(Received for publication, September 10, 1996, and in revised form, February 10, 1997)
From the Departments of Molecular Biology and Pharmacology and of Neurology, Washington University School of Medicine, St. Louis, Missouri 63110
Depolarizing concentrations of potassium promote the survival of many neuronal cell types including cerebellar granule cells. To begin to understand the intracellular mediators of neuronal survival, we have tested whether the survival-promoting effect of potassium depolarization on cerebellar granule cells is dependent on either mitogen-activated protein (MAP) kinase or phosphatidylinositol 3-kinase (PI-3-K) activity. In 7-day cerebellar granule cell cultures, potassium depolarization activated both MAP kinase and PI-3-K. Preventing the activation of MAP kinase with the MEK1 inhibitor PD98059 did not affect potassium saving. In contrast, the survival-promoting effect of 25 mM potassium was negated by the addition of 30 µM LY 294002 or 1 µM wortmannin, two distinct inhibitors of PI-3-K. The cell death induced by PI-3-K inhibition was indistinguishable from the cell death caused by potassium deprivation; LY 294002-induced death included nuclear condensation, was blocked by cycloheximide, and had the same time course as potassium deprivation-induced cell death. Cerebellar granule cells can also be maintained in serum-free medium containing either 100 ng/ml insulin-like growth factor I (IGF-I) or 800 µM cAMP. PI-3-K inhibition completely blocked the survival-promoting activity of IGF-I, but had no effect on cAMP-mediated survival. These data indicate that the survival-promoting effects of depolarization and IGF-I, but not cAMP, require PI-3-K activity.
Cell death is a widespread event during development that, in the nervous system, is thought to match the size of neuronal populations with their targets (1). One mechanism for determining which neurons survive and which undergo programmed cell death is limiting the amount of neurotrophic factors (1). Another important mechanism for determining survival is electrical activity; removal of afferent input or pharmacological blockade of electrical activity causes the death of some types of developing neurons (2-6). In vitro, mimicking electrical activity with depolarizing concentrations of extracellular potassium promotes the survival of many types of neurons (for review, see Ref. 7). Depolarization leads to the sustained activation of voltage-gated calcium channels and subsequent elevation of intracellular calcium levels. The increase in intracellular calcium is important for survival because blockade of L-type calcium channels blocks K+-mediated survival (8-14), yet the intracellular pathways that mediate this survival are unknown.
Dissociated cerebellar granule cells from early postnatal rats can be maintained in serum-containing medium by raising the extracellular potassium concentration to 25 mM (15). Dissociated granule cells develop characteristics of mature cerebellar granule cells in vivo including an extensive neuritic network, expression of excitatory amino acid receptors, and production and release of L-glutamate (16). Removal of both potassium and serum from the culture medium triggers a cell death that is morphologically apoptotic, accompanied by DNA fragmentation, and dependent on macromolecular synthesis (17). This programmed cell death (PCD)1 presumably mimics the naturally occurring death of 20-30% of granule cells (18), thought to be important for matching the number of granule cells with Purkinje cells, that occurs during the third through fifth postnatal weeks (19, 20).
One possible mechanism by which depolarization could promote survival is by activating survival-promoting signaling pathways similar to those activated by neurotrophic factors. Tyrosine kinase growth factor receptors activate several intracellular signaling pathways. Of these, the MAP kinase and phosphatidylinositol 3-kinase (PI-3-K) pathways have been implicated in survival. In PC12 cells, overexpression of a constitutively active form of MEK1 (AP/xtracellular signal-regulated kinase inase), an activator of MAP kinase, promotes survival in the absence of NGF (21). However, MAP kinase is not required for the survival of NGF-maintained sympathetic neurons (22, 23) or PC12 cells (24). In contrast, inhibition of PI-3-K blocks the survival-promoting effects of NGF in PC12 cells (24). PI-3-K is a lipid kinase that mediates the mitogenic signal of the platelet-derived growth factor receptor (25). PI-3-K products are involved in regulating mitogenesis, membrane ruffling, glucose uptake, receptor sorting, and receptor down-regulation in response to growth factors (for review, see Ref. 26). In this report, we present evidence that PI-3-K activity is required for survival promotion by K+ depolarization and insulin-like growth factor I (IGF-I).
