Gu, Weifeng et al. (2011) International Worm Meeting "Single-worm RNA-seq as a tool for following epigenetic silencing."

Ahmed Elewa, Ahmed, Gu, Weifeng, Mello, Craig

[International Worm Meeting, 2011]

Epigenetic silencing is often unstable and thus can vary dramatically from one individual to the next. We are interested in learning more about the mechanisms that underlie this variation. For example are there small-RNAs that are differentially inherited or amplified from one individual to another? In order to explore this type of variation in C. elegans, we first needed RNA-seq cloning protocols that work on single worms. Thus far we have developed a very reliable cloning strategy to recover small RNA cDNA sequences from single animals. We have also developed a fully enzymatic approach for recovering mRNA sequences using 1 ug of total RNA and are optimizing the conditions to clone mRNA from single worms. Using these new tools, we are exploring the animal-to-animal variation in endogenous and experimentally-induced small RNA species. We will report on our findings at the meeting.

Gu, Weifeng et al. (2009) International Worm Meeting "Multiple germline 22G-RNA pathways maintain genome integrity in C. elegans."

Batista, Pedro, Shirayama, Masaki, Vasale, Jessica, Mello, Craig, Chaves, Daniel, Keys, Jennifer, Claycomb, Julie, Conte, Jr, Darryl, Gu, Weifeng

[International Worm Meeting, 2009]

While analyzing RNA prepared from various C. elegans strains, we noticed that a small RNA species visible in ethidium-bromide stained gels was absent in a number of mutant strains deficient in RNAi and transposon silencing. To explore the nature of this abundant RNA species, we prepared cDNA libraries and utilized deep-sequencing technology to obtain reads corresponding to millions of genome-matched small RNAs. We refer to these small RNAs as 22G-RNAs due to the predominance of a 22-nucleotide length and the presence of a 5'' G. Surprisingly, roughly 50% of the 22G-RNAs are antisense to over 5000 germline-expressed mRNAs, while the remaining 50% match intergenic loci, pseudogenes and transposons. A breakthrough in our understanding of the functions of 22G-RNAs came with the analysis of small RNAs associated with two distinct, germline-expressed Argonaute proteins R06C7.1 (WAGO-1) and CSR-1. Both Argonautes localize prominently to P-granules (see also abstract by Claycomb et al.). Despite their overlapping expression pattern, WAGO-1 and CSR-1 interact with non-overlapping sets of the germline-expressed 22G-RNA species. CSR-1 interacts with the majority of the mRNA-targeted 22G-RNAs, while WAGO-1 interacts with 22G-RNAs that target transposons, pseudogenes, intergenic loci, and a minority of genes. Tiling array studies suggested that only a small number of mRNAs targeted by 22G-RNAs are down-regulated. These mRNAs were exclusively the targets of WAGO-1, consistent with the role of this pathway in transposon silencing. In contrast, CSR-1 appears to utilize its small RNA co-factors to promote the proper alignment of kinetochores, rather than to down-regulate its targets (see Claycomb et al.). Our genetic studies indicate that CSR-1 and WAGO-1 share several upstream factors required for 22G-RNA biogenesis, including the dicer-related helicase DRH-3, the tudor-domain protein EKL-1, and the RNA-dependent RNA polymerases (RdRPs) RRF-1 and EGO-1. How these upstream factors properly identify distinct targets and load the resulting 22G-RNA species onto the appropriate Argonautes remains a mystery. Insight into this question may come from analyzing pathway-specific factors such as the b-nucleotidyl transferase-related protein RDE-3 and the 3''-5'' exonuclease-related protein, MUT-7, which are both required for the WAGO-1 pathway. Our findings suggest that distinct Argonaute/22G-RNA pathways exert genome-scale surveillance that is essential for germline maintenance as well as the faithful transmission of both genetic and epigenetic information. (Funded by NIH GM58800).

Weifeng Gu et al. (2007) International Worm Meeting "Analysis of a novel DRH-3 dependent RNA in C. elegans."

