Although single gene loss of function analyses can identify components of particular processes, important molecules are missed due to the robustness of biological systems. Here we show that large scale RNAi screening for suppression interactions with functionally related mutants greatly expands the repertoire of genes known to act in a shared process and reveals a new layer of functional relationships. We performed RNAi screens for 17 C. elegans cell polarity mutants, generating the most comprehensive polarity network in a metazoan, connecting 184 genes. Of these, 72% were not previously linked to cell polarity and 80% have human homologs. We experimentally confirmed functional roles predicted by the network and characterised through biophysical analyses eight myosin regulators. In addition, we discovered functional redundancy between two unknown polarity genes. Similar systematic genetic interaction screens for other biological processes will help uncover the inventory of relevant genes and their patterns of interactions.
Meiotic recombination, an essential aspect of sexual reproduction, is initiated by programmed DNA double-strand breaks (DSBs). DSBs are catalyzed by the widely-conserved Spo11 enzyme; however, the activity of Spo11 is regulated by additional factors that are poorly conserved through evolution. To expand our understanding of meiotic regulation, we have characterized a novel gene, dsb-1, that is specifically required for meiotic DSB formation in the nematode Caenorhabditis elegans. DSB-1 localizes to chromosomes during early meiotic prophase, coincident with the timing of DSB formation. DSB-1 also promotes normal protein levels and chromosome localization of DSB-2, a paralogous protein that plays a related role in initiating recombination. Mutations that disrupt crossover formation result in prolonged DSB-1 association with chromosomes, suggesting that nuclei may remain in a DSB-permissive state. Extended DSB-1 localization is seen even in mutants with defects in early recombination steps, including spo-11, suggesting that the absence of crossover precursors triggers the extension. Strikingly, failure to form a crossover precursor on a single chromosome pair is sufficient to extend the localization of DSB-1 on all chromosomes in the same nucleus. Based on these observations we propose a model for crossover assurance that acts through DSB-1 to maintain a DSB-permissive state until all chromosome pairs acquire crossover precursors. This work identifies a novel component of the DSB machinery in C. elegans, and sheds light on an important pathway that regulates DSB formation for crossover assurance.
For most eukaryotes, recombination between homologous chromosomes during meiosis is an essential aspect of sexual reproduction. Meiotic recombination is initiated by programmed double-strand breaks in DNA, which have the potential to induce mutations if not efficiently repaired. To better understand the mechanisms that govern the initiation of recombination and regulate the formation of double-strand breaks, we use the nematode Caenorhabditis elegans as a model system. Here we describe a new gene, dsb-1, that is required for double-strand break formation in C. elegans. Through analysis of the encoded DSB-1 protein we illuminate an important regulatory pathway that promotes crossover recombination events on all chromosome pairs to ensure successful meiosis.
When a duplicate gene has no apparent loss-of-function phenotype, it is commonly considered that the phenotype has been masked as a result of functional redundancy with the remaining paralog. This is supported by indirect evidence showing that multi-copy genes show loss-of-function phenotypes less often than single-copy genes and by direct tests of phenotype masking using select gene sets. Here we take a systematic genome-wide RNA interference approach to assess phenotype masking in paralog pairs in the Caenorhabditis elegans genome. Remarkably, in contrast to expectations, we find that phenotype masking makes only a minor contribution to the low knockdown phenotype rate for duplicate genes. Instead, we find that non-essential genes are highly over-represented among duplicates, leading to a low observed loss-of-function phenotype rate. We further find that duplicate pairs derived from essential and non-essential genes have contrasting evolutionary dynamics: whereas non-essential genes are both more often successfully duplicated (fixed) and lost, essential genes are less often duplicated but upon successful duplication are maintained over longer periods. We expect the fundamental evolutionary duplication dynamics presented here to be broadly applicable.
