., Stretton, A.W.O., Brenner, S. and Bolle, A. (1964) Colinearity of the gene with the peptide chain. 201, 13?17. 10. Yanofsky, C., Carlton, B.C., Guest, J.R., Helinski, D.R. and Henning, U. (1964) On the colinearity of gene structure and protein structure., 51, 266? 272. 11. Brown, D. and Brink, R. (1960) Paramutagenic action of paramutant R- r and R-g alleles in maize. 45, 1313?1316. 173 12. Beisson, J. and Sonneborn, T.M. (1965) Cytoplasmic inheritance of the organization of the cell cortex in Paramecium aurelia., 53, 275?282. 13. Cantone, I. and Fisher, A.G. (2013) Epigenetic programming and reprogramming during development. Nat Struct Mol Biol, 20, 282-289. 14. Hackett, J.A., Sengupta, R., Zylicz, J.J., Murakami, K., Lee, C., Down, T.A. and Surani, M.A. (2013) Germline DNA demethylation dynamics and imprint erasure through 5-hydroxymethylcytosine. Science, 339, 448-452. 15. Lamarck, J.B. (1801) Syst?me des animaux sans vert?bres pr?c?d? du Discours d?ouverture de l?an VIII de la R?publique 16. Darwin, C. (1868) Provisional hypothesis of pangenesis. The variation of animals and plants under domestication. . 27, 349?399. 17. Weismann, A. (1885) The Continuity of the Germ-Plasm as the Foundation of a Theory of Heredity. 18. Seydoux, G. and Braun, R.E. (2006) Pathway to totipotency: lessons from germ cells. Cell, 127, 891-904. 19. Pembrey, M.E., Bygren, L.O., Kaati, G., Edvinsson, S., Northstone, K., Sj?str?m, M., Golding, J. and Team, A.S. (2006) Sex-specific, male-line transgenerational responses in humans. Eur J Hum Genet, 14, 159-166. 20. Carone, B.R., Fauquier, L., Habib, N., Shea, J.M., Hart, C.E., Li, R., Bock, C., Li, C., Gu, H., Zamore, P.D. et al. (2010) Paternally induced transgenerational environmental reprogramming of metabolic gene expression in mammals. Cell, 143, 1084-1096. 21. Fullston, T., Ohlsson Teague, E.M., Palmer, N.O., DeBlasio, M.J., Mitchell, M., Corbett, M., Print, C.G., Owens, J.A. and Lane, M. (2013) Paternal obesity initiates metabolic disturbances in two generations of mice with incomplete penetrance to the F2 generation and alters the transcriptional profile of testis and sperm microRNA content. FASEB J, 27, 4226-4243. 174 22. Grandjean, V., Fourr?, S., De Abreu, D.A., Derieppe, M.A., Remy, J.J. and Rassoulzadegan, M. (2015) RNA-mediated paternal heredity of diet-induced obesity and metabolic disorders. Sci Rep, 5, 18193. 23. Chen, Q., Yan, M., Cao, Z., Li, X., Zhang, Y., Shi, J., Feng, G.H., Peng, H., Zhang, X., Qian, J. et al. (2016) Sperm tsRNAs contribute to intergenerational inheritance of an acquired metabolic disorder. Science, 351, 397-400. 24. Zhang, Y., Zhang, X., Shi, J., Tuorto, F., Li, X., Liu, Y., Liebers, R., Zhang, L., Qu, Y., Qian, J. et al. (2018) Dnmt2 mediates intergenerational transmission of paternally acquired metabolic disorders through sperm small non-coding RNAs. Nat Cell Biol, 20, 535-540. 25. Dias, B.G. and Ressler, K.J. (2014) Parental olfactory experience influences behavior and neural structure in subsequent generations. Nat Neurosci, 17, 89- 96. 26. Perez, M.F. and Lehner, B. (2019) Intergenerational and transgenerational epigenetic inheritance in animals. Nat Cell Biol, 21, 143-151. 27. Fridmann-Sirkis, Y., Stern, S., Elgart, M., Galili, M., Zeisel, A., Shental, N. and Soen, Y. (2014) Delayed development induced by toxicity to the host can be inherited by a bacterial-dependent, transgenerational effect. Front Genet, 5, 27. 28. Yassour, M., Jason, E., Hogstrom, L.J., Arthur, T.D., Tripathi, S., Siljander, H., Selvenius, J., Oikarinen, S., Hy?ty, H., Virtanen, S.M. et al. (2018) Strain- Level Analysis of Mother-to-Child Bacterial Transmission during the First Few Months of Life. Cell Host Microbe, 24, 146-154.e144. 29. Moeller, A.H., Suzuki, T.A., Phifer-Rixey, M. and Nachman, M.W. (2018) Transmission modes of the mammalian gut microbiota. Science, 362, 453- 457. 30. Sonnenburg, E.D., Smits, S.A., Tikhonov, M., Higginbottom, S.K., Wingreen, N.S. and Sonnenburg, J.L. (2016) Diet-induced extinctions in the gut microbiota compound over generations. Nature, 529, 212-215. 175 31. Zare, A., Johansson, A.M., Karlsson, E., Delhomme, N. and Stenberg, P. (2018) The gut microbiome participates in transgenerational inheritance of low-temperature responses in Drosophila melanogaster. FEBS Lett, 592, 4078-4086. 32. Harvey, Z.H., Chen, Y. and Jarosz, D.F. (2018) Protein-Based Inheritance: Epigenetics beyond the Chromosome. Mol Cell, 69, 195-202. 33. Wickner, R.B. (1994) [URE3] as an altered URE2 protein: evidence for a prion analog in Saccharomyces cerevisiae. Science, 264, 566-569. 34. Derkatch, I.L., Chernoff, Y.O., Kushnirov, V.V., Inge-Vechtomov, S.G. and Liebman, S.W. (1996) Genesis and variability of [PSI] prion factors in Saccharomyces cerevisiae. Genetics, 144, 1375-1386. 35. Coustou, V., Deleu, C., Saupe, S. and Begueret, J. (1997) The protein product of the het-s heterokaryon incompatibility gene of the fungus Podospora anserina behaves as a prion analog. Proc Natl Acad Sci U S A, 94, 9773-9778. 36. Rose, N.R. and Klose, R.J. (2014) Understanding the relationship between DNA methylation and histone lysine methylation. Biochim Biophys Acta, 1839, 1362-1372. 37. Bocklandt, S., Lin, W., Sehl, M.E., S?nchez, F.J., Sinsheimer, J.S., Horvath, S. and Vilain, E. (2011) Epigenetic predictor of age. PLoS One, 6, e14821. 38. Hannum, G., Guinney, J., Zhao, L., Zhang, L., Hughes, G., Sadda, S., Klotzle, B., Bibikova, M., Fan, J.B., Gao, Y. et al. (2013) Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell, 49, 359- 367. 39. Andrews, F.H., Strahl, B.D. and Kutateladze, T.G. (2016) Insights into newly discovered marks and readers of epigenetic information. Nat Chem Biol, 12, 662-668. 40. Tan, M., Luo, H., Lee, S., Jin, F., Yang, J.S., Montellier, E., Buchou, T., Cheng, Z., Rousseaux, S., Rajagopal, N. et al. (2011) Identification of 67 histone marks and histone lysine crotonylation as a new type of histone modification. Cell, 146, 1016-1028. 176 41. Bannister, A.J. and Kouzarides, T. (2011) Regulation of chromatin by histone modifications. Cell Res, 21, 381-395. 42. Huang, H., Lin, S., Garcia, B.A. and Zhao, Y. (2015) Quantitative proteomic analysis of histone modifications. Chem Rev, 115, 2376-2418. 43. Siklenka, K., Erkek, S., Godmann, M., Lambrot, R., McGraw, S., Lafleur, C., Cohen, T., Xia, J., Suderman, M., Hallett, M. et al. (2015) Disruption of histone methylation in developing sperm impairs offspring health transgenerationally. Science, 350, aab2006. 44. Teperek, M., Simeone, A., Gaggioli, V., Miyamoto, K., Allen, G.E., Erkek, S., Kwon, T., Marcotte, E.M., Zegerman, P., Bradshaw, C.R. et al. (2016) Sperm is epigenetically programmed to regulate gene transcription in embryos. Genome Res, 26, 1034-1046. 45. Heard, E. and Martienssen, R.A. (2014) Transgenerational epigenetic inheritance: myths and mechanisms. Cell, 157, 95-109. 46. Almeida, M.V., Andrade-Navarro, M.A. and Ketting, R.F. (2019) Function and Evolution of Nematode RNAi Pathways. Noncoding RNA, 5. 47. Rechavi, O., Houri-Ze'evi, L., Anava, S., Goh, W.S.S., Kerk, S.Y., Hannon, G.J. and Hobert, O. (2014) Starvation-induced transgenerational inheritance of small RNAs in C. elegans. Cell, 158, 277-287. 