Craig C. Mello, Weifeng Gu, Darryl Jr. Conte, Chun-Chieh G. Chen

[International Worm Meeting, 2007]

RNAi is highly conserved in a variety of organisms including worms, fruit flies, plants, and mammals. RNAi related mechanisms are involved in developmental gene regulation, heterochromatin formation and antivirus defense. Previously, Duchaine et al1 from our lab identified a number of Dicer interacting proteins including DRH-3, a DExH-box RNA helicase. DRH-3 is required for RNAi, transgene silencing, and chromosome segregation. Worms depleted of DRH-3 fail to accumulate several small RNAs including endogenous siRNA, tiny noncoding RNAs, and X cluster RNAs. We have recently identified a few temperature-sensitive alleles of drh-3. At non-permissive temperature, these mutants not only exhibit all the phenotypes previously observed with a deletion strain, but also exhibit a high incidence of males and a mutator phenotype. Interestingly, drh-3 mutants and the deletion strain also lack a novel class of ~22nt small RNA with a 5 triphosphate. Unlike X cluster RNAs, tncRNAs and endogenous siRNAs, the ~22nt RNA is not sensitive to the depletion of Dicer, ERGO-1 or RRF-3. We cloned and sequenced several thousand cDNAs corresponding to these ~22nt RNAs, and found ~40% are complementary to mRNA and anther ~40% map to intergenic regions. Currently, we are investigating the roles of this novel RNA in gene regulation and chromatin remodeling. 1 Duchaine et al. Cell 124, 343-354, 2006. .

Dai, Hui et al. (2021) International Worm Meeting "RNA polyphosphatase PIR-1 regulates endogenous small RNA pathways"

Dai, Hui, Chaves, Daniel, Gu, Weifeng

[International Worm Meeting, 2021]

Eukaryotic 5′-triphosphorylated RNAs (ppp-RNAs) are tightly regulated to promote cellular functions and prevent recognition by antiviral RNA sensors. For example, RNA capping enzymes utilizes their triphosphatase domains to remove the gamma phosphates of ppp-RNAs during RNA capping processes, generating capped RNAs for RNA protection, nuclear export and translation. PIR-1 (phosphatase that interacts with RNA and ribonucleoprotein particle 1) is closely related to RNA capping enzymes. However, some members of PIR-1 family may serve as an RNA polyphosphatase, removing both the beta and gamma phosphates from ppp-RNAs. Here, we show that C.elegans PIR-1 is indeed an RNA polyphosphatase, converting ppp-RNA to p-RNA. Although WT PIR-1 dephosphorylates ppp-RNA and then rapidly releases product p-RNA, the catalytic-dead PIR-1(C150S) selectively binds and remains tightly bound to ppp-RNA substrates without dephosphorylation. Our proteomics analyses revealed that PIR-1 interacts with ERI (Enhanced RNAi) complex components, including DCR-1, RRF-3, DRH-3, ERI-1b and RDE-8, all of which are required for the biogenesis of 26G-RNAs. We showed that PIR-1 is also required for the biogenesis of 26G-RNAs. For example, 26G-RNAs that depend on the ALG-3/4 Argonautes, were ~10-fold less abundant in pir-1 mutants than in WT worms. We showed that 26G-RNAs are produced in a semi-phased manner along mRNA templates via a coordinated action of RRF-3, which generates short triphosphorylated double-stranded (ds) RNAs and Dicer, which dices blunt-ended dsRNAs in a special mode. PIR-1 plays at least two roles in this process:1) promoting recruitment of Dicer to triphosphorylated dsRNAs; and 2) dephosphorylating 26G-RNA precursors. Moreover, we showed that PIR-1 may be required for the biogenesis of CSR-1-dependent small RNAs either directly or indirectly. Our findings suggest that PIR-1 likely modulates both Dicer-dependent and Dicer-independent small RNAs and provide insight into how cells utilize a conserved RNA phosphatase to respond and regulate cellular and viral ppp-RNA species. keywords: RNA phosphatase; RNAi; germline gene regulation; regulation of triphosphorylated RNA; spermatogenesis; embryogenesis; germline small RNAs; mRNA regulation; double-stranded RNAs; RNA binding proteins.

