|
 |
|
|
 |
 |
 |
  |
 |
|
|
|
|
|
|
Back to 2005 Research Ron Vale - Spatial Organization of Signaling 1. Cytoskeletal polarization and membrane domain formation in the immunological synapse. Antigen recognition by the T cell receptor (TCR) occurs in the context of an interaction between a T cell and an antigen-presenting cell (APC), and over the last few years it has become apparent that a high degree of order among proteins in the two cell membranes is required for antigen-triggered signaling to proceed normally. Within minutes of the initial binding events, the normally homogenous distribution of proteins in the T cell surface changes, with molecules involved in signal transduction becoming polarized toward the contact site and concentrated at the center of a structure referred to as the immunological synapse. This central signaling zone is approximately 5 µm in diameter and is surrounded by a ring of proteins involved in adhesion, including the integrin LFA-1, talin, and F-actin. The microtubule network is also a component of the synapse, with the MTOC polarizing and lying in close association with its center. While there has been much speculation about the function of the immune synapse, relatively little is known about how it forms. During the 2005 Physiology Course, we used a combination of single molecule and population level imaging to probe the mechanisms by which membrane patterning and MTOC polarization occur in the immune synapse. We used glass-supported, artificial membranes containing the integrin ligand, ICAM-1, and stimulatory anti-TCR antibodies to trigger synapse formation in Jurkat T cells. We saw that synapse formation in this system is preceded by microclustering of the anti-TCR antibody, followed by the directed translocation of these clusters to the center of the contact. Dual-color imaging in cells expressing actin-GFP revealed that cluster translocation rates very closely matched those of actin speckles, consistent with the idea that actin retrograde flow drives coalescence of the synapse. We also found that single molecules of ICAM-1 were recruited to the periphery of the contact via a simple diffusion-trapping mechanism, and that this localization was aided by the presence of a diffusion barrier at the outside of the central signaling cluster that prevented ICAM-1 entry (also see movie). In separate experiments, we assessed T cell microtubule dynamics by tracking growing microtubule ends in cells transfected with various plus-end binding proteins, including CLIP-170-YFP and EB1-GFP. For both of these constructs, activation of TCR signaling resulted in a significant increase in the frequency of MT nucleation, providing a potential mechanism for the polarization of the microtubule network toward the synapse.  Figure: An immunological synapse formed with Jurkat T cells interacting with planar lipid bilayers on a microscope slide. Red shows the position o the T cell receptor and green shows the adhesion molecule ICAM-1 clustering around the periphery of the T cell receptor. 2. Centrosome-independent mechanisms of microtubule nucleation and spindle pole organization in Drosophila S2 cells Centrosome is the dominant microtubule (MT) nucleation site during mitotic spindle formation. However, animal somatic cells also can generate MTs in interphase and form bipolar spindle in mitosis independently of centrosome function. In Drosophila S2 cells, it is unclear where MTs are nucleated during interphase, since mature centrosomes are not present until mitosis. To investigate the MTs nucleation sites in these cells, we observed EB1-GFP, a marker of plus-ends of growing MTs, and also gamma-tubulin, a known centrosome marker, in living cells and after immunostaining using confocal microscopy (Lori Krueger). We found that EB1-GFP exhibit several punctate signals near the nuclear envelope when we extensively depolymerize the interphase MTs, and that most of those dots are co-localized with gamma-tubulin. This result suggested that the cells have multiple MT nucleation sites in which gamma-tubulin is accumulated. On the other hand, RNAi knockdown of centrosomin (CNN) leads to loss of functional centrosomes throughout the cell cycle, but spindles MTs are generated and organized into a bipolar array by chromatin-dependent pathway, as is seen in meiotic system. Despite the absence of functional centrosomes, the spindle MTs, including kinetochore microtubule bundles (K-fiber), are reasonably well focused at their minus-ends in CNN RNAi cells. To understand the mechanism of K-fiber focusing in the absence of centrosomes, we performed double RNAi screening of CNN with ~200 known mitotic genes using high-throughput automated microscope Discovery-1 (Molecular Devices) and by automated mitotic cell detection (the Matlab code was written by students), and identified Ncd (minus-end-directed kinesin) and Asp (putative homolog of vertebrate NuMA that accumulates at minus-ends of K-fibers) as essential factors for focusing in CNN-depleted spindle. Interestingly, unlike meiotic spindle reconstituted using Xenopus egg extract, RNAi knockdowns of dynein/dynactin subunits did not produce additive focusing defect in S2 cells. Further extensive double or triple RNAi analysis among Ncd, dynein, CNN and Asp suggested that dynein’s role is restricted to transport K-fibers along centrosome MTs, and that Ncd or Asp has independent functions to crosslink K-fiber minus-ends (Andrea Kirby and Hagar Barak). We also developed a computer algorithm using Matlab that tracks K-fiber dynamics in the S2 spindle (Shiv Sivaramakrishnan) (see below and section on movies).  Figure: Matlab tracking of K-fibers in mitotic Drosophila S2 cells.  Figure: Double RNAi of centrosomin (cnn) and the kinesin ncd creates a highly disorganized spindle as seen in this GFP-tubulin imag 3. Spatial organization of signaling molecules in E. coli The signaling machinery involved in bacterial chemotaxis is clustering at one pole of the bacterium. In 2004, we examined the exchange of molecules in and out of these clusters by fluorescence recovery after photobleaching (see 2004 research). This year, we examined factors that determine the spatial localization of this chemotactic signaling cluster. First we asked if the bacterial cytoskeleton might be involved in this localization. We depolymerized filamentous MreB (an actin-like protein) with a small molecule (A22). This compound caused the bacteria to adopt a spherical shape. However, by GFP localization, chemotactic receptors (TAR-GFP) remained clustered even in these spherical cells. Thus, neither the cytoskeleton nor the elongate cell shape is required for receptor clustering. We also examined elongated filamentous bacteria (10-fold longer than normal E. coli) and we observed additional receptor clustered throughout the length of the bacteria. Moreover, at early times after TAR-GFP induction, very small microclusters of receptors were observed which were diffusing freeing within the membrane. We speculate that these microclusters are precursors of the larger, mature chemotactic clusters. |
|
|
 
 |
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
HOME | EDUCATION | RESEARCH | ABOUT THE MBL | CONTACT | SEARCH copyright © 2011 by The Marine Biological LaboratoryTM |
 |
|
||
