- Email: [email protected]
Ca2+-Dependent Induction of Intracellular Ca2+ Oscillation in Hippocampal Astrocytes During Metabotropic Glutamate Receptor Activation
Journal of Pharmacological Sciences
J Pharmacol Sci 97, 212 – 218 (2005)
©2005 The Japanese Pharmacological Society
Full Paper
Ca2+-Dependent Induction of Intracellular Ca2+ Oscillation in Hippocampal Astrocytes During Metabotropic Glutamate Receptor Activation Yoshitoku Yoshida1, Remi Tsuchiya1, Nagisa Matsumoto1, Mitsuhiro Morita1, Hiroyoshi Miyakawa1, and Yoshihisa Kudo1,* 1
Molecular Life Science Division, School of Life Science, Tokyo University of Pharmacy and Life Science, 1432-1 Horinouchi Hachioji, Tokyo 192-0392, Japan
Received October 5, 2004; Accepted December 3, 2004
Abstract. We have investigated whether the intracellular calcium concentration ([Ca2+]i) oscillations induced in astrocytes using the metabotropic glutamate-receptor agonist, (1S,3R)-1aminocyclopentane-1,3-dicarboxylic acid (t-ACPD) are Ca2+-dependent, using three different Ca2+ indicators with different affinities for Ca2+. When rat hippocampal cells in culture were loaded with fura-2 (Kd: 145 nM), two-thirds of the cells showed obvious oscillatory increase in [Ca2+]i during t-ACPD-administration. Those cells were identified as astrocytes by immunohistochemistry in our previous paper. In cells loaded with fura-2FF (Kd: 25,000 nM), a similar percentage of t-ACPD-responsive cells showed oscillatory [Ca2+]i changes. However, in cells loaded with quin-2 (Kd: 60 nM), t-ACPD induced no oscillatory responses, but some cells showed a small transient increase in the [Ca2+]i. The same small transient [Ca2+]i increase was seen in cells loaded with both fura-2FF and BAPTA, a Ca2+ chelator (Kd: 135 nM). These findings indicate the involvement of [Ca2+]i-dependent regulatory mechanisms in the induction of the t-ACPD-induced oscillatory change in the [Ca2+]i in astrocytes. Keywords: astrocyte, calcium oscillation, metabotropic glutamate receptor, calcium indicator, cell culture
an important factor in information processing in the brain (14), it is important to unravel the mechanisms of calcium dynamics in astrocytes so as to understand brain function. Since only one type of intracellular Ca2+ store site has been recognized in astrocytes, some specific controlling mechanisms may be involved in promoting the typical relatively slow rise and fall in the Ca2+ concentration inside astrocytes seen in response to the metabotropic glutamate-receptor agonist (1S,3R)-1aminocyclopentane-1,3-dicarboxylic acid (t-ACPD) (15). It is possible that the Ca2+-dependent regulation of PLC activity and IP3 receptor sensitivity may be involved in the development of [Ca2+]i oscillations in these cells, as in other cells (16 – 19). In the present study, to determine whether [Ca2+]idependent processes are involved in inducing the oscillatory responses, we examined t-ACPD-induced changes in the [Ca2+]i in hippocampal astrocytes using three different indicators, quin-2, fura-2, and fura-2FF, which show the same Ca2+ concentration-dependent
Introduction Fluorometric techniques have revealed important features of glial [Ca2+]i responses, one of which, the finding of a neurotransmitter-induced oscillatory [Ca2+]i increase in astrocytes, has received most attention because of the functional roles of astrocytes in brain information processing (1 – 3). Most receptors expressed on astrocytes are coupled to Gq-protein, which, in turn, activates phospholipase C (PLC) to produce inositol 1,4,5-triphosphate (IP3), which stimulates its receptor expressed on the endoplasmic reticulum (ER) and induces Ca2+ release (4 – 8). Furthermore, recent physiological studies have shown that astrocytes release neurotransmitters, such as glutamate and ATP, in response to the oscillatory increase in the [Ca2+]i and signal to adjacent neurons (9 – 13). Since such cross-talk between astrocytes and neurons has been suggested to be *Corresponding author. FAX: +81-426-76-8841 E-mail: [email protected]
212
Ca2+ Oscillation in Astrocytes
fluorescence dynamics (the fluorescence peak at 520 nm induced by excitation at 340 nm increases and that induced by excitation at 380 nm decreases as the Ca2+ concentration increases), but have different dissociation constants (Kds) for Ca2+. The results showed that a critical Ca2+ level is required for induction of the second and subsequent Ca2+ changes following the initial peak of [Ca2+]i release from intracellular stores. Materials and Methods Cultured hippocampal astrocytes Animal experiments were performed in accordance with The Japanese Pharmacological Society “Guiding Principles for the Care and Use of Laboratory Animals” and were approved by the Animal Care Committee of Tokyo University of Pharmacy and Life Science. Hippocampal cells were cultured according to a previously described method (8, 20). Briefly, the hippocampus was dissected from the cerebrum of 18-day-old embryonic Wistar rats (purchased as a pregnant rat from Japan SLC, Inc., Shizuoka) and incubated for 20 min at 37°C with 0.25% trypsin (Difco & BD, Franklin Lakes, NJ, USA) and DNase I (0.02%, purchased from SigmaAldrich, Tokyo). The cells were then dissociated by repetitive pipetting and plated onto glass cover slips (Matsunami #1; Matsunami Glass, Ind., Ltd., Kishiwada) with a silicon rubber wall (Flexiperm; Heraeus, Hanau, Germany) at a density of 5 ´ 105 per well (15 mm in diameter). The cultures were maintained for one week in an incubator in Dulbecco’s modified Eagle’s medium (cat. No. 430 – 2100, containing no Lglutamate and L-aspartate; GIBCO BRL, Life Technologies, Inc., Rockville, MD, USA), 5% v / v precolostrum new born calf serum (Mitsubishi Chemical Co., Tokyo), and 5% v / v heat-inactivated horse serum (GIBCO BRL). No special effort was made to separate astrocytes from neuronal and other types of glial cells so as to retain the interaction between neurons and glial cells. Intracellular Ca2+ measurement We used three different Ca2+ indicators, fura-2, fura2 / FF, and quin-2, which have the same fluorescence properties (the fluorescence peak at 520 nm induced by excitation at 340 nm increases and that induced by excitation at 380 nm decreases as the Ca2+ concentration increases), but different Kds of 145, 25,000, and 60 nM, respectively. The cultured hippocampal cells were washed twice with fresh balanced salt solution (BSS: 130 mM NaCl, 5.4 mM KCl, 2.0 mM CaCl2, 5.5 mM glucose, and 20 mM HEPES-NaOH, pH 7.3), then were treated for 45 min at 36°C with 7.5 mM fura-2 acetoxymethyl ester
213
(fura-2 / AM), fura-2 FF acetoxymethyl ester (fura2FF / AM), or quin-2 acetoxymethyl ester (quin-2 / AM), prepared from a 1 mM stock solution in DMSO by dilution in BSS. After wash-out of the dye-containing medium, the cells were incubated for 60 min and then used for image analysis. For loading BAPTA / AM with fura-2FF, 7.5 m M of BAPTA / AM was added to fura-2FF / AM containing medium. The Ca2+ indicator-loaded cells were placed on the stage of an inverted fluorescence microscope (IMT-2; Olympus Co., Tokyo) equipped with a xenon arc lamp (75 W) and interference filters (340 ± 5 and 380 ± 5 nm) and perfused with BSS at a rate of 2 ml / min. Images of fluorescent Ca2+ indicator-loaded cells were obtained through a sharp cut filter (>480 nm) and detected and analyzed using a digital cooled CCD camera (HiSCA; Hamamatsu Photonics Co., Shizuoka). Fluorescence images of the cell bodies were taken as the region of interest and the intensity induced by a set of 340 and 380 nm excitation were obtained once every 1 s. Ratio images (F340 / F380) were obtained by dividing the value for the fluorescence induced by 340 nm excitation by that obtained using 380 nm excitation, with an interval between excitation processes of 0.45 s. Identification of cell types by immuno-cytochemistry According to the previous paper, we identified the cells responsive to t-ACPD as astrocytes using antiGFAP antibody (8). Briefly, at the beginning of Ca2+ measurement, we put small marks on the bottom side of the glass culture plate using a fine marker pen. We selected the proper area for Ca2+ measurement under the inverted-fluorescence microscope taking special care to involve one of the small marks within the microscopic image. When the Ca2+ imaging experiment was finished, the cells were fixed by 4% paraformaldehyde in PBS for 30 min. It was then permeabilized with 0.1% Trion X100 and subsequently rinsed (3 times, every 10 min) with PBS containing 0.2% BSA and 0.1 mM glycine. The primary immune antibody, anti-cow GFAP raised in rabbit (DAKO A / S, Glostrup, Denmark), applied in PBS containing 0.1% BSA for 1 h. Then the antibody was revealed with a rhodamine conjugated affinity purified secondary antibody for anti-rabbit IgG (Chemicon International, Inc., Temecula, CA, USA). Those fluorescent images at the same place as that subjected to Ca2+ imaging were searched for taking the marks as a guide and then photographs were taken. Comparing the fura-2 fluorometry expressed as pseudo-color imaging with the immunofluorescence images obtained by GFAP, tACPD-responsive cells were examined to determine if they were positive for the antibody.
