Enhancement of 45 Ca 2 1 Influx and Voltage-dependent Ca 2 1 Channel Activity by b -Amyloid-(1– 40) in Rat Cortical Synaptosomes and Cultured Cortical Neurons MODULATION BY THE PROINFLAMMATORY CYTOKINE INTERLEUKIN-1 b

b -Amyloid protein is thought to underlie the neurodegeneration associated with Alzheimer’s disease by in-ducing Ca 2 1 -dependent apoptosis. Elevated neuronal expression of the proinflammatory cytokine interleu-kin-1 b is an additional feature of neurodegeneration, and in this study we demonstrate that interleukin-1 b modulates the effects of b -amyloid on Ca 2 1 homeostasis in the rat cortex. b -Amyloid-(1–40) (1 m M ) caused a significant increase in 45 Ca 2 1 influx into rat cortical synaptosomes via activation of L- and N-type voltage-de-pendent Ca 2 1 channels and also increased the amplitude of N- and P-type Ca 2 1 channel currents recorded from cultured cortical neurons. In contrast, in-terleukin-1 b (5 ng/ml) reduced the 45 Ca 2 1 influx into cortical synaptosomes and inhibited Ca 2 1 channel activity in cultured cortical neurons. Furthermore, the stimulatory effects of b -amyloid protein on Ca 2 1 influx were blocked following exposure to interleukin-1 b , suggesting that interleukin-1 b may govern neuronal responses to b -amyloid by regulating Ca 2 1 homeostasis.

␤-Amyloid protein is thought to underlie the neurodegeneration associated with Alzheimer's disease by inducing Ca 2؉ -dependent apoptosis. Elevated neuronal expression of the proinflammatory cytokine interleukin-1␤ is an additional feature of neurodegeneration, and in this study we demonstrate that interleukin-1␤ modulates the effects of ␤-amyloid on Ca 2؉ homeostasis in the rat cortex. ␤-Amyloid-(1-40) (1 M) caused a significant increase in 45 Ca 2؉ influx into rat cortical synaptosomes via activation of L-and N-type voltage-dependent Ca 2؉ channels and also increased the amplitude of N-and P-type Ca 2؉ channel currents recorded from cultured cortical neurons. In contrast, interleukin-1␤ (5 ng/ml) reduced the 45 Ca 2؉ influx into cortical synaptosomes and inhibited Ca 2؉ channel activity in cultured cortical neurons. Furthermore, the stimulatory effects of ␤-amyloid protein on Ca 2؉ influx were blocked following exposure to interleukin-1␤, suggesting that interleukin-1␤ may govern neuronal responses to ␤-amyloid by regulating Ca 2؉ homeostasis.
␤-Amyloid (A␤-(1-40)) 1 is a peptide fragment derived from proteolytic processing of ␤-amyloid precursor protein (␤APP) (1), which accumulates as an insoluble extracellular deposit around neurons, giving rise to the senile plaques associated with Alzheimer's disease (AD) (2). Increased neuronal expression of the proinflammatory cytokine interleukin-1␤ (IL-1␤) is an additional neuropathological hallmark of AD (3), and inflammatory mediators such as IL-1␤ have been proposed to contribute to the development of amyloid plaques (4). Several reports describe an interaction between IL-1␤ and A␤ at the processing level; IL-1-immunoreactive microglia are prominent components of amyloid plaques in AD (4), and ␤-amyloid promotes release of IL-1␤ by the glial cells that surround senile plaques (5). In turn, IL-1␤ increases ␤APP mRNA expression (6) and promotes processing of ␤APP to liberate A␤ peptide fragments (7). Thus a chain of events involving IL-1␤ and A␤ is involved in plaque formation; however, the nature of the interaction between IL-1␤ and A␤ at a physiological level is poorly understood. Neuronal apoptosis is the suspected causative factor of neurodegeneration in AD, and A␤ fragments have been shown to promote apoptosis in vitro in human-derived neurotypic cells (8) and cultured neurons (9). The mechanism underlying A␤-induced apoptosis is thought to involve disregulation of Ca 2ϩ homeostasis (10). In the C6 glial cell line, expression of the Ca 2ϩ -binding protein calbindin was found to suppress A␤induced apoptosis (11), providing evidence for the involvement of Ca 2ϩ fluxes in A␤-induced apoptosis. In this study we report that A␤-(1-40) (i) promotes a stimulation of 45 Ca 2ϩ influx into cortical synaptosomes via activation of L-and N-type voltagedependent Ca 2ϩ channels (VDCCs) and (ii) increases the amplitude of N-and P-type VDCC current in cultured cortical neurons. Furthermore, the A␤-(1-40)-induced increase in Ca 2ϩ influx is blocked by the proinflammatory mediator IL-1␤ in both cortical synaptosome preparations and cultured cortical neuron preparations, providing evidence for a physiological interaction between IL-1␤ and A␤-(1-40) with respect to the regulation of Ca 2ϩ homeostasis.

