DHA inhibits ER Ca²⁺ release and ER stress in astrocytes following in vitro ischemia

Abstract

Docosahexaenoic acid (DHA) has neuroprotective effects in several neurodegenerative disease conditions. However, the underlying mechanisms are not well understood. In the present study, we investigated the effects of DHA on astrocyte Ca²⁺ signaling under in vitro ischemic conditions (oxygen/glucose deprivation and reoxygenation, OGD/REOX). OGD (2 hour) triggered a Ca²⁺_ER store overload (~ 1.9 fold). Ca²⁺ uptake by the Ca²⁺_ER stores was further augmented during REOX and Ca²⁺_ER was elevated by ~ 4.7-fold at 90 min REOX. Interestingly, Ca²⁺_ER stores abruptly released Ca²⁺ at ~ 120 min REOX and emptied at 160 min REOX. Depletion of Ca²⁺_ER stores led to delayed elevation of intracellular Ca²⁺ concentration (Ca²⁺_cyt) and cell death. Activation of the purinergic receptor P2Y1 was responsible for the release of Ca²⁺_ER. Most importantly, DHA blocked the initial Ca²⁺_ER store overload, the delayed depletion of Ca²⁺_ER, and rise in Ca²⁺_cyt, which was in part via inhibiting IP₃ receptors. The DHA metabolite DiHDoHE exhibited similar effects. DHA also attenuated expression of phosphorylated eukaryotic initiation factor 2α and activating transcription factor-4, two ER stress markers, following in vitro ischemia. Taken together, these findings suggest that DHA has protective effects in astrocytes following in vitro ischemia, in part, by inhibiting Ca²⁺ dysregulation and ER stress.

INTRODUCTION

Docosahexaenoic acid (DHA, 22:6n-3) is the most abundant of the polyunsaturated fatty acids (PUFAs) in the brain (Kim 2007; Bazan 2006). DHA has been shown to promote neuronal survival and differentiation (Kim 2007; Bazan 2006). DHA or its oxidized products, such as neuroprotection D1 (NPD1, 10,17S-docosatriene), are neuroprotective in several neurodegenerative disease conditions, including experimental stroke, Alzheimer’s disease, retinal degeneration, and spinal cord injury (Bazan 2006; Bazan 2007; King et al. 2006). Although the molecular mechanisms underlying the biological activities of DHA remain to be fully understood, NPD1 has been shown to inhibit apoptotic and inflammatory signaling in human neuronal cells against Aβ42-induced cytotoxicity or in ischemic brains (Lukiw et al. 2005; Marcheselli et al. 2003; Belayev et al. 2005).

Astrocytes are the only brain cell type with the capacity to synthesize DHA (Moore et al. 1991). DHA is biosynthesized in astrocytes from α-linolenic acid through chain elongation and desaturation processes by Δ4–6 desaturases (Sprecher 2000). Neurons cannot produce DHA due to the lack of desaturase activity and thus depend on astrocytes to supply DHA (Moore et al. 1991). However, the role of DHA in astrocyte function has not been extensively studied. Many astrocyte functions are known to be regulated by changes of intracellular Ca²⁺ concentration (Ca²⁺_cyt) and Ca²⁺ release from the endoplasmic reticulum (ER) Ca²⁺ (Ca²⁺_ER) stores. DHA supplementation enhances gap junction coupling capacity in rat cortical astrocytes (Champeil-Potokar et al. 2006). Moreover, DHA suppresses thrombin-evoked Ca²⁺ response, inhibits store-operated Ca²⁺ entry (SOCE), and ATP-mediated intracellular Ca²⁺ oscillations in rat astrocytes (Sergeeva et al. 2002; Sergeeva et al. 2005). On the other hand, disturbance of Ca²⁺ homeostasis in the ER causes ER stress and activates the unfolded protein response (UPR) (Kaufman 1999). Changes in ER Ca²⁺ stores in astrocytes precede ER stress and mitochondrial dysfunction and are involved in astrocyte damage (Liu et al. 2010). However, the effect of DHA on Ca²⁺ signaling and ER stress in astrocytes is unknown.

