James Russell, DVM, Head, Section on Cell Biology and Excitability
Susanna Weerth, PhD, Postdoctoral Fellow
Lynne A. Holtzclaw, BS, Research Assistant
Stan Atkin, BS, Student Training Fellow
Jennifer L. Hopp, BS, Student Training Fellow

We study the cellular mechanisms of calcium signaling by glial cells in the nervous system. Glial cells and neurons are in intimate communication with each other during central nervous system development and normal brain function. Glial cells monitor and respond to neural activity by conditioning the extracellular milieu, signaling within glial cell networks, and sending signals back to neurons. In the brain, such signaling takes the form of propagated Ca2+ waves that spread over long distances in response to synaptic activity. Similarly, myelinating glial cells (oligodendrocytes and Schwann cells) receive signals from the axon, and such signals are essential for the maintenance of the myelin sheath. One of our objectives is to understand the processes that support temporal and spatial characteristics of Ca2+ signals within and between cells. A second objective is to understand the nature of glial cell signals in response to neuronal activity and the consequences of such signals to CNS function.

Ca²+ signaling microdomains

We have previously shown that Ca²+ wave propagation occurs through a series of Ca²+-signaling microdomains organized along glial cell processes. These microdomains act as highly specialized Ca²+ release sites that produce discretely localized spurts of cytoplasmic Ca²+ increases. Wave propagation thus occurs by a fire-diffuse-fire cascade process, whereby the signal leaps along in a saltatory fashion. The specialized signaling microdomains are found 5 to 7 μm apart, in close association at least one mitochondrion, and are characterized by high-density patches of endoplasmic reticulum (ER) proteins involved in Ca²+ signaling, such as the inositol 1,4,5-trisphosphate receptors (IP₃Rs), sarco-endoplasmic reticulum calcium pumps, and calreticulin. In addition to supporting long-distance wave propagation, the specialized sites provide for locally discrete Ca²+ signals that last for only very brief periods. Complete knowledge of the proteins that make up the signaling microdomain and the mechanisms by which they support the highly organized Ca2+ release does not exist.

Molecular characterization of signaling rafts

Weerth, Holtzclaw

One of our goals is to investigate the molecular organization of Ca²+-signaling microdomains and to describe their functional regulation in detail. Our overall aim is to test the hypothesis that, in the specialized microdomains of Ca²+ release, a number of macromolecular protein complexes are physically brought together, at times in different membrane systems, and interact to generate large, spatially restricted Ca²+ signals. We have focused on glial cell processes, particularly cells of the oligodendrocyte lineage and astrocytes, and aim to undertake a molecular characterization of these specialized Ca²+ release sites to determine the interacting protein assemblies that orchestrate such release.

Using glial cell membranes isolated from mouse brains, we seek to isolate oligomeric protein assemblies that make up specialized intact microdomains of Ca²+ signaling. Using an immuno-affinity panning, we prepared cell-specific membranes from brain homogenates. We then used antibodies against cell surface proteins to specifically immuno-pan the homogenate to obtain subfractions containing membranes enriched in astrocyte membranes or oligodendrocyte progenitor (OP) cell membranes. For this purpose, we used Ran-2, an astrocytic surface antigen, or NG-2, an OP-cell antigen. We solubilized isolated membranes in detergent and used them either for immunoprecipitation or, using isotonic density gradients, for separation of lipid raft-like structures.

We fractionated the gradients and tested each fraction in Western blots for protein content. We used caveolin-1 as a marker for raft-like membranes and, using specific antibodies, probed for a number of proteins involved in Ca²+ signaling. The proteins included the surface membrane receptors M1 and P2Y1, the surface membrane ion channel TrpC1, heterotrimeric G-protein isoform Gq, scaffolding protein, homer, and endoplasmic reticulum membrane proteins IP3R2 and RyR3. The analysis showed that most of the proteins involved in Ca²+ signaling separated into low-density regions of the gradient, which also contained caveolin-1. The data indicate that many signaling proteins might exist with each other in oligomeric complexes in lipid domains that are cholesterol-rich and raft-like.

