Mapping of IP3-Mediated Ca2⫹ Signals in Single Human
Neuroblastoma SH-SY5Y Cells: Cell Volume Shaping the
Ca2⫹ Signal
K. VAN ACKER, B. BAUTMANS, G. BULTYNCK, K. MAES, A. F. WEIDEMA, P. DE SMET, J. B. PARYS,
H. DE SMEDT, L. MISSIAEN, AND G. CALLEWAERT
Laboratory of Physiology, Katholieke Universiteit Leuven, B-3000 Leuven, Belgium
cultured hippocampal neurons (Koizumi et al. 1999). The
cellular mechanisms accounting for the spatial restriction of
elementary Ca2⫹ release signals remain uncertain. Possible
mechanisms include inhomogeneous spatial distribution of
IP3Rs and the expression of different IP3R isoforms. Weak
stimulation thus would only activate the most dense or sensitive sites and produce an elementary Ca2⫹ release event. Stronger stimulation would excite all release sites, causing a global
Ca2⫹ signal (Bootman and Berridge 1995).
In the present study we have used fast laser-scanning microscopy to look for IP3-mediated Ca2⫹ release events in
human SH-SY5Y neuroblastoma cells, which express only
type I IP3R (Wojcikiewicz and Luo 1998). Stimulation of these
cells with the muscarinic agonist carbachol (CCh) causes a
transient rise in intracellular Ca2⫹ through IP3-mediated Ca2⫹
release (Lambert and Nahorski 1990; Murphy et al. 1991). Our
findings demonstrate that these cells can generate spatially and
temporally distinct Ca2⫹ signals. We further examined how
geometric features contribute to the spatiotemporal properties
of these Ca2⫹ signals.
INTRODUCTION
An increase in cytosolic Ca2⫹ concentration is a ubiquitous
signal that can trigger or regulate a diversity of cellular activities including muscular contraction, neurotransmitter release,
synaptic plasticity, cell proliferation, and cell apoptosis. In
many cell types, agonists that stimulate the production of the
second-messenger inositol 1,4,5-trisphosphate (IP3) increase
the global intracellular Ca2⫹ concentration by releasing Ca2⫹
from IP3-sensitive stores (reviewed in Berridge 1997). It has
been proposed that these global IP3-mediated Ca2⫹ signals
arise from the synchronized activation of localized clusters of
IP3Rs (Bootman and Berridge 1995) similar to those generated
by ryanodine receptor activation in muscle cells (Cheng et al.
1993). Local IP3-mediated Ca2⫹ release signals have now been
visualized and measured in a number of cells including Xenopus oocytes (Parker and Yao 1991; Sun et al. 1998; Yao et al.
1995), HeLa cells (Bootman et al. 1997), RBL cells (Horne and
Meyer 1997), astrocytes (Yagodin et al. 1994), PC12 cells
(Koizumi et al. 1999; Reber and Schindelholz 1996), vascular
endothelial cells (Huser and Blatter 1997), cerebellar Purkinje
neurons (Finch and Augustine 1998; Takechi et al. 1998), and
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1052
METHODS
Cell culture and solutions
Experiments were performed on single cultured SH-SY5Y neuroblastoma cells (European Collection of Cell Cultures, ECACC
94030304). Cells were cultured in DMEM-HamF12 medium (GIBCO
Laboratories), supplemented with 4 mM L-glutamine, 100 IU/ml
penicillin-streptomycin, 1% nonessential amino acids, and 15% fetal
calf serum at 37°C in a humidified atmosphere of 95% O2-5% CO2.
Cells were plated on gelatin-coated coverslips at a density of 104
cells/coverslip and used between 5 and 13 days after plating. For
experimentation cells were continuously perfused with a standard salt
solution containing (in mM) 150 NaCl, 6 KCl, 1.5 CaCl2, 1 MgCl2, 10
HEPES, and 10 glucose, pH 7.4. All experiments were performed at
room temperature (20 –22°C).
For rapid exchange of the extracellular solution or application of
agonist, a multibarreled pipette with a common outlet placed at a distance
of ⬃40 ␮m (Cell MicroControls, Virginia Beach, VA) from the cells
under study was used. The dead space-time of the system was ⬃4 s.