In the present study, we report that OGD/REOX triggered a biphasic change in ER Ca²⁺ stores (overload and depletion). DHA blocked the ER Ca²⁺ overload and depletion of Ca²⁺_ER stores, as well as, the subsequent delayed rise in Ca²⁺_cyt in astrocytes following in vitro ischemia. In addition, DHA attenuated ER stress and astrocyte death. Taken together, these findings suggest that the neuroprotective effects of DHA following ischemia may be, in part, through inhibiting astrocyte Ca²⁺ dysregulation and ER stress.

METHODS

Materials

Eagle's modified essential medium (EMEM) and Hanks balanced salt solution (HBSS) were from Mediatech Cellgro (Manassas, VA). Fetal bovine serum (FBS) was obtained from Valley Biomedical (Winchester, VA). Collagen-type I, 4-bromo-A23187, adenosine 5'-Triphosphate (ATP), cis-4,7,10,13,16,19-docosahexaenoic acid (DHA), 2-Methylthioadenosine- 5'- O- diphosphate (2Me-S-ADP), 6-Methyl-2-(phenylethynyl) pyridine hydrochloride (MPEP), and saponin were purchased from Sigma Chemicals (St. Louis, MO). D-myo-inositol 1,4,5-triphosphate (IP₃), mag fura-2 acetoxymethyl ester (Mag-fura 2-AM), and fura-2 acetoxymethyl ester (Fura 2-AM) were from Invitrogen (Carlsbad, CA). 2¢-Deoxy-N6-methyladenosine 3¢,5¢-bisphosphate (MRS 2179), 2-Aminoethoxydiphenylborane (2-APB), and ryanodine were from Tocris (Ellisville, MO). 2-phenyl-1,2-benzisoselenazol-d(2H)-one (Ebselen) and 10(S),17(S)-dihydroxy-4Z,7Z,11E,13Z,15E,19Z-docosahexaenoic acid (DiHDoHE) were from Cayman Chemical (Ann Arbor, MI). Mouse anti α-tubulin monoclonal antibody and rabbit anti phosphor-eIF2α monoclonal antibodies was purchased from Cell Signaling Technology (Danvers, MA). Rabbit anti-activating transcription factor 4 (ATF-4) antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Primary culture of mouse cortical astrocytes

Dissociated cortical astrocyte cultures were established as described before (Su et al. 2002). Cerebral cortices were removed from 2–3 day-old mice. The cortices were incubated in a trypsin solution (0.25 mg/ml of HBSS) for 20 min at 37°C. The dissociated cells were rinsed and re-suspended in EMEM containing 10% FBS. Viable cells (1×10⁴ cells /well) were plated in 6-well plates containing type 1 collagen coated glass cover slips (22 mm × 22 mm) or on 75 cm² flasks. Cultures were maintained in a 5% CO₂ atmosphere at 37°C. To obtain morphologically differentiated astrocytes, confluent cultures (7 days in culture, DIV 7) were treated with EMEM containing 0.25 mM dibutyryl cyclic AMP (dBcAMP) to induce differentiation. Experiments were routinely performed in DIV 10–15 cultures.

In vitro hypoxia/ischemia

Astrocytes on coverslips were rinsed twice with an isotonic OGD solution (pH 7.4) containing (in mM, at 37°C): 0 glucose, 21 NaHCO₃, 120 NaCl, 5.36 KCl, 0.33 Na₂HPO₄, 0.44 KH₂PO4, 1.27 CaCl₂, 0.81 MgSO₄ (Kintner et al. 2007). Cells were incubated in 1.0 ml of the OGD solution in a hypoxic incubator for 2 h (Forma Scientific Inc., model 3130, Marietta, OH), containing 94% N₂, 1% O₂ and 5% CO₂. An orbital shaker (Thermolyne Inc, model M48215, Dubuque, IA) in the hypoxic chamber was used to facilitate equilibration of the hypoxic gases during the initial 30 min OGD. For REOX, OGD-treated cells were superfused with HEPES-MEM buffer containing 5.5 mM glucose for 0–180 min. Normoxic control cells were incubated in 5% CO₂ and atmospheric air in an isotonic control solution for 2 h.

