In the serum-free culture of rat embryonic neurons, most neurons rapidly died by necrosis, which was revealed by propidium iodide (PI)-positive staining as early as 3 h after the start of culture and by marked membrane disruption and mitochondrial swelling in transmission electron microscopic (TEM) analysis. However, neither nuclear condensation/fragmentation stained with Hoechst 33342 nor activated caspase-3-like immunoreactivity was observed. In the serum-deprived culture, on the other hand, neurons showed apoptotic features, such as caspase-3 activation and nuclear damages in TEM analysis. Insulin at relatively higher concentrations, up to 100 μg/ml, ameliorated the rapid decrease in survival activity measured with 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt WST-8 assay and PI staining in the serum-free culture, despite the fact that brain-derived neurotrophic factor and insulin-like growth factor-I had no survival effect even at concentrations up to 100 μg/ml. Insulin-induced survival effects were abolished by the protein kinase C (PKC) inhibitor calphostin C but not by the phosphatidyl inositol-3-OH-kinase inhibitor wortmannin or the mitogen-activated protein kinase inhibitors PD98059 or U0126. Insulin significantly stimulated the PKC activity in cell lysates and suppressed the mitochondrial swelling and membrane disruption in TEM analysis in a calphostin C-reversible manner. All of these findings suggest that insulin inhibited the neuronal necrosis resistant to known neurotrophic factors under the serum-free culture through PKC mechanisms.
We have previously found that cortical neurons showed a cell density-dependent survival in the serum-free culture without supplements (Fujita et al., 2001). Further characterization of the cell death under such conditions revealed that neurons in the low-density culture die by necrosis, whereas those in the high-density culture die by apoptosis. The former necrotic cell death, however, was resistant to various neurotrophic factors, such as BDNF, basic fibroblast growth factor, and nerve growth factor (Hamabe et al., 2000; Fujita et al., 2001).
Necrosis as well as apoptosis have been identified as major modes of cell death within the central nervous system, and both cell death modes contribute to cell loss and related functional deficits after acute focal brain ischemia (Colicos and Dash, 1996; Fox et al., 1998; Dirnagl et al., 1999). Following focal brain ischemia/reperfusion, neuronal cells in the ischemic core die by necrosis, which is associated with inflammatory reactions and usually impacts large groups of adjacent cells, followed by apoptosis in the penumbra. Compared with apoptosis, the molecular events in necrosis are rather undefined. The inhibition or targeting deletion of several proapoptotic molecules, such as Bax or caspases, however, did not completely rescue the retinal damage or ischemic brain injury (Hara et al., 1997; Friedlander et al., 1997; Mosinger Ogilvie et al., 1998). The inhibition of the initial necrosis steps could be an important issue for complete rescue from brain injury.
Insulin as well as other growth-promoting factors, such as putrescine, transferrin, selenium, and progesterone, have been widely used for serum-free culture of various kinds of cells, including neurons. Previous studies revealed that only insulin is enough to keep neurons surviving for a longer period, but the concentration required for this purpose is much higher than that used for conventional growth promotion of cells in the serum deprivation (Snyder and Kim, 1980; Huck, 1983; Skaper et al., 1984; Salamanca and Mathers, 1987). However, little is known of the mechanisms underlying this survival activity of insulin for serum-free culture. In the course of characterization, we found that insulin inhibits the necrosis of cortical neurons in the serum-free and low-density culture. Here, we attempted to characterize the mechanisms underlying this action.