48. Sharma, U., Sun, F., Conine, C.C., Reichholf, B., Kukreja, S., Herzog, V.A., Ameres, S.L. and Rando, O.J. (2018) Small RNAs Are Trafficked from the Epididymis to Developing Mammalian Sperm. Dev Cell, 46, 481-494.e486. 49. Chen, K.H., Boettiger, A.N., Moffitt, J.R., Wang, S. and Zhuang, X. (2015) RNA imaging. Spatially resolved, highly multiplexed RNA profiling in single cells. Science, 348, aaa6090. 50. Arroyo, J.D., Chevillet, J.R., Kroh, E.M., Ruf, I.K., Pritchard, C.C., Gibson, D.F., Mitchell, P.S., Bennett, C.F., Pogosova-Agadjanyan, E.L., Stirewalt, D.L. et al. (2011) Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma. Proc Natl Acad Sci U S A, 108, 5003-5008. 177 51. Ogawa, Y., Taketomi, Y., Murakami, M., Tsujimoto, M. and Yanoshita, R. (2013) Small RNA transcriptomes of two types of exosomes in human whole saliva determined by next generation sequencing. Biol Pharm Bull, 36, 66-75. 52. Ogawa, Y., Tsujimoto, M. and Yanoshita, R. (2016) Next-Generation Sequencing of Protein-Coding and Long Non-protein-Coding RNAs in Two Types of Exosomes Derived from Human Whole Saliva. Biol Pharm Bull, 39, 1496-1507. 53. Vojtech, L., Woo, S., Hughes, S., Levy, C., Ballweber, L., Sauteraud, R.P., Strobl, J., Westerberg, K., Gottardo, R., Tewari, M. et al. (2014) Exosomes in human semen carry a distinctive repertoire of small non-coding RNAs with potential regulatory functions. Nucleic Acids Res, 42, 7290-7304. 54. Chim, S.S., Shing, T.K., Hung, E.C., Leung, T.Y., Lau, T.K., Chiu, R.W. and Lo, Y.M. (2008) Detection and characterization of placental microRNAs in maternal plasma. Clin Chem, 54, 482-490. 55. Kosaka, N., Izumi, H., Sekine, K. and Ochiya, T. (2010) microRNA as a new immune-regulatory agent in breast milk. Silence, 1, 7. 56. Mitchell, P.S., Parkin, R.K., Kroh, E.M., Fritz, B.R., Wyman, S.K., Pogosova- Agadjanyan, E.L., Peterson, A., Noteboom, J., O'Briant, K.C., Allen, A. et al. (2008) Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad Sci U S A, 105, 10513-10518. 57. Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E. and Mello, C.C. (1998) Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 391, 806-811. 58. Palauqui, J.C., Elmayan, T., Pollien, J.M. and Vaucheret, H. (1997) Systemic acquired silencing: transgene-specific post-transcriptional silencing is transmitted by grafting from silenced stocks to non-silenced scions. EMBO J, 16, 4738-4745. 59. Brosnan, C.A. and Voinnet, O. (2011) Cell-to-cell and long-distance siRNA movement in plants: mechanisms and biological implications. Curr Opin Plant Biol, 14, 580-587. 178 60. Kizana, E., Cingolani, E. and Marb?n, E. (2009) Non-cell-autonomous effects of vector-expressed regulatory RNAs in mammalian heart cells. Gene Ther, 16, 1163-1168. 61. Lim, P.K., Bliss, S.A., Patel, S.A., Taborga, M., Dave, M.A., Gregory, L.A., Greco, S.J., Bryan, M., Patel, P.S. and Rameshwar, P. (2011) Gap junction- mediated import of microRNA from bone marrow stromal cells can elicit cell cycle quiescence in breast cancer cells. Cancer Res, 71, 1550-1560. 62. Katakowski, M., Buller, B., Wang, X., Rogers, T. and Chopp, M. (2010) Functional microRNA is transferred between glioma cells. Cancer Res, 70, 8259-8263. 63. Morales, C.R., Lefrancois, S., Chennathukuzhi, V., El-Alfy, M., Wu, X., Yang, J., Gerton, G.L. and Hecht, N.B. (2002) A TB-RBP and Ter ATPase complex accompanies specific mRNAs from nuclei through the nuclear pores and into intercellular bridges in mouse male germ cells. Dev Biol, 246, 480- 494. 64. Jose, A.M. (2015) Movement of regulatory RNA between animal cells. Genesis, 53, 395-416. 65. Skog, J., W?rdinger, T., van Rijn, S., Meijer, D.H., Gainche, L., Sena-Esteves, M., Curry, W.T., Carter, B.S., Krichevsky, A.M. and Breakefield, X.O. (2008) Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nat Cell Biol, 10, 1470-1476. 66. Montecalvo, A., Larregina, A.T., Shufesky, W.J., Stolz, D.B., Sullivan, M.L., Karlsson, J.M., Baty, C.J., Gibson, G.A., Erdos, G., Wang, Z. et al. (2012) Mechanism of transfer of functional microRNAs between mouse dendritic cells via exosomes. Blood, 119, 756-766. 67. Turchinovich, A., Weiz, L., Langheinz, A. and Burwinkel, B. (2011) Characterization of extracellular circulating microRNA. Nucleic Acids Res, 39, 7223-7233. 68. Vickers, K.C., Palmisano, B.T., Shoucri, B.M., Shamburek, R.D. and Remaley, A.T. (2011) MicroRNAs are transported in plasma and delivered to recipient cells by high-density lipoproteins. Nat Cell Biol, 13, 423-433. 179 69. Wolfrum, C., Shi, S., Jayaprakash, K.N., Jayaraman, M., Wang, G., Pandey, R.K., Rajeev, K.G., Nakayama, T., Charrise, K., Ndungo, E.M. et al. (2007) Mechanisms and optimization of in vivo delivery of lipophilic siRNAs. Nat Biotechnol, 25, 1149-1157. 70. Zimmermann, T.S., Lee, A.C., Akinc, A., Bramlage, B., Bumcrot, D., Fedoruk, M.N., Harborth, J., Heyes, J.A., Jeffs, L.B., John, M. et al. (2006) RNAi-mediated gene silencing in non-human primates. Nature, 441, 111-114. 71. Akinc, A., Querbes, W., De, S., Qin, J., Frank-Kamenetsky, M., Jayaprakash, K.N., Jayaraman, M., Rajeev, K.G., Cantley, W.L., Dorkin, J.R. et al. (2010) Targeted delivery of RNAi therapeutics with endogenous and exogenous ligand-based mechanisms. Mol Ther, 18, 1357-1364. 72. Gilleron, J., Querbes, W., Zeigerer, A., Borodovsky, A., Marsico, G., Schubert, U., Manygoats, K., Seifert, S., Andree, C., St?ter, M. et al. (2013) Image-based analysis of lipid nanoparticle-mediated siRNA delivery, intracellular trafficking and endosomal escape. Nat Biotechnol, 31, 638-646. 73. Adams, D., Gonzalez-Duarte, A., O'Riordan, W.D., Yang, C.C., Ueda, M., Kristen, A.V., Tournev, I., Schmidt, H.H., Coelho, T., Berk, J.L. et al. (2018) Patisiran, an RNAi Therapeutic, for Hereditary Transthyretin Amyloidosis. N Engl J Med, 379, 11-21. 74. Hawkins, P.N., Ando, Y., Dispenzeri, A., Gonzalez-Duarte, A., Adams, D. and Suhr, O.B. (2015) Evolving landscape in the management of transthyretin amyloidosis. Ann Med, 47, 625-638. 75. Zhang, J., Khan, S.A., Hasse, C., Ruf, S., Heckel, D.G. and Bock, R. (2015) Pest control. Full crop protection from an insect pest by expression of long double-stranded RNAs in plastids. Science, 347, 991-994. 76. Baum, J.A., Bogaert, T., Clinton, W., Heck, G.R., Feldmann, P., Ilagan, O., Johnson, S., Plaetinck, G., Munyikwa, T., Pleau, M. et al. (2007) Control of coleopteran insect pests through RNA interference. Nat Biotechnol, 25, 1322- 1326. 77. Mao, Y.B., Cai, W.J., Wang, J.W., Hong, G.J., Tao, X.Y., Wang, L.J., Huang, Y.P. and Chen, X.Y. (2007) Silencing a cotton bollworm P450 180 monooxygenase gene by plant-mediated RNAi impairs larval tolerance of gossypol. Nat Biotechnol, 25, 1307-1313. 78. Koch, A. and Kogel, K.H. (2014) New wind in the sails: improving the agronomic value of crop plants through RNAi-mediated gene silencing. Plant Biotechnol J, 12, 821-831. 79. Timmons, L. and Fire, A. (1998) Specific interference by ingested dsRNA. Nature, 395, 854. 80. Maillard, P.V., Ciaudo, C., Marchais, A., Li, Y., Jay, F., Ding, S.W. and Voinnet, O. (2013) Antiviral RNA interference in mammalian cells. Science, 342, 235-238. 