- Email: [email protected]
G Protein and cAMP-Dependent Protein Kinase Mediate Amyloid β-Peptide Inhibition of Neuronal Glucose Uptake
Experimental Neurology 167, 59 – 64 (2001) doi:10.1006/exnr.2000.7519, available online at http://www.idealibrary.com on
G Protein and cAMP-Dependent Protein Kinase Mediate Amyloid -Peptide Inhibition of Neuronal Glucose Uptake 1 Teerasak Prapong, Etsuro Uemura, and Walter H. Hsu Department of Biomedical Sciences, 2008 Veterinary Medicine Building, Iowa State University, Ames, Iowa 50011 Received July 12, 1999; accepted July 19, 2000
utilization ranges from 19% in mild cases to 40% and more in severe cases of AD (11, 17). Immunohistochemical studies revealed a decrease in glucose transporters (GLUT1, GLUT3) in several brain regions, including the hippocampus, the area commonly affected in AD patients (9, 36). Reduced glucose uptake by individuals genetically at risk for developing AD also supports the notion that glucose deprivation precedes neuronal degeneration (16, 27, 30). In healthy, nonstarved mammalian brain, glucose is the only substrate for the formation of energy in the form of ATP (5). Glucose deprivation results in increased use of endogenous substrates, depletion of ATP, membrane depolarization, extracellular accumulation of excitatory amino acids, loss of neuronal homeostasis, and ultimately neuronal cell death (21). Neurons interfered with glucose uptake or exposed to A result in impaired mitochondrial activity and suppressed production of ATP (21). Based on the rates of oxidized glucose and oxygen, it was estimated that the rate of cerebral ATP formation decreases by 7% in early-onset AD and 20% in late-onset AD (10). Such metabolic impairment increases neuronal vulnerability to glutamate toxicity and oxidative insults (21). There are numerous published reports linking A to neurotoxicity in vivo and in vitro. However, how A inhibits glucose uptake is not known. It was shown that A induces conjugation of 4-hydroxynonenal (HNE), the product of lipid peroxidation, with neuronal transporter GLUT3, resulting in impaired glucose transport (20). Subsequent studies in synaptosomes from the rat cerebral cortex showed that A and HNE impair glucose transport and severely compromise mitochondrial function (15). These studies suggest that A-mediated oxidative stress inhibits neuronal glucose uptake. In the present study, we found that A inhibition on neuronal glucose uptake involves Gs–adenylyl cyclase coupling and activation of cAMP-dependent protein kinase (PKA), suggesting the presence of alternative mechanism by which A inhibits glucose uptake in cultured hippocampal neurons.
The mechanism by which amyloid -peptide (A) inhibits glucose uptake in cultured cells is not known. Here we demonstrated a signaling pathway in which A25-35, a neurotoxic portion of the A peptide corresponding to amino acids 25–35, inhibits neuronal glucose uptake by hippocampal neurons. The GP antagonist-2, which blocks Gs, prevented the inhibitory effect of A on the glucose uptake. Exposure of cells to A resulted in a transitory increase in intracellular levels of cAMP. To assess the role of cAMP in neuronal glucose uptake, cultured neurons were exposed to dibutyryl cAMP (Bt 2cAMP) or an adenylyl cyclase activator, forskolin. Both Bt 2cAMP and forskolin inhibited neuronal glucose uptake, and cAMP-dependent protein kinase (PKA) inhibitor KT5720 blocked the A-mediated inhibition of glucose uptake. Cholera toxin, which stimulates adenylyl cyclase by activating Gs protein, also inhibited neuronal glucose uptake, and A potentiated this inhibitory effect of cholera toxin on glucose uptake. Thus, our findings suggest that A inhibits glucose uptake by activating the Gscoupled receptors and involves the cAMP–PKA system. © 2001 Academic Press Key Words: Alzheimer’s disease; -amyloid; glucose uptake; hippocampal neuron; G protein; cAMP.
INTRODUCTION
Alzheimer’s disease (AD) is characterized by deposition of amyloid -peptide (A), a 39- to 43-amino-acid protein (40). It has been shown that A impairs glucose uptake in cultured hippocampal neurons (20) and astrocytes (26); however, it is not known whether A affects glucose uptake in the brain of AD patients. There is some evidence suggesting that glucose metabolism is changed in the brains of AD patients (4, 12, 29, 31). It was calculated that reduction of cerebral glucose 1
This study was supported by a special research initiation grant from the Iowa State University Graduate College and the Thai government. 59
0014-4886/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
60
PRAPONG, UEMURA, AND HSU
MATERIALS AND METHODS
Hippocampal neurons from 17E Sprague–Dawley rats were cultured according to Abe and Kimura (1); however, we used N2 instead of N1 supplement in our cultures. A25-35 and scrambled control peptide were prepared according to Parpura-Gill et al. (26). Neurons plated onto 24-well tissue culture plates were exposed to A25-35 or scrambled control peptide by using coverslips coated with the peptides. The coated-side was placed on top of the cultured neurons at postplating day 4. To examine whether A inhibition of neuronal glucose uptake is mediated by a G protein, cultured hippocampal neurons at day 4 postplating were exposed to synthetic A25-35 for 24 h with or without 10 M GP antagonist-2 (Biomol) in the culture medium. To rule out the possibility of nonspecific effects of GP antagonist-2, neuronal glucose uptake was also determined in the presence of 10 ⫺4 M cytochalasin B, an inhibitor of glucose transporters. Cholera toxin, which stimulates adenylyl cyclase by activating Gs (32), was used to further examine the role of Gs on the A inhibition on neuronal glucose uptake. Neuronal glucose uptake was assessed at day 4 postplating following exposure to cholera toxin (0.01, 0.1, or 1 g/ml) in the medium for 24 h. Two low doses of cholera toxin (0.01, 0.1 g/ml) were further used to study whether cholera toxin potentiates the inhibitory effect of A in neuronal glucose uptake. To examine whether PKA plays a regulatory role in neuronal glucose uptake, neurons at day 4 postplating were exposed to a PKA inhibitor KT-5720 (0.1 and 1 M) in the culture medium and A25-35 on coverslips for 24 h. To rule out the possibility of nonspecific effects of KT-5720, neuronal glucose uptake was also determined in the presence of 10 ⫺4 M cytochalasin B. Neuronal uptake of glucose was assayed by using 14 D-[ C]glucose (313 Ci/mmol, ICN) at final glucose concentrations of 600 nM (0.1 Ci/500l medium in well). The culture was incubated in the Hepes medium at 37°C for 1 h before adding radioactive glucose into the culture (38). Glucose uptake was stopped after 0 and 60 min of incubation by three washes with ice-cold Hepes medium prepared from 0.2383 g. of Hepes in 100 ml MEM. Time 0 was used to determine the background uptake of D-[ 14C]glucose. Cell lysates were used for determination of radioactivity by a beta liquid scintillation counter (Packard, 1600TR) and for assay of total cell protein (Bio-Rad protein assay kit). Glucose uptake was expressed as pmol of glucose/mg of cellular protein. The effect of A on intracellular cAMP levels was studied in neurons at day 4 postplating by exposing them to A25-35 or scrambled control peptide for 2 h, 2 days, or 3 days. To assess the role of cAMP in neuronal glucose uptake, neurons at day 4 postplating were also assessed for glucose uptake at 0 and 1 h following