Materials and Methods
Cell Culture. Primary culture of the cerebral cortex from 17-day old embryonic rats was performed according to the previously reported protocol (Sasaki et al., 1998; Hamabe et al., 2000; Fujita et al., 2001). In brief, cortical tissues were minced into pieces in sterile PBS and dissociated with 0.25% trypsin (Invitrogen, Carlsbad, CA) and 0.01% DNase I (Sigma-Aldrich, St. Louis, MO) in PBS, pH 7.4, for 12 min at 37°C. The reaction was terminated by adding 0.25% soybean trypsin inhibitor (Sigma-Aldrich). For serum-free culture, the cell suspension was centrifuged at 1,000g for 5 min, and the pellet was resuspended in a serum-free D/F medium (1:1 Dulbecco's modified Eagle's medium/Ham's F-12 medium). Dissociated neurons were seeded at a density of 1 × 105 cells/cm2 onto 96-well plates for WST-8 assay, 24-well plates for trypan blue exclusion test, 6-well plates for ATP measurements, and 8-well Lab-Tek chamber slides (Nalge Nunc International, Naperville, IL) for morphological assessment, which had been all coated with poly-dl-ornithine (Sigma-Aldrich) and cultured at 37°C in a 5% CO2 atmosphere for the indicated periods. For serum-deprived culture, the pellet was resuspended in a serum-containing D/F medium [1:1 Dulbecco's modified Eagle's medium/Ham's F-12 medium (Invitrogen) plus 5% fetal bovine serum, 5% horse serum, and 1% 2-mercaptoethanol (Invitrogen)]. Dissociated neurons were seeded at a density of 1 × 105 cells/cm2 onto 96-well plates for WST-8 assay, 6-well plates for ATP measurements, and 8-well Lab-Tek chamber slides for morphological assessment, which had been all coated with poly-dl-ornithine and cultured at 37°C in a 5% CO2 atmosphere. Cytosine β-d-arabinofuranoside (Sigma-Aldrich) at 1 μM was added to the culture 24 h after seeding, followed by another 48-h culture. The culture medium was then replaced by fresh D/F medium deficient of serum and further cultured at 37°C in a 5% CO2 atmosphere for the indicated periods.
WST-8 Assay. The viability of cells was assessed by WST-8 reduction assay kit (Dojin Laboratories, Tokyo, Japan) according to the instructions of the manufacturer. WST-8 was added to the culture and incubated for 3 h at 37°C prior to the colorimetry. Percentage of WST-8 reduction activity was represented as the ratio of activity at different time points to that in the beginning of culture.
Trypan Blue Exclusion Test. To assess the neuronal cell death, an equal volume of 0.4% trypan blue (Wako Pure Chemicals, Osaka, Japan) in saline was added to the culture, followed by two washings with ice-cold PBS and fixation with 4% paraformaldehyde (PFA) in PBS for 30 min at 25°C. Results were represented by the percentage of stained cells among the 100 to 500 cells each in five separate fields.
Hoechst 33342-Staining and PI-Staining Assay. Apoptotic or necrotic cell death was characterized by use of Hoechst 33342 (Molecular Probes, Eugene, OR) and propidium iodide (PI; Sigma-Aldrich) double staining. Twelve hours after seeding, cells were stained with 10 μg/ml Hoechst 33342 and 10 μg/ml PI for 30 min at 37°C. After being washed with PBS twice, cells were fixed with 4% PFA in PBS for 30 min at 25°C. Cells were mounted with Aqua Poly/Mount (Polysciences, Inc., Warrington, PA), and imaged on an AxioVision 3.0 (Carl Zeiss, Tokyo, Japan) attached to a fluorescence microscope (BX50; Olympus, Tokyo, Japan). The nuclei stained with Hoechst 33342, which were markedly bright and small or divided into several homologous chromatin clumps, were defined as being condensed or fragmented nuclei, respectively. The percentage of condensed or fragmented nuclei or PI-positive nuclei was expressed as the ratio of total nuclei of 100 to 500 cells each in five separate fields.
Immunocytochemistry of Active Caspase-3. Cortical cells on 8-well Lab-Tek chamber slides were fixed with 4% PFA in PBS for 30 min at 25°C, followed by permeabilization using 50 and 100% methanol each for 5 min at 25°C. Cells were then rinsed twice with PBS and preincubated for1hat25°C in blocking buffer (2% bovine serum albumin and 0.1% Triton X-100 in PBS). Cells were incubated overnight at 4°C in blocking buffer containing anti-rabbit cleaved caspase-3 antibody (1:50; Cell Signaling Technology, Inc., Tokyo, Japan), rinsed in PBS, and incubated for2hat25°C with fluorescein isothiocyanate-conjugated anti-rabbit immunoglobulin (1:200; ICN/Cappel, Aurora, OH). Immunolabeled cells were mounted with Aqua Poly/Mount (Polysciences, Inc.) and imaged on an AxioVision 3.0 (Carl Zeiss) attached to a fluorescence microscopy (BX50; Olympus).
