The Anemonia sulcata Toxin BDS-I Protects Astrocytes Exposed to Aβ1–42 Oligomers by Restoring [Ca2+]i Transients and ER Ca2+ Signaling

Intracellular calcium concentration ([Ca2+]i) transients in astrocytes represent a highly plastic signaling pathway underlying the communication between neurons and glial cells. However, how this important phenomenon may be compromised in Alzheimer’s disease (AD) remains unexplored. Moreover, the involvement of several K+ channels, including KV3.4 underlying the fast-inactivating currents, has been demonstrated in several AD models. Here, the effect of KV3.4 modulation by the marine toxin blood depressing substance-I (BDS-I) extracted from Anemonia sulcata has been studied on [Ca2+]i transients in rat primary cortical astrocytes exposed to Aβ1–42 oligomers. We showed that: (1) primary cortical astrocytes expressing KV3.4 channels displayed [Ca2+]i transients depending on the occurrence of membrane potential spikes, (2) BDS-I restored, in a dose-dependent way, [Ca2+]i transients in astrocytes exposed to Aβ1–42 oligomers (5 µM/48 h) by inhibiting hyperfunctional KV3.4 channels, (3) BDS-I counteracted Ca2+ overload into the endoplasmic reticulum (ER) induced by Aβ1–42 oligomers, (4) BDS-I prevented the expression of the ER stress markers including active caspase 12 and GRP78/BiP in astrocytes treated with Aβ1–42 oligomers, and (5) BDS-I prevented Aβ1–42-induced reactive oxygen species (ROS) production and cell suffering measured as mitochondrial activity and lactate dehydrogenase (LDH) release. Collectively, we proposed that the marine toxin BDS-I, by inhibiting the hyperfunctional KV3.4 channels and restoring [Ca2+]i oscillation frequency, prevented Aβ1–42-induced ER stress and cell suffering in astrocytes.


Introduction
Astrocytes, the most abundant glial cells in the central nervous system (CNS), may support neuronal homeostasis, not only releasing trophic factors, but also regulating neurotransmission and synaptic plasticity [1][2][3][4][5][6][7]. Moreover, astrocytes continuously handle gliotransmitters like purines, D-serine, and glutamate, whose uptake and release are regulated by the frequency of their intracellular calcium concentration ([Ca 2+ ] i ) transients, a complex ionic phenomenon involving both extracellular and intracellular compartments including the endoplasmic reticulum (ER).
[Ca 2+ ] i transients in astrocytes spread throughout gap junctions belonging to the astrocytic syncytium [8][9][10]. Clearly, the extent of this electric phenomenon is driven by

Exposure to Aβ 1-42 Oligomers Upregulated K V 3.4 Protein Expression and Activity in Activated Rat Primary Cortical Astrocytes
To examine cytoskeleton rearrangement after exposure to Aβ 1-42 oligomers, F-actin cytoskeleton was stained in primary cortical astrocytes with the actin-binding dye phalloidin. The filament network was brightly stained in control astrocytes ( Figure 1A). However, actin bundle disassembly and body extroflections revealed pronounced astrogliosis in  Figure 1A). Confocal double immunofluorescence experiments revealed that the K V 3.4 immunoreactivity was mainly confined along the plasma membrane of control GFAP-positive cortical astrocytes ( Figure 1B). Moreover, primary cortical astrocytes exposed to 5 µM Aβ 1-42 oligomers for 48 h showed a more pronounced immunoreactivity for both GFAP and K V 3.4 as compared to controls ( Figure 1B). Interestingly, the two immunosignals intensely overlapped within the filament bundles ( Figure 1B). Furthermore, Western blotting analysis revealed that after 48 h exposure to Aβ 1-42 oligomers, K V 3.4 protein expression significantly increased in primary cortical astrocytes as compared to controls ( Figure 1C). Accordingly, patch-clamp recordings in whole-cell configuration showed a significant upregulation of the fast-inactivating currents (I A ) in primary cortical astrocytes exposed for 48 h to Aβ 1-42 oligomers compared with control astrocytes (Figure 2A). In contrast, 5 µM of the scramble Aβ peptide (Aβ 42-1 ) did not induce any significant modification of I A (Figure 2A). Moreover, the contribution of K V 3.4 to I A upregulation in primary cortical astrocytes exposed to Aβ 1-42 oligomers was further tested by blocking this channel with the selective blocker BDS-I. Treatment with BDS-I (50 nM) reduced I A (Figure 2A) in primary cortical astrocytes exposed to Aβ 1-42 oligomers ( Figure 2A). Interestingly, electrophysiological patch-clamp recordings in primary cortical astrocytes exposed to Aβ 1-42 oligomers (5 µM, 48 h) revealed that firing frequency was lower than in controls ( Figure 2B), whereas resting membrane potential was more negative than in controls ( Figure 2C). Moreover, BDS-I treatment counteracted either the spike frequency decrease or the membrane hyperpolarization in primary cortical astrocytes exposed to Aβ 1-42 oligomers ( Figure 2B,C, respectively).