214
Y Yoshida et al
Drugs used Trans-ACPD (t-ACPD) was obtained from Sigma Chemicals Co. (St. Louis, MO, USA). The Ca2+ indicators, quin-2 / AM (Kd of the free acid form 60 nM) and fura-2 / AM (Kd of the free acid form about 145 nM), were obtained from Dojindo Co. (Kumamoto). Fura2FF / AM (Kd of the free acid form 25,000 nM) was obtained from Tef Labs (Austin, TX, USA). BAPTA / AM, a Ca2+ chelating agent that can be loaded into cells (Kd of the free acid form 135 nM), was purchased from Molecular Probes Inc. (Eugene, Oregon, USA). The Kd values for quin-2, fura-2, and BAPTA are those cited in The Handbook of Molecular Probes Inc. (8th Edition) and that for fura-2FF is quoted from the Tef Labs product data. Results Characteristics of t-ACPD-induced [Ca2+]i responses in cultured hippocampal cells Fura-2-loaded hippocampal cells in culture were incubated with 0.1 mM t-ACPD, the concentration shown in our previous papers to be the most effective in producing oscillatory [Ca2+]i responses in astrocytes (8, 21), and the cells were classified into three types on the basis of the response (Table 1): 1) a transient increase in the [Ca2+]i (2.2 ± 1.1% of t-ACPD-responsive cells) (T-type) (Fig. 1A); 2) a transient increase in the [Ca2+]i, followed by a persistent increase in the [Ca2+]i (31.8 ± 4.9%) (P-type) (Fig. 1B); and 3) oscillatory changes in the [Ca2+]i (66.0 ± 5.4%) (O-type) (Fig. 1C). The percentage of cells showing oscillatory responses to t-ACPD varied from 55.4% to 72.5% of the t-ACPDresponsive cells in the observed field of each tested culture dish. We confirmed by immunostaining for glial fibrillary acidic protein (GFAP), an astrocyte marker, that the t-ACPD-responsive cells were astrocytes (data not shown here, refer our previous paper) (8). The [Ca2+]i was calculated from the maximum fluorescence ratio obtained by treating the cells with 10-6 M ionomycin and the minimum fluorescence ratio obtained by subsequent treatment with 1 mM EGTA, using a Kd for fura-2 of 145 nM (22). The [Ca2+]i at the peak of
Fig. 1. Three typical t-ACPD-induced responses in fura-2-loaded hippocampal cells in culture. A few cells showed only a transient [Ca2+]i increase (T-type) (A), while almost all t-ACPD responsive cells showed a sustained [Ca2+]i increase throughout drug administration and were classified as persistent or the “P-type” (B). Cells showing oscillatory [Ca2+]i changes during t-ACPD administration were called the “O-type” (C).
Table 1. Three typical patterns of [Ca2+]i response to t-ACPD in cells loaded with quin-2, fura-2, or fura-2 FF with different Kds for Ca2+ Agents Quin2 Fura-2 Fura-2FF
Kd (nM)
Responsive cells
O-type
P-type
T-type
60
170
0
0
100
145
239
66.0 ± 5.4
31.8 ± 4.9
2.2 ± 1.1
25,000
220
69.1 ± 11.2
5.0 ± 1.8
25.9 ± 11.0
Percentage of each pattern of response was calculated based on the t-ACPD-positive cells in four culture dishes.
Ca2+ Oscillation in Astrocytes
215
Fig. 2. Typical t-ACPD-induced oscillatory responses in cells loaded with two different calcium indicators with different Kds. Simultaneously recorded [Ca2+]i responses in four t-ACPD-responsive O-type cells in fura-2-loaded (A) or fura-2FF-loaded (B) cell cultures. Note that the scales for the 340 /380 ratio are different.
the oscillatory response was estimated as 186 ± 35 nM (n = 30) and that at the bottom of the first peak as 63 ± 24 nM (n = 30), while the resting [Ca2+]i was 56 ± 28 nM (n = 30). Although [Ca2+]i estimation by this method is not sufficiently accurate to allow discussion of its physiological meaning, the results showed that the estimated peak concentration of Ca2+ during the oscillation was very close to the binding constant of fura-2. During continuous t-ACPD treatment of fura-2loaded cells, the amplitudes of the [Ca2+]i oscillation peaks remained relatively constant for the first two or three peaks, then decreased, while the [Ca2+] level required to initiate subsequent peaks increased slightly (Figs. 1C and 2A). [Ca2+]i oscillation profiles in astrocytes detected using Ca2+ indicators with different Kds We then determined if the [Ca2+]i responses obtained are real differences or an artifact of the indicator used by using Ca2+ indicators other than fura-2 with different Kds. Representative t-ACPD-induced responses of cells loaded with fura-2FF, which has a lower affinity for Ca2+ (Kd = 25,000 nM), are shown in Fig. 2B. A similar oscillatory response to that in fura-2-loaded cells was seen, but the peak amplitude decreased much more rapidly than in fura-2-loaded cells and the minimum Ca2+ level during oscillation apparently remained constant. The difference could conceivably be ascribed to the higher sensitivity of fura-2 to Ca2+. However, if the sensitivity of fura-2 accounted for the detection of the small increase at the minimum of the oscillatory changes, the peak amplitudes should have been much larger than those observed in the present study (Figs. 1
and 2A). The results indicated that the peak [Ca2+]i responses in the fura-2 loaded cells might be under the restraints of the high Ca2+ affinity of fura-2. As shown in Table 1, in fura-2FF-loaded cells, the profile of the cellular response to t-ACPD (0.1 mM) was different from that in fura-2-loaded cells. The percentage of O-type cells was almost the same in both, but the percentage of T-type cells was higher in fura-2FFloaded cells and the percentage of P-type cells was lower. In contrast, in cells loaded with quin-2, which has a very low Kd of <60 nM, only cells with a very small transient increase in the [Ca2+]i were seen in the presence of 0.1 mM t-ACPD (Fig. 3A). Although we classified the response as T-type, its peak amplitude was much smaller than that seen in fura-2- or fura-2FF-loaded cells. No oscillatory or sustained responses were ever seen during prolonged administration of t-ACPD (Table 1). Effect of BAPTA, a Ca2+ chelating agent, on the oscillatory responses in fura-2FF-loaded cells We examined the effects of 0.1 mM t-ACPD on cells treated with both fura-2FF / AM and the Ca2+ chelating agent, BAPTA / AM (7.5 m M for 45 min), which has a high Ca2+ affinity (Kd = 135 nM) and binds Ca2+ rapidly. As shown in Fig. 3B, in these cells, t-ACPD caused only a small transient increase in the [Ca2+]i, the pattern of the increase being similar to that in quin-2-loaded cells. Discussion Recent studies have demonstrated that astrocytes can
216
Y Yoshida et al
Fig. 3. Typical t-ACPD-induced [Ca2+]i responses in cells loaded with quin-2, a high Ca2+ affinity indicator, or fura-2 FF, a lower Ca2+ affinity but co-loaded with BAPTA, a high affinity Ca2+ chelator. A: [Ca2+]i response in a cell loaded with quin2, an indicator with a very high affinity for Ca2+. B: The response in a cell loaded with fura-2FF, with a lower affinity for Ca2+, plus a Ca2+ chelator with a high Ca2+ affinity. The only response seen in both was a transient small increase in the [Ca2+]i.