Transmission Electron Microscopy. Cultured cortical neurons were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, for1hat25°C. The fixed cortical neurons were postfixed with 1% osmium tetroxide for 1 h at 25°C, dehydrated in a graded alcohol series, and embedded in Epon 812 (TAAB Laboratories Equipment Ltd., Berkshire, England). Ultrathin sections (80-nm thick) were cut with an Ultracut S (Leica Microsystems, Tokyo, Japan) and then stained with uranyl acetate and lead citrate for 30 and 5 min, respectively. The stained sections were observed under an electron microscope (JEM-1210; JEOL, Tokyo, Japan).
Measurement of Intracellular ATP Levels. Intracellular ATP was extracted from cells in the exponential phase of growth and measured by the luciferin/luciferase method using the ATP Determination Kit (Molecular Probes) (Ha and Snyder, 1999). Cells were seeded onto 6-well plates at the density of 1 × 105 cells/cm2 and cultured for the indicated periods. The entire cells, including any floating cells, were subjected to the assay. The reaction buffer (200 μl) containing 0.5 μM luciferin, 1.25 μg/ml luciferase, and 1 mM dl-dithiothreitol (DTT) was mixed with cell lysates (20 μl). Luminescence was analyzed using Lumat LB 9507 from EG&G Berthold (Bad Wildbad, Germany).
2-Deoxy-d-[3H]Glucose Uptake. 2-Deoxy-d-[3H]glucose ([3H]2-DG; PerkinElmer Life Sciences, Boston, MA) uptake was measured using a modified version of the methods of Koivisto (Koivisto et al., 1991). Briefly, [3H]2-DG (1 μCi/well, 10 nM) was added to the 6-well plates in the beginning of the culture and incubated for 2 h at 37°C in a 5% CO2 atmosphere. Uptake of [3H]2-DG was terminated by a rapid removal of medium followed by two washings with ice-cold PBS. The cells were lysed in 100 μl of 0.5 M NaOH, followed by neutralization with 0.5 M HCl. The radioactivity of collected cell lysates was determined by a liquid scintillation counter.
Measurement of PKC Activity. Cultured cells were harvested using a solution containing 0.3% β-mercaptoethanol, 5 mM EDTA, 10 mM EGTA, 50 mg/ml phenylmethylsulfonyl fluoride, 10 mM benzamidine, and 50 mM Tris-HCl, pH 7.5, homogenized on ice, and added to an equal volume of glycerol. PKC assay was performed using the Protein Kinase C enzyme assay system (Amersham Biosciences, Piscataway, NJ) according to the instructions of the manufacturer, with modifications. Briefly, samples [S] were incubated with the substrate peptide to be phosphorylated by PKC in 50 mM Tris-HCl (pH 7.5), 30 mM DTT, and [γ-32P]ATP in magnesium ATP buffer (1.2 mM ATP, 30 mM Hepes, and 72 mM MgCl2, pH 7.4) for 15 min at 37°C. Maximum PKC activation [Max] was obtained by further addition of 0.3 mg/ml La-phosphatidyl-l-serine, 24 μg/ml phorbol 12-myristate 13-acetate, and 12 mM calcium acetate in 50 mM Tris-HCl, pH 7.5. Background activation [B] was obtained by omitting substrate peptide from the [S] reaction mixture. After the addition of stop solution, the reaction mixture was transferred onto peptide-binding paper. These papers were washed twice with 5% acetic acid, and the incorporation of 32P into peptide was determined by a liquid scintillation counter. PKC activation was calculated by the following equation: PKC activation (%) = ([S] – [B])/([Max] – [B]) × 100.