Activated Rat Primary Cortical Astrocytes
To examine cytoskeleton rearrangement after exposure to Aβ1-42 oligomers, F-actin cytoskeleton was stained in primary cortical astrocytes with the actin-binding dye phalloidin. The filament network was brightly stained in control astrocytes ( Figure 1A). However, actin bundle disassembly and body extroflections revealed pronounced astrogliosis in astrocytes treated with 5 µM Aβ1-42 oligomers for 48 h ( Figure 1A). Confocal double immunofluorescence experiments revealed that the KV3.4 immunoreactivity was mainly confined along the plasma membrane of control GFAP-positive cortical astrocytes ( Figure  1B). Moreover, primary cortical astrocytes exposed to 5 µM Aβ1-42 oligomers for 48 h showed a more pronounced immunoreactivity for both GFAP and KV3.4 as compared to controls ( Figure 1B). Interestingly, the two immunosignals intensely overlapped within the filament bundles ( Figure 1B). Furthermore, Western blotting analysis revealed that after 48 h exposure to Aβ1-42 oligomers, KV3.4 protein expression significantly increased in primary cortical astrocytes as compared to controls ( Figure 1C). Accordingly, patch-clamp recordings in whole-cell configuration showed a significant upregulation of the fast-inactivating currents (IA) in primary cortical astrocytes exposed for 48 h to Aβ1-42 oligomers compared with control astrocytes (Figure 2A). In contrast, 5 µM of the scramble Aβ peptide (Aβ42-1) did not induce any significant modification of IA ( Figure 2A). Moreover, the contribution of KV3.4 to IA upregulation in primary cortical astrocytes exposed to Aβ1-42 oligomers was further tested by blocking this channel with the selective blocker BDS-I. Treatment with BDS-I (50 nM) reduced IA (Figure 2A) in primary cortical astrocytes exposed to Aβ1-42 oligomers ( Figure 2A). Interestingly, electrophysiological patch-clamp recordings in primary cortical astrocytes exposed to Aβ1-42 oligomers (5 µM, 48 h) revealed that firing frequency was lower than in controls ( Figure 2B), whereas resting membrane potential was more negative than in controls ( Figure 2C). Moreover, BDS-I treatment counteracted either the spike frequency decrease or the membrane hyperpolarization in primary cortical astrocytes exposed to Aβ1-42 oligomers ( Figure 2B and 2C, respectively).  under control conditions or 48 h after 5 µM Aβ1-42 exposure. Data were normalized on the bas β-actin and expressed as fold increase compared to control's expression of three different prep tions. * p < 0.05 vs control.

Effects of BDS-I on [Ca 2+ ]i Transients and ER Ca 2+ Signaling in Rat Primary Astrocytes Exposed to Aβ1-42 Oligomers
Rat primary cortical astrocytes displayed spontaneous [Ca 2+ ]i transients in the pe of control recordings characterized by two main oscillatory patterns with differen grees of irregularity ( Figure 3

