Ebbing, A.L.P. et al. (2017) International Worm Meeting "Serial-section RNA sequencing: a novel method for genome-wide gene expression pattern analysis in C. elegans."

Ebbing, A.L.P., Korswagen, H.C., van Oudenaarden, A., Betist, M.C., Vertesy, A.

[International Worm Meeting, 2017]

To advance our understanding of development and physiology, it is important to gain detailed insight into the spatial and temporal expression patterns of all genes. Such a comprehensive overview of gene expression patterns will identify cell type and organ specific genes. It will also provide a framework for studying how gene expression patterns change upon perturbation. RNA sequencing (RNA-seq) is ideally suited for genome-wide analysis of gene expression. However, RNA-seq by itself does not provide information on where genes are expressed in the organism. A solution to this problem is to include spatial information by sectioning the organism along one of the body axes and to perform RNA-seq on each of the separate sections. We have developed a serial-section RNA-seq (serial-seq) method for C. elegans. In brief, we cryo-section a young adult animal from the head to the tail into 20 micrometer thick slices, resulting in a total of 40 to 50 slices. Each slice is then sequenced using the sensitive CEL-seq method and the sequencing data is aligned to create an anteroposterior gene expression map. Using this approach, an average of 8794 genes (43% of coding genes) (n=8) can be detected along the length of the animal. These include genes that are expressed in major tissues such as the germline, but also genes that are known to be expressed in only single neuron pairs in the head, demonstrating the high sensitivity of our approach. Furthermore, we found that these expression patterns are highly reproducible among independently sectioned animals. What makes serial-seq particularly powerful in C. elegans is its relatively simple and invariant anatomy. Because each slice contains only a limited number of cells, we can predict which cells are in the different slices and use this information to identify cell type specific genes. Furthermore, we have developed an algorithm that enables us to search for genes with specific expression patterns. Using known marker genes as a starting point, we have identified genes that are specifically expressed in the germline and sperm, in organ structures such as the pharynx, the intestine, the vulva, the spermatheca and the male reproductive tract, and in neurons of the nerve ring and tail ganglia. The ability to identify cell type specific genes is an important advantage over RNA-seq of isolated cells, which does not distinguish between specific and more generally expressed genes. In summary, we have generated high-resolution anteroposterior gene expression maps of C. elegans, including maps of hermaphrodites, males and germline deficient mutants. In addition to providing an expression pattern resource, serial-seq is a powerful method for studying gene expression pattern changes at the whole-genome level.

Ebbing, A.L.P. et al. (2017) International Worm Meeting "Serial-section RNA sequencing: genome-wide gene expression pattern analysis in C. elegans males and hermaphrodites."

Vertesy, A., Betist, M.C., Korswagen, H.C., van Oudenaarden, A., Ebbing, A.L.P.

[International Worm Meeting, 2017]

RNA sequencing (RNA-seq) is ideally suited for genome-wide analysis of gene expression. However, when a whole organism is used for RNA isolation and sequencing, the data will show the overall level of gene expression, but not where such genes are expressed in the organism. A solution to this problem is to provide spatial information by sectioning the organism along one (or more) of the body axes and to perform RNA-seq on each of the separate sections. We have developed a serial-section RNA-seq (serial-seq) method for C. elegans (see also the other abstract of Ebbing, Korswagen and co-workers). In brief, we cryo-section a young adult animal from the head to the tail into 20 micrometer thick slices, resulting in a total of 40 to 50 slices. Each slice is then sequenced using the sensitive CEL-Seq method and the sequencing data is aligned to create a high-resolution anteroposterior gene expression map. We have generated expression maps for 4 young adult hermaphrodites, 4 young adult males and 2 germline deficient glp-1 mutants. In these datasets, an average of 8794 genes (43% of coding genes) can be detected along the length of the animal. These include genes that are expressed in major tissues such as the germline, but also genes that are known to be expressed in only single neuron pairs in the head, demonstrating the high sensitivity of our approach. Furthermore, we found that these expression patterns are highly reproducible among independently sectioned animals. To explore these gene expression maps, we have developed an algorithm that enables us to search for genes with specific expression patterns. Using the expression patterns of known marker genes as a starting point, we have identified genes that are specifically expressed in the germline and sperm, in organ structures such as the pharynx, the intestine, the vulva, the spermatheca and the male reproductive tract, and in neurons of the nerve ring and tail ganglia. Furthermore, we have found extensive sex-specific differences in gene expression in the germline and sperm, in the male reproductive tract and in the male specific CEM neurons. We will present an overview of the method and discuss our findings on cell, organ and sex-specific gene expression patterns.