81. Li, Y., Lu, J., Han, Y., Fan, X. and Ding, S.W. (2013) RNA interference functions as an antiviral immunity mechanism in mammals. Science, 342, 231-234. 82. Li, Y., Basavappa, M., Lu, J., Dong, S., Cronkite, D.A., Prior, J.T., Reinecker, H.C., Hertzog, P., Han, Y., Li, W.X. et al. (2016) Induction and suppression of antiviral RNA interference by influenza A virus in mammalian cells. Nat Microbiol, 2, 16250. 83. Saleh, M.C., van Rij, R.P., Hekele, A., Gillis, A., Foley, E., O'Farrell, P.H. and Andino, R. (2006) The endocytic pathway mediates cell entry of dsRNA to induce RNAi silencing. Nat Cell Biol, 8, 793-802. 84. Rechavi, O., Minevich, G. and Hobert, O. (2011) Transgenerational inheritance of an acquired small RNA-based antiviral response in C. elegans. Cell, 147, 1248-1256. 85. Wilkins, C., Dishongh, R., Moore, S.C., Whitt, M.A., Chow, M. and Machaca, K. (2005) RNA interference is an antiviral defence mechanism in Caenorhabditis elegans. Nature, 436, 1044-1047. 86. Schott, D.H., Cureton, D.K., Whelan, S.P. and Hunter, C.P. (2005) An antiviral role for the RNA interference machinery in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 102, 18420-18424. 181 87. Greer, E.L., Maures, T.J., Hauswirth, A.G., Green, E.M., Leeman, D.S., Maro, G.S., Han, S., Banko, M.R., Gozani, O. and Brunet, A. (2010) Members of the H3K4 trimethylation complex regulate lifespan in a germline-dependent manner in C. elegans. Nature, 466, 383-387. 88. Greer, E.L., Maures, T.J., Ucar, D., Hauswirth, A.G., Mancini, E., Lim, J.P., Benayoun, B.A., Shi, Y. and Brunet, A. (2011) Transgenerational epigenetic inheritance of longevity in Caenorhabditis elegans. Nature, 479, 365-371. 89. Ni, J.Z., Kalinava, N., Chen, E., Huang, A., Trinh, T. and Gu, S.G. (2016) A transgenerational role of the germline nuclear RNAi pathway in repressing heat stress-induced transcriptional activation in C. elegans. Epigenetics Chromatin, 9, 3. 90. Klosin, A., Casas, E., Hidalgo-Carcedo, C., Vavouri, T. and Lehner, B. (2017) Transgenerational transmission of environmental information in. Science, 356, 320-323. 91. Schott, D., Yanai, I. and Hunter, C.P. (2014) Natural RNA interference directs a heritable response to the environment. Sci Rep, 4, 7387. 92. Jobson, M.A., Jordan, J.M., Sandrof, M.A., Hibshman, J.D., Lennox, A.L. and Baugh, L.R. (2015) Transgenerational Effects of Early Life Starvation on Growth, Reproduction, and Stress Resistance in Caenorhabditis elegans. Genetics, 201, 201-212. 93. Jose, A.M., Garcia, G.A. and Hunter, C.P. (2011) Two classes of silencing RNAs move between Caenorhabditis elegans tissues. Nat Struct Mol Biol, 18, 1184-1188. 94. Devanapally, S., Ravikumar, S. and Jose, A.M. (2015) Double-stranded RNA made in C. elegans neurons can enter the germline and cause transgenerational gene silencing. Proc Natl Acad Sci U S A, 112, 2133-2138. 95. Corsi, A.K., Wightman, B. and Chalfie, M. (2015) A Transparent Window into Biology: A Primer on Caenorhabditis elegans. Genetics, 200, 387-407. 182 96. Olson, S.K., Greenan, G., Desai, A., M?ller-Reichert, T. and Oegema, K. (2012) Hierarchical assembly of the eggshell and permeability barrier in C. elegans. J Cell Biol, 198, 731-748. 97. Tavernarakis, N., Wang, S.L., Dorovkov, M., Ryazanov, A. and Driscoll, M. (2000) Heritable and inducible genetic interference by double-stranded RNA encoded by transgenes. Nat Genet, 24, 180-183. 98. Marr?, J., Traver, E.C. and Jose, A.M. (2016) Extracellular RNA is transported from one generation to the next in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 113, 12496-12501. 99. Wang, E. and Hunter, C.P. (2017) SID-1 Functions in Multiple Roles To Support Parental RNAi in. Genetics, 207, 547-557. 100. Jose, A.M., Smith, J.J. and Hunter, C.P. (2009) Export of RNA silencing from C. elegans tissues does not require the RNA channel SID-1. Proc Natl Acad Sci U S A, 106, 2283-2288. 101. Jose, A.M., Kim, Y.A., Leal-Ekman, S. and Hunter, C.P. (2012) Conserved tyrosine kinase promotes the import of silencing RNA into Caenorhabditis elegans cells. Proc Natl Acad Sci U S A, 109, 14520-14525. 102. Timmons, L., Tabara, H., Mello, C.C. and Fire, A.Z. (2003) Inducible systemic RNA silencing in Caenorhabditis elegans. Mol Biol Cell, 14, 2972- 2983. 103. Ravikumar, S., Devanapally, S. and Jose, A.M. (2018) Gene silencing by double-stranded RNA from C. elegans neurons reveals functional mosaicism of RNA interference. 104. Grishok, A. (2013) Biology and Mechanisms of Short RNAs in Caenorhabditis elegans. Adv Genet, 83, 1-69. 105. Bernstein, E., Caudy, A.A., Hammond, S.M. and Hannon, G.J. (2001) Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature, 409, 363-366. 183 106. Parrish, S. and Fire, A. (2001) Distinct roles for RDE-1 and RDE-4 during RNA interference in Caenorhabditis elegans. RNA, 7, 1397-1402. 107. Tabara, H., Sarkissian, M., Kelly, W.G., Fleenor, J., Grishok, A., Timmons, L., Fire, A. and Mello, C.C. (1999) The rde-1 gene, RNA interference, and transposon silencing in C. elegans. Cell, 99, 123-132. 108. Pak, J. and Fire, A. (2007) Distinct populations of primary and secondary effectors during RNAi in C. elegans. Science, 315, 241-244. 109. Sijen, T., Fleenor, J., Simmer, F., Thijssen, K.L., Parrish, S., Timmons, L., Plasterk, R.H. and Fire, A. (2001) On the role of RNA amplification in dsRNA-triggered gene silencing. Cell, 107, 465-476. 110. Sijen, T., Steiner, F.A., Thijssen, K.L. and Plasterk, R.H. (2007) Secondary siRNAs result from unprimed RNA synthesis and form a distinct class. Science, 315, 244-247. 111. Phillips, C.M., Montgomery, T.A., Breen, P.C. and Ruvkun, G. (2012) MUT- 16 promotes formation of perinuclear mutator foci required for RNA silencing in the C. elegans germline. Genes Dev, 26, 1433-1444. 112. Wan, G., Fields, B.D., Spracklin, G., Shukla, A., Phillips, C.M. and Kennedy, S. (2018) Spatiotemporal regulation of liquid-like condensates in epigenetic inheritance. Nature, 557, 679-683. 113. Ishidate, T., Ozturk, A.R., Durning, D.J., Sharma, R., Shen, E.Z., Chen, H., Seth, M., Shirayama, M. and Mello, C.C. (2018) ZNFX-1 Functions within Perinuclear Nuage to Balance Epigenetic Signals. Mol Cell, 70, 639-649.e636. 114. Tsai, H.Y., Chen, C.C., Conte, D., Moresco, J.J., Chaves, D.A., Mitani, S., Yates, J.R., Tsai, M.D. and Mello, C.C. (2015) A ribonuclease coordinates siRNA amplification and mRNA cleavage during RNAi. Cell, 160, 407-419. 115. Yigit, E., Batista, P.J., Bei, Y., Pang, K.M., Chen, C.C., Tolia, N.H., Joshua- Tor, L., Mitani, S., Simard, M.J. and Mello, C.C. (2006) Analysis of the C. elegans Argonaute family reveals that distinct Argonautes act sequentially during RNAi. Cell, 127, 747-757. 184 116. Guang, S., Bochner, A.F., Pavelec, D.M., Burkhart, K.B., Harding, S., Lachowiec, J. and Kennedy, S. (2008) An Argonaute transports siRNAs from the cytoplasm to the nucleus. Science, 321, 537-541. 117. Guang, S., Bochner, A.F., Burkhart, K.B., Burton, N., Pavelec, D.M. and Kennedy, S. (2010) Small regulatory RNAs inhibit RNA polymerase II during the elongation phase of transcription. Nature, 465, 1097-1101. 118. Burkhart, K.B., Guang, S., Buckley, B.A., Wong, L., Bochner, A.F. and Kennedy, S. (2011) A pre-mRNA-associating factor links endogenous siRNAs to chromatin regulation. PLoS Genet, 7, e1002249. 