Duplicate genes occur in all organisms. It has been found that mutations in duplicate genes cause defects much less often than when single copy genes are mutated. It is widely believed that this is due to functional redundancy—that is, the two genes can carry out similar functions so that the non-mutated duplicate gene can cover for or “mask” the phenotype of the mutation in the first duplicate. To determine whether this hypothesis is true, it is necessary to test systematically whether defects indeed occur in the organism when both duplicate genes are inhibited. We have for the first time carried out such an analysis in a multicellular organism, the nematode Caenorhabditis elegans. In contrast to expectations, we observed that when both copies of duplicate genes are inhibited deleterious effects are very rare. We show that this is because duplicate genes are much more often non-essential compared to genes where there is only a single copy. Non-essential genes are also lost from the genome much more often than essential genes. However, when essential genes are duplicated, they remain present in the genome over longer periods. Our results give a framework to explain the evolutionary dynamics of duplications in the genome.
The Caenorhabditis elegans dosage compensation complex (DCC) equalizes X-chromosome gene dosage between XO males and XX hermaphrodites by two-fold repression of X-linked gene expression in hermaphrodites. The DCC localizes to the X chromosomes in hermaphrodites but not in males, and some subunits form a complex homologous to condensin. The mechanism by which the DCC downregulates gene expression remains unclear. Here we show that the DCC controls the methylation state of lysine 20 of histone H4, leading to higher H4K20me1 and lower H4K20me3 levels on the X chromosomes of XX hermaphrodites relative to autosomes. We identify the PR-SET7 ortholog SET-1 and the Suv4-20 ortholog SET-4 as the major histone methyltransferases for monomethylation and di/trimethylation of H4K20, respectively, and provide evidence that X-chromosome enrichment of H4K20me1 involves inhibition of SET-4 activity on the X. RNAi knockdown of set-1 results in synthetic lethality with dosage compensation mutants and upregulation of X-linked gene expression, supporting a model whereby H4K20me1 functions with the condensin-like C. elegans DCC to repress transcription of X-linked genes. H4K20me1 is important for mitotic chromosome condensation in mammals, suggesting that increased H4K20me1 on the X may restrict access of the transcription machinery to X-linked genes via chromatin compaction.
In many animals, males have one X chromosome and females have two. However, the same amount of gene expression from X chromosomes is needed in the two sexes. The process of dosage compensation (DC) globally regulates X-chromosome gene expression to make it equal between the sexes, and it occurs in different ways in different animals. In mammals, one X chromosome in females is randomly inactivated, leaving one active X chromosome. In contrast, in the nematode worm C. elegans, the two X chromosomes in hermaphrodites are repressed two-fold to match gene expression to the single X chromosome in males. Previous work in C. elegans identified proteins required for DC that bind to the X chromosome, but their mode of action is not known. Here we show that DC proteins lead to higher levels of histone H4 lysine 20 monomethylation (H4K20me1) on hermaphrodite X chromosomes and that H4K20me1 functions in repressing X-chromosome gene expression. This shows that histone modification is an important aspect of the mechanism of dosage compensation. Together with previous work linking H4K20me1 to chromatin structure regulation, our results suggest that dosage compensation might lower gene expression on hermaphrodite X chromosomes by compacting them.
During animal development, a complex of Par3, Par6 and atypical protein kinase C (aPKC) plays a central role in cell polarisation. The small G protein Cdc42 also functions in cell polarity and has been shown in some cases to act by regulating the Par3 complex. However, it is not yet known whether Cdc42 and the Par3 complex widely function together in development or whether they have independent functions. For example, many studies have implicated Cdc42 in cell migrations, but the Par3 complex has only been little studied, with conflicting results. Here we examine the requirements for CDC-42 and the PAR-3/PAR-6/PKC-3 complex in a range of different developmental events. We found similar requirements in all tissues examined, including polarised growth of vulval precursors and seam cells, migrations of neuroblasts and axons, and the development of the somatic gonad. We also propose a novel role for primordial germ cells in mediating coalescence of the C. elegans gonad. These results indicate that CDC-42 and the PAR-3/PAR-6/aPKC complex function together in diverse cell types.