Dai, Hui et al. (2021) International Worm Meeting "A convenient strategy for generating small RNA/mRNA high-throughput sequencing libraries"

Gu, Weifeng, Nguyen, An-Phong, Dai, Hui

[International Worm Meeting, 2021]

High-throughput sequencing has become a standard and powerful tool for analyzing nucleic acids in this big data era. To sequence RNA/DNA, a cDNA/DNA library is usually generated with target sequences ligated with specific 5' and 3' linkers. Unlike mRNA, small RNA often contains multiple modifications including 5' cap or triphosphate and 2'-O- methyl, requiring additional processing steps before linker additions during the cloning processes. And due to low expression levels, it is difficult to clone small RNA starting with a small amount of total RNA. To simplify the cloning process, we have developed a new strategy to clone 5' modified or unmodified small RNA in a one-pot reaction (inside a single PCR tube) with all the steps carried out sequentially using liquid manipulation and as little as 20 ng total RNA (at a single worm level). The 7-hour cloning process only needs ~1-hour labor, a mutiple-fold labor reduction as compared to the canonical cloning methods. Moreover, our method can efficiently clone purified mRNA, simplifying the need to prepare two cloning systems, one for small RNA and the other for mRNA; the barcoded PCR primers utilized in the cDNA amplication process (PCR) are also compatible with those in the genomic DNA cloning protocols, basically unifying all the PCR primers required for amlifying cDNA and DNA (a significant cost reduction since multiple barcoded primers are needed in PCR for pooling samples in the final step). Our method is more convenient for cloning modified RNA, sensitive, versatile and cost-effective than available methods. Moreover, we are developing a low cost home-made Ribo-minus method for enriching mRNA to replace the cost-prohibitive commercial kits. These methods will greatly promote nucleic acid related research.

Vasale, Jessica J. et al. (2009) International Worm Meeting "A two-step RdRP pathway for the Biogenesis of Endogenous Small RNAs."

Conte Jr., Darryl, Gu, Weifeng, Mello, Craig C., Vasale, Jessica J.

[International Worm Meeting, 2009]

Several classes of Argonaute-associated small RNAs have been shown to play an important role in gene regulation. These include micro RNAs (miRNAs), small interfering RNAs (siRNAs) and Piwi interacting RNAs (piRNAs). Recently, we have characterized a class of 22-nucleotide small RNAs, termed 22G-RNAs, that are expressed abundantly in the germline of C. elegans (see Gu et al.). RNA-dependent RNA polymerase (RdRP) proteins are required for the de novo synthesis of 22G-RNAs, which bear a 5'' tri-phosphate and are thought to be loaded directly onto Argonaute proteins without further processing. We noticed that several loci (termed 22G/26-RNA loci) are targeted by both 22G-RNAs and by a second 26-nucleotide RNA species. Others have shown that the 26-nt RNAs are likely to be 5'' mono-phosphorylated. Of the four RdRPs in C. elegans, RRF-1 and EGO-1 are required for both RNAi and the accumulation of essentially all endogenous 22G-RNA species. RRF-3 was identified as a Dicer interacting protein and as a component of Eri-pathway, an endogenous RNAi pathway that competes with the exogenous (exo-) RNAi pathway. The function of the remaining RdRP, RRF-2, is still unknown. RRF-3 and other Eri-pathway factors target a number of endogenous loci for silencing. We have found that 22G/26-RNA loci are targets of the Eri pathway and thus depend on both Dicer and RRF-3 for silencing. Interestingly, both the 26 nt and 22G-RNA species were dependent on RRF-3, while a second RdRP (RRF-1, EGO-1 or both, depending on the locus) was required for the production of the 22G-RNA species at these loci. Given the mono-phosphorylated nature of the 26 nt RNA and its dependence on both RRF-3 and Dicer, we hypothesize that these small RNAs arise from the Dicer-dependent processing of RRF-3 dsRNA products. We speculate that these 26 nt Dicer products are loaded into the Eri-pathway Argonaute ERGO-1, which cleaves an mRNA target and leads to the recruitment of RRF-1 or EGO-1. The second RdRP step produces the 22G-RNAs that are loaded onto members of the WAGO family of secondary Argonautes to mediate silencing. These findings indicate that worm RdRPs function with Dicer or independently to produce small RNAs. Why the Eri-dependent Dicer products are 26 nt long, while other Dicer products such as miRNAs are only ~22 nt in length remains a mystery.

