Froehlich, Jonathan et al. (2017) International Worm Meeting "Targeted, high-throughput interrogation of regulatory elements in C. elegans."

Rajewsky, Nikolaus, Herzog, Margareta, Froehlich, Jonathan

[International Worm Meeting, 2017]

Gene regulatory elements determine when, where and how much a protein is expressed. C. elegans allows tracking of cell-, tissue- or developmental stage specific expression in vivo. Furthermore it is possible to evaluate viability or other phenotypes and to observe several generations in a short time. Regulatory regions controlling gene expression have been studied in C. elegans by cloning different regions and injection of each construct ("bashing"). Few elements have been isolated in forward genetic screens and studied for effects on phenotype. Here we present our approach for unbiased interrogation of regulatory regions for a given locus, with CRISPR/Cas9 based systematic mutation. Using strains with inducible Cas9, staged embryos, and pools of sgRNAs, we are able to obtain thousands of independent mutant F1s in parallel, requiring very few injections. We have set up a targeted sequencing approach using large 3-6 kb amplicons to analyze indels with deep sequencing. As a proof of concept we applied this method to the his-72 locus. We focused on the 3' untranslated region, also targeting promoter-, coding-, and intron sequences at different developmental stages. In addition, we cover the lin-41 3'UTR with different sets of sgRNAs and analyze the depletion of genotypes over several generations. We will report about these ongoing experiments.

Froehlich, Jonathan et al. (2021) International Worm Meeting "Parallel genetics of regulatory sequences using induced genome editing"

Herzog, Margareta, Uyar, Bora, Rajewsky, Nikolaus, Akalin, Altuna, Theil, Kathrin, Froehlich, Jonathan, Glazar, Petar

[International Worm Meeting, 2021]

Understanding how regulatory sequences control gene expression is fundamental to explain how phenotypes arise in health and disease. The functions of regulatory elements must ultimately be understood within their genomic environment and developmental- or tissue-specific contexts. Here, we used induced Cas9 expression and multiplexed guide RNAs in C. elegans to create hundreds of mutations in enhancers, promoters and 3′ UTRs of 16 genes in parallel. We then analyzed the resulting complex populations by either selecting for phenotypic traits or reporter expression, or by DNA- or RNA amplicon sequencing of bulk samples. We developed a software pipeline, crispr-DART, to analyze targeted sequencing and describe the characteristics of >12,000 dsDNA break-induced indel mutations. We also analyzed the lin-41 3′ UTR and found that the two let-7 miRNA binding sites can function independently and that one of the sites is more important for mRNA repression and phenotype. In summary, our approach enables highly parallelized functional analysis of regulatory sequences in vivo.

Froehlich, Jonathan et al. (2019) International Worm Meeting "Conditional, targeted and multiplexed mutagenesis of regulatory sequences in animals."

Froehlich, Jonathan, Schilling, Marcel, Theil, Kathrin, Rajewsky, Nikolaus, Uyar, Bora, Herzog, Margareta, Akalin, Altuna

[International Worm Meeting, 2019]

High-throughput mutagenesis using CRISPR/Cas9 has been successfully applied in cell lines to interrogate regulatory elements or protein coding variants. However, regulation and function is often highly cell-type- and development specific and therefore only appears when tested in an animal. Here we use transient Cas9 induction and pooled guide RNAs in C. elegans to generate thousands of mutant animals in parallel. For example, when we target a 250 bp region of interest, indel mutations create hundreds of unique genotypes in the analyzed population. We apply this method to target regulatory regions, in particular 3'UTRs, of more than a dozen genes and isolate mutations which lead to very clear loss-of-function phenotypes. For one 3'UTR we found that a surprising number of insertions in different regions can have strong repressive effects and consequent phenotypes, while deletions are well tolerated. We also show that with targeted RNA sequencing we can directly measure the effect of different 3'UTR mutations on mRNA levels at particular developmental stages within a population. Altogether we created several thousand mutations which allow us to look at DNA repair patterns created by Cas9 in the germline and to compare efficiencies of different sgRNAs. Our method is useful to understand genotype to phenotype relations and to discover new functional elements directly in an animal.

Stoeckius, Marlon et al. (2011) International Worm Meeting "Transcriptome and proteome profiling during oocyte-to-embryo transition in Caenorhabditis elegans suggests paternal contribution to early development."