In another experiment, we used solubilized cell-specific membranes as starting material and, to determine the protein complement of the oligomeric complexes, immunoprecipitated the membranes by using specific antibodies against IP3R2, Homer-1, and TrpC1. This experiment showed that IP3R2 antibodies specifically immunoprecipitated both TrpC1 and Homer-1. Similarly, both TrpC1 antibodies and Homer-1 antibodies immunoprecipitated the other two antigens, suggesting that all three proteins were present in the complex. PAGE visualized a number of protein bands; we then analyzed the bands further by using high-resolution proteomic techniques for protein identification.

Ca2+ signaling between axons and myelinating glia

Atkin; in collaboration with Ellisman

In a new project, we are investigating aspects of signaling between axons and their myelinating glia (Schwann cells and oligodendrocytes) in the central and peripheral nervous system (CNS and PNS). Initially, we focused on characterizing the distribution of Ca²+-signaling proteins in the axoglial apparatus of the nodes of Ranvier. The myelin sheath functions as an electrical insulator, reducing current flow across the axonal membrane in the internode. It also reduces capacitance and increases resistance, thereby facilitating saltatory conduction. Myelin is formed by a highly specialized extension of myelinating glial cell (oligodendrocytes in the central nervous system and Schwann cells in the peripheral nervous system) membranes intimately associated with axons

Beyond the concept of saltatory conduction, a number of studies support the view that the axon/myelin/glial cell ensemble or the "nodal complex" operates in an integrated manner during conduction of the nerve impulse. Over the last two decades, several anatomical and physiological studies have shown that an intricate set of signals might exist between the components of the nodal complex. Details of such signaling between axons and the myelinating glia are unknown. We initiated a study aimed at describing the distribution of proteins involved in Ca²+ signaling in the axoglial apparatus and believed that the analysis would point us to the spatial discreteness, if any, in signaling between axons and Schwann cells. Our initial hypothesis posited that the signaling is localized to specific contact sites between the two cell types where the insulating layer of compact myelin gives way to non-compact myelin and Schwann cell membrane. We selected antibodies against proteins involved in Ca²+ signaling (e.g., IP3Rs and RyRs) and known structural proteins unique to different regions of the node and investigated their discrete localization with immunocytochemistry followed by high-resolution fluorescence microscopy.

We carried out the analysis in isolated, teased sciatic nerve preparations, using both fixed and unfixed tissues. The distribution of myelin basic protein (MBP) and S100β showed the previously described intricate architecture of the nodes of Ranvier and highlighted the intimate relationship between the axon and the myelinating glia. We focused on the arrangement of proteins in the axoglial apparatus and selected specific proteins uniquely expressed in the node (NaV1.2), paranode (MAG, connexin 32), juxtaparanode (Kv 1.2), and Schwann cell cytoplasm (S100β). Immunohistochemical analysis using antibodies against these selected proteins provided us with the anatomical arrangements of the two cell types in the nodes of Ranvier. We examined the distribution of proteins involved in Ca²+ signaling within the architecture of the myelinated axon, using specific antibodies against IP3Rs, RyRs, P2Y1, M1, and Gq. The study is now complete and a manuscript is in preparation. In future experiments, we plan to investigate the distribution of the signaling proteins at higher resolution with electron microscopic immunocytochemical techniques. We recently started these experiments in collaboration with Mark Ellisman and will focus on co-localization of specific proteins in discrete regions of the Schwann cells.

In situ imaging of Schwann cell calcium signals

Atkin; in collaboration with Pickel

We devoted a significant amount of time to a project begun in 2004 to measure directly Schwann cell Ca²+ signals associated with action potential traffic along axons. To this end, we developed a transgenic mouse line that expressed a fluorescent Ca²+ indicator in a specific manner in Schwann cells. We used the S-100b promoter to specifically target YC3.60, a mutant chameleon protein, to Schwann cells of peripheral nerves and to astrocytes in the brain. We obtained a YC3.60 construct from Dr. Miyawaki at RIKEN, Tokyo, Japan, and engineered a plasmid construct containing the S100b promoter in tandem. We produced two such plasmid constructs: S100-YC3.60.6 and S100-YC3.60.9. Transfection of C6 glioma cell lines and HEK 293 cell lines showed that YC3.60 was expressed by the glioma line only, which expresses S100b, though not in HEK293 cells. In collaboration with James Pickel, we used the plasmid DNA to generate transgenic mouse founders. Eleven founder mice were initially found by PCR genotyping and by using the level of specific YC3.60 expression in astrocytes in the brain as criteria. We have established four different founder lines expressing YC3.60 fluorescence in glial cells and are breeding them to homozygosity.

