We have three broad interests: (1) How is do "positional cues" along the anteroposterior and dorsoventral axes lead to distinct cell fates within the neuroectoderm? (2) What are the mechanisms regulating asymmetric cell division in the CNS: either the physically asymmetric divisions of CNS stem cells or the developmentally asymmetric divisions that produce sibling neurons with different fates? and (3) How do CNS stem cells produce a different daughter cell at every cell division: are daughter cells born naive into an environment with ever-changing cues, or are daughter cells born with a specific identity due to cell cycle-dependent intrinsic factors?

The Drosophila CNS develops from a ventral neuroectoderm (the neuroectoderm is created by the same genes that establish the vertebrate neural plate). Individual neural stem cells delaminate from the neuroectoderm into the embryo at precise times and positions; there are initially 9 neuroblasts and a single MP2 precursor per bilateral hemisegment. Each neuroblast divides asymmetrically to bud off a series of smaller daughter cells called ganglion mother cells (GMCs); each GMC makes two distinct neurons or glia. MP2 divides once to produce the dMP2 and vMP2 neurons.

To investigate how positional cues in the neuroectoderm lead to the unique specification of individual neuroblasts, we have generated or collected almost two dozen molecular markers to produce a "neuroblast map" in which each neuroblast can be uniquely identified. More recently, we have used the vital fluorescent tracer DiI to trace the cell lineage of all embryonic neuroblasts. Every neuroblast makes a characteristic clone, which can include a mixture of motoneurons, interneurons, and glia. We are currently characterizing genes that are expressed in anteroposterior (AP) or dorsoventral (DV) "stripes" within the neuroectoderm, testing whether they function as "positional cues" to specify neuroblast fate (e.g. the wingless and hedgehog pathways along the AP axis, and three homeobox genes -- vnd, ind, and msh -- along the DV axis).

Each of these genes is necessary (and in some cases sufficient) to specify the fate of the neuroblasts that form within its expression domain. Open questions include: How is the pattern of each homeobox gene established? and What are the transcriptional targets of each homeobox gene within the CNS?

To investigate the second question, how does a cell divide asymmetrically to produce unique daughter cells, we are focusing on neuroblast and MP2 divisions. Neuroblasts repeatedly divide to produce a large neuroblast and a smaller GMC. This division is asymmetric with regard to sibling cell size, mitotic potential and gene expression. Recent work from our lab and others has identified a number of molecules that show a polarized distribution during neuroblast mitosis: prospero RNA and Inscuteable, Miranda, Prospero, Staufen, and Numb proteins. Miranda is a scaffolding protein that binds Prospero and Staufen proteins; Staufen is an RNA-binding protein that binds prospero RNA. Inscuteable is required for the basal localization of Miranda and associated molecules as well as for the apical/basal orientation of the mitotic spindle. Thus, Inscuteable coordinates protein/RNA localization with cell division to ensure that only the GMC inherits Miranda, Prospero, Staufen, and prospero RNA. The process of asymmetric localization of proteins and RNAs is cell cycle dependent, microfilament dependent, coordinated with the positioning of the mitotic spindle, and results in the unequal distribution of cell fate determinants to a specific daughter cell at cytokinesis. Open questions include: What is the original cell polarity cue that regulates apical localization of Insc? How does apical Insc trigger basal localization of Miranda and its cargo -- by regulating polarized movement along cortical microfilament? or by regulating degradation of Miranda at the apical cortex? We are using both genetic screens and yeast 2 hybrid assays to discover new players in this process.

The simple MP2 lineage has been a useful model for studying how sibling neurons become different. We find that the membrane-associated Numb protein is localized at MP2 mitosis into dMP2 and excluded from vMP2. We also have shown that: (1) Numb is necessary and sufficient for dMP2 fate; (2) the Delta-Notch signaling pathway is necessary for vMP2 fate; and (3) that Numb blocks Notch-mediated signaling which leads to dMP2 cell fate. In addition, we have shown that mutations in sanpodo give the same sibling neuron phenotype as Notch mutations. Sanpodo is an actin-binding protein that also binds the intra-cellular domain of Notch. Open questions include: How does Sanpodo regulate Notch signaling? Are actin-binding and Notch-regulation separable or linked functions of Sanpodo? Does the same mechanism control sibling cell fate in vertebrates?

To investigate the third question, how does a neuroblast produce a different GMC at each cell cycle, we are doing cell ablation and in vitro culture experiments to see if early-born GMCs signal back to the neuroblast to regulate the fate of later-born GMCs. In addition, we are taking a genetic approach, studying the function of genes whose expression is limited to early- or late-born GMCs in a lineage. Our hypothesis is that this class of genes is involved in specifying differences between GMCs within a lineage. Open questions include just about every aspect of this important process, since we know so little about how a series of different daughter GMCs are produced from a single neuroblast.

Questions about/problems with this site? Please e-mail the webmaster.

© Copyright 2007 Kinship Foundation. All rights reserved.