Nitya Viswanathan ‘07
We have used the confocal microscope to analyze the subcellular distribution of the APH-1 protein in early embryos of C. elegans. APH-1 is one of four protein members of the gamma-secretase complex, which achieves intramembranous cleavage of a variety of target membrane proteins. Although the effect in protein cleavage is clear, the site of gamma-secretase assembly and function within the cell remains elusive. We use the early C. elegans embryo as the context in which to analyze protein localization because the cells are relatively large, and because the four-cell embryo is the site of a well-defined event of Notch signaling that has been shown to be entirely dependent on gamma-secretase activity. Our analysis of wild type C. elegans embryos shows that the APH-1 protein is present at the plasma membrane in early embryos. This location is consistent with the role of gamma-secretase in targeting the Notch receptor after it has interacted with its ligand on an adjacent cell, and is also consistent with the plasma-membrane localization of the APH-2 protein, another gamma-secretase component. We have recently focused on the analysis of a mutant form of aph-1, termed “zu147” which is predicted to encode a truncated version of the APH-1 protein. The evolutionarily conserved aph-1 gene is predicted to encode a seven-pass transmembrane protein. In C. elegans, the transmembrane domains are followed by a 50-residue hydrophilic tail, but the zu147 nonsense mutation is predicted to prevent translation of this tail. Embryos that only contain the zu147 version of aph-1 show very little APH-1 activity, as the vast majority of embryos are defective in Notch signaling. We have used immunofluorescence and confocal microscopy to investigate the state of the APH-1 protein within these defective embryos, and have found a striking difference in the subcellular localization of the APH-1 protein. The aph-1(zu147) embryos contain detectable levels of APH-1 protein, however, it appears to localize predominantly to perinuclear regions of each cell rather than to the plasma membrane, as in wild type embryos (Figure 14B). This finding is intriguing in light of the similar perinuclear localization that we have observed for the APH-2 protein in embryos that lack other gamma-secretase components. Our working model is that the abnormal localization of gamma-secretase components to perinuclear regions may reflect failed assembly of complete and functional gamma-secretase complexes. Indeed, work from several labs supports the notion that gamma-secretase complex assembly is a prerequisite for successful translocation to the plasma membrane. In order to better understand the relationship between the activity and the subcellular localization of APH-1(zu147) we have begun to analyze the localization of APH-1 by confocal microscopy in a variety of genetic backgrounds which dramatically improve aph-1(zu147) function. Our preliminary analysis of these embryos reveal that sufficient levels of APH-1 activity can be achieved through a variety of effects on protein localization and or levels. In one case (Fig. 14C) a mutation in a new gene, sao-1, appears to disperse the detectable APH-1 throughout the cell (although the decrease in perinuclear staining is unambiguous, we plan to repeat this confocal analysis with mixed embryos to ascertain whether or not cytoplasmic levels of APH-1 are truly increased relative to the aph-1(zu147) embryos). In a second example (Fig 14D) we analyzed embryos, that we predicted would contain higher levels of aph-1(zu147) mRNA due to a mutation that interferes with the cell’s usual nonsense-mediated decay machinery. Interestingly, we again found loss of perinuclear staining, and a detectable punctate cytoplasmic staining of APH-1 protein. In all cases, increasing aph-1(zu147) activity seems to be correlated with a decrease in the perinuclear localization of APH-1. We were surprised to find that wild type function can be attained in embryos that have far less plasma-membrane localized APH-1 than is observed in wild type embryos (for example, compare Fig. 14C and 14D with 14A). We have used a classic genetic approach to generate a collection of single and double mutant embryos that differ in their levels of gamma-secretase activity. Molecular analysis and predictions of the involved mutations has allowed us to make molecular models as to the functional state of the gamma-secretase components. Now, the addition of confocal immunofluorescence analysis is allowing us to look inside the cells to determine where the modified proteins are found. As the subcellular localization of the gamma-secretase complex may prove to be highly dynamic, this approach is likely to yield insights into the regulation of this important activity.