119. Gu, S.G., Pak, J., Guang, S., Maniar, J.M., Kennedy, S. and Fire, A. (2012) Amplification of siRNA in Caenorhabditis elegans generates a transgenerational sequence-targeted histone H3 lysine 9 methylation footprint. Nat Genet, 44, 157-164. 120. Mao, H., Zhu, C., Zong, D., Weng, C., Yang, X., Huang, H., Liu, D., Feng, X. and Guang, S. (2015) The Nrde Pathway Mediates Small-RNA-Directed Histone H3 Lysine 27 Trimethylation in Caenorhabditis elegans. Curr Biol, 25, 2398-2403. 121. Buckley, B.A., Burkhart, K.B., Gu, S.G., Spracklin, G., Kershner, A., Fritz, H., Kimble, J., Fire, A. and Kennedy, S. (2012) A nuclear Argonaute promotes multigenerational epigenetic inheritance and germline immortality. Nature, 489, 447-451. 122. Chirn, G.W., Rahman, R., Sytnikova, Y.A., Matts, J.A., Zeng, M., Gerlach, D., Yu, M., Berger, B., Naramura, M., Kile, B.T. et al. (2015) Conserved piRNA Expression from a Distinct Set of piRNA Cluster Loci in Eutherian Mammals. PLoS Genet, 11, e1005652. 123. Lewis, S.H., Quarles, K.A., Yang, Y., Tanguy, M., Fr?zal, L., Smith, S.A., Sharma, P.P., Cordaux, R., Gilbert, C., Giraud, I. et al. (2018) Pan-arthropod analysis reveals somatic piRNAs as an ancestral defence against transposable elements. Nat Ecol Evol, 2, 174-181. 124. Shirayama, M., Seth, M., Lee, H.C., Gu, W., Ishidate, T., Conte, D. and Mello, C.C. (2012) piRNAs initiate an epigenetic memory of nonself RNA in the C. elegans germline. Cell, 150, 65-77. 185 125. Luteijn, M.J., van Bergeijk, P., Kaaij, L.J., Almeida, M.V., Roovers, E.F., Berezikov, E. and Ketting, R.F. (2012) Extremely stable Piwi-induced gene silencing in Caenorhabditis elegans. EMBO J, 31, 3422-3430. 126. Ashe, A., Sapetschnig, A., Weick, E.M., Mitchell, J., Bagijn, M.P., Cording, A.C., Doebley, A.L., Goldstein, L.D., Lehrbach, N.J., Le Pen, J. et al. (2012) piRNAs can trigger a multigenerational epigenetic memory in the germline of C. elegans. Cell, 150, 88-99. 127. Bagijn, M.P., Goldstein, L.D., Sapetschnig, A., Weick, E.M., Bouasker, S., Lehrbach, N.J., Simard, M.J. and Miska, E.A. (2012) Function, targets, and evolution of Caenorhabditis elegans piRNAs. Science, 337, 574-578. 128. 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. et al. (2008) PRG-1 and 21U-RNAs interact to form the piRNA complex required for fertility in C. elegans. Mol Cell, 31, 67-78. 129. Han, T., Manoharan, A.P., Harkins, T.T., Bouffard, P., Fitzpatrick, C., Chu, D.S., Thierry-Mieg, D., Thierry-Mieg, J. and Kim, J.K. (2009) 26G endo- siRNAs regulate spermatogenic and zygotic gene expression in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 106, 18674-18679. 130. Winston, W.M., Molodowitch, C. and Hunter, C.P. (2002) Systemic RNAi in C. elegans requires the putative transmembrane protein SID-1. Science, 295, 2456-2459. 131. Sheth, U., Pitt, J., Dennis, S. and Priess, J.R. (2010) Perinuclear P granules are the principal sites of mRNA export in adult C. elegans germ cells. Development, 137, 1305-1314. 132. Zhang, H. and Fire, A.Z. (2010) Cell autonomous specification of temporal identity by Caenorhabditis elegans microRNA lin-4. Dev Biol, 344, 603-610. 133. Zhang, C., Montgomery, T.A., Fischer, S.E., Garcia, S.M., Riedel, C.G., Fahlgren, N., Sullivan, C.M., Carrington, J.C. and Ruvkun, G. (2012) The Caenorhabditis elegans RDE-10/RDE-11 complex regulates RNAi by promoting secondary siRNA amplification. Curr Biol, 22, 881-890. 186 134. Yang, H., Zhang, Y., Vallandingham, J., Li, H., Florens, L. and Mak, H.Y. (2012) The RDE-10/RDE-11 complex triggers RNAi-induced mRNA degradation by association with target mRNA in C. elegans. Genes Dev, 26, 846-856. 135. Tijsterman, M., May, R.C., Simmer, F., Okihara, K.L. and Plasterk, R.H. (2004) Genes required for systemic RNA interference in Caenorhabditis elegans. Curr Biol, 14, 111-116. 136. Elhassan, M.O., Christie, J. and Duxbury, M.S. (2012) Homo sapiens systemic RNA interference-defective-1 transmembrane family member 1 (SIDT1) protein mediates contact-dependent small RNA transfer and microRNA-21-driven chemoresistance. J Biol Chem, 287, 5267-5277. 137. Duxbury, M.S., Ashley, S.W. and Whang, E.E. (2005) RNA interference: a mammalian SID-1 homologue enhances siRNA uptake and gene silencing efficacy in human cells. Biochem Biophys Res Commun, 331, 459-463. 138. Shih, J.D. and Hunter, C.P. (2011) SID-1 is a dsRNA-selective dsRNA-gated channel. RNA, 17, 1057-1065. 139. Jialin, G., Xuefan, G. and Huiwen, Z. (2010) SID1 transmembrane family, member 2 (Sidt2): a novel lysosomal membrane protein. Biochem Biophys Res Commun, 402, 588-594. 140. Longtine, C., Frenk, S. and Ahmed, S. (2018) Small RNA-mediated genomic silencing promotes telomere stability in the absence of telomerase. 141. Kadekar, P. and Roy, R. (2018) AMPK regulates germline stem cell quiescence and integrity through an endogenous small RNA pathway. 142. Winston, W.M., Sutherlin, M., Wright, A.J., Feinberg, E.H. and Hunter, C.P. (2007) Caenorhabditis elegans SID-2 is required for environmental RNA interference. Proc Natl Acad Sci U S A, 104, 10565-10570. 143. McEwan, D.L., Weisman, A.S. and Hunter, C.P. (2012) Uptake of extracellular double-stranded RNA by SID-2. Mol Cell, 47, 746-754. 187 144. Nuez, I. and F?lix, M.A. (2012) Evolution of susceptibility to ingested double- stranded RNAs in Caenorhabditis nematodes. PLoS One, 7, e29811. 145. Hinas, A., Wright, A.J. and Hunter, C.P. (2012) SID-5 is an endosome- associated protein required for efficient systemic RNAi in C. elegans. Curr Biol, 22, 1938-1943. 146. Harris, K.P. and Tepass, U. (2010) Cdc42 and vesicle trafficking in polarized cells. Traffic, 11, 1272-1279. 147. Raman, P., Zaghab, S.M., Traver, E.C. and Jose, A.M. (2017) The double- stranded RNA binding protein RDE-4 can act cell autonomously during feeding RNAi in C. elegans. Nucleic Acids Res, 45, 8463-8473. 148. Cai, Q., He, B., Kogel, K.H. and Jin, H. (2018) Cross-kingdom RNA trafficking and environmental RNAi-nature's blueprint for modern crop protection strategies. Curr Opin Microbiol, 46, 58-64. 149. Calixto, A., Chelur, D., Topalidou, I., Chen, X. and Chalfie, M. (2010) Enhanced neuronal RNAi in C. elegans using SID-1. Nat Methods, 7, 554- 559. 150. Tomoyasu, Y., Miller, S.C., Tomita, S., Schoppmeier, M., Grossmann, D. and Bucher, G. (2008) Exploring systemic RNA interference in insects: a genome- wide survey for RNAi genes in Tribolium. Genome Biol, 9, R10. 151. Tabara, H., Yigit, E., Siomi, H. and Mello, C.C. (2002) The dsRNA binding protein RDE-4 interacts with RDE-1, DCR-1, and a DExH-box helicase to direct RNAi in C. elegans. Cell, 109, 861-871. 152. Parker, G.S., Eckert, D.M. and Bass, B.L. (2006) RDE-4 preferentially binds long dsRNA and its dimerization is necessary for cleavage of dsRNA to siRNA. RNA, 12, 807-818. 153. Burton, N.O., Burkhart, K.B. and Kennedy, S. (2011) Nuclear RNAi maintains heritable gene silencing in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 108, 19683-19688. 188 154. Blanchard, D., Parameswaran, P., Lopez-Molina, J., Gent, J., Saynuk, J.F. and Fire, A. (2011) On the nature of in vivo requirements for rde-4 in RNAi and developmental pathways in C. elegans. RNA Biol, 8, 458-467. 