Motion as a phenotype: the use of live-cell imaging and machine visual screening to characterize transcription-dependent chromosome dynamics.

BACKGROUND: Gene transcriptional activity is well correlated with intra-nuclear position, especially relative to the nuclear periphery, which is a region classically associated with gene silencing. Recently however, actively transcribed genes have also been found localized to the nuclear periphery in the yeast Saccharomyces cerevisiae. When genes are activated, they become associated with the nuclear pore complex (NPC) at the nuclear envelope. Furthermore, chromosomes are not static structures, but exhibit constrained diffusion in real-time, live-cell studies of particular loci. The relationship of chromosome motion with transcriptional activation and active-gene recruitment to the nuclear periphery has not yet been investigated. RESULTS: We have generated a yeast strain that enables us to observe the motion of the galactose-inducible GAL gene locus relative to the nuclear periphery in real-time under transcriptionally active and repressed conditions. Using segmented geometric particle tracking, we show that the repressed GAL locus undergoes constrained diffusive movement, and that transcriptional induction with galactose is associated with an enrichment in cells with GAL loci that are both associated with the nuclear periphery and much more constrained in their movement. Furthermore, we report that the mRNA export factor Sac3 is involved in this galactose-induced enrichment of GAL loci at the nuclear periphery. In parallel, using a novel machine visual screening technique, we find that the motion of constrained GAL loci correlates with the motion of the cognate nuclei in galactose-induced cells. CONCLUSION: Transcriptional activation of the GAL genes is associated with their tethering and motion constraint at the nuclear periphery. We describe a model of gene recruitment to the nuclear periphery involving gene diffusion and the mRNA export factor Sac3 that can be used as a framework for further experimentation. In addition, we applied to the analysis of chromosome motion a machine visual screening approach that used unbiased visual data rather than segmented geometric data. This novel analytical approach will allow for high-throughput study of processes that can be monitored via alterations in chromosome motion and connectivity with the nuclear periphery.

DEG/ENaC but not TRP channels are the major mechanoelectrical transduction channels in a C. elegans nociceptor.

Many nociceptors detect mechanical cues, but the ion channels responsible for mechanotransduction in these sensory neurons remain obscure. Using in vivo recordings and genetic dissection, we identified the DEG/ENaC protein, DEG-1, as the major mechanotransduction channel in ASH, a polymodal nociceptor in Caenorhabditis elegans. But DEG-1 is not the only mechanotransduction channel in ASH: loss of deg-1 revealed a minor current whose properties differ from those expected of DEG/ENaC channels. This current was independent of two TRPV channels expressed in ASH. Although loss of these TRPV channels inhibits behavioral responses to noxious stimuli, we found that both mechanoreceptor currents and potentials were essentially wild-type in TRPV mutants. We propose that ASH nociceptors rely on two genetically distinct mechanotransduction channels and that TRPV channels contribute to encoding and transmitting information. Because mammalian and insect nociceptors also coexpress DEG/ENaCs and TRPVs, the cellular functions elaborated here for these ion channels may be conserved.

New methods for automated phenotyping of complex cellular behaviors.

Cellular shape change and movement are central to biologic processes that range from normal embryonic development to inflammatory diseases and cancer. Quantitative visual phenotyping of dynamic cellular behaviors creates unique challenges for image capture, analysis and storage. Despite substantial technological advances in molecular biology, biochemistry, genomics and proteomics, investigating cellular processes remains tremendously challenging and labor-intensive. We have developed algorithms and software implementations that allow for fully-automated analysis of experiments designed to investigate a range of cellular and organismal behaviors. By enabling cellular phenotyping, this automated approach creates a unique opportunity for investigators to perform large-scale experiments designed to determine gene function or to screen for small molecule modulators of important cellular behaviors.