Chen, Wei, Kirchner, Marieluise, Thierfelder, Nadine, Selbach, Matthias, Rajewsky, Nikolaus, Stoeckius, Marlon, Grun, Dominic, Piano, Fabio

[International Worm Meeting, 2011]

The oocyte-to-embryo transition is a fundamental event in early development and one of the most dynamic processes in biology. During this transition the fusion of two highly specialized cells, sperm and oocyte, triggers the development of a mitotically-dividing totipotent embryo. This entire process is thought to be driven by post-transcriptional gene regulation in most animals. In C. elegans not only the oocyte-to-embryo transition, but also the first two cell divisions are thought to be entirely post-transcriptionally driven. During these divisions stereotypical lineage and axis specification is already occurring. It is believed that these processes rely on maternally provided mRNAs and proteins. Potential critical functions of sperm components are still a matter of debate. Although a number of key players are known, the oocyte-to-embryo transition remains poorly understood. Therefore, we sought to obtain a global characterization of changes on coding and non-coding transcriptome and proteome during this fundamental developmental process in C. elegans. While oocytes can be obtained in large quantities, collecting many precisely staged embryos has been impractical. Thus, C. elegans embryogenesis has not been amenable to most high-throughput genomics or biochemistry assays. We devised a method to collect staged C. elegans embryos by fluorescence-activated cell sorting (eFACS). A single eFACS run routinely yields tens of thousands of almost perfectly staged one-cell stage embryos. To this end, we developed a method that uses stable isotope labeling by amino acids to quantify changes in the abundance of thousands of proteins between any two C. elegans samples. We combined these techniques to profile the expression of coding and non-coding RNAs, small RNAs and the amount of up to 4,800 proteins in oocytes, one-cell-, two-cell stage embryos and sperm. We discern highly orchestrated and decoupled changes of proteins and mRNAs, indicating a high level of post-transcriptional regulation upon fertilization. We also discovered dynamic expression between and within almost all classes of small RNAs. We have evidence for a large number of putatively sperm-derived mRNAs, 21U-RNAs and proteins in the zygote, suggesting an unexplored substantial paternal contribution to early development. Moreover the data challenges the notion that the early C. elegans embryo is transcriptionally silent. We will report on the currently ongoing analysis and validations of our data.

Kirchner, Marieluise et al. (2010) C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany "Quantitative proteome and transcriptome analysis of C. elegans and C. briggsae development"

Stoeckius, Marlon, Selbach, Matthias, Dieterich, Christoph, Kirchner, Marieluise, Thierfelder, Nadine, Chen, Wei, Maaskola, Jonas, Rajewsky, Nikolaus, Gunsalus, Kris C

[C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany, 2010]

Development can be interpreted as the regulated expression of genes over lifetime of an organism. Transcription profiling has been extensively used to measure changes in mRNA levels during animal development. However, the final products of most genes are proteins and it is not clear how informative mRNA expression patterns are for the protein level. Another fundamental open problem for understanding development is the question how mRNA and protein levels are conserved during evolution and how conservation of these levels relate to each other. These questions have been impossible to address on a genome-wide level since accurate and yet global quantification of proteins in vivo is technically challenging. Thus, a developmental proteome has not yet been obtained for any multicellular organism. To overcome these problems, we developed a new method which allowed us to accurately measure changes in abundance of thousands of proteins in C. elegans and C. briggsae. Similar to SILAC (Stable Isotope Labelling by Amino acids in Cell culture), our method is based on the metabolic incorporation of stable isotope containing amino acids. We generated heavy SILAC worms by cultivating them on a diet of SILAC labelled E. coli. A mixture of heavy SILAC worms from different developmental stages contained a wide range of isotopically labeled proteins that served as an internal standard for accurate mass spectrometry-based quantification. This approach enabled us to identify and quantify changes in expression levels of thousands of proteins (FDR<1%) across six major developmental stages (mixed embryos, larval stages L1 through L4, adult hermaphrodites) for C. elegans and C. briggsae. In parallel, we quantified changes in mRNA levels by massive paired-end next generation sequencing in the same samples. We will report on the currently ongoing analyses of our proteome and transcriptome profiling, including cross-species comparisons of developmental regulation of gene expression

Stoeckius, Marlon et al. (2010) C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany "High resolution transcriptome and proteome profiling during oocyte to embryo transition in Caenorhabditis elegans"

Piano, Fabio, Gunsalus, Kris C, Stoeckius, Marlon, Chen, Wei, Dieterich, Christoph, Ivanov, Andranik, Grun, Dominic, Anna-Carina, Jungkamp, Nadine, Thierfelder, Selbach, Matthias, Kirchner, Marieluise, Maaskola, Jonas, Rajewsky, Nikolaus

[C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany, 2010]