155. Brenner, S. (1974) The genetics of Caenorhabditis elegans. Genetics, 77, 71- 94. 156. Semple, J.I., Biondini, L. and Lehner, B. (2012) Generating transgenic nematodes by bombardment and antibiotic selection. Nat Methods, 9, 118- 119. 157. Harfe, B.D. and Fire, A. (1998) Muscle and nerve-specific regulation of a novel NK-2 class homeodomain factor in Caenorhabditis elegans. Development, 125, 421-429. 158. Arribere, J.A., Bell, R.T., Fu, B.X., Artiles, K.L., Hartman, P.S. and Fire, A.Z. (2014) Efficient marker-free recovery of custom genetic modifications with CRISPR/Cas9 in Caenorhabditis elegans. Genetics, 198, 837-846. 159. Kamath, R.S., Fraser, A.G., Dong, Y., Poulin, G., Durbin, R., Gotta, M., Kanapin, A., Le Bot, N., Moreno, S., Sohrmann, M. et al. (2003) Systematic functional analysis of the Caenorhabditis elegans genome using RNAi. Nature, 421, 231-237. 160. Pavlidis, P. and Noble, W.S. (2003) Matrix2png: a utility for visualizing matrix data. Bioinformatics, 19, 295-296. 161. Cao, X. and Aballay, A. (2016) Neural Inhibition of Dopaminergic Signaling Enhances Immunity in a Cell-Non-autonomous Manner. Curr Biol, 26, 2398. 162. Taylor, R.C. and Dillin, A. (2013) XBP-1 is a cell-nonautonomous regulator of stress resistance and longevity. Cell, 153, 1435-1447. 163. McCallum, K.C., Liu, B., Fierro-Gonz?lez, J.C., Swoboda, P., Arur, S., Miranda-Vizuete, A. and Garsin, D.A. (2016) TRX-1 Regulates SKN-1 Nuclear Localization Cell Non-autonomously in Caenorhabditis elegans. Genetics, 203, 387-402. 189 164. Liu, Y., Samuel, B.S., Breen, P.C. and Ruvkun, G. (2014) Caenorhabditis elegans pathways that surveil and defend mitochondria. Nature, 508, 406-410. 165. Durieux, J., Wolff, S. and Dillin, A. (2011) The cell-non-autonomous nature of electron transport chain-mediated longevity. Cell, 144, 79-91. 166. Bishop, N.A. and Guarente, L. (2007) Two neurons mediate diet-restriction- induced longevity in C. elegans. Nature, 447, 545-549. 167. Napoli, C., Lemieux, C. and Jorgensen, R. (1990) Introduction of a Chimeric Chalcone Synthase Gene into Petunia Results in Reversible Co-Suppression of Homologous Genes in trans. Plant Cell, 2, 279-289. 168. Hellwig, S. and Bass, B.L. (2008) A starvation-induced noncoding RNA modulates expression of Dicer-regulated genes. Proc Natl Acad Sci U S A, 105, 12897-12902. 169. Le, H.H., Looney, M., Strauss, B., Bloodgood, M. and Jose, A.M. (2016) Tissue homogeneity requires inhibition of unequal gene silencing during development. J Cell Biol, 214, 319-331. 170. Lee, R.C., Hammell, C.M. and Ambros, V. (2006) Interacting endogenous and exogenous RNAi pathways in Caenorhabditis elegans. RNA, 12, 589-597. 171. Zhuang, J.J. and Hunter, C.P. (2011) Tissue specificity of Caenorhabditis elegans enhanced RNA interference mutants. Genetics, 188, 235-237. 172. Billi, A.C., Fischer, S.E. and Kim, J.K. (2014) Endogenous RNAi pathways in C. elegans. WormBook, 1-49. 173. Wu, X., Shi, Z., Cui, M., Han, M. and Ruvkun, G. (2012) Repression of germline RNAi pathways in somatic cells by retinoblastoma pathway chromatin complexes. PLoS Genet, 8, e1002542. 174. Kim, J.K., Gabel, H.W., Kamath, R.S., Tewari, M., Pasquinelli, A., Rual, J.F., Kennedy, S., Dybbs, M., Bertin, N., Kaplan, J.M. et al. (2005) Functional genomic analysis of RNA interference in C. elegans. Science, 308, 1164-1167. 190 175. Blumenfeld, A.L. and Jose, A.M. (2016) Reproducible features of small RNAs in C. elegans reveal NU RNAs and provide insights into 22G RNAs and 26G RNAs. RNA, 22, 184-192. 176. Qadota, H., Inoue, M., Hikita, T., K?ppen, M., Hardin, J.D., Amano, M., Moerman, D.G. and Kaibuchi, K. (2007) Establishment of a tissue-specific RNAi system in C. elegans. Gene, 400, 166-173. 177. Kelly, W.G., Xu, S., Montgomery, M.K. and Fire, A. (1997) Distinct requirements for somatic and germline expression of a generally expressed Caernorhabditis elegans gene. Genetics, 146, 227-238. 178. Yochem, J. and Herman, R.K. (2003) Investigating C. elegans development through mosaic analysis. Development, 130, 4761-4768. 179. Stinchcomb, D.T., Shaw, J.E., Carr, S.H. and Hirsh, D. (1985) Extrachromosomal DNA transformation of Caenorhabditis elegans. Mol Cell Biol, 5, 3484-3496. 180. Mello, C.C., Kramer, J.M., Stinchcomb, D. and Ambros, V. (1991) Efficient gene transfer in C.elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J, 10, 3959-3970. 181. Kelly, W.G. and Fire, A. (1998) Chromatin silencing and the maintenance of a functional germline in Caenorhabditis elegans. Development, 125, 2451-2456. 182. Feinberg, E.H. and Hunter, C.P. (2003) Transport of dsRNA into cells by the transmembrane protein SID-1. Science, 301, 1545-1547. 183. Zeller, P., Padeken, J., van Schendel, R., Kalck, V., Tijsterman, M. and Gasser, S.M. (2016) Histone H3K9 methylation is dispensable for Caenorhabditis elegans development but suppresses RNA:DNA hybrid- associated repeat instability. Nat Genet, 48, 1385-1395. 184. Wu, C.W., Deonarine, A., Przybysz, A., Strange, K. and Choe, K.P. (2016) The Skp1 Homologs SKR-1/2 Are Required for the Caenorhabditis elegans SKN-1 Antioxidant/Detoxification Response Independently of p38 MAPK. PLoS Genet, 12, e1006361. 191 185. Kennedy, S., Wang, D. and Ruvkun, G. (2004) A conserved siRNA-degrading RNase negatively regulates RNA interference in C. elegans. Nature, 427, 645- 649. 186. Simmer, F., Tijsterman, M., Parrish, S., Koushika, S.P., Nonet, M.L., Fire, A., Ahringer, J. and Plasterk, R.H. (2002) Loss of the putative RNA-directed RNA polymerase RRF-3 makes C. elegans hypersensitive to RNAi. Curr Biol, 12, 1317-1319. 187. Yoon, J.S., Mogilicherla, K., Gurusamy, D., Chen, X., Chereddy, S.C.R.R. and Palli, S.R. (2018) Double-stranded RNA binding protein, Staufen, is required for the initiation of RNAi in coleopteran insects. Proc Natl Acad Sci U S A, 115, 8334-8339. 188. Ivashuta, S., Zhang, Y., Wiggins, B.E., Ramaseshadri, P., Segers, G.C., Johnson, S., Meyer, S.E., Kerstetter, R.A., McNulty, B.C., Bolognesi, R. et al. (2015) Environmental RNAi in herbivorous insects. RNA, 21, 840-850. 189. Abdellatef, E., Will, T., Koch, A., Imani, J., Vilcinskas, A. and Kogel, K.H. (2015) Silencing the expression of the salivary sheath protein causes transgenerational feeding suppression in the aphid Sitobion avenae. Plant Biotechnol J, 13, 849-857. 190. Devanapally, S., Allgood, S. and Jose, A.M. (2017) Mating can cause transgenerational gene silencing in Caenorhabditis elegans 191. Gushchanskaia, E.R., Esse, R., Ma, Q., Lau, N. and Grishok, A. (2018) The interplay between small RNA pathways shapes chromatin landscape in C. elegans. 192. Egelhofer, T.A., Minoda, A., Klugman, S., Lee, K., Kolasinska-Zwierz, P., Alekseyenko, A.A., Cheung, M.S., Day, D.S., Gadel, S., Gorchakov, A.A. et al. (2011) An assessment of histone-modification antibody quality. Nat Struct Mol Biol, 18, 91-93. 193. Towbin, B.D., Gonz?lez-Aguilera, C., Sack, R., Gaidatzis, D., Kalck, V., Meister, P., Askjaer, P. and Gasser, S.M. (2012) Step-wise methylation of histone H3K9 positions heterochromatin at the nuclear periphery. Cell, 150, 934-947. 