In some experiments a nominally Ca2⫹-free saline supplemented
with 2 mM EGTA was used. Ryanodine was purchased from Molecular Probes.
Confocal Ca2⫹ measurements
Neuroblastoma cells were loaded with the fluorescent Ca2⫹ indicator fluo-4 by incubating the cells in standard salt solution containing
0022-3077/00 $5.00 Copyright © 2000 The American Physiological Society
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on July 31, 2017
Van Acker, K., B. Bautmans, G. Bultynck, K. Maes, A. F. Weidema, P. De Smet, J. B. Parys, H. De Smedt, L. Missiaen, and G.
Callewaert. Mapping of IP3-mediated Ca2⫹ signals in single human
neuroblastoma SH-SY5Y cells: cell volume shaping the Ca2⫹ signal.
J. Neurophysiol. 83: 1052–1057, 2000. Fast confocal laser-scanning
microscopy was used to study spatiotemporal properties of IP3-mediated Ca2⫹ release signals in human SH-SY5Y neuroblastoma cells.
[Ca2⫹]i increases were not affected by ryanodine (30 ␮M) or caffeine
(10 mM) and largely insensitive to removal of external Ca2⫹, indicating predominance of IP3-induced Ca2⫹ release. Ca2⫹ signals
evoked by high concentration (10 ␮M) of the muscarinic agonist
carbachol appeared as self-propagating waves initiating in cell processes. At low carbachol concentrations (500 nM) Ca2⫹ changes in
most cells displayed striking spatiotemporal heterogeneity. The Ca2⫹
response in the cell body was delayed and had a smaller amplitude and
a slower rise time than that in processes. Ca2⫹ changes in processes
either occurred in a homogeneous manner throughout the whole
process or were sometimes confined to hot spots. Regional differences
in surface-to-volume ratio appear to be critical clues that determine
the spatiotemporal pattern of intracellular Ca2⫹ release signals.
CA2⫹ RELEASE SIGNALS IN SH-SY5Y CELLS
1053
RESULTS
Carbachol and caffeine responses in SH-SY5Y cells
When SH-SY5Y cells were exposed to 10 ␮M CCh for 5 s,
nearly all cells (92%) exhibited a large and transient change in
intracellular [Ca2⫹] (Fig. 1). To check whether the carbachol
response involves Ca2⫹ release from IP3 or ryanodine-sensitive
stores, SH-SY5Y cells were exposed to 10 mM caffeine. Figure 1 shows that in a small number of cells caffeine elicited a
fast and large Ca2⫹ transient, but most cells failed to respond
to caffeine. When those cells that failed to respond to caffeine
were subsequently exposed to 10 ␮M CCh, nearly all cells
responded, suggesting that their intracellular Ca2⫹ stores were
not depleted before the caffeine challenge. Caffeine-sensitive
cells generally also responded to CCh. For comparison, the
CCh- and caffeine-induced Ca2⫹ signal in Fig. 1A were obtained in the same cell. Pretreatment of cells with 30 ␮M
ryanodine had no effect on CCh-induced Ca2⫹ signals (data not
shown). Removal of extracellular Ca2⫹ (data not shown) had
no apparent effect on the amplitude of the CCh-induced Ca2⫹
transient but accelerated its relaxation. Subsequent reapplication of Ca2⫹ caused a secondary [Ca2⫹]i rise (data not shown)
indicating that the decay of the Ca2⫹ transient is controlled by
capacitive Ca2⫹ influx (Grudt et al. 1996) in concert with Ca2⫹
removal and reuptake mechanisms. The results thus indicate
that the CCh response is predominantly mediated by Ca2⫹
release from IP3-sensitive stores. The slower onset and decay
of the CCh response versus the caffeine response (Fig. 1A) can
be accounted for by the intermediate steps required in the
production and degradation of IP3.