Western Blot Analysis. Cells were lysed in SDS-sample buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, and 10% glycerol) immediately after the stimulation with or without insulin or BDNF. Cell lysates were boiled and reduced with 10 mM of DTT. Samples and molecular weight standards (Invitrogen) were electrophoresed in 10% of SDS-polyacrylamide gel electrophoresis acrylamide gels and transferred onto polyvinylidene diflouride membranes (Millipore Corp., Tokyo, Japan). The membranes were blocked for 2 h at 25°C with 5% skim milk in Tris-buffered saline containing 0.1% Tween 20 for Akt, ERK, and phosphorylated ERK1/2 and with 1% bovine serum albumin in Tris-buffered saline containing 0.05% Tween 20 for phosphorylated Akt. Proteins were detected with rabbit anti-phospho-Akt (1:200; Promega, Tokyo, Japan), rabbit anti-phospho-ERK1/2 (1:200; New England Biolabs, Tokyo, Japan), rabbit anti-Akt (1:200; New England Biolabs), or rabbit anti-ERK1/2 antibody (1:200; New England Biolabs) overnight at 4°C. The blots were then incubated for 2 h at 25°C with anti-rabbit antibody conjugated with horseradish peroxidase. All visualization of immunoreactive bands was performed by the Light Capture (AE-6960/C/FC; Atto Bioscience, Tokyo, Japan) with an enhanced chemiluminescent substrate for the detection of horseradish peroxidase, SuperSignal West Pico chemiluminescent substrate (Pierce Biotechnology Inc., Rockford, IL). As this Light Capture can repeatedly expose at intervals of 1 min and automatically superimpose these signals to obtain high contrast, we can detect enhanced signals with less background signals.
Statistical Analysis. The data were analyzed using Student's t test after multiple comparisons of analysis of variance. The criterion of significance was set at p < 0.05. All results are expressed as the mean ± S.E.M.
Distinct Characteristics of Neuronal Death Modes Caused in the Serum-Free and Serum-Deprived Culture. Cortical neurons of embryonic (E17) rats were seeded at a density of 1 × 105 cells/cm2. The population of bromodeoxyuridine-positive cells was below 5%, indicating that the cells were mostly neurons (data not shown). Although neurons looked healthy in the serum-containing culture [serum(+)] and have short and long neurites at 12 h and 3 days, respectively, after the start of the culture (Fig. 1, Aa and Ac), neurons in the serum-free culture [serum(–)] lost normal thickness and neurites 12 h after the start of culture (Fig. 1Ab). On the other hand, in the serum-deprived culture, neurons looked slightly condensed and lost neurites 12 h after serum deprivation (Fig. 1Ad). PI was used for detection of the plasma membrane damage (Bal-Price and Brown, 2000). Most of neurons were PI positive in the serum-free culture (68.8%; Fig. 1, Af and B) compared with little in the serum-containing (1.7%, Fig. 1, Ae and B and 2.3%, Fig. 1, Ag and B) and serum-deprived cultures (9.4%; Fig. 1, Ah and B). Hoechst 33342 was used as an apoptosis marker, which detects apoptotic nuclei with condensed and/or fragmented DNA. Significant intense Hoechst 33342-stained neurons were observed in serum-deprived culture (75.2%; Fig. 1, Al and B) but not in serum-containing (4.9%; Fig. 1, Ai and B; 5.6%; Fig. 1, Ak and B) and serum-free culture (10.0%; Fig. 1, Aj and B). The conversion of procaspase-3 to active caspase-3 is generally accepted to be the most reliable indicator of apoptosis (Green and Reed, 1998). As shown in Fig. 1Ap, significant active caspase-3 signal was observed in serum-deprived culture (57.9%; Fig. 1, Ap and B) but not in serum-containing (1.4%; Fig. 1, Am and B; 3.3%; Fig. 1, Ao and B) and serum-free culture (6.6%; Fig. 1, An and B). There were no significant differences in PI-, Hoechst 33342-, and active caspase-3 staining between neurons in serum-containing culture at 12 h after the start of culture and neurons in the same condition at 3 days after the start of culture (Fig. 1A).
As shown in Fig. 2A, most neurons in serum-free culture rapidly became PI positive by 3 h after the start of the culture. Density of PI-positive neurons reached plateau at 12 to 24 h. On the other hand, approximately 10% of Hoechst 33342-positive neurons were found throughout the time course up to 24 h (Fig. 2A). In the serum-deprived culture, the population of significant Hoechst 33342-positive neurons was gradually increased from 6 h and reached to 93.6% at 36 h. PI-positive neurons were as low as 1.9 to 5.2% throughout the time course up to 36 h (Fig. 2B).