192 194. Lev, I., Seroussi, U., Gingold, H., Bril, R., Anava, S. and Rechavi, O. (2017) MET-2-Dependent H3K9 Methylation Suppresses Transgenerational Small RNA Inheritance. Curr Biol, 27, 1138-1147. 195. Almeida, M.V., Dietz, S., Redl, S., Karaulanov, E., Hildebrandt, A., Renz, C., Ulrich, H.D., K?nig, J., Butter, F. and Ketting, R.F. (2018) GTSF-1 is required for formation of a functional RNA-dependent RNA Polymerase complex in. EMBO J, 37. 196. Thivierge, C., Makil, N., Flamand, M., Vasale, J.J., Mello, C.C., Wohlschlegel, J., Conte, D. and Duchaine, T.F. (2011) Tudor domain ERI-5 tethers an RNA-dependent RNA polymerase to DCR-1 to potentiate endo- RNAi. Nat Struct Mol Biol, 19, 90-97. 197. Duchaine, T.F., Wohlschlegel, J.A., Kennedy, S., Bei, Y., Conte, D., Pang, K., Brownell, D.R., Harding, S., Mitani, S., Ruvkun, G. et al. (2006) Functional proteomics reveals the biochemical niche of C. elegans DCR-1 in multiple small-RNA-mediated pathways. Cell, 124, 343-354. 198. Pavelec, D.M., Lachowiec, J., Duchaine, T.F., Smith, H.E. and Kennedy, S. (2009) Requirement for the ERI/DICER complex in endogenous RNA interference and sperm development in Caenorhabditis elegans. Genetics, 183, 1283-1295. 199. Chen, C.C., Simard, M.J., Tabara, H., Brownell, D.R., McCollough, J.A. and Mello, C.C. (2005) A member of the polymerase beta nucleotidyltransferase superfamily is required for RNA interference in C. elegans. Curr Biol, 15, 378-383. 200. Brangwynne, C.P., Eckmann, C.R., Courson, D.S., Rybarska, A., Hoege, C., Gharakhani, J., J?licher, F. and Hyman, A.A. (2009) Germline P granules are liquid droplets that localize by controlled dissolution/condensation. Science, 324, 1729-1732. 201. Kalinava, N., Ni, J.Z., Peterman, K., Chen, E. and Gu, S.G. (2017) Decoupling the downstream effects of germline nuclear RNAi reveals that H3K9me3 is dispensable for heritable RNAi and the maintenance of endogenous siRNA-mediated transcriptional silencing in. Epigenetics Chromatin, 10, 6. 193 202. Snyder, M.J., Lau, A.C., Brouhard, E.A., Davis, M.B., Jiang, J., Sifuentes, M.H. and Csankovszki, G. (2016) Anchoring of Heterochromatin to the Nuclear Lamina Reinforces Dosage Compensation-Mediated Gene Repression. PLoS Genet, 12, e1006341. 203. Spracklin, G., Fields, B., Wan, G., Becker, D., Wallig, A., Shukla, A. and Kennedy, S. (2017) The RNAi Inheritance Machinery of. Genetics, 206, 1403-1416. 204. Perales, R., Pagano, D., Wan, G., Fields, B.D., Saltzman, A.L. and Kennedy, S.G. (2018) Transgenerational Epigenetic Inheritance Is Negatively Regulated by the HERI-1 Chromodomain Protein. Genetics, 210, 1287-1299. 205. Besseling, J. and Bringmann, H. (2016) Engineered non-Mendelian inheritance of entire parental genomes in C. elegans. Nat Biotechnol, 34, 982- 986. 206. Artiles, K.L., Fire, A.Z. and Fr?kj?r-Jensen, C. (2019) Assessment and Maintenance of Unigametic Germline Inheritance for C. elegans. Dev Cell, 48, 827-839.e829. 207. Sulston, J.E. and Horvitz, H.R. (1977) Post-embryonic cell lineages of the nematode, Caenorhabditis elegans. Dev Biol, 56, 110-156. 208. Vastenhouw, N.L., Brunschwig, K., Okihara, K.L., M?ller, F., Tijsterman, M. and Plasterk, R.H. (2006) Gene expression: long-term gene silencing by RNAi. Nature, 442, 882. 209. Minkina, O. and Hunter, C.P. (2018) Intergenerational Transmission of Gene Regulatory Information in Caenorhabditis elegans. Trends Genet, 34, 54-64. 210. Bosher, J.M., Dufourcq, P., Sookhareea, S. and Labouesse, M. (1999) RNA interference can target pre-mRNA: consequences for gene expression in a Caenorhabditis elegans operon. Genetics, 153, 1245-1256. 211. Montgomery, M.K., Xu, S. and Fire, A. (1998) RNA as a target of double- stranded RNA-mediated genetic interference in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 95, 15502-15507. 194 212. Sapetschnig, A., Sarkies, P., Lehrbach, N.J. and Miska, E.A. (2015) Tertiary siRNAs mediate paramutation in C. elegans. PLoS Genet, 11, e1005078. 213. Grishok, A., Pasquinelli, A.E., Conte, D., Li, N., Parrish, S., Ha, I., Baillie, D.L., Fire, A., Ruvkun, G. and Mello, C.C. (2001) Genes and mechanisms related to RNA interference regulate expression of the small temporal RNAs that control C. elegans developmental timing. Cell, 106, 23-34. 214. McMurchy, A.N., Stempor, P., Gaarenstroom, T., Wysolmerski, B., Dong, Y., Aussianikava, D., Appert, A., Huang, N., Kolasinska-Zwierz, P., Sapetschnig, A. et al. (2017) A team of heterochromatin factors collaborates with small RNA pathways to combat repetitive elements and germline stress. Elife, 6. 215. Koonin, E.V., Makarova, K.S. and Wolf, Y.I. (2017) Evolutionary Genomics of Defense Systems in Archaea and Bacteria. Annu Rev Microbiol, 71, 233- 261. 216. Schlee, M. and Hartmann, G. (2016) Discriminating self from non-self in nucleic acid sensing. Nat Rev Immunol, 16, 566-580. 217. Yang, P., Wang, Y. and Macfarlan, T.S. (2017) The Role of KRAB-ZFPs in Transposable Element Repression and Mammalian Evolution. Trends Genet, 33, 871-881. 218. Borges, F. and Martienssen, R.A. (2015) The expanding world of small RNAs in plants. Nat Rev Mol Cell Biol, 16, 727-741. 219. Holoch, D. and Moazed, D. (2015) RNA-mediated epigenetic regulation of gene expression. Nat Rev Genet, 16, 71-84. 220. Ozata, D.M., Gainetdinov, I., Zoch, A., O'Carroll, D. and Zamore, P.D. (2019) PIWI-interacting RNAs: small RNAs with big functions. Nat Rev Genet, 20, 89-108. 221. Swarts, D.C., Makarova, K., Wang, Y., Nakanishi, K., Ketting, R.F., Koonin, E.V., Patel, D.J. and van der Oost, J. (2014) The evolutionary journey of Argonaute proteins. Nat Struct Mol Biol, 21, 743-753. 195 222. Chavez, J., Murillo-Maldonado, J.M., Bahena, V., Cruz, A.K., Casta?eda- Sortibr?n, A., Rodriguez-Arnaiz, R., Zurita, M. and Valadez-Graham, V. (2017) dAdd1 and dXNP prevent genome instability by maintaining HP1a localization at Drosophila telomeres. Chromosoma, 126, 697-712. 223. Fulcher, N., Derboven, E., Valuchova, S. and Riha, K. (2014) If the cap fits, wear it: an overview of telomeric structures over evolution. Cell Mol Life Sci, 71, 847-865. 224. Palmer, W.H., Hadfield, J.D. and Obbard, D.J. (2018) RNA-Interference Pathways Display High Rates of Adaptive Protein Evolution in Multiple Invertebrates. Genetics, 208, 1585-1599. 225. Parhad, S.S., Tu, S., Weng, Z. and Theurkauf, W.E. (2017) Adaptive Evolution Leads to Cross-Species Incompatibility in the piRNA Transposon Silencing Machinery. Dev Cell, 43, 60-70.e65. 226. Olovnikov, I., Chan, K., Sachidanandam, R., Newman, D.K. and Aravin, A.A. (2013) Bacterial argonaute samples the transcriptome to identify foreign DNA. Mol Cell, 51, 594-605. 227. Swarts, D.C., Jore, M.M., Westra, E.R., Zhu, Y., Janssen, J.H., Snijders, A.P., Wang, Y., Patel, D.J., Berenguer, J., Brouns, S.J.J. et al. (2014) DNA-guided DNA interference by a prokaryotic Argonaute. Nature, 507, 258-261. 228. Makarova, K.S., Wolf, Y.I., van der Oost, J. and Koonin, E.V. (2009) Prokaryotic homologs of Argonaute proteins are predicted to function as key components of a novel system of defense against mobile genetic elements. Biol Direct, 4, 29. 229. H?ck, J. and Meister, G. (2008) The Argonaute protein family. Genome Biol, 9, 210. 230. Hutvagner, G. and Simard, M.J. (2008) Argonaute proteins: key players in RNA silencing. Nat Rev Mol Cell Biol, 9, 22-32. 