no longer be classified as non-excitable cells (14, 23) and that they express many types of neurotransmitter receptors and can respond to neurotransmitters by a characteristic oscillatory increase in the [Ca2+]i (12). Since synapses in the central nervous system are covered with the lamella or filopodia of astrocytes (24, 25), the neurotransmitters released during synaptic activities will diffuse out from the synaptic cleft and stimulate neighboring glial cells, causing an oscillatory increase in the [Ca2+]i (14). As a recent study revealed that a physiological increase in the [Ca2+]i in astrocytes can stimulate glutamate release and thus modulate the activity of adjacent neurons (12), attention is now being paid to this characteristic increase in the [Ca2+]i of astrocytes as an important mechanism in information processing by the brain. It is therefore important to determine the mechanisms responsible for inducing such [Ca2+]i dynamics in astrocytes. The oscillatory [Ca2+]i changes induced through the activation of Gq-protein coupled receptors in many types of cells, such as smooth muscle and heart muscle cells and fertilized eggs, have been explained on the basis of two different intracellular calcium components, IP3-induced calcium release (IICR) and Ca2+-induced Ca2+ release (CICR) (26 – 29). However, since IICR has been shown to be a major Ca2+ store site in astrocytes, some specific controlling mechanisms may be involved in promoting the typical temporal pattern of a relatively slow rise and fall in the Ca2+ concentration inside astrocytes (15). Ca2+-dependent regulation of PLC activity and IP3 receptor sensitivity may be involved in [Ca2+]i oscillation, as in other cells (16 – 19, 30, 31).
[Ca2+]i oscillation induced by t-ACPD in hippocampal astrocytes was observed in cells loaded with fura-2, which has a Kd of 145 nM, and also in cells loaded with fura-2FF, with a Kd of 25,000 nM, but not in cells loaded with quin-2, with a Kd estimated at about 60 nM. These findings indicated that some threshold level of the [Ca2+]i was required for induction of the oscillatory response. This Ca2+ requirement was confirmed by loading the cells with fura-2FF and BAPTA (Kd of 135 nM); in these conditions, only a single small peak occurred in the [Ca2+]i during prolonged t-ACPD administration. For discussing the influence of those chelators, we should take into account not only their affinity to Ca2+, but also the total amount of chelating agents loaded inside cells. We loaded every Ca2+-chelating agent as the acetoxymethylester form in a concentration of 7.5 m M for 45 min. This procedure may allow the infiltration of rather higher concentration of those Ca2+-chelating agents into the cells and thus show the effects as Ca2+ buffers. From these results, we can estimate the threshold Ca2+ level for the initiation of the initial large [Ca2+]i peak seen after administration of t-ACPD as greater than 135 nM and less than 145 nM. The Ca2+ binding properties of Ca2+ indicators estimated in artificial aqueous solution differ from those in the cellular environment, where they may bind to soluble protein, and their apparent Kds may be higher by a factor of two or three (32 – 34). On this basis, the Ca2+ threshold level for causing the initial large transient [Ca2+]i increase can be estimated as 270 – 435 nM. If the initial partial opening of the IP3 receptor-
Ca2+ Oscillation in Astrocytes
coupled Ca2+ channels can release this amount of Ca2+, the Ca2+ released can fully open the IP3 receptor-coupled Ca2+ channels (31) and the increased [Ca2+]i may activate PKC, which may, in turn, activate PLC to produce more IP3 and cause massive Ca2+ release from the ER (35). On the other hand, the [Ca2+]i regulates IP3-stimulated Ca2+ release from the ER by activating the IP3 receptor both directly and indirectly (17, 31, 36, 37). Recently, Nakamura et al. (38, 39) reported that the neuronal IP3 receptor requires preconditioning activation to cause Ca2+ release from its store site. If the [Ca2+]i falls to a level lower than that required for inactivation of the IP3 receptor channel, IP3-induced Ca2+ release will be again activated and thus induce an increase in the [Ca2+]i until it again reaches the inactivation level. The reactivation level may differ for different cell types; in some of them, the level is found to shift to higher level as shown in fura-2 loaded cells. In fura-2FF-loaded cells, each peak of [Ca2+]i during the oscillation was sharper than that obtained in fura-2loaded cells. Since due to the very low affinity of fura2FF, the property as a Ca2+ chelator of this agent may not be distorted by intracellular environmental factors, and thus the pattern of oscillation observed here may be close to the real response inside cells. The peak amplitude of the Ca2+ oscillation observed in fura-2FF loaded cells fell gradually during prolonged administration of t-ACPD. The result suggests at least two possibilities, one being that the Ca2+ level required for IP3-receptor inactivation decreases use-dependently, and the other that the opening time of the IP3-coupled Ca2+ channel may be constant and thus the Ca2+ released during channel activation is dependent on the size of the Ca2+ store in the ER. Our previous study suggested that the oscillatory response in astrocytes is dependent upon the size of the ER Ca2+ store (21), but the first possibility cannot be excluded. In the fura-2-loaded cells, we sometimes encountered P-type cells in which a transient increase in the [Ca2+]i was followed by a persistent increase during t-ACPD administration. Such cells were shown to be astrocytes by GFAP immuno-reactivity (8). In these cells, the IP3receptor inactivation process may be lacking or the level for inactivation may be much higher than in cells with oscillatory responses. In fura-2FF-loaded cells, the population of P-type cells was lower, and that of T-type cells higher, than in fura-2-loaded cells, but this may be due to the sensitivity of fura-2FF being too low to detect the lower persistent part of the [Ca2+]i increase. In conclusion, the release of Ca2+ from the ER in astrocytes is strictly controlled by Ca2+-dependent processes, which are involved both in the facilitation and suppression of release, and the oscillatory [Ca2+]i
217