TEM Analysis of Neurons in the Serum-Free and Serum-Deprived Culture. In the TEM analysis, the neuron 12 h after the start of serum-containing culture looked intact in the nucleus and mitochondria (Fig. 3, Aa and Ab); the neuron in serum-free culture, on the other hand, had disrupted plasma membranes, grossly swollen mitochondria that had no clear cristae structures (Fig. 3Bb), and decreased electron density in the cytoplasm (Fig. 3, Ba and Bb), as previously reported (Fujita and Ueda, 2003). Good contrast was observed with the neuron in the serum-deprived culture at 12 h after deprivation, which had intact mitochondria without swelling (Fig. 3Cb) but marked nuclear fragmentation with some condensed parts (Fig. 3Ca).
Insulin-Induced Neuroprotection in the Serum-Free Culture. The viability of neurons measured with WST-8 assay, a marker of mitochondrial reduction activity, decreased to 28.2% of the control (0 time) at 12 h after the start of culture in serum-free condition. When insulin at concentrations of 0.1 to 100 μg/ml was added to the culture in the beginning of culture, the survival activity was concentration-dependently increased (Fig. 4A). Most neurons with insulin at 100 μg/ml appeared healthy and had almost normal thickness under the phase contrast microscope (data not shown). However, neither BDNF nor IGF-I up to 100 μg/ml showed significant increase (Fig. 4A). These neurotrophic factors (insulin, BDNF, and IGF-I), however, protected neuronal death in the serum-deprived culture at as low as 100 ng/ml (data not shown), as reported elsewhere (Carlson et al., 1999; Hetman et al., 1999). In the trypan blue exclusion test, another indicator of survival activity, significant survival activity of insulin and recombinant insulin at 100 μg/ml was also observed (Fig. 4B).
Insulin-induced neuroprotection was also observed when the PI staining was evaluated. As mentioned above, PI staining was used for evaluation of membrane disruption, a death (necrosis) marker. Insulin and recombinant insulin at a concentration of 100 μg/ml significantly inhibited the PI staining (Fig. 4, Ca-Cc and D). The population of PI-positive cells in the serum-free culture was 70.8% of 500 to 600 cells assessed at 12 h after seeding (Fig. 4D). As shown in Fig. 4D, insulin (100 μg/ml) reduced the PI staining by approximately 50%. However, insulin and recombinant insulin had no effect on Hoechst 33342 staining (Fig. 4, Cd–Cf and D).
Lack of Effect of Insulin on Glucose Uptake and Cellular ATP Levels. To examine whether the potential cellular mechanisms related to energy failure are involved in the insulin-induced neuroprotection in the serum-free culture, [3H]2-DG uptake and cellular ATP levels were assessed (Fig. 5, A and B). In the serum-free culture, [3H]2-DG uptake within 2 h after the start of culture was decreased to as low as 33.5% of the serum-containing culture. Insulin did not affect [3H]2-DG uptake at concentrations up to 100 μg/ml (Fig. 5A). On the other hand, cellular ATP levels in the serum-free culture also rapidly decreased, whereas there was no decrease in ATP levels in the presence of serum, as shown in Fig. 5B. Insulin at 100 μg/ml did not affect the cellular ATP levels (Fig. 5B).
Insulin-Induced Neuroprotection in the Serum-Free Culture through PKC Mechanisms. The insulin-induced survival activity measured with WST-8 assay was significantly inhibited by the PKC inhibitors calphostin C or GF109203X at 1 μM, tyrosine kinase inhibitors herbimycin A (10 μM) or genistein (1 μM), and phospholipase C (PLC) inhibitor U73122 (1 μM) 12 h after the start of culture. However, the PI3-K inhibitor wortmannin (1 μM), MAPKK inhibitors PD98059 or U0126 (1 μM), and an inactive analog of U73122, U73343 (1 μM) failed to inhibit the insulin-induced neuroprotection (Fig. 6A). These inhibitors had no effect on the basal survival activity of neurons in the serum-free culture (data not shown). As shown in Fig. 6Ba, insulin at as high as 100 μg/ml significantly increased the PKC activity by 13% of the maximum activation induced by excess amounts of phorbol 12-myristate 13-acetate. This effect was also significantly inhibited by 1 μM of calphostin C, 3 μM of herbimycin A, and 1 μM of U73122 at 2 h after the start of culture, whereas 1 μM of U73343 and 1 μM of wortmannin showed no effects (Fig. 6Bb). Similar results were also observed 12 h after the start of culture (data not shown). As shown in Fig. 6C, insulin-induced decrease in the population of PI-positive neurons was also blocked by 1 μM of calphostin C, 3 μM of herbimycin A, and 1 μM of U73122 but not affected by 1 μM of U73343 and 1 μM of wortmannin.