231. Kwak, P.B. and Tomari, Y. (2012) The N domain of Argonaute drives duplex unwinding during RISC assembly. Nat Struct Mol Biol, 19, 145-151. 196 232. Sheng, G., Zhao, H., Wang, J., Rao, Y., Tian, W., Swarts, D.C., van der Oost, J., Patel, D.J. and Wang, Y. (2014) Structure-based cleavage mechanism of Thermus thermophilus Argonaute DNA guide strand-mediated DNA target cleavage. Proc Natl Acad Sci U S A, 111, 652-657. 233. Chapman, E.J. and Carrington, J.C. (2007) Specialization and evolution of endogenous small RNA pathways. Nat Rev Genet, 8, 884-896. 234. Ambros, V., Lee, R.C., Lavanway, A., Williams, P.T. and Jewell, D. (2003) MicroRNAs and other tiny endogenous RNAs in C. elegans. Curr Biol, 13, 807-818. 235. Ruby, J.G., Jan, C., Player, C., Axtell, M.J., Lee, W., Nusbaum, C., Ge, H. and Bartel, D.P. (2006) Large-scale sequencing reveals 21U-RNAs and additional microRNAs and endogenous siRNAs in C. elegans. Cell, 127, 1193-1207. 236. Cecere, G., Zheng, G.X., Mansisidor, A.R., Klymko, K.E. and Grishok, A. (2012) Promoters recognized by forkhead proteins exist for individual 21U- RNAs. Mol Cell, 47, 734-745. 237. Weng, C., Kosalka, J., Berkyurek, A.C., Stempor, P., Feng, X., Mao, H., Zeng, C., Li, W.J., Yan, Y.H., Dong, M.Q. et al. (2019) The USTC co-opts an ancient machinery to drive piRNA transcription in. Genes Dev, 33, 90-102. 238. Gent, J.I., Schvarzstein, M., Villeneuve, A.M., Gu, S.G., Jantsch, V., Fire, A.Z. and Baudrimont, A. (2009) A Caenorhabditis elegans RNA-directed RNA polymerase in sperm development and endogenous RNA interference. Genetics, 183, 1297-1314. 239. Vasale, J.J., Gu, W., Thivierge, C., Batista, P.J., Claycomb, J.M., Youngman, E.M., Duchaine, T.F., Mello, C.C. and Conte, D. (2010) Sequential rounds of RNA-dependent RNA transcription drive endogenous small-RNA biogenesis in the ERGO-1/Argonaute pathway. Proc Natl Acad Sci U S A, 107, 3582- 3587. 240. Ryazansky, S., Kulbachinskiy, A. and Aravin, A.A. (2018) The Expanded Universe of Prokaryotic Argonaute Proteins. MBio, 9. 197 241. Conine, C.C., Batista, P.J., Gu, W., Claycomb, J.M., Chaves, D.A., Shirayama, M. and Mello, C.C. (2010) Argonautes ALG-3 and ALG-4 are required for spermatogenesis-specific 26G-RNAs and thermotolerant sperm in Caenorhabditis elegans. Proc Natl Acad Sci U S A, 107, 3588-3593. 242. Conine, C.C., Moresco, J.J., Gu, W., Shirayama, M., Conte, D., Yates, J.R. and Mello, C.C. (2013) Argonautes promote male fertility and provide a paternal memory of germline gene expression in C. elegans. Cell, 155, 1532- 1544. 243. Kim, K.M., Abdelmohsen, K., Mustapic, M., Kapogiannis, D. and Gorospe, M. (2017) RNA in extracellular vesicles. Wiley Interdiscip Rev RNA, 8. 244. Valadi, H., Ekstr?m, K., Bossios, A., Sj?strand, M., Lee, J.J. and L?tvall, J.O. (2007) Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol, 9, 654-659. 245. Thomou, T., Mori, M.A., Dreyfuss, J.M., Konishi, M., Sakaguchi, M., Wolfrum, C., Rao, T.N., Winnay, J.N., Garcia-Martin, R., Grinspoon, S.K. et al. (2017) Adipose-derived circulating miRNAs regulate gene expression in other tissues. Nature, 542, 450-455. 246. Ashley, J., Cordy, B., Lucia, D., Fradkin, L.G., Budnik, V. and Thomson, T. (2018) Retrovirus-like Gag Protein Arc1 Binds RNA and Traffics across Synaptic Boutons. Cell, 172, 262-274.e211. 247. Lucas, W.J., Ham, B.K. and Kim, J.Y. (2009) Plasmodesmata - bridging the gap between neighboring plant cells. Trends Cell Biol, 19, 495-503. 248. Maule, A.J., Benitez-Alfonso, Y. and Faulkner, C. (2011) Plasmodesmata - membrane tunnels with attitude. Curr Opin Plant Biol, 14, 683-690. 249. Mermigka, G., Verret, F. and Kalantidis, K. (2016) RNA silencing movement in plants. J Integr Plant Biol, 58, 328-342. 250. Lefebvre, F.A., Benoit Bouvrette, L.P., Perras, L., Blanchet-Cohen, A., Garnier, D., Rak, J. and L?cuyer, ?. (2016) Comparative transcriptomic analysis of human and Drosophila extracellular vesicles. Sci Rep, 6, 27680. 198 251. Simon, M., Spichal, M., Heestand, B., Frenk, S., Hedges, A., Godwin, M., Wellman , A.S., Sakaguchi, A. and Ahmed, S. (2018) DAF-16/Foxo suppresses the transgenerational sterility of prg-1 piRNA mutants via a systemic small RNA pathway. 252. Ow, M.C., Borziak, K., Nichitean, A.M., Dorus, S. and Hall, S.E. (2018) Early experiences mediate distinct adult gene expression and reproductive programs in Caenorhabditis elegans. PLoS Genet, 14, e1007219. 253. Palominos, M.F., Verdugo, L., Gabaldon, C., Pollak, B., Ort?z-Sever?n, J., Varas, M.A., Ch?vez, F.P. and Calixto, A. (2017) Transgenerational Diapause as an Avoidance Strategy against Bacterial Pathogens in. MBio, 8. 254. Chow, F.W., Koutsovoulos, G., Ovando-V?zquez, C., Neophytou, K., Berm?dez-Barrientos, J.R., Laetsch, D.R., Robertson, E., Kumar, S., Claycomb, J.M., Blaxter, M. et al. (2019) Secretion of an Argonaute protein by a parasitic nematode and the evolution of its siRNA guides. Nucleic Acids Res, 47, 3594-3606. 255. Wang, M., Weiberg, A., Lin, F.M., Thomma, B.P., Huang, H.D. and Jin, H. (2016) Bidirectional cross-kingdom RNAi and fungal uptake of external RNAs confer plant protection. Nat Plants, 2, 16151. 256. Liu, H., Wang, X., Wang, H.D., Wu, J., Ren, J., Meng, L., Wu, Q., Dong, H., Kao, T.Y., Ge, Q. et al. (2012) Escherichia coli noncoding RNAs can affect gene expression and physiology of Caenorhabditis elegans. Nat Commun, 3, 1073. 257. Garcia-Silva, M.R., das Neves, R.F., Cabrera-Cabrera, F., Sanguinetti, J., Medeiros, L.C., Robello, C., Naya, H., Fernandez-Calero, T., Souto-Padron, T., de Souza, W. et al. (2014) Extracellular vesicles shed by Trypanosoma cruzi are linked to small RNA pathways, life cycle regulation, and susceptibility to infection of mammalian cells. Parasitol Res, 113, 285-304. 258. Zhang, T., Zhao, Y.L., Zhao, J.H., Wang, S., Jin, Y., Chen, Z.Q., Fang, Y.Y., Hua, C.L., Ding, S.W. and Guo, H.S. (2016) Cotton plants export microRNAs to inhibit virulence gene expression in a fungal pathogen. Nat Plants, 2, 16153. 199 259. Liu, S., da Cunha, A.P., Rezende, R.M., Cialic, R., Wei, Z., Bry, L., Comstock, L.E., Gandhi, R. and Weiner, H.L. (2016) The Host Shapes the Gut Microbiota via Fecal MicroRNA. Cell Host Microbe, 19, 32-43. 260. Whangbo, J.S. and Hunter, C.P. (2008) Environmental RNA interference. Trends Genet, 24, 297-305. 261. Mayoral, J.G., Hussain, M., Joubert, D.A., Iturbe-Ormaetxe, I., O'Neill, S.L. and Asgari, S. (2014) Wolbachia small noncoding RNAs and their role in cross-kingdom communications. Proc Natl Acad Sci U S A, 111, 18721- 18726. 262. Koch, A., Biedenkopf, D., Furch, A., Weber, L., Rossbach, O., Abdellatef, E., Linicus, L., Johannsmeier, J., Jelonek, L., Goesmann, A. et al. (2016) An RNAi-Based Control of Fusarium graminearum Infections Through Spraying of Long dsRNAs Involves a Plant Passage and Is Controlled by the Fungal Silencing Machinery. PLoS Pathog, 12, e1005901. 263. Zhang, J., Khan, S.A., Heckel, D.G. and Bock, R. (2017) Next-Generation Insect-Resistant Plants: RNAi-Mediated Crop Protection. Trends Biotechnol. 