Effects of BDS-I on [Ca 2+ ] i Transients and ER Ca 2+ Signaling in Rat Primary Astrocytes Exposed to Aβ 1-42 Oligomers
Rat primary cortical astrocytes displayed spontaneous [Ca 2+ ] i transients in the period of control recordings characterized by two main oscillatory patterns with different degrees of irregularity ( Figure 3  In order to assess the involvement of the major intracellular Ca 2+ -storing organ the modulation of [Ca 2+ ]i transients by Aβ1-42 oligomers, Ca 2+ depletion from the E induced in cortical astrocytes by ATP and thapsigargin in a nominal Ca 2+ -free sol Interestingly, after 48 h exposure to Aβ1-42 oligomers, ER Ca 2+ content was higher astrocytes than in controls ( Figure 4A,B). Moreover, when astrocytes were incubated BDS-I together with Aβ1-42 oligomers, ER Ca 2+ content was restored to control level (F 4A,B), thus suggesting a putative, and possibly compensatory, involvement of t creased KV3.4-mediated outward K + currents in the ER Ca 2+ overload. In order to assess the involvement of the major intracellular Ca 2+ -storing organelle in the modulation of [Ca 2+ ] i transients by Aβ 1-42 oligomers, Ca 2+ depletion from the ER was induced in cortical astrocytes by ATP and thapsigargin in a nominal Ca 2+ -free solution. Interestingly, after 48 h exposure to Aβ 1-42 oligomers, ER Ca 2+ content was higher in AD astrocytes than in controls ( Figure 4A,B). Moreover, when astrocytes were incubated with BDS-I together with Aβ 1-42 oligomers, ER Ca 2+ content was restored to control level ( Figure 4A,B), thus suggesting a putative, and possibly compensatory, involvement of the increased K V 3.4-mediated outward K + currents in the ER Ca 2+ overload.

Effects of BDS-I on ER Stress Markers in Rat Primary Astrocytes Exposed to Aβ1-42 Oligomers
Current evidence suggests that Ca 2+ dysregulation and ER stress still represent two relevant features of Aβ accumulation [24,42,43]. Therefore, the expression of the ER chaperone GRP78/BiP has been studied in astrocytes exposed to Aβ1-42 oligomers (5 µM, 48 h), considering this chaperone as one of the main markers of early AD stage [43]. After 48 h exposure to Aβ1-42 oligomers, GRP78/BiP protein expression peaked in AD astrocytes compared with controls ( Figure 4C). Importantly, BDS-I prevented GRP78/BiP overexpression when incubated with Aβ1-42 oligomers ( Figure 4C). Moreover, the expression of active caspase 12 was also upregulated in astrocytes by a long exposure to Aβ1-42 oligomers (5 µM, 48 h) ( Figure 4D), while it was prevented in astrocytes simultaneously exposed to BDS-I and Aβ1-42 oligomers ( Figure 4D).

Effects of BDS-I on ER Stress Markers in Rat Primary Astrocytes Exposed to Aβ 1-42 Oligomers
Current evidence suggests that Ca 2+ dysregulation and ER stress still represent two relevant features of Aβ accumulation [24,42,43]. Therefore, the expression of the ER chaperone GRP78/BiP has been studied in astrocytes exposed to Aβ 1-42 oligomers (5 µM, 48 h), considering this chaperone as one of the main markers of early AD stage [43]. After 48 h exposure to Aβ 1-42 oligomers, GRP78/BiP protein expression peaked in AD astrocytes compared with controls ( Figure 4C). Importantly, BDS-I prevented GRP78/BiP overexpression when incubated with Aβ 1-42 oligomers ( Figure 4C). Moreover, the expression of active caspase 12 was also upregulated in astrocytes by a long exposure to Aβ 1-42 oligomers (5 µM, 48 h) ( Figure 4D), while it was prevented in astrocytes simultaneously exposed to BDS-I and Aβ 1-42 oligomers ( Figure 4D). Considering that ER stress and dysregulated calcium homeostasis may lead to reactive oxygen species (ROS) formation [42], they were detected in rat primary astrocytes using the specific fluorescent dye 2 ,7 -dichlorodihydrofluorescein diacetate (DCFH-DA). Furthermore, dysfunction in lactate dehydrogenase (LDH) release and mitochondrial activity were detected as markers of cell suffering in AD astrocytes. ROS production was significantly increased after exposure to Aβ 1-42 oligomers (5 µM, 48 h) compared with control cells (Figure 5A), while BDS-I prevented this event in AD astrocytes ( Figure 5A). Consistently, BDS-I counteracted the significant reduction in mitochondrial activity and the increased level of LDH release induced by Aβ 1-42 oligomers in astrocytes ( Figure 5B,C, respectively).