MATERIALS AND METHODS
Preparation of Synaptosomes-The cortices of male Wistar rats (200 -250 g) were dissected and homogenized in ice-cold 0.32 M sucrose (10 strokes, Teflon-glass homogenizer). The homogenate was centrifuged at 1,500 ϫ g for 5 min at 4°C. The resulting supernatant was centrifuged at 15,000 ϫ g for 15 min at 4°C. The resultant pellet, which yielded impure synaptosomal preparation P 2 , was resuspended in incubation buffer containing the following (in mM): NaCl 128, KCl  Culture of Cortical Neurons-The effects of A␤-(1-40) and IL-1␤ on the Ca 2ϩ channel current in primary cultures of rat cortical neurons were investigated using the whole-cell configuration of the patch-clamp technique. Cells were obtained by enzymatic and mechanical dissociation as described previously (12). Briefly, cortices were removed from 16 -18-day fetal rats and triturated following trypsin digestion. Suspended cells were plated out at a density of 0.25 ϫ 10 6 cells/mm 3 on circular coverslips (10-mm diameter) and incubated in a humidified atmosphere containing 5% CO 2 , 95% air at 37°C. After 48 h 80 M fluorodeoxyuridine was included in the culture medium to prevent proliferation of non-neuronal cells. The culture medium was exchanged every 3 days, and cells were grown in culture for up to 14 days. All recordings were made from cells between days 5 and 12.
Measurement of 45 Ca 2ϩ Influx-20 l of the P 2 synaptosomal preparation Ϯ A␤ (1 M) was suspended in 60 l of oxygenated incubation buffer containing 45 Ca (final concentration, 1 Ci/ml; specific activity, 2.1 mCi/ml; Amersham Pharmacia Biotech) and either 4.8 mM KCl or 50 mM KCl for analysis of unstimulated and stimulated 45 Ca 2ϩ influx, respectively. The synaptosomes were then incubated for 5 s at 37°C. In some experiments IL-1␤ (5 ng/ml) or vehicle (Krebs-buffered saline) was included in the 4.8 mM KCl and 50 mM KCl incubation buffers. In * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Determination of [Ca 2ϩ ] i in Synaptosomes-Fura-2/AM (Molecular Probes, The Netherlands) was stored at Ϫ20°C in a 1 mM stock solution in dimethyl sulfoxide. The synaptosomal suspension Ϯ A␤ (1 M) was incubated with Fura-2/AM (2 M) for 1 h at 30°C in oxygenated Krebs buffer (composition in mM: NaCl 139, KCl 2.5, KH 2 PO 4 1.2, MgSO 4 1.2, NaHCO 3 16, glucose 10, CaCl 2 2). The Fura-2-loaded synaptosomes were centrifuged at 5,000 ϫ g for 1 min, and the pellet was resuspended in Krebs to a final protein concentration of 1 mg/ml. The [Ca 2ϩ ] i measurements were recorded as described previously (13). Briefly, 2-ml aliquots of Fura-2-loaded synaptosomal suspension were placed in the cuvette holder of a Cairn spectrophotometer (Cairn, Faversham, UK). Experiments were performed at 37°C. Using the ratio of fluorescence at 340 and 380 nm, with a constant emission at 510 nm, the change in [Ca 2ϩ ] i was determined as described previously (14). The maximum fluorescence ratio (R max ) was achieved by lysing the synaptosomes with 5 l of 10% Triton 100, and the minimum fluorescence ratio (R min ) was achieved by addition of 50 l of 0. Electrophysiological Recording-For recording Ca 2ϩ channel currents, patch pipettes were filled with solution containing (in