Intracellular Ca²⁺ ([Ca²⁺]_cyt) measurement

Astrocytes were incubated with 5 µM fura-2 AM for 30–120 min at 37°C (Lenart et al. 2004). The cells were washed and the coverslips placed in the imaging chamber with HEPES-MEM at 37°C. Using a Nikon TiE 300 inverted epifluorescence microscope (40X oil immersion objective lens), astrocytes were excited every 10–30 seconds at 340 and 380 nm and the emission fluorescence at 510 nm was recorded. Images were collected and analyzed with the MetaFluor (Molecular Devices, Sunnyvale, CA) image-processing software. At the end of each experiment, the cells were exposed to 1 mM MnCl₂ in Ca²⁺-free HEPES-MEM. The MnCl₂-corrected 340/380 emission ratios were converted to concentration as described previously (Lenart et al. 2004).

ER Ca²⁺ ([Ca²⁺]_ER) measurement

Astrocytes on coverslips were incubated with 4 µM mag-fura-2 AM and 0.02 % pluronic acid during the last 30 min of 2 h OGD. For REOX, the coverslip was quickly (< 2 min) placed on an open-bath imaging chamber in HEPES-MEM at 37°C. Cells were excited every 10 sec at 345 and 385 nm and the emission fluorescence images were collected at 510 nm using Nikon TiE 300 inverted epifluorescence microscope and a 40X objective lens. Ca²⁺_ER were determined as described previously (Chen et al. 2008). Briefly, the cytosolic mag-fura-2 signal was eliminated when the plasma membrane was permeablized with a 30 sec exposure to saponin (3.0 µg/ml) (Chen et al. 2008). The Ca²⁺_ER values were then calculated using the equation $\left[{\mathit{\text{Ca}}}_{\mathit{\text{ER}}}^{2+}\right]=K\times (R_{\mathit{\text{base}}}-R_{\text{min}})/(R_{\text{max}}-R_{\mathit{\text{base}}})$. K was determined as 72 µM in astrocytes using solutions of known Ca²⁺ concentrations (Calcium Calibration Buffer Kit, Invitrogen, Carlsbad, CA). R_min was obtained with a minimum F_345/385 ratio in a Ca²⁺-free solution. R_max was the maximum F_345/385 ratio in a high Ca²⁺ solution (10 mM CaCl₂).

Gel Electrophoresis and Western Blotting

Cells were washed with ice cold phosphate-buffered saline (PBS, pH 7.4) and lysed with 30 sec sonication at 4°C in anti-phosphatase buffer (pH 7.4) containing (mM): 145 NaCl, 1.8 NaH₂PO₄, 8.6 Na₂HPO₄, 100 NaF, 10 Na₄P₂O₇, 2 Na₃VO₄, 2 EDTA and 0.2 µM microcystin and protease inhibitors as described previously (Luo et al. 2005). Protein content was determined by the bicinchoninic acid method. Lysate protein samples (30 µg) and pre-stained molecular mass markers (Bio-Rad, Hercules, CA) were resolved by SDS-PAGE (4–15% gradient gel or 10% gel). The resolved proteins were transferred onto a nitrocellulose membrane. The blots were blocked with 7.5% nonfat dry milk in Tris-buffered saline containing Tween 20 (0.1%) for 2 h at room temperature and then incubated with a primary antibody at 4°C overnight. After rinsing, the blots were incubated with horseradish peroxidase-conjugated secondary IgG for 1 h at room temperature. Bound antibody was visualized using an enhanced chemiluminescence assay (Amersham Corp, Piscataway, NJ). Relative changes in protein expression were estimated from the mean pixel density of each protein band using Image J software. The following antibodies and dilutions were used: rabbit anti-phospho-eIF2α (p-eIF2α, 1:1000, polyclonal), rabbit anti-ATF-4 (1:500, polyclonal) and mouse anti-α tubulin (1:8000, monoclonal).