Carey, James F. et al. (2011) International Worm Meeting "Condensin II is required for RNAi-mediated transcriptional gene silencing."

Carey, James F., Mello, Craig C., Hagstrom, Kirsten A., Gu, Weifeng

[International Worm Meeting, 2011]

Condensin protein complexes are conserved regulators of chromosome architecture and are essential for chromosome segregation. Here we define a new role for condensin in RNA interference. We first establish that exogenous double-stranded RNA triggers both transcriptional and post-transcriptional silencing of the complimentary target gene during "classic RNAi" in C. elegans. In condensin II mutants, RNAi-mediated transcriptional silencing is not effective and dsRNA exposure fails to reduce target pre-mRNA and mRNA levels. Nevertheless, deep sequencing revealed that condensin II mutants produce robust levels of secondary small interfering RNAs (siRNAs) in response to dsRNA. Thus condensin II mutants produce siRNAs but these siRNAs cannot execute transcriptional silencing. Similarly, RNAi-mediated silencing of repetitive transgenes and of certain germline transposons is defective in condensin II mutants yet siRNAs are produced. We show that condensin is required for RNAi in non-mitotic cells, and gene expression profiling suggests that condensin does not act by mis-regulating RNAi pathway genes. Together, our results demonstrate that classic exogenous RNAi in C. elegans has a transcriptional component that requires condensin II downstream of secondary siRNA production. We speculate that secondary siRNAs feed back to the target locus and regulate transcription by creating repressive chromatin architecture in a condensin-dependent manner.

Chaves, Daniel et al. (2009) International Worm Meeting "PIR-1 is a 5' RNA phosphatase that interacts with Dicer and is essential for C. elegans development."

Mitani, Shohei, Chaves, Daniel, Yates III, John, Moresco, James, Mello, Craig, Gu, Weifeng

[International Worm Meeting, 2009]

Proteomics screens have identified PIR-1 as a major component of C. elegans Dicer complexes (Duchaine et al., 2006). PIR-1 is strongly conserved in metazoans and is related to the phosphatase domain of the mRNA capping enzyme. Human PIR1 sequentially removes the 5''-end g- and b-phosphates from triphosphorylated RNA molecules in vitro, leaving a 5''-monophosphate (Yuan et al., 1998; Deshpande et al., 1999). A recent study has established human PIR1 as a p53 transcription factor target, and suggests that changes in PIR1 levels affect cell proliferation (Helin et al., 2008). C. elegans pir-1 deletion mutants are sterile and arrest development as young adults. Their germline is smaller than normal and exhibits no signs of oogenesis. Malformed vulvae and occasional bursting also suggest a role in somatic development. Despite the severity of the phenotype, these animals remain active for the duration of a normal life span. These defects are fully rescued by GFP and 3xFlag fusions of pir-1. PIR-1::GFP is expressed in all tissues throughout development where it localizes in nuclei. Immunoprecipitation of these functional PIR-1 fusion proteins confirmed Dicer as a robust interactor and, through a combination of Western blotting and MudPIT proteomics, led to the identification of several other RNAi-related interactors. These novel interactions are now under further investigation. Northern blotting and deep sequencing experiments suggest that arrested pir-1 animals are wild-type for all known endogenous small RNA pathways, including Dicer-dependent pathways such as miRNA biogenesis. Exogenously triggered RNAi is also unaffected in pir-1 mutants, as exo-siRNAs can be detected and are accompanied by downregulation of their target mRNAs. Furthermore, these small RNAs have the same 5'' phosphorylation status as those of wild-type animals. Current efforts are centered on exploring whether and how PIR-1 functions in concert with Dicer to promote development, perhaps through a hitherto uncharacterized small RNA pathway.

Seth, Meetu et al. (2013) International Worm Meeting "A gene-activating pathway mediated by small RNAs (RNAa) protects self-transcripts from epigenetic silencing in C. elegans."

