[
Worm Breeder's Gazette,
2001]
RNAi is being used routinely to determine loss-of-function phenotypes and recently large-scale RNAi analyses have been reported (1,2,3). Although there is no question about the value of this approach in functional genomics, there has been little opportunity to evaluate reproducibility of these results. We are engaged in RNAi analysis of a set of 762 genes that are differentially expressed in the germline as compared to the soma (4 -- "Germline"), and have reached a point in our analysis that allows us to look at the issue of reproducibility. We have compared the RNAi results of genes in our set that were also analyzed by either Fraser et al. (1 -- Chromosome 1 set "C1") or Gonczy et al. (2 -- Chromosome 3 set "C3"). In making the comparison we have taken into account the different operational definition of "embryonic lethal" used by the three groups. In the C3 study, lethal was scored only if there were fewer than 10 surviving larva on the test plate, or roughly 90% lethal. In our screen and the C1 screen the percent survival was determined for each test. To minimize the contribution of false positives from our set, in our comparison with the C1 set we defined our genes as "embryonic lethal" if at least 30% of the embryos did not hatch, but included all lethals defined by Fraser et al. (> 10%). For our comparison with the C3 set, we used a more restrictive definition of "embryonic lethal" that required that 90% of the embryos did not hatch. (This means that in Table 1, five genes from our screen that gave lethality between 30-90% were included in the not lethal category; one of these was scored as lethal by Gonczy et al.). We have analyzed 149 genes from the germline set that overlap with the C1 set and 132 genes that overlap with the C3 set. The table below shows the number of genes scored as embryonic lethal (EL) or not embryonic lethal (NL) in each study. (Note that these comparisons do not include data from our published collection of ovary-expressed cDNAs.) Table 1. Comparing RNAi analysis of the same genes in different studies. Germline Chromosome 1 Germline Chromosome 3 NL (117) EL (32) NL (97) EL (35) NL (104) 100 4 NL (89) 87 2 EL (45) 17 28 EL (43) 10 33 Overall, the degree of reproducibility is high. The concordance between our results and the published results was 86% with C1 (128/149 genes) and 90% with C3 (120/132). However, we scored a larger number of genes as giving rise to embryonic lethal phenotypes than the other studies did. What does this mean? One possibility is that we are generating a large number of false positives (God forbid!). The other interpretation is that there is a fairly high frequency of false negatives in each screen (4-8% in our screen (2/45; 4/49); 22% in the C3 screen (10/45); and 35% (17/49) in the C1 screen). It is no surprise that the different methods used by the three groups resulted in slightly different outcomes and we can only speculate on which methodological variation contributed most. In comparing our methods to those used in the C3 study we note that our two groups used different primer pairs for each gene; that we tested genes individually while they tested genes in pairs; and that the operational definition of "embryonic lethal" differed. Considering the latter two differences, we speculate that even with pools of two, the competition noted by Gonczy et al. in dsRNA pools could reduce levels of lethality below the 90% cutoff. The major difference between our approach and the C1 approach is feeding vs. injection, raising the possibility that for some genes feeding may be a less effective means of administering dsRNA. Whatever the basis for the difference, these comparisons indicate that genes scored as "non-lethal" in any single study may show an embryonic lethal RNAi phenotype when reanalyzed. It therefore seems useful to have more than one pass at analyzing C. elegans genes via RNAi. We are indebted to P. Gonczy for very useful comments. Fraser, A. G., Kamath, R. S., Zipperlen, P., Martinez-Campos, M., Sohrmann, M. and Ahringer, J. (2000). Functional genomic analysis of C. elegans chromosome I by systematic RNA interference. Nature 408 , 325-330. Gonczy, P., Echeverri, G., Oegema, K., Coulson, A., Jones, S. J., Copley, R. R., Duperon, J., Oegema, J., Brehm, M., Cassin, E. et al. (2000). Functional genomic analysis of cell division in C. elegans using RNAi of genes on chromosome III. Nature 408 , 331-336. Piano, F., Schetter, A. J., Mangone, M., Stein, L. and Kemphues, K. J. (2000). RNAi analysis of genes expressed in the ovary of Caenorhabditis elegans. Curr Biol 10 , 1619-1622. Reinke, V., Smith, H. E., Nance, J., Wang, J., Van Doren, C., Begley, R., Jones, S. J., Davis, E. B., Scherer, S., Ward, S. et al. (2000). A global profile of germline gene expression in C. elegans. Mol Cell 6 , 605-616.