exposure for 1 h to either Bt 2cAMP (0, 0.5, 1, 5, 10, 50 M) or forskolin (0, 10, 25, 50, 100 M). The cAMP content of cultured neurons was determined with the acetylation protocol of 125I cAMP radioimmunoassay (Amersham). For acute effect of -amyloid, neurons were cultured on PEI-coated plates for 24 h, and then neurons were exposed to 10 M A25-35 or scrambled peptide for 24 h. Before harvesting cells, 10 M of forskolin was added in some of the wells as positive control. The wells were washed with ice cold 0.1 M PBS for two times, and then 0.3 ml of 0.1 N NaOH was added. The culture plates were frozen at ⫺20°C for at least 2 h before the next process was performed. The frozen cell culture plates were thawed on an ice bath and were scraped with scrapers. The cell lysate was transferred into new glass tubes and was boiled at 70°C for 10 min to inactivate phosphodiesterase. The sample tubes were centrifuged at 3000 rpm for 5 min at 4°C. The supernatants were transferred into new tubes. Supernatants were neutralized with 0.04 N HCl at the ratio of 200 l of supernatant:50 l of HCl. Neutralized supernatants were acetylated with acetylation reagent (acetic anhydride ⫹ triethylamine). Standards of cAMP (4, 8, 16, 32, 64, 128, 256 fmol) were prepared. Anti-cAMP (1:70,000) and 125I cAMP tracer (5000 cpm/tube) were added to the acetylated samples (100 l) and incubated at 4°C overnight. To the samples 100 l of normal rabbit plasma and 100 l of second antibody were added. Cold PEG was added and samples were spun at 3000 rpm for 20 min. The supernatant was aspirated and tube was counted with ␥-counter. Total cell protein was assayed (Bio-Rad protein assay kit) and cAMP levels were expressed as femtomoles per milligram of protein. Each experiment was replicated three times (n ⫽ 3) with four wells per condition per replication. All pups from each pregnant rat were used as an experimental unit. Data were analyzed for treatment differences by analysis of variance and intertreatment differences were determined by post hoc test (Tukey’s). RESULTS
To examine whether a G protein mediates A inhibition of neuronal glucose uptake, hippocampal neurons were exposed to GP antagonist-2 (25), which blocks the receptor–Gs interaction. In this study, cultured hippocampal neurons at day 4 postplating were exposed to synthetic A25-35 for 24 h with or without 10 M GP antagonist-2 in the culture medium. The neuronal glucose uptake was measured and expressed as picomoles of D-[ 14C]glucose per milligram of cellular protein (26). A25-35 significantly inhibited neuronal glucose uptake [F(3, 8) ⫽ 66.8, P ⫽ 0.0001] (Fig. 1a). Those neurons exposed to A25-35 showed 40% less glucose uptake compared with the control. However, the GP antagonist-2 prevented this inhibitory effect of
AMYLOID -PEPTIDE INHIBITION OF NEURONAL GLUCOSE UPTAKE
FIG. 1. G protein mediates A inhibition of neuronal glucose uptake. A, amyloid -peptide; Ga, GP antagonist-2; Ctx1, 1 g/ml cholera toxin; Ctx.1, 0.1 g/ml cholera toxin; Ctx.01, 0.01 g/ml cholera toxin. (a) Cultured hippocampal neurons at day 4 postplating were exposed to synthetic A25-35 for 24 h with or without 10 M GP antagonist-2 in the culture medium. Those neurons exposed to A25-35 showed 40% less glucose uptake compared with the control. However, GP antagonist-2 prevented this inhibitory effect of A (P ⫽ 0.001). (b) Neuronal glucose uptake was assessed at day 4 postplating following exposure to cholera toxin (0.01, 0.1, or 1 g/ml) in the medium for 1 h. Cholera toxin at 0.01 and 0.1 g/ml had no effect on neuronal glucose uptake, but 1 g/ml cholera toxin significantly inhibited glucose uptake (P ⫽ 0.001). (c) Two low doses of cholera toxin (0.01, 0.1 g/ml) were used to study whether cholera toxin potentiates the inhibitory effect of A in neuronal glucose uptake. Neurons at day 4 postplating were exposed for 24 h and cholera toxin at either 0.01 or 0.1 g/ml in the culture medium. As noted above (b), neurons exposed to A25-35 showed significantly less glucose uptake than the control. Neurons exposed to both A25-35 and 0.01 g/ml cholera toxin did not differ in their glucose uptake from those exposed to A25-35 alone. However, those neurons exposed to both A25-35 and 0.1 g/ml cholera toxin showed 26% less glucose uptake than those exposed to A25-35 alone (P ⫽ 0.0004).
A, suggesting that G protein mediates A inhibition of neuronal glucose uptake. The effect of GP antagonist-2 was specific, as it did not prevent the inhibition of glucose uptake by cytochalasin B. To further assess the role of Gs on the inhibitory effect of A on neuronal glucose uptake, the effect of cholera toxin, which stimulates adenylyl cyclase by
61
activating Gs (32), was studied in cultured hippocampal neurons. In this study, neuronal glucose uptake was assessed at day 4 postplating following exposure to cholera toxin (0.01, 0.1, or 1 g/ml) in the medium for 1 h. Cholera toxin at 0.01 and 0.1 g/ml had no effect on neuronal glucose uptake, but 1 g/ml cholera toxin significantly inhibited glucose uptake [F(3, 8) ⫽ 38.63, P ⫽ 0.001] (Fig. 1b). Based on this observation, the two low doses of cholera toxin (0.01, 0.1 g/ml) were used to study whether this toxin potentiates the inhibitory effect of A. In this study, neurons at day 4 postplating were exposed for 24 h to both A25-35 coated on coverslips and cholera toxin at either 0.01 or 0.1 g/ml in the culture medium. Those neurons exposed to A25-35 showed 38% less glucose uptake than the control (t ⫽ 3.78, P ⫽ 0.0001). When neurons were exposed to both A25-35 and 0.1 g/ml cholera toxin, their glucose uptake was 53% less than the control (t ⫽ 5.99, P ⫽ 0.001) and 26% less than those exposed to A25-35 alone (t ⫽ 4.87, P ⫽ 0.0004) (Fig. 1c). Thus, cholera toxin potentiated the effect of A2535, suggesting that A inhibits glucose uptake by activating the Gs-coupled receptors. Since Gs activates adenylyl cyclase (33), we studied the effect of A on intracellular cAMP levels in cultured hippocampal neurons. In this study, neurons at day 4 postplating were exposed to A25-35 for 2 h, 2 days, or 3 days. Neurons were analyzed for cAMP content using radioimmunoassay (Amersham). A significant change in intracellular cAMP levels was observed in neurons exposed to A [F(1, 12) ⫽ 75.5, P ⫽ 0.0001] and this change was time dependent [F(2, 12) ⫽ 19.3, P ⫽ 0.0002] (Fig. 2a). The highest cAMP level was found in neurons that were exposed to A for 2 h, followed by those neurons exposed to A for 48 h cAMP levels at 72 h postexposure were not different from the control. An elevation of intracellular cAMP levels by A25-35 within 2 h of application suggests that A stimulates Gs–adenylyl cyclase coupling. To further assess the role of cAMP in neuronal glucose uptake, we studied the effect of cAMP on neuronal glucose uptake in hippocampal neurons. At postplating day 4, cultured neurons were exposed for 1 h to either Bt 2cAMP (0, 0.5, 1, 5, 10, 50 M) or forskolin (0, 10, 25, 50, 100 M). Neurons were assessed for glucose uptake and cellular protein at 0 and 1 h posttreatment. Neuronal glucose uptake was significantly inhibited by Bt 2cAMP [F(5, 18) ⫽ 9.07, P ⫽ 0.0002] (Fig. 2b) or forskolin [F(4, 15) ⫽ 4.68, P ⫽ 0.0006] (Fig. 2c). This inhibitory effect was dose dependent for both Bt 2cAMP (P ⫽ 0.0001, r 2 ⫽ 0.54) and forskolin (P ⫽ 0.0001, r 2 ⫽ 0.58). Neurons exposed to either 50 M Bt 2cAMP or 100 M forskolin showed only 13% of glucose uptake by the control neurons. Since A elevated intracellular cAMP levels within 2 h, we also assessed the inhibitory effect of A on glucose uptake following exposure for
62
PRAPONG, UEMURA, AND HSU
FIG. 2. cAMP mediates A inhibition of neuronal glucose uptake. A, amyloid -peptide. (a) Neurons at day 4 postplating were analyzed for cAMP content following exposure to A25-35 or scrambled control peptide for 2 h, 2 days, or 3 days. A significant change in intracellular cAMP levels was observed in neurons exposed to A (P ⫽ 0.0001) and this change was time dependent (P ⫽ 0.0002). The highest cAMP level was found in neurons exposed to A for 2 h. The cAMP level had fallen to the control level by 72 h. (b, c) Neurons at day 4 postplating were assessed for glucose uptake and cellular protein at 0 and 1 h following exposure for 1 h to either Bt 2cAMP (0, 0.5, 1, 5, 10, 50 M) or forskolin (0, 10, 25, 50, 100 M). Neuronal glucose uptake was significantly inhibited by Bt 2 cAMP (P ⫽ 0.0002) (b) or forskolin (P ⫽ 0.0006) (c). This inhibitory effect was dose dependent for both Bt 2 cAMP and forskolin. (d) Since A25-35 elevated cAMP by 2 h (a), neurons at day 4 postplating were assessed for glucose uptake at 0 and 1 h following exposure to either A25-35 or forskolin (10 M) for 3 h. Glucose uptake was significantly inhibited by A25-35 (P ⫽ 0.0025), suggesting an acute inhibitory effect of A25-35.