C. elegans; cell migration; polarity; vulva; anchor cell; gonad; Cdc42; Par3; Par6; aPKC
We systematically generated large-scale data sets to improve genome annotation for the nematode Caenorhabditis elegans, a key model organism. These data sets include transcriptome profiling across a developmental time course, genome-wide identification of transcription factor–binding sites, and maps of chromatin organization. From this, we created more complete and accurate gene models, including alternative splice forms and candidate noncoding RNAs. We constructed hierarchical networks of transcription factor–binding and microRNA interactions and discovered chromosomal locations bound by an unusually large number of transcription factors. Different patterns of chromatin composition and histone modification were revealed between chromosome arms and centers, with similarly prominent differences between autosomes and the X chromosome. Integrating data types, we built statistical models relating chromatin, transcription factor binding, and gene expression. Overall, our analyses ascribed putative functions to most of the conserved genome.
We report testing of the specificity and utility of over 200 antibodies raised against 57 different histone modifications, in Drosophila melanogaster, Caenorhabditis elegans and human cells. While most antibodies performed well, over 25% failed specificity tests by dot blot or western blot. Among specific antibodies, over 20% failed in chromatin immunoprecipitation experiments. We advise rigorous testing of histone-modification antibodies before use and provide a website for posting new test results.
Genomic sequences obtained through high-throughput sequencing are not uniformly distributed across the genome. For example, sequencing data of total genomic DNA show significant, yet unexpected enrichments on promoters and exons. This systematic bias is a particular problem for techniques such as chromatin immunoprecipitation, where the signal for a target factor is plotted across genomic features. We have focused on data obtained from Illumina’s Genome Analyser platform, where at least three factors contribute to sequence bias: GC content, mappability of sequencing reads, and regional biases that might be generated by local structure. We show that relying on input control as a normalizer is not generally appropriate due to sample to sample variation in bias. To correct sequence bias, we present BEADS (bias elimination algorithm for deep sequencing), a simple three-step normalization scheme that successfully unmasks real binding patterns in ChIP-seq data. We suggest that this procedure be done routinely prior to data interpretation and downstream analyses.
Here we describe a toolkit for the production of fluorescently tagged proteins in
the C. elegans germline and early embryo using Mos1-mediated
single copy insertion (MosSCI) transformation. We have generated promoter and
3′UTR fusions to sequences of different fluorescent proteins yielding
constructs for germline expression that are compatible with MosSCI MultiSite
Gateway vectors. These vectors allow tagged transgene constructs to be inserted
as single copies into known sites in the C. elegans genome
using MosSCI. We also show that two C. elegans heat shock
promoters (Phsp-16.2 and Phsp-16.41) can be
used to induce transgene expression in the germline when inserted via MosSCI
transformation. This flexible set of new vectors, available to the research
community in a plasmid repository, should facilitate research focused on the
C. elegans germline and early embryo.
Attenuation of the EGFR (Epidermal Growth Factor Receptor) signalling cascade is crucial to control cell fate during development. A candidate-based RNAi approach in C. elegans identified CDT-2 as an attenuator of LET-23 (EGFR) signalling. Human CDT2 is a component of the conserved CDT2/CUL4/DDB1 ubiquitin ligase complex that plays a critical role in DNA replication and G2/M checkpoint. Within this complex, CDT2 is responsible for substrate recognition. This ubiquitin ligase complex has been shown in various organisms, including C. elegans, to target the replication-licensing factor CDT1, and the CDK inhibitor p21. However, no previous link to EGFR signalling has been identified.
We have characterised CDT-2's role during vulva development and found that it is a novel attenuator of LET-23 signalling. CDT-2 acts redundantly with negative modulators of LET-23 signalling and CDT-2 or CUL-4 downregulation causes persistent expression of the egl-17::cfp transgene, a marker of LET-23 signalling during vulva development. In addition, we show that CDT-2 physically interacts with SEM-5 (GRB2), a known negative modulator of LET-23 signalling that directly binds LET-23, and provide genetic evidence consistent with CDT-2 functioning at or downstream of LET-23. Interestingly, both SEM-5 and CDT-2 were identified independently in a screen for genes involved in receptor-mediated endocytosis in oocytes, suggesting that attenuation of LET-23 by CDT-2 might be through regulation of endocytosis.