Effects of BDS-I on ROS Production, Mitochondrial Activity, and Lactate Dehydrogenase (LDH) Release in Rat Primary Astrocytes Exposed to Aβ1-42 Oligomers
Considering that ER stress and dysregulated calcium homeostasis may lead to reactive oxygen species (ROS) formation [42], they were detected in rat primary astrocytes using the specific fluorescent dye 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA). Furthermore, dysfunction in lactate dehydrogenase (LDH) release and mitochondrial activity were detected as markers of cell suffering in AD astrocytes. ROS production was significantly increased after exposure to Aβ1-42 oligomers (5 µM, 48 h) compared with control cells (Figure 5A), while BDS-I prevented this event in AD astrocytes ( Figure 5A). Consistently, BDS-I counteracted the significant reduction in mitochondrial activity and the increased level of LDH release induced by Aβ1-42 oligomers in astrocytes ( Figure 5B,C, respectively).

Discussion
The present study shows that the Anemonia sulcata toxin BDS-I restored spontaneous [Ca 2+ ]i transients in rat primary astrocytes exposed to Aβ1-42 oligomers. This may occur via

Discussion
The present study shows that the Anemonia sulcata toxin BDS-I restored spontaneous [Ca 2+ ] i transients in rat primary astrocytes exposed to Aβ 1-42 oligomers. This may occur via the inhibition of the hyperfunctional K V 3.4 potassium channel, whose expression and activity resulted to be significantly upregulated in Aβ 1-42 -treated astrocytes. Accordingly, the expression and function of K V 3.4 channel subunits have been previously found to be upregulated in reactive astrocytes in AD Tg2576 mouse brains [30]. The chosen time-point of BDS-I incubation comes from the previous study [30] showing that, nevertheless the upregulation of both K V 3.4 potassium channel expression and activity starts before, the GFAP-monitored astrogliosis occurs only at 48 h exposure to Aβ 1-42 oligomers [30].
In line with our results, both astrocytes associated with senile plaques in APP/PS1 mice [44] and astrocytes from APP Swe mice in an early stage of the disease display atypical Ca 2+ waves [45].
Moreover, the mechanisms of spontaneous astrocytic [Ca 2+ ] i transients, mainly involved in important brain functions, are still unclear. In this study, we have identified: (i) a novel mechanism controlling the electrical phenomenon represented by K V 3.4 channel subunits and (ii) a new putative molecule with a therapeutic profile.
Furthermore, K V 3.4-mediated I A currents are dysfunctional in AD not only in neurons but also in glial cells [30,[35][36][37]. In this respect, and in consideration that K V 3.4 overexpression intervenes in the neuro-inflammation underlying AD development [30], BDS-I may assume a putative neuroprotective profile. Therefore, controlling the function of K V 3.4 channels in AD brain with BDS-I might represent a novel therapeutic approach for slowing down the progression of the disease.
At higher concentrations than that used in the present manuscript, BDS-I is also an efficient modulator of Na v 1.7 channel [46]. In fact, BDS-I slows down the inactivation Na + channels, but slightly more than the specific toxins AsI, AsII, and AxI from A. sulcata and A. xanthogrammica [47,48]. Of note, BDS-I has only a small effect on tetrodotoxin (TTX)-sensitive Na + channels [49] and no action on voltage-sensitive Na + channels in cardiac cells or in skeletal muscle myotubes [39]. Therefore, putative downstream effects associated to its action on Na + channels can be hypothesized, although restricted to the site of Na v 1.7 localization.
The treatment of cultured astrocytes with exogenous Aβ oligomers induces a pathological remodeling of Ca 2+ signaling [50] by affecting both neurotransmission machinery and ER Ca 2+ handling [51]. In particular, under these conditions, changes in inositol 1,4,5-triphosphate receptor (IP 3 R) expression with consequent defects in store-operated calcium entry (SOCE) were recorded in astrocytes and in organotypic slices [24]. Of note, astrocytes isolated from 3xTg-AD mice displayed increased SOCE [52] and altered [Ca 2+ ] i oscillations [53]. Therefore, previous evidence showed the importance of astrocytic Ca 2+ signaling in AD progression with particular respect to the intracellular component of Ca 2+ machinery. In line with the burgeon literature on this issue, the present data show that BDS-I reduced ER Ca 2+ overload in astrocytes induced by Aβ 1-42 oligomers with the consequent prevention of the expression of ER stress markers, including active caspase 12 and GRP78/BiP. Indeed, calcium dysregulation, ER stress, and overexpression of unfolded protein response (UPR) elements like GRP78/BiP have been identified as common pathways in neurodegenerative diseases, including AD [24,54,55]. In particular, GRP78/BiP protein expression levels change differently according to the stage of AD: in chronic AD patients, GRP78/BiP expression is very low, while in the early AD stages, GRP78/BiP has been found to be overexpressed [43]. This highlights the putative role of the Ca 2+ -dependent ER chaperone as an early marker of the disease [43].
Of note, Ca 2+ homeostasis and ER Ca 2+ signaling in glia are under the control of other ionic players, including K + channels as with Ca 2+ signaling in a classical model of excitable cells [56]. Accordingly, deregulation of potassium homeostasis may underlie gliosis in AD. Mechanistically, the intermediate-conductance Ca 2+ -activated K + channel K Ca 3.1 may induce astrogliosis and microglia activation in the disease [57]. Furthermore, K Ca 3.1, through the regulation of the membrane potential hyperpolarization, induces SOCE potentiation, ER Ca 2+ overload, and the consequent ER stress in AD astrocytes [58]. Interestingly, gene deletion or pharmacological blockade of astrocytic K Ca 3.1 reduce ER stress and prevent downstream neuronal loss in APP/PS1 mice [58], thus highlighting the importance of the astrocytic component in neuronal fate during AD progression.
Moreover, in the present study, the new role of astrocytic K V 3.4 channel in the modulation of ER Ca 2+ signaling has been highlighted in an in vitro model of AD.
In this respect, the mechanistic inquiry into AD pathogenesis and progression have recently switched from the neuron-centric doctrine to an astrocyte-centered theory [59]. Besides the evidence on the role of astrocytic K + dysregulation during AD, with the present study, we could only partially answer to this enormous issue. In fact, many efforts should be made in the future to address this important question.
In conclusion, by investigating the relationship between K V 3.4 channel and calcium signaling in astrocytes exposed to Aβ 1-42 oligomers, the present study showed that: (i) dysregulation of K V 3.4 channel induced an altered Ca 2+ transient activity and (ii) the Anemonia sulcata toxin BDS-I prevented ER stress by reducing K V 3.4 potassium channel hyperfunction and restoring spontaneous [Ca 2+ ] i transients, thus revealing a putative neuroprotective role for this marine toxin.

