Cell Calcium 39 (2006) 197–208 Anti-phase calcium oscillations in astrocytes via inositol (1, 4, 5)-trisphosphate regeneration Ghanim Ullah a, , Peter Jung a , A.H. Cornell-Bell b a Department of Physics and Astronomy and Quantitive Biology Institute, Ohio University, Athens, OH 45701, USA b Anscans LLC, Ivoryton, CT, USA Received 18 July 2005; received in revised form 22 October 2005; accepted 26 October 2005 Available online 2 December 2005 Abstract Observations in cultured mouse astrocytes suggest anti-phase synchronization of cytosolic calcium concentrations in nearest neighbor cells that are coupled through gap junctions. A mathematical model is used to investigate physiologic conditions under which diffusion of the second messenger inositol (1, 4, 5)-trisphosphate (IP 3 ) through gap junctions can facilitate synchronized anti-phase Ca 2+ oscillations. Our model predicts anti-phase oscillations in both cytosolic calcium and IP 3 concentrations if (a) the gap junction permeability is within a window of values and (b) IP 3 is regenerated in the astrocytes via, e.g. phospholipase C . This result sheds new light on the current dispute on the mechanism of intercellular calcium signaling. It provides indirect evidence for a partially regenerative mechanism as the model excludes anti-phase synchrony in the absence of IP 3 regeneration. © 2005 Elsevier Ltd. All rights reserved. 1. Introduction Calcium signaling is one of the most frequently observed intracellular signaling mechanisms (for a comprehensive review see [1]). Calcium is released from internal stores, most notably the endoplasmic reticulum, through inositol (1, 4, 5)-trisphosphate receptors (IP 3 Rs) activated by inositol (1, 4, 5)-trisphosphate (IP 3 ). As Ca 2+ is released, it activates nearby IP 3 Rs through Ca 2+ induced Ca 2+ release, triggering an intracellular Ca 2+ wave propagating through the cell. In many cases, Ca 2+ signals can also propagate across the cell membrane to form what has been termed an intercellular Ca 2+ wave. The coupling mechanism, however, varies between the cell-types. Pancreatic acinar cells are connected by Cx32- and Cx26-built gap-junctional channels. Yule et al. [2] have shown that these gap junctions are permeable to IP 3 and Ca 2+ . Through mathematical modeling it has been Corresponding author. Tel.: +1 740 593 9671; fax: +1 740 593 0433. E-mail address: ghanimul@helios.phy.ohiou.edu (G. Ullah). shown that the Ca 2+ coupling, enhanced by IP 3 diffusion leveling the IP 3 concentrations between the cells, can lead to in-phase and anti-phase synchronization of Ca 2+ and propagating Ca 2+ waves [3]. In astrocytes, however, the situation is different. In astro- cytes the gap junctions are build by Cx43 [4,5]. Numerous observations indicate that Ca 2+ cannot be the main messenger of the Ca 2+ waves across the cells although there may exists a small conductance for Ca 2+ . Carter et al. [6], have shown that the gap junctions composed of Cx43 in aortic endothe- lial cells are extensively permeable to IP 3 . Intercellular Ca 2+ waves propagate with or without exhibiting a Ca 2+ response; if the intracellular Ca 2+ response is reduced pharmacologi- cally, its propagation range is not affected [7]. Intracellular Ca 2+ waves propagate through cells where calcium-induced calcium release is inhibited [7]. In the absence of IP 3 loading astrocytes with an elevated Ca 2+ concentration do not initiate a Ca 2+ wave. Another observation suggesting that Ca 2+ is not the main signaling molecule is that intercellular Ca 2+ waves only travel a finite distance (about 300 m). If Ca 2+ would be the messenger, such an abortive wave would not be expected. 0143-4160/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ceca.2005.10.009