STATISTICS

Statistical significance was determined by student’s t-test or an ANOVA (Bonferroni post-hoc test) in the case of multiple comparisons. A P-value < 0.05 was considered statistically significant. N values represent the number of different cultures used in each experiment.

RESULTS

Effects of DHA on P2Y receptor-mediated Ca²⁺ transient in normoxic astrocytes

Astrocytes express abundant purinergic P2Y receptors (Fumagalli et al. 2003). When ATP (100 µM), the physiological ligand of P2Y receptors, was applied to astrocytes under extracellular Ca²⁺-free conditions, there was an immediate transient increase in Ca²⁺_cyt with a peak value of 1048 ±173 nM that resolved by 3–4 min (Figure 1 A). The mean integrated Ca²⁺ release (IA_Ca2+) was 240 ± 50 nM*min (data not shown). A similar response occurred when cells were exposed to the specific P2Y1 agonist 2Me-S-ADP (10 µM, Figure 1 A). Activation of the P2Y1 receptor by 2Me-S-ADP triggered a similar increase in the Ca²⁺ transient (IA_Ca2+, 187 ± 17 nM*min). This implies that the ATP-mediated response can be largely attributable to activation of the P2Y1 receptor. This was confirmed by the effects of the specific P2Y1 receptor antagonist MRS-2179. As shown in Figure 1 (A, B), MRS-2179 (20 µM) blocked ~ 80% of the 2Me-S-ADP -induced transient increase in Ca²⁺_cyt (IA_Ca2+, 37 ± 12 nM*min, p < 0.05).

Figure 1. DHA suppresses P2Y1 receptor-mediated Ca²⁺ transients in normoxic astrocytes.

A. Representative tracing of changes in [Ca²⁺]_cyt induced by ATP (100 µM) and P2Y1 agonist 2Me-S-ADP (10 µM) under Ca²⁺ free conditions. Effects of P2Y1 antagonist MRS-2179 (20 µM) on 2MeSADP-mediated Ca²⁺ changes were tested. Data are means ± SEM, n = 3. B. Summary data. Integrated area (IA) under the Ca²⁺_cyt transient in response to 2Me-S-ADP was determined. Data are means ± SEM, n = 3. * p < 0.05 vs. 2Me-S-ADP. C. Effects of DHA (10 µM) on 2Me-S-ADP -induced Ca²⁺_cyt transient. Inset: Representative tracings showing pretreatment (3 min) of astrocytes with DHA (10 µM) prior to application of 2Me-S-ADP (10 µM) blocked the Ca²⁺_cyt transient. Data are means ± SEM, n = 3. * p < 0.05 vs. 2Me-S-ADP

Interestingly, DHA affected the P2Y1-mediated rise in Ca²⁺_cyt. As shown in Figure 1 C (inset), in the presence of 10 µM DHA, 2Me-S-ADP (10 µM) failed to trigger the Ca²⁺_cyt transient under extracellular Ca²⁺-free conditions. IA_Ca2+ following 2Me-S-ADP application was significantly suppressed in a dose-dependent manner (2.5, 5, and 10 µM DHA). 10 µM DHA blocked nearly ~90% of the P2Y1 receptor-mediated increase in Ca²⁺_cyt (Figure 1 C).