changes in astrocytes seen during metabotropic glutamate-receptor activation are the overall results of the activation of these different processes. Acknowledgments This work is supported by Grant-in Aid #15082207 for Scientific Research on Priority Area on “Elucidation of Glia-Neuron Network Mediated Information Processing System”. References 1 Smith SJ. Do astrocytes process neuronal information? Prog Brain Res. 1992;94:119–136. 2 Nedergaard M. Direct signaling from astrocytes to neurons in cultures of mammalian brain cells. Science. 1994;263:1768– 1771. 3 Verkhratsky A, Kettenmann H. Calcium signalling in glial cells. Trends Neurosci. 1996;19:346–352. 4 Cornell-Bell AH, Finkbeiner SM, Cooper MS, Smith SJ. Glutamate induces calcium waves in cultured astrocytes: longrange glial signaling. Science. 1990;247:470–473. 5 Cornell-Bell AH, Finkbeiner SM. Ca2+ waves in astrocytes. Cell Calcium. 1991;12:185–204. 6 De Barry J, Ogura A, Kudo Y. Ca2+ mobilization in cultured rat cerebellar cells: astrocytes are activated by t-ACPD. Eur J Neurosci. 1991;3:1146–1154. 7 Mennerick S, Zorumski CF. Glial contributions to excitatory neurotransmission in cultured hippocampal cells. Nature. 1994; 368:59–62. 8 Yoshida Y, Matsumoto N, Tsuchiya R, Morita M, Miyakawa H, Kudo Y. Expression of group I metabotropic glutamate receptors in rat hippocampal cells in culture and their characterization by intracellular calcium ion dynamics. J Pharmacl Sci. 2003;92: 245–251. 9 Nakahara K, Okada M, Nakanishi S. The metabotropic glutamate receptor mGluR induces calcium oscillations in cultured astrocytes via protein kinase C. J Neurochem. 1997; 69:1467–1475. 10 Araque A, Parpura V, Sanzgiri RP, Haydon PG. Glutamatedependent astrocyte modulation of synaptic transmission between cultured hippocampal neurons. Eur J Neurosci. 1998;10:2129–2142. 11 Sanzgiri RP, Araque A, Haydon PG. Prostaglandin E2 stimulates glutamate receptor-dependent astrocyte neuromodulation in cultured hippocampal cells. J Neurobiol. 1999;41:221–229. 12 Parpura V, Haydon PG. Physiological astrocytic calcium levels stimulate glutamate release to modulate adjacent neurons. Proc Natl Acad Sci USA. 2000;97:8629–8634. 13 Mazzanti M, Sul JY, Haydon PG. Glutamate on demand: astrocytes as a ready source. Neuroscientist. 2001;7:396–405. 14 Haydon PG. Glia: listening and talking to the synapse. Nat Rev Neurosci. 2001;2:185–193. 15 Charles AC, Dirksen ER, Merrill JE, Saderson MJ. Mechanisms of intracellular calcium signaling in glial cells studied with dantrolene and thapsigarigin. Glia. 1993;7:135–145. 16 Harootunian AT, Kao JP, Paranjape S, Tsien RY. Generation of
218
17 18
19
20
21
22
23 24
25
26
27
Y Yoshida et al calcium oscillation in fibroblasts by positive feedback between calcium and IP3. Science. 1991;251:75–78. Keizer J, De Young GW. Two roles of Ca2+ in agonist stimulated Ca2+ oscillations. Biophys J. 1992;61:649–660. Hansen M, Boitano S, Dirksen ER, Sanderson MJ. A role for phospholipase C activity but not ryanodine receptors in the initiation and propagation of intercellular calcium waves. J Cell Sci. 1995;108:2583–2590. Thomas AP, Bird GS, Hajnoczky G, Robb-Gaspers LD, Putney JW Jr. Spatial and temporal aspects of cellular calcium signaling. FASEB J. 1996;10:1505–1517. Hirasawa T, Nakamura T, Morita M, Ezawa I, Miyakawa H, Kudo Y. Activation of dihydropyridine sensitive Ca2+ channels in rat hippocampal neurons in culture by parathyroid hormone. Neurosci Lett. 1998;256:139–142. Morita M, Higuchi C, Moto T, Kozuka N, Susuki J, Itofusa R, et al. Dual regulation of calcium oscillation in astrocytes by growth factors and pro-inflammatory cytokines via the mitogenactivated protein kinase cascade. J Neurosci. 2003;23:10944– 10952. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440–3450. Vernadakis A. Glia-neuron intercommunications and synaptic plasticity. Prog Neurobiol. 1996;49:185–214. Hama K, Arii T, Kosaka T. Three dimensional organization of neuronal and glial processes: high voltage electronmicroscopy. Micros Res Tech. 1994;29:357–367. Deouiche A, Fortscher M. Peripheral astrocyte processes: monitoring by selective immunostaining for actin-binding ERM proteins. Glia. 2001;36:330–341. Miyazaki S, Shirakawa H, Nakada K, Honda Y. Essential role of the inositol 1,4,5-trisphosphate receptor /Ca2+ release channel in Ca2+ waves and Ca2+ oscillations at fertilization of mammalian eggs. Dev Biol. 1993;158:62–78. Albrecht MA, Colegrove SL, Hongpaisan J, Pivovarova NB, Andrew SB, Friel DD. Multiple modes of calcium-induced calcium release in sympathetic neurons I: attenuation of endo-
28 29 30
31
32
33
34
35 36
37
38
39
plasmic reticulum Ca2+ accumulation at low [Ca2+]i during weak depolarization. J Gen Physiol. 2001;118:83–100. Berridge MJ. The endoplasmic reticulum: a multifunctional signaling organelle. Cell Calcium. 2002;32:235–249. Verkhratsky A. The endoplasmic reticulum and neuronal calcium signaling. Cell Calcium. 2002;32:393–404. Cockcroft S, Thomas GMH. Inositol-lipid-specific phospholipase C isoenzymes and their differential regulation by receptors. Biochem J. 1992;288:1–4. Laurent M, Claret M. Signal-induced Ca2+ oscillations through the regulation of the inositol 1,4,5-trisphosphate-gated Ca2+ channel: an allosteric model. J Theor Biol. 1997;186:307–326. Hove-Madsen L, Bers DM. Indo-1 binding to protein in permeabilized ventricular myocytes alters its spectral and Ca binding properties. Biophys J. 1992;63:89–97. Harkins AB, Kurebayashi N, Baylor SM. Resting myoplasmic free calcium in frog skeletal muscle fibers estimated with fluo-3. Biophys J. 1993;65:865–881. Bassani JW, Bassani RA, Bers DM. Calibration of indo-1 and resting intracellular [Ca]i in intact rabbit cardiac myocytes. Biophys J. 1995;86:1453–1460. Meyer T, Stryer L. Molecular model for receptor-stimulated calcium spiking. Proc Natl Acad Sci USA. 1988;85:5051–5055. Finch EA, Turner TJ, Goldin SM. Calcium as a coagonist of inositol 1,4,5-trisphosphate-induced calcium release. Science. 1991;252:443–446. Keizer J, Li YX, Stojikovic S, Rinzel J. InsIp3-induced Ca2+ excitability of endoplasmic reticulum. Mol Biol Cell. 1995; 6:945–951. Nakamura T, Barabra JG, Nakamura K, Ross WN. Synergistic release of Ca2+ from IP3-sensitive stores evoked by synaptic activation of mGluRs paired backpropagating action potentials. Neuron. 1999;24:727–737. Nakamura T, Nakamura K, Lasser-Ross N, Barbara JG, Sandler VM, Ross WN. Inositol 1,4,5-triphosphate (IP3)-mediated Ca2+ release evoked by metabotropic agonists and backpropagating action potentials in hippocampal CA1 pyramidal neurons. J Neurosci. 2000;20:8365–8376.