The following cell culture media were
used: K25+S (Eagle's basal medium (Life Technologies, Inc.) containing
10% dialyzed fetal bovine serum (Sigma; Mr
cutoff = 10,000), 25 mM KCl, 100 units/ml penicillin,
and 100 µg/ml streptomycin), K25S (Eagle's basal medium containing
no serum, 25 mM KCl, 100 units/ml penicillin, and 100 µg/ml streptomycin), K5
S (Eagle's basal medium containing no
serum, 5 mM KCl, 100 units/ml penicillin, and 100 µg/ml
streptomycin).
The cell culture protocol used is extensively detailed by Miller and Johnson (27). In brief, cerebella were dissected from P7 Harlan Sprague Dawley rats (Harlan Sprague Dawley, Inc., Indianapolis, IN), sliced into 1-mm pieces, and incubated at 37 °C for 15 min in 0.30 mg/ml trypsin (Worthington). The tissue was then triturated in K25+S medium with 0.5 mg/ml trypsin inhibitor (Sigma) using a flame-polished Pasteur pipette. The resulting cell suspension was spun at 500 × g for 6 min. The pellet was gently triturated in fresh K25+S medium and filtered through a Nitex filter (Tetko Inc., Elmsford, NY). Trypan blue exclusion was used to count the living neurons before plating 2-2.5 × 105 cells/cm2 in either four-well (Nunc, Roskilde, Denmark) or 35-mm (Corning Inc.) dishes. Prior to plating, dishes were coated with 0.1 mg/ml poly-L-lysine (Sigma, P2636). The granule cells were kept at 35 °C in a humidified incubator with 5% CO2 and 95% air for 7 days. To reduce the number of non-neuronal cells, aphidicolin (3.3 µg/ml; Sigma) was added to the medium 24 h after plating. The cultures were 98-99% pure granule cells; the culture conditions do not support the survival of other neuronal cell types (28, 29), and the non-neuronal contamination was 1-2% (27).
All cultures were treated 7 days after plating. For survival studies,
culture dishes were rinsed twice with K5S medium before adding medium
containing 25 mM potassium, LY 294002 (BIOMOL Research Laboratories Inc., Plymouth Meeting, PA), wortmannin (Sigma), PD98059
(New England Biolabs Inc., Beverly, MA), 10 µg/ml cycloheximide (Sigma), 800 µM chlorophenylthio-cAMP (CPT-cAMP;
Boehringer Mannheim), or 100 ng/ml IGF-I (a generous gift from Monsanto
Co., St. Louis, MO) or a combination of these reagents. Cell viability
was quantified from photomicrographs of representative fields of cells
labeled with calcein AM (Molecular Probes, Inc., Eugene, OR). Calcein AM is an acetoxymethyl ester fluorescein derivative that is cleaved and
trapped inside viable cells that have nonspecific esterase activity
(30). The photomicrographs were both taken and scored by a naive
observer. Eight photomicrographs at a magnification × 200 were
taken for each condition, and the number of calcein AM-positive cells
was counted. Cellular nuclei were visualized by fixing cells with 4%
paraformaldehyde and staining with 1 µg/ml Hoechst 33258 (Molecular
Probes, Inc.).
Cells were washed three times with K5S
medium, incubated at 35 °C in 5% CO2 for 3 h, and
then stimulated by adding concentrated factors directly to the culture
medium. Cells were lysed in PI-3-K lysis buffer as described by
Newberry and Pike (31). Approximately 400 µg of lysate protein and 35 µl of anti-phosphotyrosine antibody conjugated to agarose
(Calbiochem) were added to the samples and incubated for 2 h at
4 °C. Immune complexes were isolated by centrifugation and washed
twice with phosphate-buffered saline (2.7 mM KCl, 140 mM NaCl, 1.5 mM KH2PO4,
and 10 mM Na2HPO4, pH 7.4) plus 1%
Triton X-100; twice with 0.5 mM LiCl and 10 mM
Tris, pH 7.5; and twice with 10 mM Tris, pH 7.5, 100 mM NaCl, and 1 mM EDTA. PI-3-K activity was
assayed in an in vitro kinase reaction, and the products
were separated by thin layer chromatography (32). Phosphatidylinositol 3-phosphate was identified based on its migration relative to that of a
radiolabeled phosphatidylinositol phosphate marker (33). Radioactivity
was visualized and quantified with a PhosphorImager (Molecular
Dynamics, Inc., Sunnyville, CA).