Effects of DHA on IP₃-mediated ER Ca²⁺ release in normoxic astrocytes

Effect of DHA on IP₃ receptor function was further investigated in normoxic conditions. In these experiments, the basal level of Ca²⁺_ER in astrocytes was 28.1 ± 1.1 µM (Figure 2 A, B). Application of 10 µM IP₃ to these cells triggered an immediate release of Ca²⁺_ER. Ca²⁺_ER decreased steadily to ~ 5 µM over 2–3 min in the presence of IP₃ (Figure 2 A, B). Astrocytes were treated with 10 µM DHA (prior to permeabilization) for 3 min. The basal level of Ca²⁺_ER was not affected by the DHA treatment (29.1 ± 0.6 µM). However, in the presence of DHA (10 µM), addition of IP₃ caused a much slower initial release of Ca²⁺_ER (9.5 ± 2.6 µM/min vs. 39 ± 6 µM/min, p < 0.05). At steady state, Ca²⁺_ER release was only 37 % in the presence of IP₃ and DHA (10 µM), compared to ~ 80% release in controls (Figure 2 B). Increasing DHA to 20 µM nearly blocked the IP₃-mediated Ca²⁺_ER release (23.5 ± 0.2 µM vs. control value of 28.7 ± 1.3 µM p < 0.05, Figure 2 A, B). These results clearly imply that DHA can directly inhibit IP₃ receptor-mediated Ca²⁺_ER release.

Figure 2. Effects of DHA on IP₃-mediated changes in [Ca²⁺]_ER in normoxic astrocytes.

A. Representative tracings of Ca²⁺_ER release in response to IP₃. IP₃ (10 µM) was applied to saponin-permeablized astrocytes to trigger the Ca²⁺_ER release. DHA (10 or 20 µM) was added 2–3 min prior to permeabilization and present in subsequent solutions. B. Summary data. Data are means ± SEM, n = 3. * p < 0.001 vs. corresponding baseline, # p < 0.001 vs. Con IP₃.

DHA does not affect store-operated Ca²⁺ entry following depletion of Ca²⁺_ER stores in normoxic astrocytes

It has been reported that DHA inhibits SOCE and prevents refilling of Ca²⁺_ER stores following their depletion in undifferentiated rat astrocytes (Sergeeva et al. 2002). In the current study, depletion of the Ca²⁺ ER stores was induced by G-protein coupled P2Y1 receptor activation. In order to test whether DHA affects SOCE in mouse cortical astrocytes, Ca²⁺_ER stores were emptied with P2Y1 specific agonist 2Me-S-ADP (10 µM) in the absence of extracellular Ca²⁺. Following return of extracellular Ca²⁺, an increase in Ca²⁺_cyt (a peak value of 47 ± 12 nM) was detected upon activation of SOCE (Figure 3 A, B). Adding DHA (2.5–10 µM) did not significantly affect SOCE in mouse astrocytes (p > 0.05). To further investigate the effects of DHA on SOCE, thapsigargin (2 µM) was used to deplete intracellular Ca²⁺ stores in the absence of extracellular Ca²⁺. As shown in Figure 3C, [Ca²⁺]_cyt in astrocytes were elevated resulting from activation of SOCE after thapsigargin-induced Ca²⁺_ER release. In the presence of DHA (10 µM), activation of SOCE remained unchanged (a Ca²⁺_cyt peak value of 122.7 ±10.6 nM vs. control value of 120.9 ±13.1nM). These findings clearly suggest that DHA selectively modulates IP₃ receptor-mediated Ca²⁺_ER release but not the SOCE in mouse cortical astrocytes.

Figure 3. DHA does not affect store-operated Ca²⁺ entry in astrocytes.

A. Representative tracing showing release of Ca²⁺_ER stores and store-operated Ca²⁺ entry (SOCE) in astrocytes. SOCE was determined after Ca²⁺_ER stores were emptied by application of 10 µM 2-MeSADP in the absence of extracellular Ca²⁺ (arrow). SOCE-mediated increase in [Ca²⁺]_cyt was detected when Ca²⁺ was reintroduced at 3–4 min. Inset: effects of DHA (10 µM) on SOCE were tested following 2-MeSADP application, added 3–4 min prior to reintroduction of Ca²⁺. B. Summary data of experiments in A. C. Representative traces showing thapsigargin-mediated release of Ca²⁺_ER stores and activation of SOCE in astrocytes. SOCE was determined after Ca²⁺_ER stores were emptied by application of 2 µM thapsigargin in the absence of extracellular Ca²⁺. DHA (10 µM) failed to block SOCE. Data are means ± SEM, n = 3–4.