J Pharmacol Sci 97, 212 – 218 (2005)
©2005 The Japanese Pharmacological Society
Full Paper
Ca2+-Dependent Induction of Intracellular Ca2+ Oscillation in Hippocampal Astrocytes During Metabotropic Glutamate Receptor Activation Yoshitoku Yoshida1, Remi Tsuchiya1, Nagisa Matsumoto1, Mitsuhiro Morita1, Hiroyoshi Miyakawa1, and Yoshihisa Kudo1,* 1
Molecular Life Science Division, School of Life Science, Tokyo University of Pharmacy and Life Science, 1432-1 Horinouchi Hachioji, Tokyo 192-0392, Japan
Received October 5, 2004; Accepted December 3, 2004
Abstract. We have investigated whether the intracellular calcium concentration ([Ca2+]i) oscillations induced in astrocytes using the metabotropic glutamate-receptor agonist, (1S,3R)-1aminocyclopentane-1,3-dicarboxylic acid (t-ACPD) are Ca2+-dependent, using three different Ca2+ indicators with different affinities for Ca2+. When rat hippocampal cells in culture were loaded with fura-2 (Kd: 145 nM), two-thirds of the cells showed obvious oscillatory increase in [Ca2+]i during t-ACPD-administration. Those cells were identified as astrocytes by immunohistochemistry in our previous paper. In cells loaded with fura-2FF (Kd: 25,000 nM), a similar percentage of t-ACPD-responsive cells showed oscillatory [Ca2+]i changes. However, in cells loaded with quin-2 (Kd: 60 nM), t-ACPD induced no oscillatory responses, but some cells showed a small transient increase in the [Ca2+]i. The same small transient [Ca2+]i increase was seen in cells loaded with both fura-2FF and BAPTA, a Ca2+ chelator (Kd: 135 nM). These findings indicate the involvement of [Ca2+]i-dependent regulatory mechanisms in the induction of the t-ACPD-induced oscillatory change in the [Ca2+]i in astrocytes. Keywords: astrocyte, calcium oscillation, metabotropic glutamate receptor, calcium indicator, cell culture
an important factor in information processing in the brain (14), it is important to unravel the mechanisms of calcium dynamics in astrocytes so as to understand brain function. Since only one type of intracellular Ca2+ store site has been recognized in astrocytes, some specific controlling mechanisms may be involved in promoting the typical relatively slow rise and fall in the Ca2+ concentration inside astrocytes seen in response to the metabotropic glutamate-receptor agonist (1S,3R)-1aminocyclopentane-1,3-dicarboxylic acid (t-ACPD) (15). It is possible that the Ca2+-dependent regulation of PLC activity and IP3 receptor sensitivity may be involved in the development of [Ca2+]i oscillations in these cells, as in other cells (16 – 19). In the present study, to determine whether [Ca2+]idependent processes are involved in inducing the oscillatory responses, we examined t-ACPD-induced changes in the [Ca2+]i in hippocampal astrocytes using three different indicators, quin-2, fura-2, and fura-2FF, which show the same Ca2+ concentration-dependent
Introduction Fluorometric techniques have revealed important features of glial [Ca2+]i responses, one of which, the finding of a neurotransmitter-induced oscillatory [Ca2+]i increase in astrocytes, has received most attention because of the functional roles of astrocytes in brain information processing (1 – 3). Most receptors expressed on astrocytes are coupled to Gq-protein, which, in turn, activates phospholipase C (PLC) to produce inositol 1,4,5-triphosphate (IP3), which stimulates its receptor expressed on the endoplasmic reticulum (ER) and induces Ca2+ release (4 – 8). Furthermore, recent physiological studies have shown that astrocytes release neurotransmitters, such as glutamate and ATP, in response to the oscillatory increase in the [Ca2+]i and signal to adjacent neurons (9 – 13). Since such cross-talk between astrocytes and neurons has been suggested to be *Corresponding author. FAX: +81-426-76-8841 E-mail: [email protected]
212
Ca2+ Oscillation in Astrocytes
fluorescence dynamics (the fluorescence peak at 520 nm induced by excitation at 340 nm increases and that induced by excitation at 380 nm decreases as the Ca2+ concentration increases), but have different dissociation constants (Kds) for Ca2+. The results showed that a critical Ca2+ level is required for induction of the second and subsequent Ca2+ changes following the initial peak of [Ca2+]i release from intracellular stores. Materials and Methods Cultured hippocampal astrocytes Animal experiments were performed in accordance with The Japanese Pharmacological Society “Guiding Principles for the Care and Use of Laboratory Animals” and were approved by the Animal Care Committee of Tokyo University of Pharmacy and Life Science. Hippocampal cells were cultured according to a previously described method (8, 20). Briefly, the hippocampus was dissected from the cerebrum of 18-day-old embryonic Wistar rats (purchased as a pregnant rat from Japan SLC, Inc., Shizuoka) and incubated for 20 min at 37°C with 0.25% trypsin (Difco & BD, Franklin Lakes, NJ, USA) and DNase I (0.02%, purchased from SigmaAldrich, Tokyo). The cells were then dissociated by repetitive pipetting and plated onto glass cover slips (Matsunami #1; Matsunami Glass, Ind., Ltd., Kishiwada) with a silicon rubber wall (Flexiperm; Heraeus, Hanau, Germany) at a density of 5 ´ 105 per well (15 mm in diameter). The cultures were maintained for one week in an incubator in Dulbecco’s modified Eagle’s medium (cat. No. 430 – 2100, containing no Lglutamate and L-aspartate; GIBCO BRL, Life Technologies, Inc., Rockville, MD, USA), 5% v / v precolostrum new born calf serum (Mitsubishi Chemical Co., Tokyo), and 5% v / v heat-inactivated horse serum (GIBCO BRL). No special effort was made to separate astrocytes from neuronal and other types of glial cells so as to retain the interaction between neurons and glial cells. Intracellular Ca2+ measurement We used three different Ca2+ indicators, fura-2, fura2 / FF, and quin-2, which have the same fluorescence properties (the fluorescence peak at 520 nm induced by excitation at 340 nm increases and that induced by excitation at 380 nm decreases as the Ca2+ concentration increases), but different Kds of 145, 25,000, and 60 nM, respectively. The cultured hippocampal cells were washed twice with fresh balanced salt solution (BSS: 130 mM NaCl, 5.4 mM KCl, 2.0 mM CaCl2, 5.5 mM glucose, and 20 mM HEPES-NaOH, pH 7.3), then were treated for 45 min at 36°C with 7.5 mM fura-2 acetoxymethyl ester
213
(fura-2 / AM), fura-2 FF acetoxymethyl ester (fura2FF / AM), or quin-2 acetoxymethyl ester (quin-2 / AM), prepared from a 1 mM stock solution in DMSO by dilution in BSS. After wash-out of the dye-containing medium, the cells were incubated for 60 min and then used for image analysis. For loading BAPTA / AM with fura-2FF, 7.5 m M of BAPTA / AM was added to fura-2FF / AM containing medium. The Ca2+ indicator-loaded cells were placed on the stage of an inverted fluorescence microscope (IMT-2; Olympus Co., Tokyo) equipped with a xenon arc lamp (75 W) and interference filters (340 ± 5 and 380 ± 5 nm) and perfused with BSS at a rate of 2 ml / min. Images of fluorescent Ca2+ indicator-loaded cells were obtained through a sharp cut filter (>480 nm) and detected and analyzed using a digital cooled CCD camera (HiSCA; Hamamatsu Photonics Co., Shizuoka). Fluorescence images of the cell bodies were taken as the region of interest and the intensity induced by a set of 340 and 380 nm excitation were obtained once every 1 s. Ratio images (F340 / F380) were obtained by dividing the value for the fluorescence induced by 340 nm excitation by that obtained using 380 nm excitation, with an interval between excitation processes of 0.45 s. Identification of cell types by immuno-cytochemistry According to the previous paper, we identified the cells responsive to t-ACPD as astrocytes using antiGFAP antibody (8). Briefly, at the beginning of Ca2+ measurement, we put small marks on the bottom side of the glass culture plate using a fine marker pen. We selected the proper area for Ca2+ measurement under the inverted-fluorescence microscope taking special care to involve one of the small marks within the microscopic image. When the Ca2+ imaging experiment was finished, the cells were fixed by 4% paraformaldehyde in PBS for 30 min. It was then permeabilized with 0.1% Trion X100 and subsequently rinsed (3 times, every 10 min) with PBS containing 0.2% BSA and 0.1 mM glycine. The primary immune antibody, anti-cow GFAP raised in rabbit (DAKO A / S, Glostrup, Denmark), applied in PBS containing 0.1% BSA for 1 h. Then the antibody was revealed with a rhodamine conjugated affinity purified secondary antibody for anti-rabbit IgG (Chemicon International, Inc., Temecula, CA, USA). Those fluorescent images at the same place as that subjected to Ca2+ imaging were searched for taking the marks as a guide and then photographs were taken. Comparing the fura-2 fluorometry expressed as pseudo-color imaging with the immunofluorescence images obtained by GFAP, tACPD-responsive cells were examined to determine if they were positive for the antibody.