Samples were homogenized in SDS sample buffer and subjected to 10% SDS-polyacrylamide gel electrophoresis. After electrophoresis, the gels were transferred to polyvinylidene difluoride membranes (Millipore Corp., Bedford, MA). Blots were incubated for 1 h at 25 °C in Buffer A (phosphate-buffered saline containing 0.1% Tween 20 and 5.0% (w/v) nonfat dry milk) and then for 16 h at 4 °C with a phospho-specific MAP kinase antibody (New England Biolabs Inc.) diluted 1:1000 in phosphate-buffered saline plus 0.05% Tween 20 and 5% bovine serum albumin. The membranes were then washed in Buffer A (3 × 15 min) and incubated for 1 h in Buffer A containing a 1:1000 dilution of alkaline phosphatase-conjugated anti-rabbit antibody (New England Biolabs Inc.). This was followed by three washes (15 min/wash) in Buffer A; two washes (10 min/wash) in phosphate-buffered saline plus 0.1% Tween 20; and two washes (10 min/wash) in 1 mM MgCl2, 10 mM NaCl, and 10 mM Tris-HCl, pH 9.5. The blot was then incubated for 5 min in CDP-Star (Tropix Inc., Bedford, MA) and exposed to x-ray film.
Equal loading of samples was checked by stripping the blots and reprobing with an antibody that detects total MAP kinase (New England Biolabs Inc.). The membranes were stripped by incubation for 1 h at 25 °C in 1.0% Tween 20, 0.1% SDS, and 0.2 M glycine, pH 2.2. To verify that the blots were completely stripped, they were washed as described above and developed using the CDP-Star reagent. The blots were returned to Buffer A for 1 h and then probed as described above with the total MAP kinase antibody at a 1:1000 dilution.
Determination of Intracellular Calcium ConcentrationCultures were incubated for 45-60 min in K25S
medium with 4 µM fura-2 (Molecular Probes, Inc.) (34) and
then washed three times with Locke's solution (154 mM
NaCl, 5.6 mM KCl, 3.6 mM NaHCO3, 2.7 mM CaCl2, 1.2 mM
MgCl2, 5.6 mM D-glucose, and 5 mM Hepes, pH 7.4) containing 5 mM potassium, 25 mM potassium, or 25 mM potassium plus 30 µM LY 294002. After ~1 h at 35 °C, the cultures were
transferred to the heated stage (35 °C) of an inverted fluorescence
microscope and illuminated with a 75-watt xenon lamp. Fields containing
~50 neurons were viewed using a 1.3NA ×40 epifluorescent objective (Nikon Inc.). The cultures were alternately illuminated at 340 and 380 nm by switching the excitation filters. The fluorescent signal was
detected by an intensified charge-coupled device camera that was
connected to a personal computer for image acquisition and analysis
(Metafluor, Universal Imaging Corp.). Data represent means ± S.D.
for three independent experiments in which ratios were determined for
150-200 cells/condition/ experiment.
To
determine whether depolarizing concentrations of extracellular
potassium increase MAP kinase or PI-3-K activity, cerebellar granule
cells were switched to K5S medium for 3 h before the medium was
raised to 25 mM by the addition of 2.0 M
K+. Serum-free medium was used in all experiments to
isolate the survival-promoting activity of K+ alone; in
granule cells, serum provides survival-promoting activity of unknown
origin (27). For PI-3-K activity, neuronal cultures were lysed,
immunoprecipitated with a phosphotyrosine antibody, and assayed for
PI-3-K activity. Thirty min after treatment, the cultures in 25 mM potassium showed increased PI-3-K activity (Fig. 1A). In three independent experiments, PI-3-K
activity from depolarized samples was 3.0 ± 0.2-fold above the
controls. To verify that the observed signal was attributable to PI-3-K
activity, 30 µM LY 294002, a PI-3-K inhibitor, was
included in the in vitro labeling reaction (Fig.