3 h to A25-35. Glucose uptake was significantly inhibited by A25-35 (t ⫽ 2.97, P ⫽ 0.025) (Fig. 2d), suggesting an acute inhibitory effect of A25-35. Inhibition of neuronal glucose uptake by Bt 2cAMP and forskolin suggests that glucose uptake is regulated by cAMP-dependent PKA. To examine whether PKA plays a regulatory role in neuronal glucose uptake, the effect of a PKA inhibitor KT-5720 (0.1 and 1 M) was studied in cultured hippocampal neurons. Neuronal glucose uptake and cellular protein were assessed at 1 h post-sampling time. Glucose uptake by control neurons was 40 pmol/mg cellular protein (Fig. 3). In contrast, A25-35 significantly reduced glucose uptake to 15 pmol/mg cellular protein, i.e., 38% of the control neurons. This inhibitory effect of A25-35 was prevented by 1 M KT-5720, but not by 0.1 M KT-5720. This dose-dependent effect of the PKA inhibitor was highly significant [F(3, 16) ⫽ 8.62, P ⫽ 0012]. The result suggests an inhibitory role by PKA on neuronal glucose uptake. The effect of KT-5720 was specific, as 1 M KT-5720 did not prevent the inhibition of glucose uptake by cytochalasin B. DISCUSSION
The present study suggests that a G protein and cAMP mediate A25-35 inhibition of neuronal glucose
uptake. A role played by Gs in neuronal glucose uptake was shown by the use of GP antagonist-2 and two adenylyl cyclase activators, cholera toxin and forskolin, all of which inhibited neuronal glucose uptake. Cholera toxin also potentiated the inhibitory effect of A. Although, GP antagonist-2 interacts with a part of the receptor-binding domain on G proteins to block the effects of receptor in reconstituted vesicles (25), its effect on cultured cells is not known. Thus, use of this agent does not present a definitive, but provides a compelling case for Gs-coupled-receptor mechanisms mediating the A effect. Gs elevates intracellular cAMP by activating adenylyl cyclase (7, 33). In our study, A acutely increased intracellular cAMP and an addition of Bt 2cAMP in culture medium inhibited neuronal glucose uptake, suggesting the importance of cAMP in neuronal glucose uptake. To date the role of cAMP in the regulation of neuronal glucose is not known. Elevation of intracellular cAMP by forskolin was shown to inhibit the glucose uptake by neurons (8). Similarly, Bt 2cAMP inhibits glucose uptake in L6 myoblast (19) and hamster ovary cells (28). A inhibition of neuronal glucose uptake was acute and such inhibition was not associated with changes in the number of GLUT3 in cells (20). cAMP is known to activate PKA and the course of events that follows in a given cell is determined by the specific proteins phosphorylated by
AMYLOID -PEPTIDE INHIBITION OF NEURONAL GLUCOSE UPTAKE
63
the importance of glucose in the CNS function, the effect of A on neuronal glucose metabolism in AD warrants further study. REFERENCES 1.
2.
3. FIG. 3. PKA mediates A inhibition of neuronal glucose uptake. A, amyloid -peptide; PKAi, PKA inhibitor. Neurons at day 4 postplating were analyzed for glucose uptake and cellular protein following exposure to KT-5720 (0.1 or 1 M) in the culture medium and A25-35 or scrambled control peptide on coverslips for 24 h. Glucose uptake by control neurons was 40 pmol/mg cellular protein. In contrast, A25-35 significantly reduced glucose uptake to 15 pmol/mg cellular protein, i.e., 38% of the control neurons. This inhibitory effect of A25-35 was prevented by 1 M KT-5720, but not by 0.1 M KT-5720. This dose-dependent effect of PKA was highly significant (P ⫽ 0.0012).
4.
5. 6.
7.
this enzyme (14). In adipocytes, PKA-triggered phosphorylation of GLUT4 inhibits insulin-stimulated glucose transport (13, 18). In the present study, PKA inhibitor KT5720 blocked the A-mediated inhibition of glucose uptake. Whether A alters phosphorylation of GLUT3 is not known. How these in vitro data relate to the decreased glucose metabolism in AD is not certain. It is possible that the decreased cerebral glucose uptake shown in AD (4, 12, 29, 31) may reflect, in part, the inhibitory effect of A on neuronal and astrocytic glucose uptake (20, 26). Such a hypothesis infers the presence of functional G-protein-stimulated activity of adenylyl cyclase and cAMP in AD patients. Postmortem studies of AD patients have shown normal G protein levels (3, 22, 32, 39), catalytic activity of cAMP, and PKA (2, 6, 23, 34). However, the regulatory function of G protein on adenylyl cyclase was reported as either normal (32) or reduced (3, 6, 35). This inconsistency may reflect differences in postmortem instability of the enzyme at different stages of Alzheimer’s disease (2). The regulatory function of G protein on adenylyl cyclase could also be affected only at the terminal stage of AD disease. A inhibition of neuronal glucose uptake could also be mediated by Ca 2⫹/calmodulin-sensitive adenylyl cyclase and an increase cAMP levels, the process known to occur in the hippocampus (24). However, Ca 2⫹/calmodulin-mediated inhibition of glucose uptake is less likely because it was shown that A-induced apoptosis was not inhibited by KN-62, an inhibitor of Ca 2⫹/calmodulin-dependent protein kinase (37). Considering
8.
9.
10.
11.
12.
13.
14. 15.
16.