[
International Worm Meeting,
2003]
Complete genome sequences of model organisms lead to large numbers of predicted genes, of which many are poorly or un-characterized functionally. In recent years various functional genomic projects have been carried out to characterize large numbers of genes at a time, and a large data sets emerging from these projects have been deposited into publicly available databases. However, due to the volume and error-prone property of such data sets, it is not obvious how they can help in elucidating biological processes of interest. Here we demonstrate an application of combined experimental and computational functional genomic approaches to C. elegans early embryogenesis. Proteins required for early embryogenesis were subjected to yeast two-hybrid analysis and putative protein-protein interactions were identified. For a significant number of these putative interactions, both proteins correspond to genes that gave rise to embryonic lethality phenotypes in high-throughput RNAi assays. In addition, gene pairs corresponding to some interactions have been found to exhibit very similar RNAi phenotypes during early embryogenesis, as recorded by differential interference contrast microscopy. In addition, we computationally generated a network of functional associations for these proteins by searching through publicly available data for genes that are either co-expressed across many different conditions, or share similar RNAi phenotypes, or both. Gene products that are specifically linked to proteins that are required for early embryogenesis were identified as potentially involved in this process. By examining the neighborhood cohesiveness of this network, we identified protein pairs of functional similarity for early embryogenesis. We propose that such integration of functional genomic information can be applied to other biological processes and to other organisms as well.
[
International C. elegans Meeting,
2001]
Since the complete genome sequence of C. elegans became available at the end of 1998, several functional genomics projects have been initiated to functionally annotate the many predicted genes that had remained previously uncharacterized. The data generated by these projects can be viewed as hypotheses until validated further. We have proposed that the likelihood of such hypotheses to be biologically meaningful might increase as the data from different projects are integrated. For example, genes that cluster in expression profiling, protein interaction maps and phenotypic analysis might have a relatively high likelihood to functionally interact in the same biological process. Recently, microarray analysis has led to the identification of a set of genes that show increased expression in the germline (Reinke et al, Mol. Cell. 6, 605-616, 2000). We have cloned these ~750 open reading frames and transferred them into yeast two-hybrid vectors using Gateway cloning. Subsequently, we have generated a 750 X 750 protein interaction map. The data obtained and its integration with the RNAi data obtained by Fabio Piano and colleagues will be discussed.
[
Development & Evolution Meeting,
2006]
Genes with similar loss-of-function phenotypes often act in similar developmental processes. Coupled with genome-scale RNAi studies, this premise suggests that starting from RNAi-induced phenotypes, previously uncharacterized genes required for specific processes can be identified by grouping them with genes of known function that confer similar RNAi phenotypes.
Successfully using this strategy for analysis of the early embryo, Piano et al. (2002) defined 47 discrete characters to characterize complex cell biological and developmental phenotypes resulting from RNAi of genes required for embryogenesis. Phenotypic signatures - combinations of explicit outcomes for each character - were generated for each RNAi experiment, and genes were clustered by phenotypic similarity (phenoclusters). These data provided hypotheses about the function of previously uncharacterized genes. Primary data, scoring criteria and results are publicly available in RNAiDB
(http://www.RNAi.org) and can be analyzed and queried using additional tools (e.g. PhenoBlast; Gunsalus et al., 2004).
We have applied the same approach to RNAi-induced sterility. In previous large-scale screens, over 800 genes conferred a sterile (Ste) or sterile progeny (Stp) phenotype, based on scoring of brood size (Gonczy et al., 2000; Fraser et al. 2001; Piano et al., 2000; Maeda et al., 2001; Hanazawa et al. 2001; Piano et al., 2002; Kamath et al., 2003; Simmer et al., 2003). To characterize specific mophological defects underlying the observed sterility, we have developed within RNAiDB a scoring system for ~300 discrete phenotypic features visible at high-magnification in adult hermaphrodites, providing data amenable to phenoclustering and PhenoBlast. As a pilot study, we have chosen a set of genes previously identified as Stp and/or Ste by RNAi (Fraser et al. 2001; Kamath et al., 2003) that confer a wide range of RNAi-induced defects. Analysis of phenotypic signatures (~100 characters) and phenoclustering are in progress. Data (Z-plane stacks) will be available at RNAiDB, together with the phenotypic scoring results. The scoring system and analysis tools will also be made available for general use.
[
International Worm Meeting,
2005]
A fundamental premise of genetic analysis is that genes with similar loss-of-function phenotypes likely act in similar developmental processes. Extending this concept to RNAi-based genome-scale functional characterization, Piano et al. (2002) defined 47 embryonic phenotypic characters in C. elegans and generated a digital phenotypic signature for each RNAi experiment. Genes conferring similar signatures were clustered (phenoclusters), placing previously uncharacterized genes in clusters with genes of known function. The primary data (high-magnification time-lapse movies), scoring criteria and results are publicly available and can be analyzed and queried using additional tools (e.g. Phenoblast) (Gunsalus et al., 2004;
http://www.RNAi.org).We have applied this approach to gonadogenesis and germline development in C. elegans. In previous large-scale RNAi screens, over 800 genes are reported to confer a sterile (Ste) or sterile progeny (Stp) phenotype, based on low-magnification screening (Gonczy et al., 2000; Fraser et al. 2001; Piano et al., 2001; Maeda et al., 2001; Hanazawa et al. 2001; Piano et al., 2002; Kamath et al., 2003; Simmer et al., 2003). A subset of these have been examined under high magnification (Maeda et al., 2001; Hanazawa et al., 2001; Colaiacovo et al., 2002), but these data are not available in a systematic digital format amenable to analyses such as phenoclustering or Phenoblast, nor to phenotypic search criteria.We developed and are implementing a web-based digital scoring system (currently 262 characters) for high-magnification phenotypes in adult hermaphrodites. Characters include both quantitative and qualitative features of the gonad, germ line and gametes as well as several gonad-related characters. We generate a series of digital Z-plane images at high magnification from each RNAi experiment. All data will be made publicly available on the RNAi.org website. Our test set is 147 Stp genes and 260 Ste genes conferring a wide range of RNAi-induced defects.