1A, lane 5). LY 294002 does not inhibit
phosphatidylinositol 4-kinase, which could contribute to the observed
signal, nor does it have any effect on protein kinase C, protein kinase
A, MAP kinase, S6 kinase, c-Src kinase, or diacylglycerol kinase
(35).
Potassium depolarization also caused an activation of the MAP kinase pathway. Samples lysed 15 min after treatment with 25 mM potassium were subjected to Western blot analysis. Probing the blot with a phospho-MAP kinase antibody showed that depolarization led to MAP kinase phosphorylation (Fig. 1B, upper panel). The blot was stripped and reprobed with an antibody that recognizes both the phosphorylated and non-phosphorylated forms of the ERK1 and ERK2 (xtracellular signal-egulated inase) isoforms of MAP kinase to verify that different lanes contained comparable amounts of sample (Fig. 1B, lower panel). The finding that K+ increased both PI-3-K and MAP kinase activities raised the possibility that one or both of these pathways are important for granule cell survival.
MAP Kinase Activation Is Not Required for Potassium-mediated SurvivalTo determine whether the MAP kinase pathway is important
for granule cell survival, cultures were treated with the MEK inhibitor PD98059 (36, 37). Seven-day cultures were switched to K5S medium for
2.5 h, treated for 30 min with PD98059, and then depolarized by
the addition of potassium to 25 mM. To assure that the
inhibition was sustained during the time the neurons would become
irreversibly committed to die (27), fresh medium and MEK inhibitor were
added at 12 and 24 h. Forty-eight h after treatment, neuronal
survival was quantified from photomicrographs of calcein AM-stained
cultures. Inhibiting the activation of MAP kinase with either 30 or 100 µM PD98059 had no apparent effect on the ability of
potassium to promote survival (Fig. 2A).
Phospho-MAP kinase Western blots revealed that these doses of PD98059
were sufficient to inhibit MAP kinase activation (Fig. 2B),
and this inhibition was sustained even 24 h after
treatment.2 Thus, the events downstream of
MAP kinase activation were not required for potassium-mediated survival
of granule neurons.
Inhibition of PI-3-K Kills Depolarization-maintained Neurons
Since LY 294002 can block PI-3-K in intact cells (24,
35), we also used this compound to determine whether PI-3-K activity is
involved in the survival-promoting effect of K+
depolarization. Cultures were switched to K25S medium either with or
without LY 294002, and cell survival was assayed 48 h later by
counting the number of cells in photomicrographs of calcein AM-stained
cultures. LY 294002 induced a dose-dependent cell death of
granule cells with an EC50 of ~20 µM (Fig.
3A, closed squares). This
dose-response curve is consistent with that of LY 294002 for inhibition
of PI-3-K (35).
When granule cells are switched to K5S medium, the neurons die by
apoptosis (17). One property of this PCD is its inhibition by
cycloheximide. LY 294002-induced cell death was also blocked by
cycloheximide (Fig. 3A, open squares). The
addition of cycloheximide concurrent with the PI-3-K inhibitor
prevented the dose-dependent killing, indicating that
inhibition of PI-3-K triggered PCD.
Similar results were obtained with wortmannin, a fungal metabolite that
also blocks PI-3-K. Wortmannin caused a dose-dependent killing (Fig. 3B, closed squares) that was
blocked by cycloheximide (open squares). Because wortmannin
is a labile compound (38), the culture medium was changed every 6 h during the first 24 h of treatment. Thus, two inhibitors of
PI-3-K that act via distinct mechanisms block the survival-promoting
ability of potassium depolarization. If this killing is truly caused by
a disruption in a specific survival pathway, then the death induced by
the PI-3-K inhibitors should share features of that induced by
switching to K5S medium. Because of the lability of wortmannin, these
experiments were conducted with LY 294002.