As previously described (Larsson et al. 1998) individual
cells differed in amplitude, rate of rise, and latency time of
the Ca2⫹ response. Differences were not only detected
among cells but also within a single cell. In cells with
well-developed cellular processes, it was consistently observed that the CCh response initiated in the process and
2⫹
FIG. 1. Ca
signals evoked by 10 ␮M carbachol (CCh) or 10 mM
caffeine in SH-SY5Y cells. A: the Ca2⫹ response of a single SH-SY5Y cell
stimulated 1st with 10 mM caffeine (green trace) and then with 10 ␮M CCh
(red trace). Between the 2 applications there was a 5-min wash out period.
Solid bar indicates the time of drug application. B: 10 ␮M CCh caused a
Ca2⫹ response in nearly all SH-SY5Y cells. By contrast, only a small
population of CCh-sensitive cells responded to 10 mM caffeine. Data
collected from 120 cells. C: comparison of the time course of CCh-induced
Ca2⫹ changes in the cell body (green trace) and process (red trace). Contour
image shows the corresponding regions over which the relative changes in
fluorescence were integrated. The kinetics and amplitude of the increase in
[Ca2⫹]i are the same in both regions, but it is clear that the Ca2⫹ signal
initiates in the cell process. Solid bar indicates the time of application of 10
␮M CCh.
then took the form of a whole cell Ca2⫹ wave (Fig. 1C). In
cells without well-defined cell processes, the Ca2⫹ response
apparently started at the cell surface and then propagated
toward the center of the soma. With 10 ␮M CCh, the
amplitude, rate of rise, and decay of the Ca2⫹ response were
much alike in all parts of the cell. This clearly suggests that
the Ca2⫹ wave reflects regenerative Ca2⫹ release rather than
Ca2⫹ diffusion from an active region. The speed of wave
propagation amounted to 89 ⫾ 8 ␮m s⫺1 (mean ⫾ SE, n ⫽
13). But in many instances the cell size was too small to
accurately resolve the speed of propagation.
Ca2⫹ Responses at low agonist concentration
With 10 ␮M CCh the concentration of IP3 in the cell is likely
to be supramaximal for Ca2⫹ release at all sites (EC50 for CCh
is ⬃38 ␮M) (Larsson et al. 1998). With lower agonist concentrations the level of IP3 would only be sufficiently high at
production sites, and therefore it was expected that regional
differences in Ca2⫹ responses would be more pronounced.
With CCh concentrations ⬍1 ␮M, only a small proportion
(36%) of the cells responded. Below 250 nM CCh, cells failed
to respond.
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on July 31, 2017
10 ␮M of the acetoxymethyl (AM) ester of the dye (fluo-4 AM;
Molecular Probes) for 30 – 45 min at room temperature. After loading,
the cells were washed for at least 15 min with standard saline and
transferred to a perfusion chamber on an upright microscope (Optiphot-2, Nikon) equipped with a ⫻40 water-immersion objective
(N.A. 0.8, Zeiss). The microscope was attached to a confocal laserscanning system (Odyssey, Noran Instruments). Fluo-4 was excited
with the 488-nm line of an argon ion laser, and emitted fluorescence
was measured at wavelengths ⬎505 nm. Fluorescence images were
acquired at 25 Hz (50-Hz frame rate after off-line deinterlacing) and
digitized by a frame-grabber board connected to a personal computer
(Kinetic Imaging, Liverpool, UK). Fluorescence changes were analyzed off-line by choosing small regions of interest (ROI). Background-corrected fluorescence traces were calculated as %⌬F/F, i.e.,
fluorescence increase divided by average prestimulus fluorescence.
Contour images were reconstructed from the average of acquired
images.
Under the present culturing conditions, cells bodies and processes of neighboring cells often made numerous contacts. To
ascertain that Ca2⫹ signals in selected ROI originated from the
same cell, kinetics of Ca2⫹ signals in all neighboring cells were
carefully analyzed. The latency time of the CCh response varied
greatly among cells making this dissection feasible. Only those
cells, where the morphological overlay was well confined and the
differences in Ca2⫹ response with adjacent cells were obvious,
were further analyzed.
1054
VAN ACKER ET AL.