Effects of Insulin and BDNF on the Phosphorylation of Akt and ERK1/2. As shown in Fig. 7A, the phosphorylation of Akt but not ERK1/2 was enhanced by insulin or BDNF at 100 μg/ml in the serum-free culture. In the serum-deprived culture, on the other hand, the application of insulin or BDNF at as low as 100 ng/ml induced activation of both of Akt and ERK1/2 (Fig. 7B), as reported elsewhere (Hetman et al., 1999; Ryu et al., 1999).
PKC Involvement in the Insulin-Induced Inhibition of Necrosis in TEM Analysis. The mitochondrial swelling and loss of electron density in the cytoplasm due to membrane disruption in the serum-free culture, as shown in Fig. 3, were completely abolished by the treatment with insulin (100 μg/ml), in a calphostin C (1 μM)-reversible manner (Fig. 8, Aa and Ab and Ba and Bb). As shown in Fig. 8, Ca and Cb, wortmannin (1 μM) did not affect the insulin-induced necrosis inhibition.
Necrosis is characterized as a more passive process due to energy failure, which leads to loss of ionic homeostasis, membrane disruption, and mitochondrial swelling. Compared with the studies of apoptosis, little is known of the mechanisms for necrosis, which may give us the strategies to develop antinecrotic drugs. In the previous studies, we developed a new approach to causing neuronal necrosis by performing serum-free cultures of cerebral cortex without any supplements (Fujita et al., 2001). We have proposed that necrosis might be caused by rapid decreases in glucose uptake and cellular ATP levels (Fujita and Ueda, 2003), which might be related to the disappearance of glucose transporters (GLUT1 and GLUT4) from plasma membranes (R. Fujita and H. Ueda, manuscript in preparation). From these findings, we initiated the experiments to examine whether the neuronal necrosis might be protected by insulin, which translocates glucose transporters to the plasma membrane of peripheral cells, such as muscle cells or adipocytes (Pessin and Saltiel, 2000; Saltiel, 2001; Bose et al., 2002). As expected, insulin showed marked neuroprotective actions (Fig. 4). However, this peptide did not ameliorate the rapid decrease in glucose uptake and cellular ATP levels (Fig. 5, A and B). These findings suggest that the neuroprotective actions of insulin in this culture system are supposed to use different cellular mechanisms.
In the present study, we characterized necrosis by the use of selective staining with PI or Hoechst 33342 as well as TEM analysis. In the serum-free culture, most neurons are PI positive, but very few are intensely stained with Hoechst 33342 or anti-active caspase-3 antibody. Good contrast is observed with neurons in the serum-deprived culture, which are intensely stained with Hoechst 33342. As previously reported (Raff, 1998; Hetman et al., 2000; Barber et al., 2001), serum deprivation is one of the proapoptotic stresses. Differentiated cells, including neurons in the central nervous system, require the presence of survival factors to support the intrinsic cell death machinery and thereby avoid apoptosis. From the time course study, the PI staining was mostly completed at 3 h after the start of serum-free culture, whereas the Hoechst 33342 staining showing apoptosis was gradually increased until 36 h after serum deprivation (Fig. 2B). Thus, it is evident that PI and Hoechst 33342 staining are good tools to characterize necrosis and apoptosis, respectively. The validity of this view was clearly proved by TEM analysis. Neurons in the serum-free culture showed loss of electron density in the plasma due to membrane disruption and mitochondrial swelling (Fig. 3, Ba and Bb). The former phenomena may link to the intracellular penetration of hydrophilic PI (Bal-Price and Brown, 2000). On the other hand, neurons in the serum-deprived culture showed intact mitochondria and intact plasma membrane but with nuclear fragmentation (Fig. 3, Ca and Cb).
As shown in Fig. 4A, insulin increased the survival activity in a concentration-dependent manner, with the effective concentrations as high as 10 and 100 μg/ml, which is approximately 1,000 times higher than the concentrations required for antiapoptotic actions in different culture systems (Carlson et al., 1999; Hetman et al., 1999; Barber et al., 2001). However, it is unlikely that insulin exerts such antinecrotic actions through the IGF-I receptor, which has low affinity to insulin, since IGF-I at as high as 100 μg/ml had no significant effect (Fig. 4A). It is also unlikely that small amounts of unidentified molecules contained in the purified insulin (derived from bovine) are in charge of this action, since the recombinant peptide showed neuroprotective and antinecrotic actions (Fig. 4, B, C, and D).