Switching to K5S medium causes
the programmed cell death of 60% of granule cells in 48 h (17,
27). To determine whether the cell death induced by PI-3-K inhibition
had properties similar to PCD in K5
S medium, we used 30 µM LY 294002, a dose that caused a comparable amount of
cell loss. We then compared the cell death caused by LY 294002 with
that caused by K5
S medium based on several different criteria
including dependence on macromolecular synthesis, morphology of the
nuclei of dying cells, and time course of cell loss.
Similar to PCD induced by K5S medium, the protein synthesis inhibitor
cycloheximide (10 µg/ml) blocked the cell death induced by LY 294002 (Fig. 3A, open squares). Fig. 4
shows the phase-contrast (panels a-d) and the corresponding
calcein AM-stained (panels e-h) images after 48 h and
the bisbenzimide-stained nuclei (panels i-l) after 18 h of granule cells maintained in K25
S medium (panels a,
e, and i), in K25
S medium plus 30 µM LY 294002 (panels b, f, and
j), in K5
S medium (panels c, g, and
k), or in K25
S medium plus 30 µM LY 294002 and 10 µg/ml cycloheximide (panels d, h, and
l). LY 294002 blocked the survival-promoting effect of 25 mM potassium (compare panels a and e
and panels b and f) and led to a cell death that
resembled the programmed cell death induced by K5
S medium
(panels c and g). Cycloheximide prevented the LY 294002-induced cell death (panels d and h) as it
prevents PCD in K5
S medium (17). The nuclear morphology of granule
cells treated with LY 294002 also resembled the nuclei of neurons
undergoing K5
S medium-induced PCD. In contrast to the round and
homogeneously stained nuclei of cells maintained in 25 mM
K+ (panel i), the nuclei of neurons in K25
S
medium plus LY 294002 (panel j) and those in K5
S medium
(panel k) were condensed and fragmented, features typical of
apoptotic nuclei. Again, however, cycloheximide was able to block the
effects of PI-3-K inhibition. The nuclei of granule cells in K25
S
medium plus LY 294002 and cycloheximide (panel l) were
indistinguishable from controls (panel i).
Further evidence that inhibition of PI-3-K is similar to removal of
K+ was found in the time course of death. After 6, 12, 24, or 48 h, the number of remaining cells was assessed by counting
photomicrographs of calcein AM-stained cultures. The resulting time
course of cell loss was identical in cultures switched to K5S medium
or to K25
S medium plus LY 294002 (Fig. 5). Together,
these results suggest that inhibiting PI-3-K mimicked the programmed
cell death that accompanies a switch to low K+.
LY 294002 Does Not Interfere with the Depolarization-induced Increase in Intracellular Calcium
Since the survival-promoting
effect of potassium depolarization is critically dependent on an
increase in intracellular calcium (15), we tested whether LY 294002 had
any effect on the sustained rise in intracellular calcium in
depolarized cells. Intracellular calcium was measured by determining
340/380 nm ratios for the calcium-sensitive dye fura-2. An increase in
the 340/380 nm ratio is indicative of an increase in intracellular free
calcium. K+ depolarization of granule cells caused an
~2-fold increase in the 340/380 nm ratio that was not blocked by 30 µM LY 294002 (Fig. 6). Thus, blocking
calcium channels or the sustained rise in intracellular calcium was not
the mechanism by which LY 294002 blocked the survival-promoting activity of depolarizing concentrations of K+.
Neurons Saved by IGF-I or cAMP Differ in PI-3-K Activation and Their Susceptibility to Cell Death Caused by LY 294002
In
addition to potassium depolarization, cerebellar granule cells can be
maintained in vitro in low K+ medium
supplemented with either IGF-I or CPT-cAMP (17). We found that, similar
to K+ (Fig. 1), IGF-I (100 ng/ml) activated PI-3-K, whereas
CPT-cAMP (800 µM) did not (Fig.
7A). In accordance with their ability to stimulate PI-3-K, LY 294002 blocked the survival-promoting effect of
IGF-I and K+, but not that of CPT-cAMP (Fig.