214
Y Yoshida et al
Drugs used Trans-ACPD (t-ACPD) was obtained from Sigma Chemicals Co. (St. Louis, MO, USA). The Ca2+ indicators, quin-2 / AM (Kd of the free acid form 60 nM) and fura-2 / AM (Kd of the free acid form about 145 nM), were obtained from Dojindo Co. (Kumamoto). Fura2FF / AM (Kd of the free acid form 25,000 nM) was obtained from Tef Labs (Austin, TX, USA). BAPTA / AM, a Ca2+ chelating agent that can be loaded into cells (Kd of the free acid form 135 nM), was purchased from Molecular Probes Inc. (Eugene, Oregon, USA). The Kd values for quin-2, fura-2, and BAPTA are those cited in The Handbook of Molecular Probes Inc. (8th Edition) and that for fura-2FF is quoted from the Tef Labs product data. Results Characteristics of t-ACPD-induced [Ca2+]i responses in cultured hippocampal cells Fura-2-loaded hippocampal cells in culture were incubated with 0.1 mM t-ACPD, the concentration shown in our previous papers to be the most effective in producing oscillatory [Ca2+]i responses in astrocytes (8, 21), and the cells were classified into three types on the basis of the response (Table 1): 1) a transient increase in the [Ca2+]i (2.2 ± 1.1% of t-ACPD-responsive cells) (T-type) (Fig. 1A); 2) a transient increase in the [Ca2+]i, followed by a persistent increase in the [Ca2+]i (31.8 ± 4.9%) (P-type) (Fig. 1B); and 3) oscillatory changes in the [Ca2+]i (66.0 ± 5.4%) (O-type) (Fig. 1C). The percentage of cells showing oscillatory responses to t-ACPD varied from 55.4% to 72.5% of the t-ACPDresponsive cells in the observed field of each tested culture dish. We confirmed by immunostaining for glial fibrillary acidic protein (GFAP), an astrocyte marker, that the t-ACPD-responsive cells were astrocytes (data not shown here, refer our previous paper) (8). The [Ca2+]i was calculated from the maximum fluorescence ratio obtained by treating the cells with 10-6 M ionomycin and the minimum fluorescence ratio obtained by subsequent treatment with 1 mM EGTA, using a Kd for fura-2 of 145 nM (22). The [Ca2+]i at the peak of
Fig. 1. Three typical t-ACPD-induced responses in fura-2-loaded hippocampal cells in culture. A few cells showed only a transient [Ca2+]i increase (T-type) (A), while almost all t-ACPD responsive cells showed a sustained [Ca2+]i increase throughout drug administration and were classified as persistent or the “P-type” (B). Cells showing oscillatory [Ca2+]i changes during t-ACPD administration were called the “O-type” (C).
Table 1. Three typical patterns of [Ca2+]i response to t-ACPD in cells loaded with quin-2, fura-2, or fura-2 FF with different Kds for Ca2+ Agents Quin2 Fura-2 Fura-2FF
Kd (nM)
Responsive cells
O-type
P-type
T-type
60
170
0
0
100
145
239
66.0 ± 5.4
31.8 ± 4.9
2.2 ± 1.1
25,000
220
69.1 ± 11.2
5.0 ± 1.8
25.9 ± 11.0
Percentage of each pattern of response was calculated based on the t-ACPD-positive cells in four culture dishes.
Ca2+ Oscillation in Astrocytes
215
Fig. 2. Typical t-ACPD-induced oscillatory responses in cells loaded with two different calcium indicators with different Kds. Simultaneously recorded [Ca2+]i responses in four t-ACPD-responsive O-type cells in fura-2-loaded (A) or fura-2FF-loaded (B) cell cultures. Note that the scales for the 340 /380 ratio are different.
the oscillatory response was estimated as 186 ± 35 nM (n = 30) and that at the bottom of the first peak as 63 ± 24 nM (n = 30), while the resting [Ca2+]i was 56 ± 28 nM (n = 30). Although [Ca2+]i estimation by this method is not sufficiently accurate to allow discussion of its physiological meaning, the results showed that the estimated peak concentration of Ca2+ during the oscillation was very close to the binding constant of fura-2. During continuous t-ACPD treatment of fura-2loaded cells, the amplitudes of the [Ca2+]i oscillation peaks remained relatively constant for the first two or three peaks, then decreased, while the [Ca2+] level required to initiate subsequent peaks increased slightly (Figs. 1C and 2A). [Ca2+]i oscillation profiles in astrocytes detected using Ca2+ indicators with different Kds We then determined if the [Ca2+]i responses obtained are real differences or an artifact of the indicator used by using Ca2+ indicators other than fura-2 with different Kds. Representative t-ACPD-induced responses of cells loaded with fura-2FF, which has a lower affinity for Ca2+ (Kd = 25,000 nM), are shown in Fig. 2B. A similar oscillatory response to that in fura-2-loaded cells was seen, but the peak amplitude decreased much more rapidly than in fura-2-loaded cells and the minimum Ca2+ level during oscillation apparently remained constant. The difference could conceivably be ascribed to the higher sensitivity of fura-2 to Ca2+. However, if the sensitivity of fura-2 accounted for the detection of the small increase at the minimum of the oscillatory changes, the peak amplitudes should have been much larger than those observed in the present study (Figs. 1
and 2A). The results indicated that the peak [Ca2+]i responses in the fura-2 loaded cells might be under the restraints of the high Ca2+ affinity of fura-2. As shown in Table 1, in fura-2FF-loaded cells, the profile of the cellular response to t-ACPD (0.1 mM) was different from that in fura-2-loaded cells. The percentage of O-type cells was almost the same in both, but the percentage of T-type cells was higher in fura-2FFloaded cells and the percentage of P-type cells was lower. In contrast, in cells loaded with quin-2, which has a very low Kd of <60 nM, only cells with a very small transient increase in the [Ca2+]i were seen in the presence of 0.1 mM t-ACPD (Fig. 3A). Although we classified the response as T-type, its peak amplitude was much smaller than that seen in fura-2- or fura-2FF-loaded cells. No oscillatory or sustained responses were ever seen during prolonged administration of t-ACPD (Table 1). Effect of BAPTA, a Ca2+ chelating agent, on the oscillatory responses in fura-2FF-loaded cells We examined the effects of 0.1 mM t-ACPD on cells treated with both fura-2FF / AM and the Ca2+ chelating agent, BAPTA / AM (7.5 m M for 45 min), which has a high Ca2+ affinity (Kd = 135 nM) and binds Ca2+ rapidly. As shown in Fig. 3B, in these cells, t-ACPD caused only a small transient increase in the [Ca2+]i, the pattern of the increase being similar to that in quin-2-loaded cells. Discussion Recent studies have demonstrated that astrocytes can
216
Y Yoshida et al
Fig. 3. Typical t-ACPD-induced [Ca2+]i responses in cells loaded with quin-2, a high Ca2+ affinity indicator, or fura-2 FF, a lower Ca2+ affinity but co-loaded with BAPTA, a high affinity Ca2+ chelator. A: [Ca2+]i response in a cell loaded with quin2, an indicator with a very high affinity for Ca2+. B: The response in a cell loaded with fura-2FF, with a lower affinity for Ca2+, plus a Ca2+ chelator with a high Ca2+ affinity. The only response seen in both was a transient small increase in the [Ca2+]i.