mM) Cs-HEPES 100, EGTA 30, CaCl 2 3, MgCl 2 2.5, K 2 ATP 3.25; osmolarity was 320 mosM with sucrose; pH was adjusted to 7.2 with CsOH. Cells were bathed in a solution that consisted of (in mM) tetraethylammonium acetate 70, N-methyl-D-glucamine 70, KOH 3, magnesium acetate 0.6, glucose 4, barium acetate 10, HEPES 10, and tetrodotoxin 0.0005; pH was 7.4 with acetic acid; osmolarity was 320 mosM with sucrose. To record Ca 2ϩ current-voltage relationships, cells were held at a potential of either Ϫ90 or Ϫ50 mV and depolarized to potentials ranging from Ϫ60 to ϩ60 mV. The steps were repeated every 10 s. For all electrophysiological recordings, five leak subtraction steps were made prior to depolarization to allow off-line removal of linear leak and residual capacity artifacts. Resting current was also measured to ensure that changes in leak did not affect the results. Current-voltage relationships for the Ca 2ϩ channels in cortical neurons were recorded following 24-h preincubation with either 1 M A␤-(1-40) or 1 M A␤-(40 -1) as controls.
No A␤ was present in the recording medium when Ca 2ϩ channel currents were measured.
Reagents-The two batches of A␤-(1-40) were obtained from Sigma and Bachem (Saffron Walden, UK). A␤ was dissolved in deoxygenated, deionized water at 100 M, aliquoted to 10 l, and stored at Ϫ20°C until needed. Stock peptide was stored as powder at Ϫ70°C. The reverse sequence peptide A␤-(40 -1) was prepared in exactly the same way. Ca 2ϩ channel antagonists -conotoxin GVIA (Peninsula Laboratories, St. Helens, UK), nimodipine (Tocris Neuramin, Bristol, UK), and -agatoxin IVA (Sigma) were made up as stock solutions in water before appropriate dilution in recording medium. All culture reagents were obtained from Life Technologies, Inc. except for chick embryo extract, which was obtained from Imperial Laboratories (UK). Fig. 1 demonstrates that the 45 Ca 2ϩ influx evoked following depolarization with 50 mM KCl is significantly enhanced in synaptosomes treated with A␤-(1-40). In control synaptosomes 45 Ca 2ϩ influx is significantly increased from 1.08 Ϯ 0.141 nmol/mg of protein (mean Ϯ S.E.) to 1.55 Ϯ 0.23 nmol/mg of protein following K ϩ depolarization (p Ͻ 0.05, paired t test, n ϭ 30). In the synaptosomes pretreated with A␤, 45 Ca 2ϩ influx was increased from 1.44 Ϯ 0.132 to 2.48 Ϯ 0.154 nmol/mg of protein following K ϩ depolarization (p Ͻ 0.01, paired t test, n ϭ 30).
Identification of the VDCC Subtype Involved in the K ϩ -induced Rise in 45 Ca 2ϩ Influx-To identify the nature of the VDCC subtype involved in the A␤-induced rise in K ϩ -stimulated 45 Ca 2ϩ influx, the cortical synaptosomes were treated with the L-type VDCC inhibitor nifedipine (10 M) and the N-type VDCC inhibitor -conotoxin GVIA (2 M). Fig. 3 illustrates the effect of these VDCC inhibitors on the K ϩ -stimulated rise in 45 Ca 2ϩ influx in control and A␤-(1-40)-treated synaptosomes. In control synaptosomes the K ϩ -stimulated rise in 45 Ca 2ϩ influx is significantly reduced from 0.386 Ϯ 0.065 to 0.252 Ϯ 0.056 nmol/mg of protein (p Ͻ 0.05, paired t test, n ϭ 10) and 0.164 Ϯ 0.08 nmol/mg of protein (p Ͻ 0.05, paired t test, n ϭ 10) following treatment with nifedipine (10 M) and -conotoxin GVIA (2 M), respectively.