Abe, K., and H. Kimura. 1996. Amyloid  toxicity consists of a Ca 2⫹-independent early phase and a Ca 2⫹-dependent late phase. J. Neurochem. 67: 2074 –2078. Bonkale, W. L., J. Fastbom, B. Wiehager, R. Ravid, B. Winblad, and R. F. Cowburn. 1996. Inpaired G-protein-stimulated adenylyl cyclase activity in Alzheimer’s disease brain is not accompanied by reduced cyclic-AMP-dependent protein kinase A activity. Brain Res. 737: 155–161. Cowburn, R. F., C. O’Neill, R. Ravid, I. Alafuzoff, B. Winblad, and C. J. Fowler. 1992. Adenylyl cyclase activity in postmortem human brain: evidence of altered G protein mediation in Alzheimer’s disease. J. Neurochem. 58: 1409 –1419. Duara, R., C. Grady, J. Haxby, M. Sundaram, N. R. Cutler, L. Heston, A. Moore, N. Schlageter, S. Larson, and S. I. Rapoport. 1986. Positron emission tomography in Alzheimer’s disease. Neurology 36: 879 – 887. Erecinska, M., and I. A. Silver. 1989. ATP and brain functon. J. Cereb. Blood Flow Metab. 9: 2–19. Fowler, C. J., R. F. Cowburn, A. Garlind, B. Winblad, and C. O’Neill. 1995. Disturbances in signal transduction mechanisms in Alzheimer’s disease. Mol. Cell. Biochem. 149: 287–292. Gilman, A. G. 1987. G proteins: Transducers of receptor-generated signals. Annu. Rev. Biochem. 56: 615– 649. Hara, M., Y. Matsuda, K. Hirai, N. Okumura, and H. Nakagawa. 1989. Characteristics of glucose transport in neuronal cells and astrocytes from rat brain in primary culture. J. Neurochem. 52: 902–908. Harr, S. D., N. A. Simonian, and B. T. Hyman. 1995. Functional alterations in Alzheimer’s disease: Decreased glucose transporter 3 immunoreactivity in the perforant pathway terminal zone. J. Neuropathol. Exp. Neurol. 54: 38 – 41. Hoyer, S. 1992. Oxidative energy metabolism in Alzheimer brain: Studies in early-onset and late onset cases. Mol. Chem. Neuropathol. 16: 207–224. Hoyer, S., R. Nitsch, and K. Oesterreich. 1991. Predominant abnormality in cerebral glucose utilization in late-onset dementia of the Alzheimer type: A cross-sectional comparison against advanced late-onset and incipient early-onset cases. J. Neural Transm. 3: 1–14. Jagus, W. J., J. P. Seab, R. H. Huesman, P. E. Valk, C. A. Mathis, B. R. Reed, P. G. Coxson, and T. F. Budinger. 1991. Diminished glucose transport in Alzheimer’s disease: Dynamic PET studies. J. Cereb. Blood Flow Metab. 11: 323–330. James, D. E., J. F. Hiken, and J. C. Lawernce, Jr. 1989. Isoproterenol stimulates phosphorylation of the insulin-regulatable glucose transporter in rat adipocytes. Proc. Natl. Acad. Sci. USA 86: 8368 – 8372. Karp, G. 1996. Cell and Molecular Biology. Wiley, New York. Keller, J. N., Z. Pang, J. W. Geddes, A. Germeyer, G. Waeg, and M. P. Mattson. 1997. Amyloid -peptide impairs glucose and glutamate transport, induces oxidative stress in mitochondria, and reduces ATP levels in synaptosomes: Role of the lipid peroxidation product 4-hydroxynonenal. J. Neurochem. 69: 273–284. Kennedy, A. M., R. S. Frackowiak, S. K. Newman, P. M. Bloomfield, J. Seaward, P. Roques, G. Lewington, V. J. Cunningham, and M. N. Rossor. 1995. Deficits in cerebral glucose metabolism demonstrated by positron emission tomography in individuals
64
17.
18.
19.
20.
21.
22.
23.
24. 25. 26.
27.
28.
29.
PRAPONG, UEMURA, AND HSU at risk of familial Alzheimer’s disease. Neurosci. Lett. 186: 17–20. Kumar, A., M. B. Schapiro, C. Grady, J. V. Haxby, E. Wagner, J. A. Salerno, R. P. Friedland, and S. I. Rapoport. 1991. Highresolution PET studies in Alzheimer’s disease. Neuropsychopharmacology 4: 35– 46. Lowrence, J. C. J., J. F. Hiken, and D. E. James. 1990. Phosphorylation of the glucose transporter in rat adipocytes. J. Biol. Chem. 265: 2324 –2332. Lowrence, J. C. J., R. C. Piper, L. J. Robinson, and D. E. James. 1992. GLUT4 facilitates insulin stimulation and cAMP-mediated inhibition of glucose transport. Proc. Natl. Acad. Sci. USA 89: 3493–3497. Mark, R. J., Z. Pang, J. W. Geddes, K. Uchida, and M. P. Mattson. 1997. Amyloid -peptide impairs glucose transport in hippocampal and cortical neurons: Involvement of membrane lipid peroxidation. J. Neurosci. 17: 1045–1054. Mattson, M. 1997. Cellular actions of -amyloid precursor protein and its soluble and fibrillogenic derivatives. Physiol. Rev. 77: 1081–1132. McLaughlin, M., B. M. Ross, G. Milligan, J. McCulloch, and J. T. Knowler. 1991. Robustness of G proteins in Alzheimer’s disease: An immunoblot study. J. Neurochem. 57: 9 –14. Meier-Ruge, W., P. Iwangoff, and K. Erichlmeier. 1984. Neurochemical enzyme changes in Alzheimer’s and Pick’s disease. Arch. Gerontol. Geriatr. 3: 161–165. Mons, N., and D. M. F. Cooper. 1995. Adenylate cyclases: Critical foci in neuronal signaling. TINS 18: 536 –542. Mukai, H., E. Munekata, and T. Higashijima. 1992. G protein antagonists. J. Biol. Chem. 267: 16237–16243. Parpura-Gill, A., D. Beitz, and E. Uemura. 1997. The inhibitory effects of -amyloid on glutamate and glucose uptakes by cultured astrocytes. Brain Res. 754: 65–71. Pettegrew, J. W., K. Panchalingam, W. E. Klunk, R. J. McLure, and L. R. Muenz. 1994. Alterations of cerebral metabolism in probable Alzheimer’s disease. Neurobiol. Aging 15: 117–132. Piper, R. C., D. E. James, J. W. Slot, C. Puri, and J. C. J. Lawrence. 1993. GLUT4 phosphorylation and inhibition of glucose transport by dibutyryl cAMP. J. Biol. Chem. 268: 16557– 16563. Rapoport, S. I., B. Horwitz, C. L. Grady, J. V. Haxby, C. DeCarli, and M. B. Schapiro. 1991. Abnormal Brain Glucose Me-
30.
31.
32.
33. 34.
35.
36.
37.
38.
39.
40.
tabolism in Alzheimer’s Disease, as Measured by Positron Emission Tomography. Plenum, New York. Reiman, E. M., R. J. Caselli, L. S. Yun, K. Chen, D. Bundy, S. Minoshima, S. N. Thibodeau, and D. Osborne. 1996. Preclinical evidence of Alzheimer’s disease in persons homozygous for the epsilon 4 allele for apolipoprotein E. N. Engl. J. Med. 334: 752–758. Riege, W. H., and E. J. Metter. 1988. Cognitive and brain imaging measures of Alzheimer’s disease. Neurobiol. Aging 9: 41– 42. Ross, B. M., M. McLaughlin, M. Roberts, G. Milligan, J. McCulloch, and J. T. Knowler. 1993. Alterations in the activity of adenylate cyclase and high affinity GTPase in Alzheimer’s disease. Brain Res. 622: 35– 42. Ross, E. M. 1989. Signal sorting and amplification through G protein-coupled receptors. Neuron 3: 141–152. Saitoh, T., and K. Dobkins. 1988. Increased in vitro phosphorylation of a Mr 60,000 protein in brain from patients with Alzheimer’s disease. Proc. Natl. Acad. Sci. USA 85: 6218 – 6221. Schnecko, A., K. Witte, J. Bohl, T. Ohm, and B. Lemmer. 1994. Adenylyl cyclase activity in Alzheimer’s disease brain: Stimulatory and inhibitory signal transduction pathways are differently affected. Brain Res. 644: 291–296. Simpson, I. A., K. R. Koteswara, T. Davies-Hill, W. G. Horner, and P. Davies. 1994. Decreased concentrations of GLUT1 and GLUT3 glucose transporters in the brain of patients with Alzheimer’s disease. Ann. Neurol. 35: 546 –551. Tokumitsu, H., T. Chijia, M. Hagiwara, A. Mizutani, M. Terasawa, and H. Hidaka. 1990. KN-62, 1-[N,O-bis(1,5-isoquinolinesulfonyl)-N-methyl-L-tyrosyl]-4-phenyl-piperazine, a specific inhibitor of Ca 2⫹/calmodulin-dependent protein Kinase II. J. Biol. Chem. 265: 4315– 4320. Walker, P. S., J. A. Donovan, B. G. Van Ness, R. E. Fellows, and J. E. Pessin. 1988. Glucose-dependent regulation of glucose transport activity, protein, and mRNA in primary cultures of rat brain glial cells. J. Biol. Chem. 263: 15594 –15601. Wang, H.-Y., and E. Friedman. 1994. Receptor-mediated activation of G protein is reduced in postmortem brains from Alzheimer’s disease patients. Neurosci. Lett. 173: 37–39. Yankner, B. A. 1996. Mechanisms of neuronal degeneration in Alzheimer’s disease. Neuron 16: 921–932.