no longer be classified as non-excitable cells (14, 23) and that they express many types of neurotransmitter receptors and can respond to neurotransmitters by a characteristic oscillatory increase in the [Ca2+]i (12). Since synapses in the central nervous system are covered with the lamella or filopodia of astrocytes (24, 25), the neurotransmitters released during synaptic activities will diffuse out from the synaptic cleft and stimulate neighboring glial cells, causing an oscillatory increase in the [Ca2+]i (14). As a recent study revealed that a physiological increase in the [Ca2+]i in astrocytes can stimulate glutamate release and thus modulate the activity of adjacent neurons (12), attention is now being paid to this characteristic increase in the [Ca2+]i of astrocytes as an important mechanism in information processing by the brain. It is therefore important to determine the mechanisms responsible for inducing such [Ca2+]i dynamics in astrocytes. The oscillatory [Ca2+]i changes induced through the activation of Gq-protein coupled receptors in many types of cells, such as smooth muscle and heart muscle cells and fertilized eggs, have been explained on the basis of two different intracellular calcium components, IP3-induced calcium release (IICR) and Ca2+-induced Ca2+ release (CICR) (26 – 29). However, since IICR has been shown to be a major Ca2+ store site in astrocytes, some specific controlling mechanisms may be involved in promoting the typical temporal pattern of a relatively slow rise and fall in the Ca2+ concentration inside astrocytes (15). Ca2+-dependent regulation of PLC activity and IP3 receptor sensitivity may be involved in [Ca2+]i oscillation, as in other cells (16 – 19, 30, 31).
[Ca2+]i oscillation induced by t-ACPD in hippocampal astrocytes was observed in cells loaded with fura-2, which has a Kd of 145 nM, and also in cells loaded with fura-2FF, with a Kd of 25,000 nM, but not in cells loaded with quin-2, with a Kd estimated at about 60 nM. These findings indicated that some threshold level of the [Ca2+]i was required for induction of the oscillatory response. This Ca2+ requirement was confirmed by loading the cells with fura-2FF and BAPTA (Kd of 135 nM); in these conditions, only a single small peak occurred in the [Ca2+]i during prolonged t-ACPD administration. For discussing the influence of those chelators, we should take into account not only their affinity to Ca2+, but also the total amount of chelating agents loaded inside cells. We loaded every Ca2+-chelating agent as the acetoxymethylester form in a concentration of 7.5 m M for 45 min. This procedure may allow the infiltration of rather higher concentration of those Ca2+-chelating agents into the cells and thus show the effects as Ca2+ buffers. From these results, we can estimate the threshold Ca2+ level for the initiation of the initial large [Ca2+]i peak seen after administration of t-ACPD as greater than 135 nM and less than 145 nM. The Ca2+ binding properties of Ca2+ indicators estimated in artificial aqueous solution differ from those in the cellular environment, where they may bind to soluble protein, and their apparent Kds may be higher by a factor of two or three (32 – 34). On this basis, the Ca2+ threshold level for causing the initial large transient [Ca2+]i increase can be estimated as 270 – 435 nM. If the initial partial opening of the IP3 receptor-
Ca2+ Oscillation in Astrocytes
coupled Ca2+ channels can release this amount of Ca2+, the Ca2+ released can fully open the IP3 receptor-coupled Ca2+ channels (31) and the increased [Ca2+]i may activate PKC, which may, in turn, activate PLC to produce more IP3 and cause massive Ca2+ release from the ER (35). On the other hand, the [Ca2+]i regulates IP3-stimulated Ca2+ release from the ER by activating the IP3 receptor both directly and indirectly (17, 31, 36, 37). Recently, Nakamura et al. (38, 39) reported that the neuronal IP3 receptor requires preconditioning activation to cause Ca2+ release from its store site. If the [Ca2+]i falls to a level lower than that required for inactivation of the IP3 receptor channel, IP3-induced Ca2+ release will be again activated and thus induce an increase in the [Ca2+]i until it again reaches the inactivation level. The reactivation level may differ for different cell types; in some of them, the level is found to shift to higher level as shown in fura-2 loaded cells. In fura-2FF-loaded cells, each peak of [Ca2+]i during the oscillation was sharper than that obtained in fura-2loaded cells. Since due to the very low affinity of fura2FF, the property as a Ca2+ chelator of this agent may not be distorted by intracellular environmental factors, and thus the pattern of oscillation observed here may be close to the real response inside cells. The peak amplitude of the Ca2+ oscillation observed in fura-2FF loaded cells fell gradually during prolonged administration of t-ACPD. The result suggests at least two possibilities, one being that the Ca2+ level required for IP3-receptor inactivation decreases use-dependently, and the other that the opening time of the IP3-coupled Ca2+ channel may be constant and thus the Ca2+ released during channel activation is dependent on the size of the Ca2+ store in the ER. Our previous study suggested that the oscillatory response in astrocytes is dependent upon the size of the ER Ca2+ store (21), but the first possibility cannot be excluded. In the fura-2-loaded cells, we sometimes encountered P-type cells in which a transient increase in the [Ca2+]i was followed by a persistent increase during t-ACPD administration. Such cells were shown to be astrocytes by GFAP immuno-reactivity (8). In these cells, the IP3receptor inactivation process may be lacking or the level for inactivation may be much higher than in cells with oscillatory responses. In fura-2FF-loaded cells, the population of P-type cells was lower, and that of T-type cells higher, than in fura-2-loaded cells, but this may be due to the sensitivity of fura-2FF being too low to detect the lower persistent part of the [Ca2+]i increase. In conclusion, the release of Ca2+ from the ER in astrocytes is strictly controlled by Ca2+-dependent processes, which are involved both in the facilitation and suppression of release, and the oscillatory [Ca2+]i
217