2⫹
With low agonist concentrations (500 nM) Ca2⫹ responses
fell into two main categories: whole cell Ca2⫹ waves and Ca2⫹
responses with a different spatial profile (Fig. 2). In the latter
group Ca2⫹ changes in cell processes appeared homogeneous
or were confined to hot spots.
2⫹
FIG. 3. Whole cell Ca
wave evoked by low concentrations of CCh. Top: pseudocolored Ca2⫹ images of a SH-SY5Y cell
exposed to 500 nM CCh at times indicated by 1– 4 in bottom panel. Line scan plot (top right) represents Ca2⫹ changes along the
red line (A-B-C) depicted in image 4. The calculated wave velocity amounted to 35 ␮m s⫺1. Changes in [Ca2⫹] are expressed as
%⌬F/F as indicated by the pseudocolor calibration bar 0 –300%. Bottom: time course of Ca2⫹ signals in 2 selected regions. Traces
were obtained by averaging %⌬F/F in a small region along the line scan as indicated by B (somatic region) and C (process region).
Solid bar indicates the time of CCh application. Ryanodine (30 ␮M) was present.
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on July 31, 2017
2. Different Ca2⫹ responses evoked by low concentrations of CCh.
Ca signals evoked by brief (5 s) pulses of low CCh concentrations (500 nM)
appeared either as a whole cell Ca2⫹ wave or was primarily restricted to cell
processes. In the latter, Ca2⫹ in the process was either homogeneously distributed or confined to hot spots. Data collected from 26 cells.
FIG.
A number of cells (19% of the responding cells) displayed a
whole cell Ca2⫹ wave with properties quite similar to Ca2⫹ waves
induced by 10 ␮M CCh; i.e., the wave initiated in the process,
propagated to the soma, but kinetics and amplitude were much
alike in all parts of the cell (Fig. 3). The speed of wave propagation was, however, much lower than with high agonist concentrations and amounted to 36 ⫾ 5 ␮m s⫺1 (n ⫽ 15).
In most responding cells (⬃80%), however, the Ca2⫹ profile
displayed marked regional differences. The essential properties
of these Ca2⫹ signals are illustrated in Figs. 4 and 5. Figure 4
shows a series of confocal images of a neuroblastoma cell
together with the time course of [Ca2⫹] changes at three
different cellular regions. Before stimulation, intracellular
[Ca2⫹] was uniformly distributed throughout the cell. When
the cell was stimulated with 500 nM CCh, a clear and fast Ca2⫹
transient was produced in the cell processes (positions A and
C). This Ca2⫹ transient reached a peak level after ⬃0.5 s and
then decayed to basal levels in different steps: after an initial
fast decay, [Ca2⫹] leveled off or even slightly increased and
finally slowly returned to resting levels. Despite CCh wash out
for ⬃12 s, a second Ca2⫹ transient initiated at that time in both
cellular processes. In the main cell body (position B), the
CCh-induced Ca2⫹ signal had a complete different temporal
profile. The increase in somatic [Ca2⫹] clearly showed two
humps: a slow rise coinciding with the fast relaxing phase of
the Ca2⫹ transient in the process was followed by a secondary
slow rise of [Ca2⫹]. The second hump coincided with the
[Ca2⫹] plateau in the processes. The decay of the somatic Ca2⫹
signal was apparently monophasic. So it is remarkable that
within a single cell CCh can evoke Ca2⫹ signals with a
completely different spatiotemporal domain.