7B). At 48 h, K5S medium plus IGF-I or plus CPT-cAMP
promoted granule cell survival comparable to K25
S medium. As with
K25
S medium, LY 294002 reversed the saving effects of IGF-I. In fact,
the dose-response curve for LY 294002 killing of IGF-maintained cells
(Fig. 7C) mirrored the effect of LY 294002 on
potassium-saved neurons (Fig. 3A). In contrast, cells
maintained in CPT-cAMP survived despite the presence of the PI-3-K
inhibitor.
Since IGF-I stimulates a tyrosine kinase receptor, the MAP kinase pathway is another candidate for survival-promoting activity. Like the potassium-maintained cells, however, MEK inhibition had no discernible effect on IGF-I saving (Fig. 7C, closed circles) or on CPT-cAMP-maintained neurons.2 The fact that IGF-I survival was blocked by LY 294002 implies that both K+ and IGF-I promote survival by signaling through PI-3-K, whereas CPT-cAMP appears to act downstream of PI-3-K or through a non-PI-3-K-dependent pathway. The failure of LY 294002 to block CPT-cAMP-mediated survival also provides additional evidence that LY 294002 is not generally toxic to cerebellar granule cells; these results support a role for PI-3-K in neuronal survival.
In this study, we provide evidence for a link between K+ depolarization and a signaling pathway that promotes the survival of neurons. Depolarizing concentrations of extracellular potassium increased PI-3-K activity; that this activity is critical for survival is indicated by the ability of inhibitors of PI-3-K to block depolarization-mediated survival. The cell death resulting from inhibition of PI-3-K was indistinguishable from the programmed cell death triggered by lowering the K+ concentration. In both cases, cell death was dependent on macromolecular synthesis and accompanied by morphologically apoptotic nuclei. In addition, the time course of cell loss was identical, and the cell death was prevented by cAMP. These data strongly suggest that PI-3-K inhibition resulted in programmed cell death by specifically blocking the survival-promoting activity of K+ depolarization.
The survival-promoting activity of IGF-I was also blocked by inhibition of PI-3-K. K+ depolarization and growth factors may therefore converge on PI-3-K to signal survival promotion. PI-3-K activity increases in response to several growth factors including platelet-derived growth factor, insulin, colony-stimulating factor 1, nerve growth factor, hepatocyte growth factor, stem cell growth factor, and epidermal growth factor (26). As previously reported in cell lines (39, 40), we found that IGF-I increased PI-3-K activity in cerebellar granule neurons. Although both IGF-I and 25 mM potassium increased PI-3-K activity and inhibiting PI-3-K blocked their ability to promote survival, we did not directly test whether the rise in PI-3-K activity per se was critical for survival. Basal activity could augment other actions of these agents that result in survival. In the absence of 25 mM potassium or IGF-I, however, the basal PI-3-K activity was not sufficient to maintain the neurons.
For growth factors, activation of PI-3-K is dependent on the protein-tyrosine kinase activity of the growth factor receptor. Autophosphorylation of specific tyrosine residues on the receptor recruits the Src homology 2 domain (SH2) of the p85 regulatory subunit of PI-3-K. This leads to relocalization of PI-3-K to the membrane and activation of the catalytic p110 subunit. In contrast to other known growth factors, phosphorylation of the insulin or IGF-I receptor leads to phosphorylation of insulin receptor substrate 1, which serves as an SH2-docking protein for p85 (39, 40-42).
One mechanism by which potassium depolarization may increase PI-3-K activity is by causing phosphorylation events similar to those caused by growth factor receptors. Some evidence supports this hypothesis. For example, PYK2, a novel protein-tyrosine kinase, is rapidly phosphorylated in response to stimuli that increase intracellular calcium (43). The p85 subunit of PI-3-K may associate with this calcium-responsive protein-tyrosine kinase. Alternatively, PYK2 may activate PI-3-K by first activating Ras, which can be an upstream activator of PI-3-K (44). Potassium depolarization of PC12 cells results in phosphorylation of the adapter protein Shc and leads to the association of Shc with Grb2, causing Ras and MAP kinase activation (45-47) and phosphorylation of the epidermal growth factor receptor (47). At this point, which of these phosphorylation events occurs in depolarized granule cells is unclear.