We imaged brain slice preparations from two of the lines by two-photon confocal microscopy. We readily visualized individual astrocytes brightly fluorescent with YC3.60 chameleon and observed cells with large spongiform process arbors as well as smaller cells containing YC 3.60. Many of the cells had elongated cell somas with several large branches radiating parallel or perpendicular to those of their neighboring glia. Numerous YC 3.60-positive astrocytic processes extended to wrap small blood vessels, and we found YC 3.60 expression in the cell soma and processes in astrocytes. Dual staining with anti-S100β and anti-GFP antibodies in the cerebellum showed that over 78 percent of S100β-positive Bergmann glial cells expressed YC 3.60.

To verify that the YC 3.60 within cells was functional as a Ca²+ indicator in situ, we used isolated acute brain slice preparations and applied previously characterized experimental paradigms to elicit neurotransmitter-induced, glial cell Ca²+ signals. A number of laboratories have demonstrated astrocytic Ca²+ signals in response to exogenously applied glutamate. Indeed, glutamate elicited increases in the YFP/CFP fluorescence ratio in a majority of the cells examined (average ratio about 65 percent). In most experiments, one or two cells showed very large YFP/CFP ratio changes (greater than 100 percent) as compared with the rest while some cells did not respond at all. In another set of experiments, we examined Bergmann glial cells in cerebellar slices obtained from heterozygous S100-YC-C mice. Glutamate elicited increases in YFP/CFP fluorescence ratios, thus indicating cellular Ca²+ rises. However, the peak amplitude of these signals was significantly lower than those recorded in astrocytes in cortical or hippocampal slices.

Stimulation of Schaffer collaterals in hippocampal slices evoked robust Ca²+ signals in astrocytes in the stratum radiatum and stratum moleculare-lacunosum. Schaffer collateral stimulation caused an immediate Ca²+ rise in a number of astrocytes in slice preparations. In particular, in slice preparations from S100-YC-C and S100-YC-H mice, the stimulus evoked a YFP/CFP ratio increase that occurred in over 70 percent of cells, with an average response of about 60 percent; in each case, we recorded extracellular field potentials. We then focused on one of the spongiform protoplasmic astrocytes using high-magnification objectives and electronic zoom. We repeated the stimulus paradigm during two-photon imaging and measured YC 3.60 fluorescence changes within the indicated regions of interest. In the cell soma, a large increase in YFP/CFP ratio followed the stimulus. Many of these regions of interest represent glial microdomains within the cell's amorphous spongiform morphology. The cellular Ca²+ response elicited by the neural stimulation spreads as a wave through the cell. Such glial microdomains of high activity are reminiscent of previously described small (less than 2 μm) astrocytic terminal sheaths that enwrap single synapses or groups of synapses. It is likely that the discrete domains of high Ca²+ signals observed here might represent spontaneous synaptic activity that excites terminal astrocytic branches ensheathing synapses.

One of the transgenic mouse lines showed abundant YC3.60 fluorescence within all Schwann cells in the peripheral nerves, whereas astrocytes in the CNS did not show appreciable fluorescence. We propose to use mice derived from this line to investigate action potential-dependent Ca²+ signals in Schwann cells. We would use both isolated nerve trunks and intact animals. We would rely on two-photon confocal microscopy to allow for the use of infrared illumination and deep tissue imaging so that signals could be recorded many hundreds of micrometers deep within tissue. Isolated nerve trunks would be positioned in imaging chambers and imaged under dual photon microscopy during electrical stimulation of the nerves.

COLLABORATORS

Mark Ellisman, PhD, University of California San Diego, La Jolla, CA
James Pickel, PhD, Laboratory of Genetics, NIMH, Bethesda, MD

For further information, contact james@helix.nih.gov.