The oocyte-to-embryo transition is a fundamental event in early development. During this transition, cell-cycle arrested oocytes are rapidly converted into the mitotically-dividing embryo. Within 40 minutes after fertilization, the C. elegans embryo has completed oocyte meiotic maturation, pronuclear fusion and the first intrinsically asymmetric division is initiated. This entire developmental process is thought to be dr iven by post-transcriptional gene regulation. Although a number of key players are known, most of the molecular underpinnings of this transition remain poorly understood. Therefore, we sought to obtain a global characterization of changes on transcriptome and proteome during this critical developmental progression. To this end, we developed a new method which uses in vivo SILAC to directly quantify changes in the abundance of thousands of proteins between two C. elegans samples (Thierfelder et al , unpublished). We combined this method with our recently developed method to obtain synchronized embryonic samples (""eFACS"", Stoeckius et al , 2009). We combined these techniques to measure mRNAs, small RNAs (microRNAs, 21URNAs, 22G-RNAs, 26G-RNAs, and others) and the fold changes of thousands of proteins in oocytes, 1-cell stage embryos, 2-4 cell stage embryos and sperm. We observe highly dynamic and orchestrated changes on all levels (mRNAs, proteins, small RNAs) within t his snapshot of early development. We will report in detail on the results of the currently ongoing analyses of these large data sets.

Mora-Martin, A. et al. (2017) International Worm Meeting "Possible regulation of gene expression by RDE-4 and ADAR through dsRNA in introns."

Grishok, A., Mora-Martin, A.

[International Worm Meeting, 2017]

RDE-4 is a double-stranded RNA (dsRNA)-binding protein and a member of the Dicer complex required to initiate silencing in response to exogenous dsRNA. Research in our lab revealed that RDE-4 cooperates with a chromatin-binding protein ZFP-1, a C. elegans homolog of mammalian AF10, in regulation of hermaphrodite-specific neuron (HSN) migration. Supporting this fact, we reported that rde-4(null) and zfp-1(lof) mutants showed: 1) similar profiles of changes in gene expression (Grishok et al., 2008) and 2) HSN undermigration phenotypes (Kennedy and Grishok, 2014). However, the mechanistic connection between ZFP-1 and RDE-4 is not clear. Our research identified DOT-1.1, a histone methyltransferase, as an essential partner of ZFP-1 in transcription regulation (Cecere et al., 2013). Moreover, we found that dot-1.1 lof mutants exhibit HSN migration defects, similar to those observed in rde-4 and zfp-1 mutants. Therefore, we considered the possibility that RDE-4 directly or indirectly regulates dot-1.1 or zfp-1 genes. RDE-4 is well known to compete with ADAR proteins in binding with dsRNAs substrates. ADARs are dsRNA-specific adenosine deaminase enzymes that convert adenosine to inosine, which is then read as G in sequencing. We found that the dot-1.1 gene contains dsRNA structures in its introns that are targeted by ADARs, according to the sites of deamination identified genome wide by the Bass (Whipple et al., 2015) and Rajewsky (Ivanov et al., 2015) labs. On the basis of these findings, we propose the possibility that RDE-4 and ADARs regulate dot-1.1 pre-mRNA splicing, which contributes to the HSN phenotype observed in the rde-4 mutant worms.

Mangone, Marco et al. (2010) C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany "The landscape of C. elegans 3'UTRs."

Sugano, Sumio, Suzuki, Yukata, Salehi-Ashtiani, Kourosh, Gunsalus, Kris C, Rajewsky, Nikolaus, Harkins, Tim, Prasad Manoharan, Arun, Thierry-Mieg, Danielle, Thierry-Mieg, Jean, Mis, Emily, Mackowiak, Sebastian, Han, Ting, Khivansara, Vishal, Kim, John, Piano, Fabio, Zegar, Charles, Gutwein, Michelle, Mangone, Marco, Kohara, Yuji, Buffard, Pascal

[C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany, 2010]

Three-prime untranslated regions (3UTRs) of metazoan mRNAs contain numerous regulatory elements, yet the structure and developmental impact of 3UTRs remain largely uncharacterized. By integrating data from all Caenorhabditis elegans (C. elegans) developmental stages obtained by polyA capture, 3RACE, full-length cDNAs, and RNAseq, we define ~26,000 distinct mRNA 3 UTRs for ~85% of the 18,328 experimentally supported protein coding genes and refine the annotation of ~30% of gene models. Alternative 3UTR isoforms display widespread differential expression during development. Surprisingly, no canonical or variant polyadenylation signal (PAS) sequence is detec ted for 13% of polyA sites, most frequently among shorter alternative isoforms. Comparing trans-spliced and non trans-spliced genes reveals a strong correlation between the processing of transcript 5 and 3 ends: trans-spliced mRNAs possess longer 3UTRs and a higher frequency of no PAS or variant PAS motifs. We also predict conserved isoform-specific microRNA (miRNA) binding sites and identify additional evolutionarily constrained sequence blocks that may mediate 3UTR regulation. Thus, our data reveal a rich complexity of 3UTRs genome-wide and throughout development in C. elegans.