A␤ caused a 2-fold increase in the K ϩ -stimulated rise in 45 Ca 2ϩ influx compared with the control synaptosomal preparation (p Ͻ 0.01, paired t test, n ϭ 10). Furthermore, in A␤treated synaptosomes the K ϩ -stimulated rise in 45 Ca 2ϩ influx is also significantly reduced from 1.22 Ϯ 0.21 to 0.186 Ϯ 0.099 nmol/mg of protein (p Ͻ 0.01, paired t test, n ϭ 10) and 0.176 Ϯ 0.06 nmol/mg of protein (p Ͻ 0.01, paired t test, n ϭ 10) following treatment with nifedipine (10 M) and -conotoxin GVIA (2 M), respectively. These results suggest a role for Land N-type VDCCs in K ϩ -stimulated 45 Ca 2ϩ influx in cortical synaptosomes under control conditions. Furthermore, since A␤ failed to evoke an increase in K ϩ -stimulated 45 Ca 2ϩ influx in the presence of nifedipine and -conotoxin GVIA, this result implicates both the L-and N-type VDCCs as targets for modulation by A␤-(1-40) in this presynaptic preparation.
Whereas K ϩ -stimulated 45 Ca 2ϩ influx is significantly greater in the A␤-treated synaptosomal preparation (1.01 Ϯ 0.31 nmol/mg of protein; p Ͻ 0.01, paired t test, n ϭ 12), a 5-s exposure to IL-1␤ (5 ng/ml) significantly reduces the K ϩ -stimulated 45 Ca 2ϩ influx to Ϫ0.14 ϩ 0.09 nmol/mg of protein (p Ͻ 0.01, paired t test, n ϭ 12). This result demonstrates that IL-1␤ has the proclivity to inhibit the A␤-induced enhancement of 45 Ca 2ϩ influx in cortical synaptosomes. At this potential only Ca 2ϩ channel currents that are relatively resistant to inactivation by depolarization would be expected to be activated. Thus the Ca 2ϩ current that is increased by A␤ in cortical neurons appears, in some degree, to be resistant to inactivation. Whole cell conductance and leak current values were also measured during recordings and were found to be unchanged. To identify the Ca 2ϩ channel subtype augmented by preincubation of cortical neurons with A␤, subtype-selective Ca 2ϩ channel blockers were included in the recording medium during measurement of Ca 2ϩ channel activity. Fig. 5 shows Ca 2ϩ channel current density evoked at the ϩ10 mV potential when recordings were made in the presence of various Ca 2ϩ channel antagonists.
The L-type Ca 2ϩ channel blocker nimodipine (2 M) slightly reduced Ca 2ϩ channel current density when compared with untreated cells but had little impact on the increase in Ca 2ϩ channel current density induced by A␤ (control, Ϫ84.9 Ϯ 11.2 pA/pF, n ϭ 13; A␤-(1-40), Ϫ128.6 Ϯ 14.4 pA/pF, n ϭ 16, p Ͻ 0.01), suggesting that L-type channels were not augmented by preincubation with A␤.
-Conotoxin GVIA (1 M) attenuated currents in untreated cells by approximately 60%, indicating the presence of a large N-type Ca 2ϩ channel component. Furthermore, A␤ appeared to interact with N-type Ca 2ϩ channels as -conotoxin GVIA greatly reduced the increase in Ca 2ϩ channel current density induced by A␤. However, a small (31%), statistically significant increase in Ca 2ϩ channel current was observed at the ϩ10 mV potential (control, Ϫ37.0 Ϯ 3.0 pA/pF, n ϭ 42; A␤-(1-40), Ϫ48.3 Ϯ 4.8 pA/pF, n ϭ 39, p Ͻ 0.05, t test). As -conotoxin GVIA was unable to completely prevent the increase in current   ; black bar). Exposure of the synaptosomes to IL-1␤ (5 ng/ml; gray bar) during the depolarization phase resulted in a significant reduction in K ϩ -stimulated 45 Ca 2ϩ (*, p Ͻ 0.05). In synaptosomes pretreated with ␤-amyloid (1 M, 1 h at 30°C) the 45 Ca 2ϩ influx resulting from a 5-s exposure to KCl (50 mM; black bar) was significantly greater than that observed in control synaptosomes (p Ͻ 0.05). In addition, exposure of A␤-(1-40)-treated synaptosomes to IL-1␤ (5 ng/ml; gray bar) during the K ϩ depolarization phase caused a significant reduction in K ϩ -stimulated 45 Ca 2ϩ influx (p Ͻ 0.01). Results are expressed as the mean Ϯ S.E. of 12 observations. caused by A␤, it is possible that another Ca 2ϩ channel subtype may be involved.