It is well known that various insulin actions are closely related to the cellular protein kinase activities. Recent reports suggest that glucose uptake by insulin is closely related to the phosphorylation of glucose transporters by PKC (Pessin and Saltiel, 2000; Watson and Pessin, 2001), whereas its cell growth or survival activity against apoptosis is related to PI3-K, Akt, or ERK1/2 (Barber et al., 2001, Watson and Pessin, 2001). In the present study, however, the neuroprotective activity of insulin was only inhibited by tyrosine kinase inhibitors, PKC inhibitors, or PLC inhibitor but not by PI3-K inhibitor and MAPKK inhibitors, which are known to block insulin-induced signaling related to antiapoptotic actions (Gauthier et al., 2001; Johnson and Denton, 2003). As the insulin-induced PKC activation was blocked by herbimycin A and U73122, it is suggested that the neuroprotection is mediated by tyrosine kinase and PLC activation followed by PKC activation, which are also known as downstream mechanisms of insulin receptor (Slaaby et al., 2000; Eichhorn et al., 2002; Lorenzo et al., 2002). The involvement of tyrosine kinase, PLC, and PKC mechanisms in the insulin actions were confirmed in the inhibition of PI staining as a necrosis marker (Fig. 6C). The lack of action by PI3-K and MAPKK inhibitors may be attributed to the fact that the neuronal death that occurred in the serum-free culture is characterized as necrosis resistance to known neurotrophic factors, such as BDNF or IGF-I (Fig. 4A). Although the necessity of high concentrations of insulin remains to be determined, it might be related to the unique finding that insulin-induced survival activity was selectively inhibited by PLC and PKC inhibitors, but not by inhibitors of PI3-K and MAPKK, in the serum-free culture system. Indeed, we observed that high concentrations of insulin were also required for the PKC activation in this system (Fig. 6Ba).
Calphostin C, a PKC inhibitor, abolished the neuroprotective and antinecrotic actions of insulin in the WST-8 assay, PI staining, and TEM analysis (Fig. 6, A and C; Fig. 8). However, the molecular targets for PKC in relation to the antinecrotic actions following insulin stimulation remain to be determined. Glucose transporters observed in muscle or adipocytes (Bose et al., 2002) are unlikely to be related, since insulin did not show significant changes in glucose uptake or in cellular ATP levels (Fig. 5, A and B). The stimulation of Na+-K+ ATPase by insulin might be a candidate for the antinecrotic mechanisms, since energy failure related to necrosis is coupled to the reduction of Na+-K+ ATPase activity, which leads to a loss of ionic homeostasis (Xiao et al., 2002), followed by mitochondrial swelling (Halestrap et al., 2002) and membrane disruption (Malis and Bonventre, 1988), and this pump is stimulated by PKC mechanisms (Chibalin et al., 2001; Sweeney et al., 2001). Molecular identification of PKC targets in relation to the insulin-induced antinecrosis would be the next important goal.
In conclusion, we provided evidence that insulin protects neuronal necrosis resistance to known neurotrophic factors under serum-free starvation stress through PKC activation.
We thank Sumitomo Pharmaceutical (Osaka, Japan) for the gift of BDNF and T. Suematsu (Nagasaki University) for technical help in TEM analysis. Parts of this study were supported by grants-in-aid and Special Coordination Funds from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
ABBREVIATIONS: BDNF, brain-derived neurotrophic factor; PI, propidium iodide; TEM, transmission electron microscopy; IGF-I, insulin-like growth factor-I; PKC, protein kinase C; PI3-K, phosphatidyl inositol-3-OH-kinase; MAPKK, mitogen-activated protein kinase kinase; ERK, extracellular signal-activated protein kinase; PLC, phospholipase C; PBS, phosphate-buffered saline; WST-8, 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt; PFA, paraformaldehyde; DTT, dithiothreitol; [3H]2-DG, 2-deoxy-d-[3H]glucose.
- The American Society for Pharmacology and Experimental Therapeutics