Rat Primary Astrocytes
Primary cultures of rat astrocytes were obtained as previously described [60][61][62]. This protocol yields 98% of GFAP-positive cells. In brief, dissected cortices from 1-to 2-day-old rat pups were first dissociated enzymatically in a solution containing 0.125% trypsin and 1.5 mg/mL DNase and then, mechanically, in DMEM supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 µg/mL streptomycin, and 2 mM L-glutamine. Cell pellets were plated on poly-L-lysine-coated plates. The medium was changed 24 h after plating and twice a week thereafter. For the mechanical dissociation, they were shaken vigorously to remove non-adherent cells and sub-cultured firstly at 1:3 dilution and then, once they reached confluency, at 1:4 dilution.

Solubilization of Aβ 1-42 Peptide and Cellular Treatment
After synthesis, high-performance liquid chromatography and mass spectrometry showed a 95% purity for the yielded Aβ peptides. Lyophilized peptides were resuspended in HFIP at 1 mM, as previously described [63]. The clear solution was dried until complete elimination of the solvent and recovery of the dried powder. Immediately prior to use, the HFIP-treated aliquots were carefully and completely resuspended to a concentration of 5 mM in anhydrous dimethyl sulfoxide (Me 2 SO). Aβ 1-42 and Aβ 42-1 oligomers were prepared by diluting the peptides at 5 mM concentration to a 100 µM solution in ice-cold cell culture medium (phenol red-free Ham's F-12, 4 • C for 24 h). Then, the solution was centrifuged at 14,000 rpm at 4 • C for 10 min and the supernatant containing Aβ 1-42 oligomers was stored at −20 • C [63,64]. Aβ 1-42 preparation was tested with monoclonal anti-Aβ antibody, which recognizes an epitope within residues 17-42 of human Aβ (see Drugs and Chemicals Section). In particular, Western blotting showed a specific band at~8 kDa, corresponding to Aβ 1-42 dimers, and a smear ranging from~8 to~15 kDa, comprising lower molecular weight intermediates (trimers), at the highest concentration of Aβ 1-42 preparation [63].