264. Bolognesi, R., Ramaseshadri, P., Anderson, J., Bachman, P., Clinton, W., Flannagan, R., Ilagan, O., Lawrence, C., Levine, S., Moar, W. et al. (2012) Characterizing the mechanism of action of double-stranded RNA activity against western corn rootworm (Diabrotica virgifera virgifera LeConte). PLoS One, 7, e47534. 265. Pentinat, T., Ramon-Krauel, M., Cebria, J., Diaz, R. and Jimenez-Chillaron, J.C. (2010) Transgenerational inheritance of glucose intolerance in a mouse model of neonatal overnutrition. Endocrinology, 151, 5617-5623. 266. Radford, E.J., Ito, M., Shi, H., Corish, J.A., Yamazawa, K., Isganaitis, E., Seisenberger, S., Hore, T.A., Reik, W., Erkek, S. et al. (2014) In utero effects. In utero undernourishment perturbs the adult sperm methylome and intergenerational metabolism. Science, 345, 1255903. 267. Kalinava, N., Ni, J.Z., Gajic, Z., Kim, M., Ushakov, H. and Gu, S.G. (2018) C. elegans Heterochromatin Factor SET-32 Plays an Essential Role in Transgenerational Establishment of Nuclear RNAi-Mediated Epigenetic Silencing. Cell Rep, 25, 2273-2284.e2273. 200 268. Woodhouse, R., Buchmann, G., Hoe, M., Harney, D., Larance, M., Boag, P.R. and Ashe, A. (2018) The chromatin modifiers SET-25 and SET-32 are required for initiation but not long-term 2 maintenance of transgenerational epigenetic inheritance. 269. Ni, J.Z., Kalinava, N., Mendoza, S.G. and Gu, S.G. (2018) The spatial and temporal dynamics of the nuclear RNAi-targeted RNA transcripts in Caenorhabditis elegans. 270. Claycomb, J.M., Batista, P.J., Pang, K.M., Gu, W., Vasale, J.J., van Wolfswinkel, J.C., Chaves, D.A., Shirayama, M., Mitani, S., Ketting, R.F. et al. (2009) The Argonaute CSR-1 and its 22G-RNA cofactors are required for holocentric chromosome segregation. Cell, 139, 123-134. 271. Brink, R.A. (1956) A Genetic Change Associated with the R Locus in Maize Which Is Directed and Potentially Reversible. Genetics, 41, 872-889. 272. Chandler, V.L., Eggleston, W.B. and Dorweiler, J.E. (2000) Paramutation in maize. Plant Mol Biol, 43, 121-145. 273. Rassoulzadegan, M., Grandjean, V., Gounon, P., Vincent, S., Gillot, I. and Cuzin, F. (2006) RNA-mediated non-mendelian inheritance of an epigenetic change in the mouse. Nature, 441, 469-474. 274. de Vanssay, A., Boug?, A.L., Boivin, A., Hermant, C., Teysset, L., Delmarre, V., Antoniewski, C. and Ronsseray, S. (2012) Paramutation in Drosophila linked to emergence of a piRNA-producing locus. Nature, 490, 112-115. 275. Brennecke, J., Malone, C.D., Aravin, A.A., Sachidanandam, R., Stark, A. and Hannon, G.J. (2008) An epigenetic role for maternally inherited piRNAs in transposon silencing. Science, 322, 1387-1392. 276. Luteijn, M.J. and Ketting, R.F. (2013) PIWI-interacting RNAs: from generation to transgenerational epigenetics. Nat Rev Genet, 14, 523-534. 277. Lee, H.C., Gu, W., Shirayama, M., Youngman, E., Conte, D. and Mello, C.C. (2012) C. elegans piRNAs mediate the genome-wide surveillance of germline transcripts. Cell, 150, 78-87. 201 278. Seth, M., Shirayama, M., Tang, W., Shen, E.Z., Tu, S., Lee, H.C., Weng, Z. and Mello, C.C. (2018) The Coding Regions of Germline mRNAs Confer Sensitivity to Argonaute Regulation in C. elegans. Cell Rep, 22, 2254-2264. 279. Fr?kj?r-Jensen, C., Jain, N., Hansen, L., Davis, M.W., Li, Y., Zhao, D., Rebora, K., Millet, J.R.M., Liu, X., Kim, S.K. et al. (2016) An Abundant Class of Non-coding DNA Can Prevent Stochastic Gene Silencing in the C. elegans Germline. Cell, 166, 343-357. 280. Lind, M.I. and Spagopoulou, F. (2018) Evolutionary consequences of epigenetic inheritance. Heredity (Edinb), 121, 205-209. 281. van der Graaf, A., Wardenaar, R., Neumann, D.A., Taudt, A., Shaw, R.G., Jansen, R.C., Schmitz, R.J., Colom?-Tatch?, M. and Johannes, F. (2015) Rate, spectrum, and evolutionary dynamics of spontaneous epimutations. Proc Natl Acad Sci U S A, 112, 6676-6681. 282. Grishok, A., Tabara, H. and Mello, C.C. (2000) Genetic requirements for inheritance of RNAi in C. elegans. Science, 287, 2494-2497. 283. Shin, I.T. and Kohara, Y. (1999) NEXTDB: the expression pattern map database for C. elegans . 10, 213-214. 284. Chisholm, A.D. and Hsiao, T.I. (2012) The C. elegans epidermis as a model skin. I: development, patterning, and growth. 1(6), 861?878. 285. Watts, J.S., Guenther, Q., Dalelio, J. and Watts, J.L. (2018) Previously characterized Caenorhabditis elegans RNA interference resistant mutants retain substantial silencing capacity. 286. Whipple, J.M., Youssef, O.A., Aruscavage, P.J., Nix, D.A., Hong, C., Johnson, W.E. and Bass, B.L. (2015) Genome-wide profiling of the C. elegans dsRNAome. RNA, 21, 786-800. 287. Cheng, X., Deng, P., Cui, H. and Wang, A. (2015) Visualizing double- stranded RNA distribution and dynamics in living cells by dsRNA binding- dependent fluorescence complementation. Virology, 485, 439-451. 202 288. Arnold, A., Rahman, M.M., Lee, M.C., Muehlhaeusser, S., Katic, I., Gaidatzis, D., Hess, D., Scheckel, C., Wright, J.E., Stetak, A. et al. (2014) Functional characterization of C. elegans Y-box-binding proteins reveals tissue-specific functions and a critical role in the formation of polysomes. Nucleic Acids Res, 42, 13353-13369. 289. Baugh, L.R., Hill, A.A., Slonim, D.K., Brown, E.L. and Hunter, C.P. (2003) Composition and dynamics of the Caenorhabditis elegans early embryonic transcriptome. Development, 130, 889-900. 290. Iwata, S., Yoshina, S., Suehiro, Y., Hori, S. and Mitani, S. (2016) Engineering new balancer chromosomes in C. elegans via CRISPR/Cas9. Sci Rep, 6, 33840. 291. Dejima, K., Hori, S., Iwata, S., Suehiro, Y., Yoshina, S., Motohashi, T. and Mitani, S. (2018) An Aneuploidy-Free and Structurally Defined Balancer Chromosome Toolkit for Caenorhabditis elegans. Cell Rep, 22, 232-241. 292. Piero Lamelza, J.M.Y., Luke M. Noble, Arielle Isakharov, Meenakshi Palanisamy, and Matthew V. Rockman, H.S.M., Michael Ailion. (2019). Biorxiv, doi: http://dx.doi.org/10.1101/588152. 293. Dominguez, A.A., Lim, W.A. and Qi, L.S. (2016) Beyond editing: repurposing CRISPR-Cas9 for precision genome regulation and interrogation. Nat Rev Mol Cell Biol, 17, 5-15. 294. Stevens, T.J., Lando, D., Basu, S., Atkinson, L.P., Cao, Y., Lee, S.F., Leeb, M., Wohlfahrt, K.J., Boucher, W., O'Shaughnessy-Kirwan, A. et al. (2017) 3D structures of individual mammalian genomes studied by single-cell Hi-C. Nature, 544, 59-64. 295. Gebhardt, J.C., Suter, D.M., Roy, R., Zhao, Z.W., Chapman, A.R., Basu, S., Maniatis, T. and Xie, X.S. (2013) Single-molecule imaging of transcription factor binding to DNA in live mammalian cells. Nat Methods, 10, 421-426. 296. Chen, B.C., Legant, W.R., Wang, K., Shao, L., Milkie, D.E., Davidson, M.W., Janetopoulos, C., Wu, X.S., Hammer, J.A., Liu, Z. et al. (2014) Lattice light- sheet microscopy: imaging molecules to embryos at high spatiotemporal resolution. Science, 346, 1257998. 203 297. Boeck, M.E., Huynh, C., Gevirtzman, L., Thompson, O.A., Wang, G., Kasper, D.M., Reinke, V., Hillier, L.W. and Waterston, R.H. (2016) The time-resolved transcriptome of C. elegans. Genome Res, 26, 1441-1450. 298. Cao, J., Packer, J.S., Ramani, V., Cusanovich, D.A., Huynh, C., Daza, R., Qiu, X., Lee, C., Furlan, S.N., Steemers, F.J. et al. (2017) Comprehensive single-cell transcriptional profiling of a multicellular organism. Science, 357, 661-667. 204