G Protein and cAMP-Dependent Protein Kinase Mediate Amyloid -Peptide Inhibition of Neuronal Glucose Uptake 1 Teerasak Prapong, Etsuro Uemura, and Walter H. Hsu Department of Biomedical Sciences, 2008 Veterinary Medicine Building, Iowa State University, Ames, Iowa 50011 Received July 12, 1999; accepted July 19, 2000
utilization ranges from 19% in mild cases to 40% and more in severe cases of AD (11, 17). Immunohistochemical studies revealed a decrease in glucose transporters (GLUT1, GLUT3) in several brain regions, including the hippocampus, the area commonly affected in AD patients (9, 36). Reduced glucose uptake by individuals genetically at risk for developing AD also supports the notion that glucose deprivation precedes neuronal degeneration (16, 27, 30). In healthy, nonstarved mammalian brain, glucose is the only substrate for the formation of energy in the form of ATP (5). Glucose deprivation results in increased use of endogenous substrates, depletion of ATP, membrane depolarization, extracellular accumulation of excitatory amino acids, loss of neuronal homeostasis, and ultimately neuronal cell death (21). Neurons interfered with glucose uptake or exposed to A result in impaired mitochondrial activity and suppressed production of ATP (21). Based on the rates of oxidized glucose and oxygen, it was estimated that the rate of cerebral ATP formation decreases by 7% in early-onset AD and 20% in late-onset AD (10). Such metabolic impairment increases neuronal vulnerability to glutamate toxicity and oxidative insults (21). There are numerous published reports linking A to neurotoxicity in vivo and in vitro. However, how A inhibits glucose uptake is not known. It was shown that A induces conjugation of 4-hydroxynonenal (HNE), the product of lipid peroxidation, with neuronal transporter GLUT3, resulting in impaired glucose transport (20). Subsequent studies in synaptosomes from the rat cerebral cortex showed that A and HNE impair glucose transport and severely compromise mitochondrial function (15). These studies suggest that A-mediated oxidative stress inhibits neuronal glucose uptake. In the present study, we found that A inhibition on neuronal glucose uptake involves Gs–adenylyl cyclase coupling and activation of cAMP-dependent protein kinase (PKA), suggesting the presence of alternative mechanism by which A inhibits glucose uptake in cultured hippocampal neurons.
The mechanism by which amyloid -peptide (A) inhibits glucose uptake in cultured cells is not known. Here we demonstrated a signaling pathway in which A25-35, a neurotoxic portion of the A peptide corresponding to amino acids 25–35, inhibits neuronal glucose uptake by hippocampal neurons. The GP antagonist-2, which blocks Gs, prevented the inhibitory effect of A on the glucose uptake. Exposure of cells to A resulted in a transitory increase in intracellular levels of cAMP. To assess the role of cAMP in neuronal glucose uptake, cultured neurons were exposed to dibutyryl cAMP (Bt 2cAMP) or an adenylyl cyclase activator, forskolin. Both Bt 2cAMP and forskolin inhibited neuronal glucose uptake, and cAMP-dependent protein kinase (PKA) inhibitor KT5720 blocked the A-mediated inhibition of glucose uptake. Cholera toxin, which stimulates adenylyl cyclase by activating Gs protein, also inhibited neuronal glucose uptake, and A potentiated this inhibitory effect of cholera toxin on glucose uptake. Thus, our findings suggest that A inhibits glucose uptake by activating the Gscoupled receptors and involves the cAMP–PKA system. © 2001 Academic Press Key Words: Alzheimer’s disease; -amyloid; glucose uptake; hippocampal neuron; G protein; cAMP.
INTRODUCTION
Alzheimer’s disease (AD) is characterized by deposition of amyloid -peptide (A), a 39- to 43-amino-acid protein (40). It has been shown that A impairs glucose uptake in cultured hippocampal neurons (20) and astrocytes (26); however, it is not known whether A affects glucose uptake in the brain of AD patients. There is some evidence suggesting that glucose metabolism is changed in the brains of AD patients (4, 12, 29, 31). It was calculated that reduction of cerebral glucose 1
This study was supported by a special research initiation grant from the Iowa State University Graduate College and the Thai government. 59
0014-4886/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
60
PRAPONG, UEMURA, AND HSU
MATERIALS AND METHODS
Hippocampal neurons from 17E Sprague–Dawley rats were cultured according to Abe and Kimura (1); however, we used N2 instead of N1 supplement in our cultures. A25-35 and scrambled control peptide were prepared according to Parpura-Gill et al. (26). Neurons plated onto 24-well tissue culture plates were exposed to A25-35 or scrambled control peptide by using coverslips coated with the peptides. The coated-side was placed on top of the cultured neurons at postplating day 4. To examine whether A inhibition of neuronal glucose uptake is mediated by a G protein, cultured hippocampal neurons at day 4 postplating were exposed to synthetic A25-35 for 24 h with or without 10 M GP antagonist-2 (Biomol) in the culture medium. To rule out the possibility of nonspecific effects of GP antagonist-2, neuronal glucose uptake was also determined in the presence of 10 ⫺4 M cytochalasin B, an inhibitor of glucose transporters. Cholera toxin, which stimulates adenylyl cyclase by activating Gs (32), was used to further examine the role of Gs on the A inhibition on neuronal glucose uptake. Neuronal glucose uptake was assessed at day 4 postplating following exposure to cholera toxin (0.01, 0.1, or 1 g/ml) in the medium for 24 h. Two low doses of cholera toxin (0.01, 0.1 g/ml) were further used to study whether cholera toxin potentiates the inhibitory effect of A in neuronal glucose uptake. To examine whether PKA plays a regulatory role in neuronal glucose uptake, neurons at day 4 postplating were exposed to a PKA inhibitor KT-5720 (0.1 and 1 M) in the culture medium and A25-35 on coverslips for 24 h. To rule out the possibility of nonspecific effects of KT-5720, neuronal glucose uptake was also determined in the presence of 10 ⫺4 M cytochalasin B. Neuronal uptake of glucose was assayed by using 14 D-[ C]glucose (313 Ci/mmol, ICN) at final glucose concentrations of 600 nM (0.1 Ci/500l medium in well). The culture was incubated in the Hepes medium at 37°C for 1 h before adding radioactive glucose into the culture (38). Glucose uptake was stopped after 0 and 60 min of incubation by three washes with ice-cold Hepes medium prepared from 0.2383 g. of Hepes in 100 ml MEM. Time 0 was used to determine the background uptake of D-[ 14C]glucose. Cell lysates were used for determination of radioactivity by a beta liquid scintillation counter (Packard, 1600TR) and for assay of total cell protein (Bio-Rad protein assay kit). Glucose uptake was expressed as pmol of glucose/mg of cellular protein. The effect of A on intracellular cAMP levels was studied in neurons at day 4 postplating by exposing them to A25-35 or scrambled control peptide for 2 h, 2 days, or 3 days. To assess the role of cAMP in neuronal glucose uptake, neurons at day 4 postplating were also assessed for glucose uptake at 0 and 1 h following