In this study, we have shown that CDT-2 and CUL-4, members of the CUL-4/DDB-1/CDT-2 E3 ubiquitin ligase complex attenuate LET-23 signalling in vulval precursor cells. In future, it will be interesting to investigate the potential link to endocytosis and to determine whether other signalling pathways dependent on endocytosis, e.g. LIN-12 (Notch) could be regulated by this ubiquitin ligase complex. This work has uncovered a novel function for the CUL-4/DDB-1/CDT-2 E3 ligase that may be relevant for its mammalian oncogenic activity.
Variation in patterns of methylations of histone tails reflects and modulates chromatin structure and function1-3. To provide a framework for the analysis of chromatin function in C. elegans, we generated a genome-wide map of histone H3 tail methylations. We find that C. elegans genes show similarities in distributions of histone modifications to those of other organisms, with H3K4me3 near transcription start sites, H3K36me3 in the body of genes, and H3K9me3 enriched on silent genes. Unexpectedly, we also observe a striking novel pattern: exons are preferentially marked with H3K36me3 relative to introns. H3K36me3 exon marking is dependent on transcription and its level is lower in alternatively spliced exons, supporting a splicing related marking mechanism. We further show that the difference in H3K36me3 marking between exons and introns is evolutionarily conserved in human and mouse. We propose that H3K36me3 exon marking in chromatin provides a dynamic link between transcription and splicing.
Cell cycle lengths vary widely among different cells within an animal, yet mechanisms of cell cycle length regulation are poorly understood. In the Caenorhabditis elegans embryo, the first cell division produces two cells with different cell cycle lengths, which are dependent on the conserved partitioning-defective (PAR) polarity proteins. We show that two key cell cycle regulators, the Polo-like kinase PLK-1 and the cyclin-dependent kinase phosphatase CDC-25.1, are asymmetrically distributed in early embryos. PLK-1 shows anterior cytoplasmic enrichment and CDC-25.1 shows PLK-1–dependent enrichment in the anterior nucleus. Both proteins are required for normal mitotic progression. Furthermore, these asymmetries are controlled by PAR proteins and the muscle excess (MEX) proteins MEX-5/MEX-6, and the latter is linked to protein degradation. Our results support a model whereby the PAR and MEX-5/MEX-6 proteins asymmetrically control PLK-1 levels, which asymmetrically regulates CDC-25.1 to promote differences in cell cycle lengths. We suggest that control of Plk1 and Cdc25 may be relevant to regulation of cell cycle length in other developmental contexts.
Spindle positioning is an essential feature of asymmetric cell division. The conserved PAR proteins together with heterotrimeric G proteins control spindle positioning in animal cells, but how these are linked is not known. In C. elegans, PAR protein activity leads to asymmetric spindle placement through cortical asymmetry of Gα regulators GPR-1/2. Here, we establish that the casein kinase 1 gamma CSNK-1 and a PIP2 synthesis enzyme (PPK-1) transduce PAR polarity to asymmetric Gα regulation. PPK-1 is posteriorly enriched in the one-celled embryo through PAR and CSNK-1 activities. Loss of CSNK-1 causes uniformly high PPK-1 levels, high symmetric cortical levels of GPR-1/2 and LIN-5, and increased spindle pulling forces. In contrast, knockdown of ppk-1 leads to low GPR-1/2 levels and decreased spindle forces. Furthermore, loss of CSNK-1 leads to increased levels of PIP2. We propose that asymmetric generation of PIP2 by PPK-1 directs the posterior enrichment of GPR-1/2 and LIN-5, leading to posterior spindle displacement.