CA2⫹ RELEASE SIGNALS IN SH-SY5Y CELLS
Figure 5 shows two other examples and emphasizes that
Ca2⫹ changes in the cell process either displayed a homogeneous spatial distribution or appeared in hot spots. Both cells
were stimulated with 500 nM CCh for 5 s. The surface plots
represent the Ca2⫹ response along the cell process up to the
central part of the cell body. As observed in most cells, the
CCh-induced Ca2⫹ increase in the process was not confined to
one particular region or spot but rather extended over the whole
distance of ⬃20 ␮m (plot A). The contour plot further indicates
a simultaneous Ca2⫹ release throughout this structure. At the
point where the cell process merged into the cell body, the
signal was markedly delayed, and its amplitude rapidly decreased with distance. The other cell in Fig. 5 (plot B) demonstrates that in some cells the CCh-induced Ca2⫹ change was
restricted to a small region in the cell process. These localized
events or hot spots dissipated over a very short distance. At
50% of Ca2⫹ decay the average dimensions amounted to 5.5 ⫾
0.7 ␮m (n ⫽ 5). In general, the time course of localized and
homogeneous Ca2⫹ transients in the process were quite similar. Hot spots were never observed in somatic regions. Finally,
the size of the CCh-induced Ca2⫹ transients in the process
showed marked variability irrespective of their spatial appearance. Figure 6 displays recordings of 17 cells exposed to low
[CCh] (1 ␮M, 800 or 500 nM). Peak amplitudes of Ca2⫹
transients in processes varied between 20 and 500% ⌬F/F. In
most cells a single clear Ca2⫹ peak appeared in the process. In
a number of cells (6 of 17), however, the Ca2⫹ transient in the
process was preceded by one or more smaller Ca2⫹ peaks. In
these cells the Ca2⫹ transient appeared along the entire process
simultaneously, and the smaller peaks had a similar spatial
distribution.
DISCUSSION
Spatiotemporal dynamics of Ca2⫹ signals were investigated
in single neuroblastoma SH-SY5Y cells by fast laser-scanning
confocal imaging.
IP3-induced Ca2⫹ release signals
As previously demonstrated (Larsson et al. 1998), stimulation of muscarinic receptors with CCh induced a rapid increase
in intracellular Ca2⫹. This Ca2⫹ increase mainly reflects Ca2⫹
2⫹
FIG. 5. Spatiotemporal properties of Ca
signals evoked by low concentrations of CCh:
homogeneous distribution vs. hot spot. A and B
illustrate 2 cells exposed to 500 nM CCh for
5 s. Surface and contour plots represent Ca2⫹
signals along a line encompassing the process
and the soma (indicated by a colored line in
contour image of the cell). A: the [Ca2⫹] rise
was seen simultaneously along the entire process. The Ca2⫹ signal in the soma was much
smaller and delayed. Solid bar indicates the
time of CCh application. B: the CCh-induced
Ca2⫹ rise, measured at the peak of the response, was confined to a small region or hot
spot in the cell process. Note different time
scale in A and B.
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on July 31, 2017
2⫹
FIG. 4. Spatiotemporal properties of Ca
signals evoked by low concentrations of CCh: cell body vs. cell processes. Top panel: confocal images of an
SH-SY5Y cell before and at the indicated times after stimulation with 500 nM
CCh. Changes in [Ca2⫹] are expressed as %⌬F/F as indicated by the pseudocolor calibration bar 0 –500%. The scale bar is 10 ␮m. Bottom panel: comparison of the time course of the Ca2⫹ signals at cell processes (red trace, ROI
A; blue trace, ROI C) and cell body (green trace, ROI B). Note the occurrence
of a secondary Ca2⫹ release event in cell processes. Solid bar indicates time of
CCh application.
1055
1056
VAN ACKER ET AL.
release from IP3-sensitive pools, because omitting extracellular
Ca2⫹ or pretreatment of the cells with ryanodine had no obvious effects on the rise in intracellular [Ca2⫹]. In addition,
almost all cells were sensitive to 10 ␮M CCh, but only a small
proportion to 10 mM caffeine. This may indicate that the
expression level of ryanodine receptors in neuroblastoma cells
is quite low and varies greatly from cell to cell. Alternatively,
the loading state of the caffeine-sensitive store may depend on
cell activity, whereas that of the IP3-sensitive store would
largely rely on a capacitive Ca2⫹ entry pathway.
Signals evoked by a high CCh concentration
In agreement with previous reports (Lambert and Nahorski
1990; Larsson et al. 1998; Murphy et al. 1991), 10 ␮M CCh
evoked a single large Ca2⫹ transient in essentially all cells.