The importance of the P-type component Ca 2ϩ channel current in cortical neurons was shown by a 30% reduction in control Ca 2ϩ channel current by 30 nM -agatoxin IVA. After increasing the Ca 2ϩ channel current with A␤, -agatoxin IVA (30 nM) blocked the current by approximately 45%. When Ca 2ϩ channel currents measured in the presence of -agatoxin IVA in A␤-(40 -1)-and A␤-(1-40)-treated cells were compared, it was found that A␤ no longer increased the Ca 2ϩ channel current (control, Ϫ66.5 Ϯ 7.0 pA/pF, n ϭ 22; A␤-(1-40), Ϫ79.5 Ϯ 9.7 pA/pF, n ϭ 20). This suggests that P-type channels are also modulated by A␤.
Attenuation of 45 Ca 2ϩ influx in synaptosomes by IL-1␤ suggested that we would be able to block the increased current in rat cortical neurons with IL-1␤. Currents from A␤-(40 -1)-and A␤-(1-40)-treated cells were measured in the presence of 5 ng/ml IL-1␤. IL-1␤ significantly inhibited the Ca 2ϩ channel current following both treatments. Furthermore, A␤ was unable to cause a significant increase in Ca 2ϩ channel current density in the presence of IL-1␤ when cells were depolarized from a holding potential of Ϫ90 mV (control, Ϫ51.9 Ϯ 5.7, n ϭ 12; A␤-(1-40), Ϫ60.4 Ϯ 4.4, n ϭ 12, Figs. 5 and 6A). There was also no effect of A␤ on Ca 2ϩ channel current density when cells were depolarized from a holding potential of Ϫ50 mV (Fig. 6B). The decrease in Ca 2ϩ channel current density observed in the presence of IL-1␤ represents a reduction of approximately 45%. These data support the previous finding that IL-1␤ can attenuate the effects of A␤ on 45 Ca 2ϩ influx in synaptosomes. DISCUSSION The aim of this study was to examine the effects of A␤-(1-40) and the proinflammatory cytokine IL-1␤ on Ca 2ϩ homeostasis in cortical synaptosomes and cultured cortical neurons. The results demonstrate that A␤-(1-40) stimulates 45 Ca 2ϩ influx into the synaptosomal preparation with a parallel rise in [Ca 2ϩ ] i . The electrophysiological data demonstrate that A␤-(1-40) also increases voltage-dependent Ca 2ϩ channel activity in cultured cortical neurons. Since the stimulatory effects of A␤-(1-40) on 45 Ca 2ϩ influx into cortical synaptosomes were blocked by nifedipine and -conotoxin GVIA we conclude that L-and N-type VDCCs are involved in the A␤-induced elevation in 45 Ca 2ϩ influx in this preparation. In the cultured cortical neurons the A␤-induced elevation in Ca 2ϩ channel current neurons was found to be sensitive to -conotoxin GVIA and -agatoxin IVA, indicating the involvement of N-and P-type VDCCs. The data also demonstrate that the proinflammatory mediator IL-1␤ attenuates 45 Ca 2ϩ influx into cortical synaptosomes and reduces Ca 2ϩ channel current density in cultured cortical neurons by approximately 50%, providing evidence that cortical voltage-sensitive Ca 2ϩ channels are substrates for modulation by IL-1␤. Furthermore, exposure to IL-1␤ was found to prevent the stimulatory effects of A␤-(1-40) on Ca 2ϩ influx, and this result is indicative of an interaction between A␤-(1-40) and IL-1␤ with respect to the regulation of Ca 2ϩ homeostasis.