In cultures of cortical neurons, depolarization promotes survival by stimulating the release of the neurotrophin brain-derived neurotrophic factor that serves as an autocrine or paracrine trophic agent (48). This contrasts with sympathetic neurons, however, in which potassium depolarization fails to induce the phosphorylation of growth factor receptors (10). In granule cell cultures, antibodies to IGF-I, neurotrophin-3, NGF, or brain-derived neurotrophic factor do not affect granule cell survival in depolarizing medium,2 suggesting that granule cells behave similarly to sympathetic neurons in this regard. Another mechanism whereby calcium may influence survival is by increasing calcium/calmodulin kinase activity. Inhibitors of calmodulin- or Ca2+/calmodulin-dependent protein kinase II activity block K+-mediated survival (15, 49). However, these compounds may affect survival by blocking the sustained rise in intracellular calcium (10).
We found that LY 294002 did not interfere with increases in
intracellular calcium after 1 h (Fig. 6), a time when inhibitors of calcium channels, such as nifedipine, will block calcium channels completely (10). It is possible that LY 294002 affected calcium at
later time points, but measurement of calcium at later time points
would be difficult to interpret because the cells begin to degenerate
within 6 h, and changes in metabolic parameters such as glucose
uptake and protein synthesis occur within 2 h of the induction of
apoptosis (27). In addition to our direct measurement of calcium at
1 h, two other pieces of evidence argue against LY 294002 affecting calcium. First, since the time course of death triggered by
K5S medium or by K25
S medium plus LY 294002 overlaps (Fig. 5), an
LY 294002 effect on calcium would have to occur rapidly. If LY 294002 caused a delayed decrease in intracellular calcium that initiated PCD,
we would expect to see a corresponding delay in the cell death time
course, but this was not the case (Fig. 5). Second, LY 294002 blocked
the survival-promoting effect of IGF-I, which does not elevate
intracellular calcium (11).
An indication that Ras is activated by depolarization is that depolarization activates the MAP kinase pathway (Fig. 6) (45-47, 50). In PC12 cells, overexpression of a constitutively active form of MEK1, an activator of MAP kinase, promotes survival in the absence of NGF (21). However, MAP kinase is not required for the survival of NGF-maintained sympathetic neurons (22, 23) or NGF-maintained PC12 cells (24). Consistent with this, our data demonstrate that MAP kinase was not required for survival because the MEK inhibitor, which blocks activation of MAP kinase, did not affect survival. The fact that PI-3-K inhibitors do not prevent the activation of MAP kinase2 yet lead to PCD (Fig. 3) indicates that MAP kinase activity was also not sufficient to promote granule neuron survival.
Several possible downstream targets of PI-3-K may be involved in neuronal survival. Protein kinase C isozymes (51-54), Rac (55, 56), p70S6K (57-59), and AKT/protein kinase B serine/threonine kinase (60, 61) are all effectors of PI-3-K. PI-3-K also affects glucose transporters (62-64) and, through this pathway, may be critical for maintaining the metabolic functions of cells. We have found that glucose uptake, protein synthesis, and RNA synthesis decrease dramatically as granule cells (27) and sympathetic neurons (65) undergo PCD and have suggested that this dramatic decrease in metabolic parameters is part of PCD in neurons. Inhibiting PI-3-K may specifically decrease metabolic functions maintained by either K+ depolarization or growth factors and thus trigger PCD.
Although peripheral neurons are particularly dependent on growth factors for survival, neurons from the central nervous system may depend on a combination of trophic factors and electrical activity (66). Our results highlight PI-3-K as an important intracellular mediator of survival promotion by either activity or trophic factors. PI-3-K is one of the pathways upon which neurons critically depend for survival in vitro and, perhaps, in vivo.
We thank Monsanto Co. for the generous gift of IGF-I; S. Handran and Dr. S. Rothman for help with calcium measurements and for the use of the calcium measurement system; Dr. L. Pike for preparing the radiolabeled phosphatidylinositol 4-phosphate; M. Deshmukh, J. L. Franklin, and P. A. Osborne for critical evaluation of this manuscript; and N. Girgis for help with cell counts.