Ebbing, A.L.P. et al. (2019) International Worm Meeting "Single cell transcriptomics of the Q neuroblast lineage reveals mechanisms in neurogenesis, directional migration and Wnt pathway regulation."

Ebbing, A.L.P., Louisse, T., Korswagen, H.C., Betist, M., Povoa, E.E.F., Soto, L.

[International Worm Meeting, 2019]

The two Q neuroblasts and their descendants migrate in opposite directions along the AP axis, in a process that can be divided into two distinct phases. In the first phase, left-right asymmetry is established, while in the second phase, Wnt signaling and Hox genes regulate the long-range migration of the Q descendants. Here, we have used mRNA sequencing of isolated Q neuroblasts to gain detailed insight into the transcriptional mechanisms that control these two phases. First, we performed single cell mRNA sequencing of the Q neuroblasts and their descendants, with a particular focus on cells in the early stage of the migration process. Cluster analysis and pseudo-temporal ordering showed that the single cell data provide unique insight into the gene expression changes that accompany the transition from an epithelial to neuronal fate, Wnt pathway activation, and neuronal differentiation. Interestingly, clustering did not reveal significant differences between the left and right Q neuroblasts, indicating that the initial asymmetric polarization of the Q cells is independent of transcriptional regulation. Second, we made use of loss- and gain-of-function mutants of the Hox gene mab-5 to focus on the second, long-range migration phase. mab-5 expression is activated by canonical Wnt signaling in the left Q neuroblast (QL) and mediates posterior migration of the QL descendants. We performed mRNA sequencing on pooled Q neuroblasts, and identified genes that are differentially expressed in the two mutant backgrounds. These results provide detailed insight into the guidance mechanisms that mediate posterior and anterior migration. Among the genes that are upregulated in mab-5 gain-of-function mutants is plr-1, an ortholog of ZNRF3/RNF43, which functions as a direct feedback inhibitor of canonical Wnt signaling in mammalian cells. We found that loss of plr-1 leads to increased mab-5 expression, indicating that it is part of a negative feedback loop that enables mab-5 to fine tune its own expression by modulating Wnt signaling. Such an indirect feedback loop has been predicted by quantitative modelling of Wnt pathway regulation in the QL lineage. Taken together, we have used transcriptomics to further elucidate the mechanisms underlying the two phases of the directional migration of the Q cells. For the first phase, our single cell resolved dataset reveals expression dynamics of genes involved in epithelial to neuronal fate change, Wnt pathway activation, and neurogenesis. Moreover, our data indicate that the initial left-right asymmetry of the Q neuroblasts is not dependent on transcriptional regulation. For the second phase, our analysis of mab-5 gain and loss-of-function mutants provides insight into the anterior and posterior migration mechanisms of the Q cell descendants, and shows a novel role of plr-1 as an indirect feedback inhibitor of the Wnt pathway.

Fernandes Povoa, E.E. et al. (2017) International Worm Meeting "A transcriptomic approach to understand complex (Wnt) signaling dynamics in Q neuroblast lineage progression and migration."

Korswagen, H.C., Ebbing, A.L.P., Rella, L., Betist, M.C., Fernandes Povoa, E.E.

[International Worm Meeting, 2017]

Wnt signaling plays a major role in regulating cell migration, not only during metazoan development, but also in the context of human disease. Frequently, these regulatory mechanisms comprise intricate interactions and cross-talk between different signaling pathways. For instance, a shift from canonical to non-canonical Wnt signaling has been reported to underlie a "phenotypical switch" that induces melanoma cells to adopt a migratory and invasive phenotype. In the nematode C. elegans, the migration of the QR neuroblast and its descendants can be divided into four sequential steps, three of which are mediated by distinct Wnt signaling mechanisms. However, little is known on how these mechanisms are regulated and how they interact to achieve the correct temporal regulation of QR lineage progression and migration. We are using this single cell migration model system to deepen our knowledge on how such types of complex interactions can be regulated. We have combined FACS-based Q neuroblast sorting and RNA-sequencing (CEL-seq) to better understand the temporal transcriptional dynamics occurring during QR lineage progression and migration. Here, we show that our approach is capable of capturing expression dynamics of multiple Wnt-related genes previously described in literature. Based on the data generated, we are currently testing and characterizing a list of prominent candidate genes for their roles in these Wnt signaling switches, and their underlying relevance for QR lineage progression and migration.