exposure for 1 h to either Bt 2cAMP (0, 0.5, 1, 5, 10, 50 M) or forskolin (0, 10, 25, 50, 100 M). The cAMP content of cultured neurons was determined with the acetylation protocol of 125I cAMP radioimmunoassay (Amersham). For acute effect of -amyloid, neurons were cultured on PEI-coated plates for 24 h, and then neurons were exposed to 10 M A25-35 or scrambled peptide for 24 h. Before harvesting cells, 10 M of forskolin was added in some of the wells as positive control. The wells were washed with ice cold 0.1 M PBS for two times, and then 0.3 ml of 0.1 N NaOH was added. The culture plates were frozen at ⫺20°C for at least 2 h before the next process was performed. The frozen cell culture plates were thawed on an ice bath and were scraped with scrapers. The cell lysate was transferred into new glass tubes and was boiled at 70°C for 10 min to inactivate phosphodiesterase. The sample tubes were centrifuged at 3000 rpm for 5 min at 4°C. The supernatants were transferred into new tubes. Supernatants were neutralized with 0.04 N HCl at the ratio of 200 l of supernatant:50 l of HCl. Neutralized supernatants were acetylated with acetylation reagent (acetic anhydride ⫹ triethylamine). Standards of cAMP (4, 8, 16, 32, 64, 128, 256 fmol) were prepared. Anti-cAMP (1:70,000) and 125I cAMP tracer (5000 cpm/tube) were added to the acetylated samples (100 l) and incubated at 4°C overnight. To the samples 100 l of normal rabbit plasma and 100 l of second antibody were added. Cold PEG was added and samples were spun at 3000 rpm for 20 min. The supernatant was aspirated and tube was counted with ␥-counter. Total cell protein was assayed (Bio-Rad protein assay kit) and cAMP levels were expressed as femtomoles per milligram of protein. Each experiment was replicated three times (n ⫽ 3) with four wells per condition per replication. All pups from each pregnant rat were used as an experimental unit. Data were analyzed for treatment differences by analysis of variance and intertreatment differences were determined by post hoc test (Tukey’s). RESULTS
To examine whether a G protein mediates A inhibition of neuronal glucose uptake, hippocampal neurons were exposed to GP antagonist-2 (25), which blocks the receptor–Gs interaction. In this study, cultured hippocampal neurons at day 4 postplating were exposed to synthetic A25-35 for 24 h with or without 10 M GP antagonist-2 in the culture medium. The neuronal glucose uptake was measured and expressed as picomoles of D-[ 14C]glucose per milligram of cellular protein (26). A25-35 significantly inhibited neuronal glucose uptake [F(3, 8) ⫽ 66.8, P ⫽ 0.0001] (Fig. 1a). Those neurons exposed to A25-35 showed 40% less glucose uptake compared with the control. However, the GP antagonist-2 prevented this inhibitory effect of
AMYLOID -PEPTIDE INHIBITION OF NEURONAL GLUCOSE UPTAKE
FIG. 1. G protein mediates A inhibition of neuronal glucose uptake. A, amyloid -peptide; Ga, GP antagonist-2; Ctx1, 1 g/ml cholera toxin; Ctx.1, 0.1 g/ml cholera toxin; Ctx.01, 0.01 g/ml cholera toxin. (a) Cultured hippocampal neurons at day 4 postplating were exposed to synthetic A25-35 for 24 h with or without 10 M GP antagonist-2 in the culture medium. Those neurons exposed to A25-35 showed 40% less glucose uptake compared with the control. However, GP antagonist-2 prevented this inhibitory effect of A (P ⫽ 0.001). (b) Neuronal glucose uptake was assessed at day 4 postplating following exposure to cholera toxin (0.01, 0.1, or 1 g/ml) in the medium for 1 h. Cholera toxin at 0.01 and 0.1 g/ml had no effect on neuronal glucose uptake, but 1 g/ml cholera toxin significantly inhibited glucose uptake (P ⫽ 0.001). (c) Two low doses of cholera toxin (0.01, 0.1 g/ml) were used to study whether cholera toxin potentiates the inhibitory effect of A in neuronal glucose uptake. Neurons at day 4 postplating were exposed for 24 h and cholera toxin at either 0.01 or 0.1 g/ml in the culture medium. As noted above (b), neurons exposed to A25-35 showed significantly less glucose uptake than the control. Neurons exposed to both A25-35 and 0.01 g/ml cholera toxin did not differ in their glucose uptake from those exposed to A25-35 alone. However, those neurons exposed to both A25-35 and 0.1 g/ml cholera toxin showed 26% less glucose uptake than those exposed to A25-35 alone (P ⫽ 0.0004).
A, suggesting that G protein mediates A inhibition of neuronal glucose uptake. The effect of GP antagonist-2 was specific, as it did not prevent the inhibition of glucose uptake by cytochalasin B. To further assess the role of Gs on the inhibitory effect of A on neuronal glucose uptake, the effect of cholera toxin, which stimulates adenylyl cyclase by
61
activating Gs (32), was studied in cultured hippocampal neurons. In this study, neuronal glucose uptake was assessed at day 4 postplating following exposure to cholera toxin (0.01, 0.1, or 1 g/ml) in the medium for 1 h. Cholera toxin at 0.01 and 0.1 g/ml had no effect on neuronal glucose uptake, but 1 g/ml cholera toxin significantly inhibited glucose uptake [F(3, 8) ⫽ 38.63, P ⫽ 0.001] (Fig. 1b). Based on this observation, the two low doses of cholera toxin (0.01, 0.1 g/ml) were used to study whether this toxin potentiates the inhibitory effect of A. In this study, neurons at day 4 postplating were exposed for 24 h to both A25-35 coated on coverslips and cholera toxin at either 0.01 or 0.1 g/ml in the culture medium. Those neurons exposed to A25-35 showed 38% less glucose uptake than the control (t ⫽ 3.78, P ⫽ 0.0001). When neurons were exposed to both A25-35 and 0.1 g/ml cholera toxin, their glucose uptake was 53% less than the control (t ⫽ 5.99, P ⫽ 0.001) and 26% less than those exposed to A25-35 alone (t ⫽ 4.87, P ⫽ 0.0004) (Fig. 1c). Thus, cholera toxin potentiated the effect of A2535, suggesting that A inhibits glucose uptake by activating the Gs-coupled receptors. Since Gs activates adenylyl cyclase (33), we studied the effect of A on intracellular cAMP levels in cultured hippocampal neurons. In this study, neurons at day 4 postplating were exposed to A25-35 for 2 h, 2 days, or 3 days. Neurons were analyzed for cAMP content using radioimmunoassay (Amersham). A significant change in intracellular cAMP levels was observed in neurons exposed to A [F(1, 12) ⫽ 75.5, P ⫽ 0.0001] and this change was time dependent [F(2, 12) ⫽ 19.3, P ⫽ 0.0002] (Fig. 2a). The highest cAMP level was found in neurons that were exposed to A for 2 h, followed by those neurons exposed to A for 48 h cAMP levels at 72 h postexposure were not different from the control. An elevation of intracellular cAMP levels by A25-35 within 2 h of application suggests that A stimulates Gs–adenylyl cyclase coupling. To further assess the role of cAMP in neuronal glucose uptake, we studied the effect of cAMP on neuronal glucose uptake in hippocampal neurons. At postplating day 4, cultured neurons were exposed for 1 h to either Bt 2cAMP (0, 0.5, 1, 5, 10, 50 M) or forskolin (0, 10, 25, 50, 100 M). Neurons were assessed for glucose uptake and cellular protein at 0 and 1 h posttreatment. Neuronal glucose uptake was significantly inhibited by Bt 2cAMP [F(5, 18) ⫽ 9.07, P ⫽ 0.0002] (Fig. 2b) or forskolin [F(4, 15) ⫽ 4.68, P ⫽ 0.0006] (Fig. 2c). This inhibitory effect was dose dependent for both Bt 2cAMP (P ⫽ 0.0001, r 2 ⫽ 0.54) and forskolin (P ⫽ 0.0001, r 2 ⫽ 0.58). Neurons exposed to either 50 M Bt 2cAMP or 100 M forskolin showed only 13% of glucose uptake by the control neurons. Since A elevated intracellular cAMP levels within 2 h, we also assessed the inhibitory effect of A on glucose uptake following exposure for
62
PRAPONG, UEMURA, AND HSU
FIG. 2. cAMP mediates A inhibition of neuronal glucose uptake. A, amyloid -peptide. (a) Neurons at day 4 postplating were analyzed for cAMP content following exposure to A25-35 or scrambled control peptide for 2 h, 2 days, or 3 days. A significant change in intracellular cAMP levels was observed in neurons exposed to A (P ⫽ 0.0001) and this change was time dependent (P ⫽ 0.0002). The highest cAMP level was found in neurons exposed to A for 2 h. The cAMP level had fallen to the control level by 72 h. (b, c) Neurons at day 4 postplating were assessed for glucose uptake and cellular protein at 0 and 1 h following exposure for 1 h to either Bt 2cAMP (0, 0.5, 1, 5, 10, 50 M) or forskolin (0, 10, 25, 50, 100 M). Neuronal glucose uptake was significantly inhibited by Bt 2 cAMP (P ⫽ 0.0002) (b) or forskolin (P ⫽ 0.0006) (c). This inhibitory effect was dose dependent for both Bt 2 cAMP and forskolin. (d) Since A25-35 elevated cAMP by 2 h (a), neurons at day 4 postplating were assessed for glucose uptake at 0 and 1 h following exposure to either A25-35 or forskolin (10 M) for 3 h. Glucose uptake was significantly inhibited by A25-35 (P ⫽ 0.0025), suggesting an acute inhibitory effect of A25-35.