Shirayama, Masaki, Conte Jr, Darryl, Ishidate, Takao, Gu, Weifeng, Mello, Craig C., Seth, Meetu

[International Worm Meeting, 2013]

Plants and animals employ a sophisticated array of mechanisms to detect and silence pathogenic nucleic acids. For example, the well-studied, sequence-specific silencing mechanism, RNA interference (RNAi), detects pathogenic activity by scanning for double-stranded (dsRNA), a hallmark of RNA-dependent RNA replication. However, many pathogens do not produce dsRNA, and it is clear that organisms can utilize additional strategies to detect and silence foreign gene expression. Recent studies on transgene silencing in C. elegans have uncovered an RNA surveillance system mediated by the Piwi Argonaute, PRG-1, and its genomically-encoded piRNA cofactors. Remarkably this system does not appear to recognize a structural or molecular feature of foreign sequences, but rather identifies foreign RNA by comparing the sequence information itself to a memory of previous gene expression. Interestingly, our findings also suggest that under some circumstances foreign sequences can be adopted (or "licensed", see Johnson and Spence (2011) Science, 333:1311) as "self." One phenomenological manifestation of licensing is the ability of an "adopted" transgene to send a sequence-specific anti-silencing signal that can convert a homologous epigenetically silent allele to an active state. We now have evidence that this transactivation mechanism is mediated by the CSR-1 Argonaute. We find that CSR-1 activity is required parentally (both maternally and paternally) and that heterozygosity for a csr-1 deletion (in either parent) abolishes transactivation. The cloning of CSR-1-associated small RNA reveals that transactivation correlates with the accumulation of CSR-1-22G-RNAs targeting the foreign (GFP) part of the transgene. We also provide evidence that a licensing signal builds up over multiple generations, such that expressed GFP transgenes become more resistant to silencing over time. Our findings are consistent with a model in which Argonaute systems function together to recognize foreign sequences by constantly comparing de novo RNA expression with a memory of RNA expression from previous generations.

Batista, Pedro J. et al. (2009) International Worm Meeting "Characterization of the small RNA populations of C. briggsae and natural isolates of C. elegans."

Hammell, Molly C., Mello, Craig C., Shirayama, Masaki, Batista, Pedro J., Youngman, Elaine M., Gu, Weifeng

[International Worm Meeting, 2009]

Small-RNA-mediated pathways are involved in a myriad of processes that regulate gene expression. In Caenorhabditis elegans several classes of endogenous small RNAs have been identified. At least two of these classes - the Piwi-associated small RNAs (21U-RNAs) and the worm argonaute (WAGO)-associated small RNAs (22G-RNAs) - are expressed in and essential for the proper development and function of the germline. However the biogenesis and functions of these small RNA classes remain almost entirely mysterious. We have taken a phylogenetic approach to explore small RNA biogenesis and function in Caenorhabditis. We have sequenced small RNA libraries generated from wild isolates of C. elegans (AB1, KR314, JU258, RW7000 and CB4856), as well as from the related species Caenorhabditis briggsae. As expected, the C. elegans wild strains exhibited globally similar levels for the various known classes of small RNAs. However, substantial variations from the N2 levels were observed at fewer than 10% of the 22G-RNA-generating loci, and at fewer than 15% of the 21U loci. In some cases one or more wild isolates exhibited a complete lack of specific 22G or 21U species that are abundant in N2. Our analysis so far suggests that these changes rarely reflect deletions that remove the corresponding sequence, although in many cases there are small insertions or deletions in nearby regions. For those cases where the corresponding regions remain structurally intact we are now searching by DNA sequencing in the wild isolates for alterations that might underlie the absence or acquisition of 22G or 21U-RNA species. Small RNAs with the hallmarks of both 22G- and 21U-RNAs are present in C. briggsae, as are the Argonautes and other factors involved in their biogenesis. As in C. elegans, 22G-RNAs are the most abundant small RNAs in C. briggsae. The C. briggsae 21U-RNAs are highly divergent but associated to the same upstream motif originally characterized in C. elegans, and like their C. elegans counterparts begin predominanty with a U and are modified at the 3'' end. Curiously, while the vast majority of 21U-RNAs in C. elegans arise from 2 clusters on chromosome IV, in C. briggsae these small RNAs are generated from at least 3 clusters: 2 on chromosome IV and 1 on chromosome I. Ultimately, we believe this examination of small RNA populations in both closely- and distantly-related nematode strains will provide important insight into the largely mysterious mechanisms that underlie the biogenesis and function of these small RNA pathways.