Microtubules deliver positional signals and are required for establishing polarity in many different organisms and cell types. In Caenorhabditis elegans embryos, posterior polarity is induced by an unknown centrosome-dependent signal. Whether microtubules are involved in this signaling process has been the subject of controversy. Although early studies supported such an involvement (O'Connell, K.F., K.N. Maxwell, and J.G. White. 2000. Dev. Biol. 222:55–70; Wallenfang, M.R., and G. Seydoux. 2000. Nature. 408:89–92; Hamill, D.R., A.F. Severson, J.C. Carter, and B. Bowerman. 2002. Dev. Cell. 3:673–684), recent work involving RNA interference knockdown of tubulin led to the conclusion that centrosomes induce polarity independently of microtubules (Cowan, C.R., and A.A. Hyman. 2004. Nature. 431:92–96; Sonneville, R., and P. Gonczy. 2004. Development. 131: 3527–3543). In this study, we investigate the consequences of tubulin knockdown on polarity signaling. We find that tubulin depletion delays polarity induction relative to wild type and that polarity only occurs when a small, late-growing microtubule aster is visible at the centrosome. We also show that the process of a normal meiosis produces a microtubule-dependent polarity signal and that the relative levels of anterior and posterior PAR (partitioning defective) polarity proteins influence the response to polarity signaling. Our results support a role for microtubules in the induction of embryonic polarity in C. elegans.
Endocytosis is involved in the regulation of many cellular events, including signalling, cell migration, and cell polarity. To begin to investigate roles for endocytosis in early C. elegans development, we examined the distribution and dynamics of early endosomes (EEs) in embryos.
EEs are primarily found at the cell periphery with an initially uniform distribution after fertilization. Strikingly, we find that during the first cell cycle, EEA-1 positive EEs become enriched at the anterior cortex. In contrast, the Golgi compartment shows no asymmetry in distribution. Asymmetric enrichment of EEs depends on acto-myosin contractility and embryonic PAR polarity. In addition to their localization at the cortex, EEs are also found around the centrosome. These EEs move rapidly (1.3um/s) from the cortex directly to the centrosome, a speed comparable to that of the minus end directed motor dynein.
We speculate that the asymmetry of early endosomes might play a role in cell asymmetries or fate decisions.
RNA interference coupled with videorecording of C. elegans embryos is a powerful method for identifying genes involved in cell division processes. Here we present a functional analysis of the gene B0511.9, previously identified as a candidate cell polarity gene in an RNAi videorecording screen of chromosome I embryonic lethal genes.
Whereas weak RNAi inhibition of B0511.9 causes embryonic cell polarity defects, strong inhibition causes embryos to arrest in metaphase of meiosis I. The range of defects induced by RNAi of B0511.9 is strikingly similar to those displayed by mutants of anaphase-promoting complex/cyclosome (APC/C) components. Although similarity searches did not reveal any obvious homologue of B0511.9 in the non-redundant protein database, we found that the N-terminus shares a conserved sequence pattern with the N-terminus of the small budding yeast APC/C subunit Cdc26 and its orthologues from a variety of other organisms. Furthermore, we show that B0511.9 robustly complements the temperature-sensitive growth defect of a yeast cdc26Δ mutant.
These data demonstrate that B0511.9 encodes the C. elegans APC/C subunit CDC-26.
A report on the Second EMBL/EMBO Symposium on Functional Genomics: 'Exploring the Edges of Omics', European Molecular Biology Laboratory (EMBL), Heidelberg, Germany, 16-19 October 2004.
A report on the Second EMBL/EMBO Symposium on Functional Genomics: 'Exploring the Edges of Omics', European Molecular Biology Laboratory (EMBL), Heidelberg, Germany, 16-19 October 2004.
Cell polarity is essential for many decisions made during development. While investigation of polarity-specific factors has yielded great insights into the polarization process, little is known on how these polarity-specific factors link to the basic cellular mechanisms that function in non-polarity aspects of the cell. To better understand the mechanisms that establish embryonic polarity, we investigated genes required for polarity in the one-cell C. elegans embryo that are also required for other non-polarity functions. This has led to the identification of the Pod-class of mutants that are characterized by osmosensitive embryos and defects in anterior-posterior polarity.