3 h to A25-35. Glucose uptake was significantly inhibited by A25-35 (t ⫽ 2.97, P ⫽ 0.025) (Fig. 2d), suggesting an acute inhibitory effect of A25-35. Inhibition of neuronal glucose uptake by Bt 2cAMP and forskolin suggests that glucose uptake is regulated by cAMP-dependent PKA. To examine whether PKA plays a regulatory role in neuronal glucose uptake, the effect of a PKA inhibitor KT-5720 (0.1 and 1 M) was studied in cultured hippocampal neurons. Neuronal glucose uptake and cellular protein were assessed at 1 h post-sampling time. Glucose uptake by control neurons was 40 pmol/mg cellular protein (Fig. 3). In contrast, A25-35 significantly reduced glucose uptake to 15 pmol/mg cellular protein, i.e., 38% of the control neurons. This inhibitory effect of A25-35 was prevented by 1 M KT-5720, but not by 0.1 M KT-5720. This dose-dependent effect of the PKA inhibitor was highly significant [F(3, 16) ⫽ 8.62, P ⫽ 0012]. The result suggests an inhibitory role by PKA on neuronal glucose uptake. The effect of KT-5720 was specific, as 1 M KT-5720 did not prevent the inhibition of glucose uptake by cytochalasin B. DISCUSSION
The present study suggests that a G protein and cAMP mediate A25-35 inhibition of neuronal glucose
uptake. A role played by Gs in neuronal glucose uptake was shown by the use of GP antagonist-2 and two adenylyl cyclase activators, cholera toxin and forskolin, all of which inhibited neuronal glucose uptake. Cholera toxin also potentiated the inhibitory effect of A. Although, GP antagonist-2 interacts with a part of the receptor-binding domain on G proteins to block the effects of receptor in reconstituted vesicles (25), its effect on cultured cells is not known. Thus, use of this agent does not present a definitive, but provides a compelling case for Gs-coupled-receptor mechanisms mediating the A effect. Gs elevates intracellular cAMP by activating adenylyl cyclase (7, 33). In our study, A acutely increased intracellular cAMP and an addition of Bt 2cAMP in culture medium inhibited neuronal glucose uptake, suggesting the importance of cAMP in neuronal glucose uptake. To date the role of cAMP in the regulation of neuronal glucose is not known. Elevation of intracellular cAMP by forskolin was shown to inhibit the glucose uptake by neurons (8). Similarly, Bt 2cAMP inhibits glucose uptake in L6 myoblast (19) and hamster ovary cells (28). A inhibition of neuronal glucose uptake was acute and such inhibition was not associated with changes in the number of GLUT3 in cells (20). cAMP is known to activate PKA and the course of events that follows in a given cell is determined by the specific proteins phosphorylated by
AMYLOID -PEPTIDE INHIBITION OF NEURONAL GLUCOSE UPTAKE
63
the importance of glucose in the CNS function, the effect of A on neuronal glucose metabolism in AD warrants further study. REFERENCES 1.
2.
3. FIG. 3. PKA mediates A inhibition of neuronal glucose uptake. A, amyloid -peptide; PKAi, PKA inhibitor. Neurons at day 4 postplating were analyzed for glucose uptake and cellular protein following exposure to KT-5720 (0.1 or 1 M) in the culture medium and A25-35 or scrambled control peptide on coverslips for 24 h. Glucose uptake by control neurons was 40 pmol/mg cellular protein. In contrast, A25-35 significantly reduced glucose uptake to 15 pmol/mg cellular protein, i.e., 38% of the control neurons. This inhibitory effect of A25-35 was prevented by 1 M KT-5720, but not by 0.1 M KT-5720. This dose-dependent effect of PKA was highly significant (P ⫽ 0.0012).
4.
5. 6.
7.
this enzyme (14). In adipocytes, PKA-triggered phosphorylation of GLUT4 inhibits insulin-stimulated glucose transport (13, 18). In the present study, PKA inhibitor KT5720 blocked the A-mediated inhibition of glucose uptake. Whether A alters phosphorylation of GLUT3 is not known. How these in vitro data relate to the decreased glucose metabolism in AD is not certain. It is possible that the decreased cerebral glucose uptake shown in AD (4, 12, 29, 31) may reflect, in part, the inhibitory effect of A on neuronal and astrocytic glucose uptake (20, 26). Such a hypothesis infers the presence of functional G-protein-stimulated activity of adenylyl cyclase and cAMP in AD patients. Postmortem studies of AD patients have shown normal G protein levels (3, 22, 32, 39), catalytic activity of cAMP, and PKA (2, 6, 23, 34). However, the regulatory function of G protein on adenylyl cyclase was reported as either normal (32) or reduced (3, 6, 35). This inconsistency may reflect differences in postmortem instability of the enzyme at different stages of Alzheimer’s disease (2). The regulatory function of G protein on adenylyl cyclase could also be affected only at the terminal stage of AD disease. A inhibition of neuronal glucose uptake could also be mediated by Ca 2⫹/calmodulin-sensitive adenylyl cyclase and an increase cAMP levels, the process known to occur in the hippocampus (24). However, Ca 2⫹/calmodulin-mediated inhibition of glucose uptake is less likely because it was shown that A-induced apoptosis was not inhibited by KN-62, an inhibitor of Ca 2⫹/calmodulin-dependent protein kinase (37). Considering
8.
9.
10.
11.
12.
13.
14. 15.
16.