Within an individual cell the profile of the Ca2⫹ transient was
much the same in all regions of the cell, but it was also
apparent that the signal initiated in cell processes before invading somatic regions. The wave speed was ⬃89 ␮m s⫺1,
which is close to the value reported for bradykinin-induced
Ca2⫹ waves in PC12 cells (Reber and Schindelholz 1996). The
uniform rate of rise and amplitude of the Ca2⫹ response suggests that the wave is self-propagating (Clapham and Sneyd
1995). Because the caffeine-sensitive store appears not to be
operative under our experimental conditions, we assume that
wave propagation entirely relies on IP3Rs. It is evident that in
cell processes, because of their small volume (or higher surface-to-volume ratio), [IP3] will raise much faster to threshold
levels, and thus the latency before IP3-mediated Ca2⫹ release
onset will be much shorter compared with the soma.
Signals evoked by low CCh concentrations
As expected, lower CCh concentrations (ⱕ1 ␮M) produced
more restricted Ca2⫹ signals. Two distinct patterns of Ca2⫹
responses were observed: whole cell Ca2⫹ waves and spatially
heterogeneous Ca2⫹ signals.
Whole cell Ca2⫹ waves had similar properties to those
induced with 10 ␮M CCh, with the exception that the wave
speed was markedly decreased (from 89 to 36 ␮m s⫺1). This
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on July 31, 2017
2⫹
FIG. 6. Different patterns of CCh-evoked Ca
signals in cell processes.
Examples of Ca2⫹ responses during application of low concentration of CCh
in cell processes of different cells. Different colors correspond to different
[CCh] (500 nM red traces, 800 nM green trace, and 1 ␮M blue traces). Arrows
indicate the occurrence of small events preceding a larger Ca2⫹ release signal.
finding suggests that under our experimental conditions diffusion of IP3 may limit the speed of wave propagation.
Most cells did not display a Ca2⫹ wave on stimulation with
a low concentration of CCh. Instead, the CCh-induced Ca2⫹
signal showed marked spatial variability.
[Ca2⫹] changes were most prominent in cell processes and
dissipated beyond the point where the cell process merged with
the soma. In the cell soma a slow component was sometimes
followed by a second, delayed component. The initial somatic
component coincided with the decay of the Ca2⫹ signal in the
process and thus mainly reflects diffusion of Ca2⫹ from the cell
process into the cell soma (Eilers et al. 1995). The second
delayed somatic component was sometimes obviously (2nd
hump in Fig. 4) outweighing the Ca2⫹ level in the process. This
resulted in Ca2⫹ diffusing back to the adjacent processes,
producing an elevated plateau in the process. The second
somatic component most likely reflects IP3-mediated Ca2⫹
release in the soma, the threshold of which can be set by the
level of Ca2⫹ reached by diffusion from the cell process.
Because the cell soma has a lower surface-to-volume ratio than
the process, CCh will produce lower levels of IP3. Properties of
Ca2⫹ signals are very sensitive to the IP3 concentration (Finch
et al. 1991; Khodakhah and Ogden 1995; Parker et al. 1996),
and therefore somatic Ca2⫹ release signals have a much
smaller amplitude, a longer latency, and a slower rise time. In
many instances the somatic Ca2⫹ release signal was not visible
as a distinct peak but only produced an elevated somatic
plateau. The ability of CCh to release Ca2⫹ in a cell compartment is thus essentially related to the rise in [IP3], and this will
be largely determined by the surface-to-volume ratio of that
compartment. For example, the surface-to-volume ratio of a
cell process with a radius of 1.5 ␮m and a length of 25 ␮m
would amount to 1.42, as compared with 0.73 for a cell soma
with a radius of 6 ␮m and a height of 5 ␮m. Therefore IP3
levels reached in the cell soma would only be about one-half of
those obtained in the cell process.
With low CCh concentrations, Ca2⫹ signals in cell processes
initiated simultaneously along the entire process, termed homogeneous Ca2⫹ signal, or were sometimes restricted to a
small area, termed “hot spot.” The diameter of hot spots was
⬃5.5 ␮m, allowing them to be identified as elementary Ca2⫹
release signals or Ca2⫹ puffs described in other cell types
(Bootman et al. 1997; Koizumi et al. 1999; Yao et al. 1995).