changes in astrocytes seen during metabotropic glutamate-receptor activation are the overall results of the activation of these different processes. Acknowledgments This work is supported by Grant-in Aid #15082207 for Scientific Research on Priority Area on “Elucidation of Glia-Neuron Network Mediated Information Processing System”. References 1 Smith SJ. Do astrocytes process neuronal information? Prog Brain Res. 1992;94:119–136. 2 Nedergaard M. Direct signaling from astrocytes to neurons in cultures of mammalian brain cells. Science. 1994;263:1768– 1771. 3 Verkhratsky A, Kettenmann H. Calcium signalling in glial cells. Trends Neurosci. 1996;19:346–352. 4 Cornell-Bell AH, Finkbeiner SM, Cooper MS, Smith SJ. Glutamate induces calcium waves in cultured astrocytes: longrange glial signaling. Science. 1990;247:470–473. 5 Cornell-Bell AH, Finkbeiner SM. Ca2+ waves in astrocytes. Cell Calcium. 1991;12:185–204. 6 De Barry J, Ogura A, Kudo Y. Ca2+ mobilization in cultured rat cerebellar cells: astrocytes are activated by t-ACPD. Eur J Neurosci. 1991;3:1146–1154. 7 Mennerick S, Zorumski CF. Glial contributions to excitatory neurotransmission in cultured hippocampal cells. Nature. 1994; 368:59–62. 8 Yoshida Y, Matsumoto N, Tsuchiya R, Morita M, Miyakawa H, Kudo Y. Expression of group I metabotropic glutamate receptors in rat hippocampal cells in culture and their characterization by intracellular calcium ion dynamics. J Pharmacl Sci. 2003;92: 245–251. 9 Nakahara K, Okada M, Nakanishi S. The metabotropic glutamate receptor mGluR induces calcium oscillations in cultured astrocytes via protein kinase C. J Neurochem. 1997; 69:1467–1475. 10 Araque A, Parpura V, Sanzgiri RP, Haydon PG. Glutamatedependent astrocyte modulation of synaptic transmission between cultured hippocampal neurons. Eur J Neurosci. 1998;10:2129–2142. 11 Sanzgiri RP, Araque A, Haydon PG. Prostaglandin E2 stimulates glutamate receptor-dependent astrocyte neuromodulation in cultured hippocampal cells. J Neurobiol. 1999;41:221–229. 12 Parpura V, Haydon PG. Physiological astrocytic calcium levels stimulate glutamate release to modulate adjacent neurons. Proc Natl Acad Sci USA. 2000;97:8629–8634. 13 Mazzanti M, Sul JY, Haydon PG. Glutamate on demand: astrocytes as a ready source. Neuroscientist. 2001;7:396–405. 14 Haydon PG. Glia: listening and talking to the synapse. Nat Rev Neurosci. 2001;2:185–193. 15 Charles AC, Dirksen ER, Merrill JE, Saderson MJ. Mechanisms of intracellular calcium signaling in glial cells studied with dantrolene and thapsigarigin. Glia. 1993;7:135–145. 16 Harootunian AT, Kao JP, Paranjape S, Tsien RY. Generation of
218
17 18
19
20
21
22
23 24
25
26
27
Y Yoshida et al calcium oscillation in fibroblasts by positive feedback between calcium and IP3. Science. 1991;251:75–78. Keizer J, De Young GW. Two roles of Ca2+ in agonist stimulated Ca2+ oscillations. Biophys J. 1992;61:649–660. Hansen M, Boitano S, Dirksen ER, Sanderson MJ. A role for phospholipase C activity but not ryanodine receptors in the initiation and propagation of intercellular calcium waves. J Cell Sci. 1995;108:2583–2590. Thomas AP, Bird GS, Hajnoczky G, Robb-Gaspers LD, Putney JW Jr. Spatial and temporal aspects of cellular calcium signaling. FASEB J. 1996;10:1505–1517. Hirasawa T, Nakamura T, Morita M, Ezawa I, Miyakawa H, Kudo Y. Activation of dihydropyridine sensitive Ca2+ channels in rat hippocampal neurons in culture by parathyroid hormone. Neurosci Lett. 1998;256:139–142. Morita M, Higuchi C, Moto T, Kozuka N, Susuki J, Itofusa R, et al. Dual regulation of calcium oscillation in astrocytes by growth factors and pro-inflammatory cytokines via the mitogenactivated protein kinase cascade. J Neurosci. 2003;23:10944– 10952. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440–3450. Vernadakis A. Glia-neuron intercommunications and synaptic plasticity. Prog Neurobiol. 1996;49:185–214. Hama K, Arii T, Kosaka T. Three dimensional organization of neuronal and glial processes: high voltage electronmicroscopy. Micros Res Tech. 1994;29:357–367. Deouiche A, Fortscher M. Peripheral astrocyte processes: monitoring by selective immunostaining for actin-binding ERM proteins. Glia. 2001;36:330–341. Miyazaki S, Shirakawa H, Nakada K, Honda Y. Essential role of the inositol 1,4,5-trisphosphate receptor /Ca2+ release channel in Ca2+ waves and Ca2+ oscillations at fertilization of mammalian eggs. Dev Biol. 1993;158:62–78. Albrecht MA, Colegrove SL, Hongpaisan J, Pivovarova NB, Andrew SB, Friel DD. Multiple modes of calcium-induced calcium release in sympathetic neurons I: attenuation of endo-
28 29 30
31
32
33
34
35 36
37
38
39
plasmic reticulum Ca2+ accumulation at low [Ca2+]i during weak depolarization. J Gen Physiol. 2001;118:83–100. Berridge MJ. The endoplasmic reticulum: a multifunctional signaling organelle. Cell Calcium. 2002;32:235–249. Verkhratsky A. The endoplasmic reticulum and neuronal calcium signaling. Cell Calcium. 2002;32:393–404. Cockcroft S, Thomas GMH. Inositol-lipid-specific phospholipase C isoenzymes and their differential regulation by receptors. Biochem J. 1992;288:1–4. Laurent M, Claret M. Signal-induced Ca2+ oscillations through the regulation of the inositol 1,4,5-trisphosphate-gated Ca2+ channel: an allosteric model. J Theor Biol. 1997;186:307–326. Hove-Madsen L, Bers DM. Indo-1 binding to protein in permeabilized ventricular myocytes alters its spectral and Ca binding properties. Biophys J. 1992;63:89–97. Harkins AB, Kurebayashi N, Baylor SM. Resting myoplasmic free calcium in frog skeletal muscle fibers estimated with fluo-3. Biophys J. 1993;65:865–881. Bassani JW, Bassani RA, Bers DM. Calibration of indo-1 and resting intracellular [Ca]i in intact rabbit cardiac myocytes. Biophys J. 1995;86:1453–1460. Meyer T, Stryer L. Molecular model for receptor-stimulated calcium spiking. Proc Natl Acad Sci USA. 1988;85:5051–5055. Finch EA, Turner TJ, Goldin SM. Calcium as a coagonist of inositol 1,4,5-trisphosphate-induced calcium release. Science. 1991;252:443–446. Keizer J, Li YX, Stojikovic S, Rinzel J. InsIp3-induced Ca2+ excitability of endoplasmic reticulum. Mol Biol Cell. 1995; 6:945–951. Nakamura T, Barabra JG, Nakamura K, Ross WN. Synergistic release of Ca2+ from IP3-sensitive stores evoked by synaptic activation of mGluRs paired backpropagating action potentials. Neuron. 1999;24:727–737. Nakamura T, Nakamura K, Lasser-Ross N, Barbara JG, Sandler VM, Ross WN. Inositol 1,4,5-triphosphate (IP3)-mediated Ca2+ release evoked by metabotropic agonists and backpropagating action potentials in hippocampal CA1 pyramidal neurons. J Neurosci. 2000;20:8365–8376.