Molly C. Hammell et al. (2007) International Worm Meeting "Computational & genetic identification of cooperating microRNAs."

Molly C. Hammell, Ye Ding, Victor Ambros, Dang Long

[International Worm Meeting, 2007]

For many of the nematode miRNA genes, single gene loss-of-function mutations appear to cause little or no obvious phenotype, whereas multiple mutants can display visible phenotypes, suggesting extensive redundancy among microRNA genes (Abbott et al., 2005). Such redundancy could occur if multiple distinct microRNAs target the same mRNAs. This view is supported by the observation that many 3 UTRs contain clusters of predicted sites for miRNAs of distinct families (Chen & Rajewsky, 2007). Non-familial redundancy has been demonstrated in Drosophila where the K-box, GY-box, and Brd-box families of microRNAs cooperate to regulate target genes in the Notch pathway (Lai et al., 2005). To conduct a computational screen for additional functionally-interacting microRNAs, we developed a novel algorithm for identifying co-targeting relationships between microRNAs. In this algorithm, co-targeting is indicated by significant overlap between the target lists for distinct microRNAs. The algorithm computes systems-level co-targeting for all C. elegans and D. melanogaster microRNAs. On the level of the individual microRNA/target site interaction, we filter the results based on parameters associated with functionally-verified microRNA target sites, including the predicted structural accessibility of the binding sites using Sfold (Ding et al., 2005; Long et al, 2007). The results of the C. elegans and Drosophila analyses are then compared to look for deeply conserved regulatory relationships. We find that some of the most significant co-targeting relationships are conserved. In addition, the lists of co-regulated target transcripts are significantly enhanced for GO terms corresponding to major developmental pathways. The availability of a full library of individual microRNA deletion strains provides a powerful means for testing these computational predictions [see abstract #...]. As each regulatory input is removed, synthetic interactions masked by functional redundancy emerge. This combined genetics & genomics approach should help elucidate the biological roles of C. elegans microRNAs.

Mangone, Marco et al. (2009) International Worm Meeting "The landscape of C.elegans 3'UTRs."

Thierry-Mieg, Danielle, Piano, Fabio, Prasad Manoharan, Arun, Rajewsky, Nikolaus, Thierry-Mieg, Jean, Chen, Kevin, MacMenamin, Philip, Kim, John, Zegar, Charles, Ting, Han, Chen, Wei, Mis, Emily, Mangone, Marco, Kohara, Yuji, Vidal, Marc, Attie, Oliver, Khivansara, Vishal, Salehi-Ashtiani, Kourosh, Gunsalus, Kristin

[International Worm Meeting, 2009]

Three-prime untranslated regions (3''UTRs) contain sequence elements used by RNA-binding proteins and regulatory RNAs such as microRNAs (miRNAs) to influence mRNA stability, translation and localization. However, the annotation of these regions within transcripts is generally incomplete. In C.elegans, which has a well annotated genome, about half of the ~20,000 curated genes in WS190 are not annotated with any 3''UTR and less than 10% are annotated with multiple or alternative 3''UTR isoforms. As part of the modENCODE project, we have developed a pipeline targeting ~7,000 genes using 3'' RACE and found at least one 3''UTR for around 90% of this targeted set. To analyze these data we have used different sequencing technologies that resulted in the identification of multiple distinct 3''UTR isoforms for over half of the targeted genes. In addition to 3''RACE, we have developed a technique to capture polyA ends followed by deep sequencing. This polyA-capture approach has resulted in over 2,000,000 polyA tags that map to ~12,000 genes across all major developmental stages and males, with the majority of these sequences mapping to full-length 3''UTRs. The depth of these data shows the remarkable complexity in the distribution of polyadenylation events in vivo. We have also manually curated 3''UTR boundaries using all available cDNAs, derived mostly from the 200,000 staged ESTs, and obtained 3''UTRs boundaries for ~11,500 genes. For about half of these (~5,000 genes), the data connect 3''UTRs to specific transcripts with known trans-spliced leader sequences at the 5'' end. Combining the results of these analyses has increased our knowledge of 3''UTR structures significantly, identifying novel 3''UTR isoforms for about half of all C.elegans protein-coding genes.