No gross differences were seen between the kinetics and amplitude of hot spots and homogeneous Ca2⫹ signals. Therefore
it is very conceivable that homogeneous Ca2⫹ signals arise
from the simultaneous activation of a small number of elementary Ca2⫹ release sites. The finding that the homogeneous
Ca2⫹ release signal was composed of multiple peaks with a
diameter of ⬃5 ␮m (Fig. 5B) supports this idea. Homogeneous
Ca2⫹ signals were more frequently observed than hot spots.
This indicates that IP3Rs are distributed along the whole length
of the process and suggests that the distance between release
sites is too short to allow activation of a single site.
In PC12 and hippocampal neuronal cells, Ca2⫹ puffs occur
repetitively in the maintained presence of threshold concentrations of IP3-generating agonists (Koizumi et al. 1999; Reber
and Schindelholz 1996). With prolonged threshold CCh stimulation, we also observed repetitive Ca2⫹ puffs in SH-SY5Y
cells (data not shown). These Ca2⫹ puffs could be observed in
both the cell body and the processes. The reason for the
CA2⫹ RELEASE SIGNALS IN SH-SY5Y CELLS
difference between brief and prolonged application of agonist
may be related to the IP3 gradient. With brief CCh application,
the difference in surface-to-volume ratio will create a large IP3
gradient between the cell process and cell soma. With prolonged CCh application, this IP3 gradient will dissipate resulting in a threshold elevation of IP3 throughout the whole cell.
Conclusions
To convey information to other cells, neurons have developed complex regulating mechanisms, many of which depend
on changes in intracellular [Ca2⫹]. Yet, the present experiments suggest that a change in the surface-to-volume ratio is a
very simple way for a neuronal cell to alter the spatiotemporal
domain of its Ca2⫹ response and thus to determine the intracellular messenger function of Ca2⫹.
Received 29 July 1999; accepted in final form 19 October 1999.
REFERENCES
BERRIDGE, M. J. Elementary and global aspects of calcium signalling.
J. Physiol. (Lond.) 499: 291–306, 1997.
BOOTMAN, M. D. AND BERRIDGE, M. J. The elemental principles of calcium
signaling. Cell 83: 675– 678, 1995.
BOOTMAN, M. D., NIGGLI, E., BERRIDGE, M. J., AND LIPP, P. Imaging the
hierarchical Ca2⫹ signalling system in HeLa cells. J. Physiol. (Lond.) 499:
307–314, 1997.
CHENG, H., LEDERER, W. J., AND CANNELL, M. B. Calcium sparks: elementary
events underlying excitation-contraction coupling in heart muscle. Science
262: 740 –744, 1993.
CLAPHAM, D. E. AND SNEYD, J. Intracellular calcium waves. Adv. Second
Messenger Phosphoprotein Res. 30: 1–24, 1995.
EILERS, J., CALLEWAERT, G., ARMSTRONG, C., AND KONNERTH, A. Calcium
signaling in a narrow somatic submembrane shell during synaptic activity in
cerebellar Purkinje neurons. Proc. Natl. Acad. Sci. USA 92: 10272–10276,
1995.
FINCH, E. A. AND AUGUSTINE, G. J. Local calcium signalling by inositol-1,4,5trisphosphate in Purkinje cell dendrites. Nature 396: 753–756, 1998.
FINCH, E. A., TURNER, T. J., AND GOLDIN, S. M. Calcium as a coagonist of
inositol 1,4,5-trisphosphate–induced calcium release. Science 252: 443–
446, 1991.
GRUDT, T. J., USOWICZ, M. M., AND HENDERSON, G. Ca2⫹ entry following store
depletion in SH-SY5Y neuroblastoma cells. Brain Res. Mol. Brain Res. 36:
93–100, 1996.
HORNE, J. H. AND MEYER, T. Elementary calcium-release units induced by
inositol trisphosphate. Science 276: 1690 –1693, 1997.
HUSER, J. AND BLATTER, L. A. Elementary events of agonist-induced Ca2⫹
release in vascular endothelial cells. Am. J. Physiol. 273 (Cell Physiol. 42):
C1775–C1782, 1997.