Our finding that A␤-(1-40) stimulated 45 Ca 2ϩ influx and elevated [Ca 2ϩ ] i at the level of the nerve terminal means that A␤-(1-40) is likely to have a significant impact on Ca 2ϩ -dependent neuronal functions such as neurotransmitter release and synaptic plasticity. In support of this hypothesis, ␤-amyloid peptides have been shown to potentiate Ca 2ϩ -dependent release of glutamate and aspartate from hippocampal slices (15), impair long term potentiation in the hippocampus (16), and alter spatial memory performance tests (17). Given that elevations in [Ca 2ϩ ] i have been shown to precede apoptotic events in several cell systems (18), the stimulation of 45 Ca 2ϩ influx and enhancement of VDCC activity by A␤-(1-40) may act as a trigger for the neuronal apoptosis associated with neurodegenerative disease. In support of this suggestion, A␤induced increases in Ca 2ϩ influx are reported to be involved in mediating the apoptotic effects of A␤ in PC12 cells (19,20), and L-type VDCCs have been demonstrated to mediate A␤-induced neurotoxicity in cultured hippocampal neurons (21). In addition, mobilization of intracellular Ca 2ϩ stores via formation of reactive oxygen species has been suggested to underlie the detrimental effects of A␤-(25-35) on synaptosomal membrane lipid structure and composition (22). Although elevation in [Ca 2ϩ ] i has been demonstrated to underlie A␤-induced neuronal apoptosis, A␤ also stimulates glia to produce growth factors that contribute to plaque development (5) and may influence neuronal viability (23). In addition, the ability of A␤ to induce expression of adhesion molecules (24) may also play a significant role in the pathogenesis of AD.
The subtype of VDCC modulated by A␤-(1-40) in the cortical synaptosomes and cultured neurons was investigated using selective Ca 2ϩ channel blockers. The L-type VDCC blocker nifedipine and the N-type VDCC inhibitor -conotoxin GVIA were both found to attenuate the K ϩ -stimulated 45 Ca 2ϩ influx in control cortical synaptosomes, indicating that L-and N-type VDCCs become activated following K ϩ depolarization. The finding that -conotoxin GVIA blocked K ϩ -stimulated 45 Ca 2ϩ influx in the synaptosomal preparation is consistent with a presynaptic role for the N-type VDCC because N-type VDCCs have been localized to nerve terminals, where they play a critical role in presynaptic events such as neurotransmitter release (25). In addition, a presynaptic role for L-type VDCCs has been described in cerebellar granule neurons (26), and our finding that nifedipine attenuated the K ϩ -induced increase in 45 Ca 2ϩ influx in the synaptosomal preparation is consistent with previous studies that demonstrate the presence of L-type VDCCs in cortical synaptosomes (27). The finding that nifedipine and -conotoxin GVIA also blocked the A␤-mediated poten- tiation of the K ϩ -induced increase in 45 Ca 2ϩ influx in cortical synaptosomes suggests that N-and L-type VDCCs are targets for presynaptic modulation by A␤- . In contrast, in the cultured cortical neurons a 20-ms, 110-mV depolarization from a holding potential of Ϫ90 mV resulted in activation of N-and P-type VDCCs, with little contribution from L-type VDCCs, demonstrating that N-and P-type VDCCs are prevalent in cultured cortical neurons. The A␤-(1-40)-induced increase in Ca 2ϩ channel current density was significantly reduced by -conotoxin GVIA and the P-type VDCC blocker -agatoxin IVA but was unaffected by the L-type VDCC blocker nimodipine, indicating that N-and P-type VDCCs are modulated by A␤-(1-40) in these cells. In cerebellar granule neurons A␤ has been shown to act via exclusive up-regulation of N-type channels (28) and L-type VDCCs in microglia (29); thus the nature of the VDCC regulated by A␤ may be dependent upon the cell type and subcellular fraction under investigation.