Immunohistochemistry
Immunostaining and confocal immunofluorescence procedures in cells were performed as previously described [56,57]. Cell cultures were fixed in 4% paraformaldehyde in PBS for 30 min and incubated in primary antisera, mouse monoclonal anti-K V 3.4 (1:500), and rabbit polyclonal anti-GFAP (1:1000). Subsequently, they were incubated in a mixture of fluorescent-labeled secondary antibodies. Control experiments were performed as previously described [66,67]. Images were observed with a Zeiss LSM510 META/laser scanning confocal microscope (Gottingen, Germany). Single images were taken with an optical thickness of 0.7 µm and a resolution of 1024 × 1024. All images were obtained with a 40× objective with identical laser power settings.

Assessment of Nuclear and Cytoskeletal Morphology
Cytoskeleton morphology was studied by the staining of actin with rhodamine phalloidin at 1:50 dilution from stock solution of 100 µg/1 mL in PBS for 15 min at 37 • C [68]. Coverslips were mounted on glass slides and observed by fluorescence microscopy on a Nikon Eclipse E400 microscope (Nikon, Torrance, CA, USA). Digital images were taken with a CoolSnap camera (Media Cybernetics Inc., Silver Spring, MD, USA), stored on the hard-disk of a Pentium III computer, and analyzed with the Image-Pro Plus 4.5 software.

Electrophysiology
K + currents were recorded from primary cortical astrocytes in control conditions, exposed to Aβ 1-42 oligomers in the presence and in the absence of BDS-I with the patchclamp technique in whole-cell configuration using a commercially available amplifier Axopatch 200B and Digidata 1322A interface (Molecular Devices, San Jose, CA, USA), as previously described [30,37]. In the same experimental conditions, current signals were acquired in gap-free modality using a Digidata 1322A interface using the protocol previously described [63]. Data were acquired using the pClamp software (version 9.0, Molecular Devices) and data analysis was performed using Clampfit software (version 9.0, Molecular Devices). Spontaneous action potential (AP) activity was measured in primary cortical astrocytes in control conditions, exposed to Aβ 1-42 oligomers in the presence and in the absence of BDS-I using the protocol previously described [63,69,70]. Importantly, sustained high-quality whole-cell recordings could be maintained for >15 min with a stable membrane potential and AP waveform, confirming that the presence of spontaneous APs was not the result of declining cell health. Spontaneous AP amplitude and frequency were determined using our own computer program written in Java computer language. Briefly, for each primary cortical astrocyte, the software calculated the AP mean ± standard deviation (SD) during the baseline recording interval. This was used to define a cutoff identifying AP, which was set at mean AP ± 2SD. Subsequently, the software identified each value higher than this cut-off pointas AP. To quantify AP features in primary cortical astrocytes in control conditions, exposed to Aβ 1-42 oligomers in the presence and in the absence of BDS-I, the following parameters were determined: the amplitude, defined as the difference between transient AP and mean basal and the frequency, defined as the number of peaks divided by the duration of observation. The pipette solution contained the following (in mM): 140 KCl, 2 MgCl 2 , 10 acido 4-2-idrossietil-1-piperaziniletansolfonico (HEPES), 10 glucose, 10 EGTA, and 1 Mg-ATP adjusted at pH 7.4 with KOH. The extracellular solution contained the following (in mM): 150 NaCl, 5.4 KCl, 3 CaCl 2 , 1 MgCl 2 , 10 HEPES, adjusted pH 7.4 with NaOH. 50 nM tetrodotoxin (TTX) and 10 µM nimodipine were added to Ringer's solution to abolish TTX-sensitive Na + , and L-type Ca 2+ current. The blood-depressing substance-I (BDS-I; synthetized by Prof. P. Grieco, Department of Pharmacy, "Federico II" University of Naples, Naples, Italy) at the concentration of 50 nM was used to block K V 3.4 currents [36,37,40].
Spontaneous action potential (AP) activity was measured in primary cortical astrocytes in control conditions, exposed to Aβ 1-42 oligomers in the presence and in the absence of BDS-I using the protocol previously described [63,69,70]. Importantly, sustained highquality whole-cell recordings could be maintained for > 15 min with a stable membrane potential and AP waveform, confirming that the presence of spontaneous APs was not the result of declining cell health. Spontaneous AP amplitude and frequency were determined using our own computer program written in Java computer language. Briefly, for each primary cortical astrocyte, the software calculated the AP mean ±standard deviation (SD) during the baseline recording interval. This was used to define a cutoff identifying AP, which was set at mean AP ± 2SD. Subsequently, the software identified each value higher than this cut-off pointas AP. To quantify AP features in primary cortical astrocytes in control conditions, exposed to Aβ 1-42 oligomers in the presence and in the absence of BDS-I, the following parameters were determined: the amplitude, defined as the difference between transient AP and mean basal and the frequency, defined as the number of peaks divided by the duration of observation. The pipette solution contained the following (in mM): 140 KCl, 2 MgCl 2 , 10 acido 4-2-idrossietil-1-piperazinil-etansolfonico (HEPES), 10 glucose, 10 EGTA, and 1 Mg-ATP adjusted at pH 7.4 with KOH. The extracellular solution contained the following (in mM): 150 NaCl, 5.4 KCl, 3 CaCl 2 , 1 MgCl 2 , 10 HEPES, adjusted pH 7.4 with NaOH. 50 nM tetrodotoxin (TTX) and 10 µM nimodipine were added to Ringer's solution to abolish TTX-sensitive Na + , and L-type Ca 2+ current. The blood-depressing substance-I (BDS-I; synthetized by Prof. P. Grieco, Department of Pharmacy, "Federico II" University of Naples, Naples, Italy) at the concentration of 50 nM was used to block K V 3.4 currents [36,37,40].