Mutation in either of two loci of this class, emb-8 and pod-2, disrupts embryonic polarization and results in osmotically-sensitive embryos. Loss of emb-8, a previously uncharacterized polarity gene, causes mislocalization of PAR-3 and PAR-2 that molecularly mark the anterior and posterior cortices. emb-8 encodes NADPH-cytochrome P450 reductase, a protein supplying electrons to cytochrome P450-family enzymes, some of which catalyze fatty acid modifications. Cloning of the previously characterized polarity gene pod-2 reveals it encodes acetyl-CoA carboxylase, an enzyme that catalyzes the first step in de novo fatty acid synthesis. Depletion of fatty acid synthase, the next enzyme in the biosynthetic pathway, by RNA-interference (RNAi) also causes similar loss of one-cell polarity. Furthermore, pod-2 polarity defects can be rescued by addition of exogenous fatty acids. By following the behavior of the pronucleus in emb-8 and pod-2 mutant embryos, we demonstrate that loss of polarity correlates with impaired interaction between the pronucleus-centrosome complex and the posterior cortex.
The characterization of emb-8 and pod-2 mutant embryos suggests that the pronucleus-centrosome complex interaction with the cortex plays a direct role in establishing polarity and that fatty acid pathways are important for this polarizing event.
polarity; Caenorhabditis elegans; embryonic development; NADPH-cytochrome P450 reductase; acetyl-CoA carboxylase; fatty acids
RNAi can be achieved by feeding worms Escherichia coli expressing dousble-stranded RNA corresponding to a specific gene. An optimized feeding method is presented that results in phenotypes at least as strong as those produced by direct injection of RNA for embryonic lethal genes, and stronger for genes with post-embryonic phenotypes.
In Caenorhabditis elegans, injection of double-stranded RNA (dsRNA) results in the specific inactivation of genes containing homologous sequences, a technique termed RNA-mediated interference (RNAi). It has previously been shown that RNAi can also be achieved by feeding worms Escherichia coli expressing dsRNA corresponding to a specific gene; this mode of dsRNA introduction is conventionally considered to be less efficient than direct injection, however, and has therefore seen limited use, even though it is considerably less labor-intensive.
Here we present an optimized feeding method that results in phenotypes at least as strong as those produced by direct injection of dsRNA for embryonic lethal genes, and stronger for genes with post-embryonic phenotypes. In addition, the interference effect generated by feeding can be titrated to uncover a series of hypomorphic phenotypes informative about the functions of a given gene. Using this method, we screened 86 random genes on consecutive cosmids and identified functions for 13 new genes. These included two genes producing an uncoordinated phenotype (a previously uncharacterized POU homeodomain gene, ceh-6, and a gene encoding a MADS-box protein) and one gene encoding a novel protein that results in a high-incidence-of-males phenotype.
RNAi by feeding can provide significant information about the functions of an individual gene beyond that provided by injection. Moreover, it can be used for special applications for which injection or the use of mutants is sometimes impracticable (for example, titration, biochemistry and large-scale screening). Thus, RNAi by feeding should make possible new experimental approaches for the use of genomic sequence information.
RNA-mediated interference (RNAi) is a method to inhibit gene function by introduction of double-stranded RNA (dsRNA). Recently, an RNAi library was constructed that consists of bacterial clones expressing dsRNA, corresponding to nearly 90% of the 19,427 predicted genes of C. elegans. Feeding of this RNAi library to the standard wild-type laboratory strain Bristol N2 detected phenotypes for approximately 10% of the corresponding genes. To increase the number of genes for which a loss-of-function phenotype can be detected, we undertook a genome-wide RNAi screen using the rrf-3 mutant strain, which we found to be hypersensitive to RNAi. Feeding of the RNAi library to rrf-3 mutants resulted in additional loss-of-function phenotypes for 393 genes, increasing the number of genes with a phenotype by 23%. These additional phenotypes are distributed over different phenotypic classes. We also studied interexperimental variability in RNAi results and found persistent levels of false negatives. In addition, we used the RNAi phenotypes obtained with the genome-wide screens to systematically clone seven existing genetic mutants with visible phenotypes. The genome-wide RNAi screen using rrf-3 significantly increased the functional data on the C. elegans genome. The resulting dataset will be valuable in conjunction with other functional genomics approaches, as well as in other model organisms.
The screen suggested functions for 393 genes for which no RNAi-mediated phenotype was known. The comparison with similar screens in worms has general implications for RNAi experiments