KHODAKHAH, K. AND OGDEN, D. Fast activation and inactivation of inositol
trisphosphate-evoked Ca2⫹ release in rat cerebellar Purkinje neurones.
J. Physiol. (Lond.) 487: 343–358, 1995.
KOIZUMI, S., BOOTMAN, M. D., BOBANOVIC, L. K., SCHELL, M. J., BERRIDGE,
M. J., AND LIPP, P. Characterization of elementary Ca2⫹ release signals in
NGF-differentiated PC12 cells and hippocampal neurons. Neuron 22: 125–
137, 1999.
LAMBERT, D. G. AND NAHORSKI, S. R. Muscarinic-receptor–mediated changes
in intracellular Ca2⫹ and inositol 1,4,5-trisphosphate mass in a human
neuroblastoma cell line, SH-SY5Y. Biochem. J. 265: 555–562, 1990.
LARSSON, C., THOMAS, A. P., AND HOEK, J. B. Carbachol-stimulated Ca2⫹
increase in single neuroblastoma SH-SY5Y cells: effects of ethanol. Alcohol
Clin. Exp. Res. 22: 637– 645, 1998.
MURPHY, N. P., VAUGHAN, P. F., BALL, S. G., AND MCCORMACK, J. G. The
cholinergic regulation of intracellular calcium in the human neuroblastoma,
SH-SY5Y. J. Neurochem. 57: 2116 –2123, 1991.
PARKER, I. AND YAO, Y. Regenerative release of calcium from functionally
discrete subcellular stores by inositol trisphosphate. Proc. R.. Soc. Lond. B
Biol. Sci. 246: 269 –274, 1991.
PARKER, I., YAO, Y., AND ILYIN, V. Fast kinetics of calcium liberation induced
in Xenopus oocytes by photoreleased inositol trisphosphate. Biophys. J. 70:
222–237, 1996.
REBER, B.F.X. AND SCHINDELHOLZ, B. Detection of a trigger zone of bradykinin-induced fast calcium waves in PC12 neurites. Pflügers Arch. 432:
893–903, 1996.
SUN, X. P., CALLAMARAS, N., MARCHANT, J. S., AND PARKER, I. A continuum
of InsP3-mediated elementary Ca2⫹ signalling events in Xenopus oocytes.
J. Physiol. (Lond.) 509: 67– 80, 1998.
TAKECHI, H., EILERS, J., AND KONNERTH, A. A new class of synaptic response
involving calcium release in dendritic spines. Nature 396: 757–760, 1998.
WOJCIKIEWICZ, R. J. AND LUO, S. G. Phosphorylation of inositol 1,4,5-trisphosphate receptors by cAMP-dependent protein kinase. Type I, II, and III
receptors are differentially susceptible to phosphorylation and are phosphorylated in intact cells. J. Biol. Chem. 273: 5670 –5677, 1998.
YAGODIN, S. V., HOLTZCLAW, L., SHEPPARD, C. A., AND RUSSELL, J. T.
Nonlinear propagation of agonist-induced cytoplasmic calcium waves in
single astrocytes. J. Neurobiol. 25: 265–280, 1994.
YAO, Y., CHOI, J., AND PARKER, I. Quantal puffs of intracellular Ca2⫹ evoked
by inositol trisphosphate in Xenopus oocytes. J. Physiol. (Lond.) 482:
533–553, 1995.
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on July 31, 2017
J. B. Parys is Research Associate and P. De Smet is Senior Research
Assistant at the Foundation for Scientific Research—Flanders. G. Bultynck is
Predoctoral Fellow at the “Vlaams Instituut voor de bevordering van het
Wetenschappelijk-Technologisch Onderzoek in de industrie.”
This work was supported by Grants 3.0238.95, 3.0207.99, and G.0186.97
from the Foundation for Scientific Research—Flanders and by Grants 94/04
and 99/08 of the Concerted Actions K. U. Leuven.
Address for reprint requests: G. Callewaert, Laboratory of Physiology, K. U.
Leuven, Campus Gasthuisberg, 3000 Leuven, Belgium.
1057