[Ca 2+ ] i Measurement
[Ca 2+ ] i was measured by single-cell computer-assisted video-imaging in astrocytes loaded with 10 µM Fura-2/AM [24,71,72]. [Ca 2+ ] i transients were identified using a software written in Java computer language, as previously reported [71]. Briefly, for each single astrocyte, the software calculated, during the time of recording, the [Ca 2+ ] i mean ± SD in order to define a cut-off point. Then, the software identified each value higher than this cutoff point as a single [Ca 2+ ] i transient. For each experiment, [Ca 2+ ] i transients were detected during the recordings and used to calculate the oscillation frequency that corresponds to the number of peaks divided by the duration of the recording (oscillation index).

Measurement of Reactive Oxygen Species
DCFH-DA, a cell membrane permeable fluorescein analogue, was used to detect ROS species production [36]. The rat primary astrocytes were pre-loaded with DCFH-DA (10 µM) for 30 min at 37 • C in PBS at the end of each pharmacological treatment. Cells were then washed with PBS, and the reaction was stopped by adding 2,6-di-tert-butyl-4methylphenol (0.2% in ethanol) and EDTA (2 mM). The cells were then viewed with a Zeiss Axioscope 2FS plus fluorescence microscope (Gottingen, Germany) using excitation and emission wavelengths of 488 and 525 nm, respectively. Digital images were taken with a CoolSnap camera (Media Cybernetics Inc., Silver Spring, MD, USA), and analyzed with the Image-Pro Plus 4.5 software (Media Cybernetics Inc., Silver Spring, MD, USA). Image acquisition and processing were performed equally for all experimental conditions; for the quantification, background fluorescence was subtracted from the data.

Determination of Mitochondrial Activity
Mitochondrial dysfunction was evaluated with theMTT test [73,74]. In this test, the MTT dye is metabolized by viable mitochondria to a colored product and can be detected photometrically. Briefly, after the experimental procedures, rat primary astrocytes were washed with normal Krebs and incubated with 1 mL of MTT solution (0.5 mg/mL in PBS), as previously described [73,74]. After 1 h incubation at 37 • C, rat primary astrocytes were dissolved in 1 mL of DMSO, in which the rate of MTT reduction was measured using a spectrophotometer at a wavelength of 540 nm. Data are expressed as percentage of mitochondrial dysfunction versus sham-treated cultures.

LDH Release Assay
Cytotoxicity was detected using a LDH release assay kit (Jiancheng Bioengineering Institute of Nanjing, Jiangsu, China) according to the manufacturer's protocol. Rat primary astrocytes were cultured in 96-well plates (1 × 10 4 cells/well), and after treatment, supernatants were transferred to clean 96-well plates. Absorbance was analyzed at 450 nm using a Bio-Rad 680 Microplate Reader (Bio-Rad Laboratorie, Hercules, CA, USA) [75].