Abe, K., and H. Kimura. 1996. Amyloid  toxicity consists of a Ca 2⫹-independent early phase and a Ca 2⫹-dependent late phase. J. Neurochem. 67: 2074 –2078. Bonkale, W. L., J. Fastbom, B. Wiehager, R. Ravid, B. Winblad, and R. F. Cowburn. 1996. Inpaired G-protein-stimulated adenylyl cyclase activity in Alzheimer’s disease brain is not accompanied by reduced cyclic-AMP-dependent protein kinase A activity. Brain Res. 737: 155–161. Cowburn, R. F., C. O’Neill, R. Ravid, I. Alafuzoff, B. Winblad, and C. J. Fowler. 1992. Adenylyl cyclase activity in postmortem human brain: evidence of altered G protein mediation in Alzheimer’s disease. J. Neurochem. 58: 1409 –1419. Duara, R., C. Grady, J. Haxby, M. Sundaram, N. R. Cutler, L. Heston, A. Moore, N. Schlageter, S. Larson, and S. I. Rapoport. 1986. Positron emission tomography in Alzheimer’s disease. Neurology 36: 879 – 887. Erecinska, M., and I. A. Silver. 1989. ATP and brain functon. J. Cereb. Blood Flow Metab. 9: 2–19. Fowler, C. J., R. F. Cowburn, A. Garlind, B. Winblad, and C. O’Neill. 1995. Disturbances in signal transduction mechanisms in Alzheimer’s disease. Mol. Cell. Biochem. 149: 287–292. Gilman, A. G. 1987. G proteins: Transducers of receptor-generated signals. Annu. Rev. Biochem. 56: 615– 649. Hara, M., Y. Matsuda, K. Hirai, N. Okumura, and H. Nakagawa. 1989. Characteristics of glucose transport in neuronal cells and astrocytes from rat brain in primary culture. J. Neurochem. 52: 902–908. Harr, S. D., N. A. Simonian, and B. T. Hyman. 1995. Functional alterations in Alzheimer’s disease: Decreased glucose transporter 3 immunoreactivity in the perforant pathway terminal zone. J. Neuropathol. Exp. Neurol. 54: 38 – 41. Hoyer, S. 1992. Oxidative energy metabolism in Alzheimer brain: Studies in early-onset and late onset cases. Mol. Chem. Neuropathol. 16: 207–224. Hoyer, S., R. Nitsch, and K. Oesterreich. 1991. Predominant abnormality in cerebral glucose utilization in late-onset dementia of the Alzheimer type: A cross-sectional comparison against advanced late-onset and incipient early-onset cases. J. Neural Transm. 3: 1–14. Jagus, W. J., J. P. Seab, R. H. Huesman, P. E. Valk, C. A. Mathis, B. R. Reed, P. G. Coxson, and T. F. Budinger. 1991. Diminished glucose transport in Alzheimer’s disease: Dynamic PET studies. J. Cereb. Blood Flow Metab. 11: 323–330. James, D. E., J. F. Hiken, and J. C. Lawernce, Jr. 1989. Isoproterenol stimulates phosphorylation of the insulin-regulatable glucose transporter in rat adipocytes. Proc. Natl. Acad. Sci. USA 86: 8368 – 8372. Karp, G. 1996. Cell and Molecular Biology. Wiley, New York. Keller, J. N., Z. Pang, J. W. Geddes, A. Germeyer, G. Waeg, and M. P. Mattson. 1997. Amyloid -peptide impairs glucose and glutamate transport, induces oxidative stress in mitochondria, and reduces ATP levels in synaptosomes: Role of the lipid peroxidation product 4-hydroxynonenal. J. Neurochem. 69: 273–284. Kennedy, A. M., R. S. Frackowiak, S. K. Newman, P. M. Bloomfield, J. Seaward, P. Roques, G. Lewington, V. J. Cunningham, and M. N. Rossor. 1995. Deficits in cerebral glucose metabolism demonstrated by positron emission tomography in individuals
64
17.
18.
19.
20.
21.
22.
23.
24. 25. 26.
27.
28.
29.
PRAPONG, UEMURA, AND HSU at risk of familial Alzheimer’s disease. Neurosci. Lett. 186: 17–20. Kumar, A., M. B. Schapiro, C. Grady, J. V. Haxby, E. Wagner, J. A. Salerno, R. P. Friedland, and S. I. Rapoport. 1991. Highresolution PET studies in Alzheimer’s disease. Neuropsychopharmacology 4: 35– 46. Lowrence, J. C. J., J. F. Hiken, and D. E. James. 1990. Phosphorylation of the glucose transporter in rat adipocytes. J. Biol. Chem. 265: 2324 –2332. Lowrence, J. C. J., R. C. Piper, L. J. Robinson, and D. E. James. 1992. GLUT4 facilitates insulin stimulation and cAMP-mediated inhibition of glucose transport. Proc. Natl. Acad. Sci. USA 89: 3493–3497. Mark, R. J., Z. Pang, J. W. Geddes, K. Uchida, and M. P. Mattson. 1997. Amyloid -peptide impairs glucose transport in hippocampal and cortical neurons: Involvement of membrane lipid peroxidation. J. Neurosci. 17: 1045–1054. Mattson, M. 1997. Cellular actions of -amyloid precursor protein and its soluble and fibrillogenic derivatives. Physiol. Rev. 77: 1081–1132. McLaughlin, M., B. M. Ross, G. Milligan, J. McCulloch, and J. T. Knowler. 1991. Robustness of G proteins in Alzheimer’s disease: An immunoblot study. J. Neurochem. 57: 9 –14. Meier-Ruge, W., P. Iwangoff, and K. Erichlmeier. 1984. Neurochemical enzyme changes in Alzheimer’s and Pick’s disease. Arch. Gerontol. Geriatr. 3: 161–165. Mons, N., and D. M. F. Cooper. 1995. Adenylate cyclases: Critical foci in neuronal signaling. TINS 18: 536 –542. Mukai, H., E. Munekata, and T. Higashijima. 1992. G protein antagonists. J. Biol. Chem. 267: 16237–16243. Parpura-Gill, A., D. Beitz, and E. Uemura. 1997. The inhibitory effects of -amyloid on glutamate and glucose uptakes by cultured astrocytes. Brain Res. 754: 65–71. Pettegrew, J. W., K. Panchalingam, W. E. Klunk, R. J. McLure, and L. R. Muenz. 1994. Alterations of cerebral metabolism in probable Alzheimer’s disease. Neurobiol. Aging 15: 117–132. Piper, R. C., D. E. James, J. W. Slot, C. Puri, and J. C. J. Lawrence. 1993. GLUT4 phosphorylation and inhibition of glucose transport by dibutyryl cAMP. J. Biol. Chem. 268: 16557– 16563. Rapoport, S. I., B. Horwitz, C. L. Grady, J. V. Haxby, C. DeCarli, and M. B. Schapiro. 1991. Abnormal Brain Glucose Me-
30.
31.
32.
33. 34.
35.
36.
37.
38.
39.
40.
tabolism in Alzheimer’s Disease, as Measured by Positron Emission Tomography. Plenum, New York. Reiman, E. M., R. J. Caselli, L. S. Yun, K. Chen, D. Bundy, S. Minoshima, S. N. Thibodeau, and D. Osborne. 1996. Preclinical evidence of Alzheimer’s disease in persons homozygous for the epsilon 4 allele for apolipoprotein E. N. Engl. J. Med. 334: 752–758. Riege, W. H., and E. J. Metter. 1988. Cognitive and brain imaging measures of Alzheimer’s disease. Neurobiol. Aging 9: 41– 42. Ross, B. M., M. McLaughlin, M. Roberts, G. Milligan, J. McCulloch, and J. T. Knowler. 1993. Alterations in the activity of adenylate cyclase and high affinity GTPase in Alzheimer’s disease. Brain Res. 622: 35– 42. Ross, E. M. 1989. Signal sorting and amplification through G protein-coupled receptors. Neuron 3: 141–152. Saitoh, T., and K. Dobkins. 1988. Increased in vitro phosphorylation of a Mr 60,000 protein in brain from patients with Alzheimer’s disease. Proc. Natl. Acad. Sci. USA 85: 6218 – 6221. Schnecko, A., K. Witte, J. Bohl, T. Ohm, and B. Lemmer. 1994. Adenylyl cyclase activity in Alzheimer’s disease brain: Stimulatory and inhibitory signal transduction pathways are differently affected. Brain Res. 644: 291–296. Simpson, I. A., K. R. Koteswara, T. Davies-Hill, W. G. Horner, and P. Davies. 1994. Decreased concentrations of GLUT1 and GLUT3 glucose transporters in the brain of patients with Alzheimer’s disease. Ann. Neurol. 35: 546 –551. Tokumitsu, H., T. Chijia, M. Hagiwara, A. Mizutani, M. Terasawa, and H. Hidaka. 1990. KN-62, 1-[N,O-bis(1,5-isoquinolinesulfonyl)-N-methyl-L-tyrosyl]-4-phenyl-piperazine, a specific inhibitor of Ca 2⫹/calmodulin-dependent protein Kinase II. J. Biol. Chem. 265: 4315– 4320. Walker, P. S., J. A. Donovan, B. G. Van Ness, R. E. Fellows, and J. E. Pessin. 1988. Glucose-dependent regulation of glucose transport activity, protein, and mRNA in primary cultures of rat brain glial cells. J. Biol. Chem. 263: 15594 –15601. Wang, H.-Y., and E. Friedman. 1994. Receptor-mediated activation of G protein is reduced in postmortem brains from Alzheimer’s disease patients. Neurosci. Lett. 173: 37–39. Yankner, B. A. 1996. Mechanisms of neuronal degeneration in Alzheimer’s disease. Neuron 16: 921–932.