We found that IL-1␤ attenuated 45 Ca 2ϩ influx in cortical synaptosomes in a similar manner to that previously reported in synaptosomes prepared from the hippocampus (30). IL-1␤ functions as a neuromodulator to reduce neurotransmitter release (30) and impair long term potentiation in the hippocampus (31) possibly as a consequence of its inhibitory effect on Ca 2ϩ influx (30). The finding that IL-1␤ reduces Ca 2ϩ channel current in cultured cortical neurons is also similar to the modulatory effects reported for IL-1␤ on hippocampal Ca 2ϩ channel currents, where the IL-1␤ effects were found to proceed via activation of a pertussis toxin-sensitive G-protein (32). Modulatory effects of IL-1␤ on Ca 2ϩ homeostasis in cortical neurons and cortical nerve terminals are likely to have a significant impact on cortical function and may underlie the modulatory effects of IL-1␤ on synaptic transmission in the neocortex (33).
The finding that IL-1␤ blocked the stimulatory effects of A␤-(1-40) on 45 Ca 2ϩ influx and Ca 2ϩ channel currents is of particular interest. Elevations in neuronal expression of both A␤ (2) and IL-1␤ (3) are pathological features of Alzheimer's disease, and several lines of evidence demonstrate an interaction between A␤ and IL-␤. ␤-Amyloid induces release of IL-1␤ from activated microglia (34) and monocytes (35) in a Ca 2ϩ -dependent manner and also promotes transcription of IL-1 mRNA (36). Aside from the association of IL-1 with amyloid plaques (37), where it is thought to contribute to gliosis, IL-1 is also believed to be influential at an earlier phase of AD pathology because this cytokine has been found to enhance ␤APP mRNA expression in neurons (6) and promote processing of ␤APP to generate ␤-amyloid fragments (7). IL-1␤ and A␤ therefore participate in a cycle of events contributing to the formation of the neuritic plaques and tangles associated with AD. Although those studies reveal an interaction between A␤ and IL-1␤ at a processing level, little information is currently available concerning a physiological interaction between A␤ and IL-1␤. The results presented in this study demonstrate that IL-1␤ modulates the A␤-(1-40)-induced stimulation of Ca 2ϩ influx in the cortex, and such neuromodulatory properties of IL-1␤ may participate in governing neuronal responses to A␤ at certain stages of the neurodegenerative disease process. Thus, whereas IL-1␤ is considered to have a significant influence on the pathophysiology of AD by increasing ␤APP expression resulting in the formation of neuritic plaques (4), the proclivity of IL-1␤ to attenuate A␤-induced Ca 2ϩ influxes at the level of the nerve terminal may impact on presynaptic function, such as the release of excitatory neurotransmitters, to limit cell dam-  , n ϭ 12). No significant differences between the currents at each test potential could be detected (Student's unpaired t test). Mean peak Ca 2ϩ current density is plotted against the test potential evoking the current when cells were depolarized from a holding potential of Ϫ90 mV. Inset, averaged current traces, which are not corrected for capacitance, from the same cells shown in A at the ϩ20 mV potential. B, mean current/ voltage relationships when Ca 2ϩ channel current was evoked by depolarizing cells from a holding potential of Ϫ50 mV (A␤-(1-40), filled triangles, n ϭ 17; A␤-(40 -1), open circles, n ϭ 17). No significant differences between the currents at each test potential could be detected (Student's unpaired t test). Inset, averaged current traces from the same cells shown in B at the ϩ20 mV potential. age caused by glutamate excitotoxicity, and as such this may reflect a neuroprotective role for IL-1␤. Furthermore, because A␤-induced neuronal apoptosis is dependent upon the activation of Ca 2ϩ -sensitive proteases (38) and Ca 2ϩ -sensitive tyrosine kinases (39), our observation that IL-1␤ occludes the stimulatory effects of A␤ on Ca 2ϩ influx suggests that IL-1␤ may serve to restrain these intracellular cascades associated with A␤-induced neurodegeneration.
In summary, the increase in Ca 2ϩ influx mediated by A␤-(1-40) in cortical synaptosomes and cultured cortical neurons is attenuated by the proinflammatory cytokine IL-1␤. This finding indicates an interaction between A␤ and IL-1␤ that serves to regulate Ca 2ϩ homeostasis, and this interaction may have implications in the manifestation of AD.