[
Development & Evolution Meeting,
2006]
Using combined network analysis of large-scale functional genomic data we mapped multi-protein modules required for distinct processes during early embryogenesis (Gunsalus et al. 2005). A basic question is how these molecular modules are coordinated through the mitotic cycle to ensure the proper unfolding of early developmental events. To identify proteins that could coordinate different modules we searched for proteins that bridged different modules. One such protein, C38D4.3, could be placed in either the nuclear pore complex module or in the chromosome maintenance module by a network clustering algorithm M-CODE (Bader et al. 2003). Consistent with its predicted roles at the nuclear pore and in chromosome segregation, GFP fusions and anti-C38D4.3 immmunolocalizations show that C38D4.3 shuttles between the nuclear envelope and the kinetochore during the cell cycle. Functionally, C38D4.3 is required for proper nuclear envelope and chromatin maintenance. C38D4.3 (RNAi) embryos, like embryos without nuclear pore components, are incapable of completely separating cytoplasm from nucleoplasm, failing to exclude microtubules and affecting the nuclear localization of PIE-1, a protein normally enriched in the P1 nucleus (Mello, 1996). Additionally, pronuclei fail to meet, and centrosomes do not remain attached to the paternal pronucleus and segregate prematurely. In these embryos, metaphase spindles are not established and chromatin neither condenses, congresses, nor segregates properly. These phenotypes are reminiscent of RNAi phenotypes of genes from the Ran GTPase cycle (Askjaer 2002). Looking for C38D4.3 (RNAi) phenotypic neighbors using PhenoBlast (Gunsalus et al 2004) or phenoclusters from large-scale RNAi analyses (Sonnichsen et al 2005; Gunsalus et al 2005) we identified ~25 other genes with similar defects when analyzed by time-lapse Nomarski microscopy. Of these, genes that are part of the RanGTPase pathway (
ran-1,
ran-2, and
npp-9) were required for proper C38D4.3 localization. Thus C38D4.3 is critical for both mitotic and interphase cell functions and is a likely target of the Ran GTPase pathway.
[
International Worm Meeting,
2007]
Natural diversity ultimately arises from the evolution of developmental programs. The molecular mechanisms underlying the evolution of developmental programs are still largely mysterious. A critical question is thus to discover mechanisms that enable evolution of developmental programs in light of the highly constrained mechanistic view that arises from developmental genetics, where altering important developmental genes is known to lead to lethality. We are using nematode early embryogenesis (EE) as a model to study these questions. When analyzed by time-lapse microscopy, nematode embryos show a remarkable array of different characteristics during EE. To analyze this diversity in more detail, we have developed a series of 40 discrete phenotypic characters describing the first two rounds of cell division among 35 Rhabditidae species. When all the characters are mapped onto a molecular phylogeny we find that they have evolved multiple times, showing a high rate of convergent evolution. This observation supports the idea that early development is plastic. The diversity in EE is surprising given the reproducible way in which each single species, like C. elegans, develop. One possible reason for this evolutionary plasticity is that early development is governed by highly modular molecular mechanisms that can direct specific events independently from one another. In support of this idea, we find almost no evidence for correlation (co-evolution) among characters across the 35 species. To obtain molecular clues that could underlie the diversity we observed, we compared the evolutionary characters with the phenotypes observed from RNAi data in C. elegans. Surprisingly, for almost every character we could find a matching RNAi phenotype that phenocopies the character of another species. One hypothesis that arose from these RNAi comparisons was that a rotation of the spindle in the AB cell seen in Protorhabditis and Diploscapter species was due to a change in the polarity network. The polarity network is defined by the regulatory interactions among the PAR proteins in C. elegans. These comparisons led us to postulate that the PAR-3/PAR-6/PKC-3 complex may be missing or inactive in Diploscapter or Protorhabditis one-cell embryos. We tested this hypothesis by localizing PAR-1 in these species and found a novel localization pattern that support the idea that the PAR network is evolving across these species.