Paul W. Sternberg
Molecular Genetics of Nematode Development and BehaviorUsing the nematode Caenorhabditis elegans, our laboratory takes molecular genetic and genomic approaches to basic questions in developmental biology and neurogenetics: What are the molecular mechanisms by which cells interact to establish a spatial pattern of cell types? How are signals among cells integrated to coordinate organ formation? How do genes control the ability to execute stereotyped behavior? How do neural circuits evolve? Many of the genes we have identified are the nematode counterparts of human genes, and we expect that some of our findings will apply to human genes as well. On the other hand, nematode-specific genes enhance our understanding of this important phylum. Our strategies include identification of genes through high throughput DNA sequencing, genetic screens, detailed observation of cell and organism behavior, and cycles of computational and experimental analyses.
A major ongoing focus has been the role of peptide growth factors in controlling cell fate patterns. We have analyzed the roles of LIN-3, a nematode homolog of human epidermal growth factor (EGF), and LET-23, its receptor, a homolog of human EGF receptor. In addition to roles in developmental fate specification, this EGF signalng pathway controls ovulation and a sleep-like state. For developmental roles, LET-23 signals via a pathway utilizing the C. elegans homolog of the RAS proto-oncogene; for ovulation and sleep LET-23 signals via phospholipase C gamma and the two second messengers it produces, inositol tris-phosphate and diacylglycerol.< /p>
Expression of LIN-3 in the anchor cell of the gonad induces the vulva. We have found a small region of the lin-3 gene that directs its expression specifically in the anchor cell; by studying this element, we are learning how the state of anchor cell differentiation is programmed, and have identified computationally other genes that have this element and are involved in anchor cell specification. After vulva induction, the vulval precursor cells generate cells that differentiate as vulval cells (of which there are seven types) and undergo morphogenesis to form the mature vulva. We have developed a panel of yellow and cyan fluorescent protein markers for these terminally differentiated cells and are elucidating how multiple signaling pathways and a number of transcriptional regulatory proteins interact to control expression of these genes.
Three WNT signaling pathways control the polarity of one of the vulval precursor cells. One of these signaling pathways involves the lin-18 gene, which we have shown encodes a receptor-type tyrosine kinase-related protein; another pathway involves a classical WNT receptor encoded by lin-17; a third pathway acts antagonstically to these two and works via the Ror tyrosine kinase CAM-1. The WNT EGL-20 is expressd in the tail and promotes a posterior orientation. Two other WNTs (MOM-2 and LIN-44) are expressed in the anchor cell and provide a localized signal to orient the polarity toward the anchor cell.
After these interactions occur, particular differentiated vulval cells connect to the anchor cell and the uterus. As part of this process, the anchor cell breaks down the basement membrane separating the gonad and vulva, and invades the vulval epithelium; this process guides the ultimate attachment of the uterus and vulva. We have found that distinct transcriptional regulators control the vulva-inducing signal and the invasion process. This separation of induction and invasion programs nicely reflects evolutionary differences in anchor cell function: several nematode species have anchor cells that invade the vulval epithelium but do not induce the vulva. Transcriptional targets of the invasion pathway include a zinc metalloprotease of the type implicated in tumor metastasis.
To examine how broadly used regulatory pathways combine to specify cell fates, and how the specificity of LET-23 signaling is determined, we are studying pattern formation during male hook and spicule development. The male hook, like the hermaphrodite vulva, develops from tripotent precursor cells. In both cases multiple signaling pathways work together to specify the precise pattern of cell fates. We have found, however, that there are differences in which pathways are more important. Induction of the vulva requires EGF and Notch signaling, with WNT playing a minor role. By contrast, during hook development, WNT and Notch signaling are crucial, and EGF plays a minor role.
We have continued to analyze the mating behavior of the C. elegans male to understand how genes control neuronal function and to identify new proteins involved in neuronal function. By ablating cells and observing mating behavior, we dissected the behavior into several steps, and we are identifying genes used at many of these steps. For example, some mutants fail to recognize the hermaphrodite; others fail to turn at the end of the hermaphrodite; others fail to locate the vulva; yet others fail to transfer sperm.
It has been known for several years that C. elegans hermaphrodites produce a soluble signal that alters male behavior. In collaboration with the laboratories of Arthur Edison (University of Florida) and Frank Schroeder (Cornell University), we have discovered by chemical fractionation that the soluble mating cue comprises several derivatives of ascarylose sugars (ascarosides) that act synergistically as a mating pheromone. Some of these compounds were previously known to act as density signals for C. elegans, although at much higher concentration. We hypothesize that ascarosides are a general class of chemical signals among nematodes. At least one of the ascarosides, ascr#3, is sensed by four male-specific sensory neurons called CEMs and two non-sex specific neurons called ASKs. (This work was funded by the Human Frontier Science Program.)
Recognition of hermaphrodite and vulva location is of interest because it involves the polycystins, proteins disrupted in autosomal dominant polycystic kidney disease. The polycystins, divergent members of the transient receptor potential (TRP) family of calcium channels, focused our interest in the classic members of this family, the TRPC proteins. We have screened for deletions of three TRPC and one TRPN genes in C. elegans and have found functions for them in fertilization (TRP-3), nicotine addiction (TRP-2) and locomotion (TRP-4).
We have developed an automated system to analyze nematode sinusoidal locomotion. The system tracks worms on a Petri plate and records their position and posture, thus allowing us to measure the parameters of their movement, such as their velocity and the velocity of the muscle contraction wave that propels the worm. The data from different Caenorhabditis species and from C. elegans mutants with defects in muscle, skeleton (cuticle), or nervous system have allowed us to construct a mathematical model of the major components of C. elegans movement. We are collaborating with Bill Schafer and colleagues at the MRC-LMB (Cambridge) to perfect the system. We are using it to perform phenotypic profiling of mutants with defects in genes relevant to nervous system function. This research is funded by NIH-NIDA.
We are using comparative genomics to study transcriptional regulation. In collaboration with Barbara Wold, we analyzed cis-regulatory sequences of particular genes in two other Caenorhabditis species to determine which would be the most immediately useful for inferring transcriptional control elements from multiple genomic sequences, given the existence of C. elegans and C. briggsae sequences. These data encouraged the National Human Genome Research Institute to initiate sequencing of the C. remanei, C. brenneri, and C. japonica genomes. We are collaborating with the Genome Sequencing Center of Washington University and many other colleagues to analyze these genomes.
We are involved in an international effort to organize information about C. elegans genomics, genetics, and biology and present this information in an Internet-accessible database, WormBase (www.wormbase.org). Our major contribution is to extract information from the literature, focusing on gene, protein, and cell function; gene expression; microarray and chromatin immunoprecipitation data and gene-gene interactions. To facilitate this process, we have developed a useful search engine for the C. elegans literature ((www.textpresso.org). We are working with other model organism databases to extend Textpresso to other domains, such as Arabidopsis, rat and Drosophila. (This effort is funded by the National Human Genome Research Institute.) We are also implementing a neuroscience literature search engine as part of the Neuroscience Information Framework (funded by NIH-NIDA).
We have established a three-organism system for symbiosis, infection, and vectorborne disease. The nematode Heterorhabditis bacteriophora infects insect larvae and regurgitates its symbiotic bacterium Photorhabdus luminescens, which kill the insect host and provide food for the production of more nematodes. We have found that H. bacteriophora will infect Drosophila melanogaster larvae, and we developed RNAi for this organism. We are involved in the genome sequencing project for H. bacteriophora, and have started to work with another insect parasitic nematode, Steinernema carpocapsae, for comparison.