Effects of DHA on Ca²⁺_cyt and Ca²⁺_ER in astrocytes following in vitro ischemia

As shown in Figure 4 A, a low level of Ca²⁺_cyt (65.4 ± 7.2 nM) was detected during 3 h normoxic incubation. 2 h OGD did not trigger any significant changes of Ca²⁺_cyt. Furthermore, Ca²⁺_cyt remained at the basal level during ~ 100 min REOX. This coincided with the sustained ER Ca²⁺ uptake during 0–120 min REOX (Figure 4 B). Interestingly, Ca²⁺_cyt started rising by 120 min REOX and reached a plateau level of 353 ± 69 nM between 150 ~ 180 min REOX. In the presence of 10 µM DHA, elevation of Ca²⁺_cyt was largely suppressed (p < 0.05). The DHA-mediated effects may, in part, result from DHA-mediated blockage of Ca²⁺_ER release during REOX (see below).

Figure 4. Effect of DHA on changes in [Ca²⁺]_cyt and [Ca²⁺]_ER in astrocytes following OGD and REOX.

A. Changes of [Ca²⁺]_cyt in astrocytes was determined with fura-2. Cells were either exposed to 3 h normoxia or 2 h OGD and 0–180 min REOX. In the DHA (10 µM) studies, DHA was present during REOX. Data are mean ± SEM, n = 3. * p < 0.05 vs. Normoxia control B. Changes of [Ca²⁺]_ER in astrocytes were determined with mag-fura-2. In the DHA (10 µM), ebselen (10 µM) or DiHDoHE (69 µM) studies, the drugs were present during 0–180 min REOX. Data are mean ± SEM, n = 3–7. * p < 0.05 vs. control, # p < 0.05 vs. OGD/REOX. C. Role of IP₃R and RyR in Ca²⁺_ER release in astrocytes following OGD/REOX. Changes of [Ca²⁺]_ER were determined at 90 and 120 min REOX following 2 h OGD. For the drug treatment, either 2-APB (100 µM), ryanodine (100 µM), or 2-APB (100 µM,) plus ryanodine (100 µM) were added at 0 min REOX. Data are mean ± SEM, n = 4–7. # p < 0.05 vs. OGD/REOX. D. Addition of DHA (10 µM) or the P2Y1 antagonist MRS-2179 (20 µM) at 80 min REOX prevents astrocytes from release of [Ca²⁺]_ER at 120, 140 or 160 min REOX. Data are mean ± SEM, n = 3. * p < 0.05 vs. corresponding REOX control.

We then investigated the possible effects of DHA on Ca²⁺_ER in ischemic astrocytes. As shown in Figure 4 B, basal [Ca²⁺]_ER was 27 ± 3 µM. It increased to 65 ± 4 µM at 2 h OGD (p < 0.05, Figure 4 B). [Ca²⁺]_ER was further augmented during 30–90 min REOX (p < 0.05) and reached a peak of 130 ± 3 µM (p < 0.05). However, an abrupt release of [Ca²⁺]_ER at a rate of 2 µM/min occurred beginning at ~ 100 min REOX. [Ca²⁺]_ER was reduced to basal levels by 140 min REOX (Figure 4 B). Interestingly, application of DHA (10 µM) at 0 min REOX prevented the overloading of Ca²⁺_ER and subsequent release at 120 min REOX (Figure 4 B).

We further investigated whether DHA itself or its oxidation products block changes of Ca²⁺_ER. We used the non-specific lipid oxygenase inhibitor ebselen to block the endogenous lipid oxygenase activity and prevent DHA oxidation (Walther et al. 1999). As shown in Figure 4 B, in the presence of ebselen (10 µM), DHA failed to block the loading of Ca²⁺_ER and the subsequent release following OGD/REOX. This suggests that the oxidized products of DHA via lipidoxygenase function were involved. Indeed, when we applied the DHA oxidation product DiHDoHE, an isoform of 10,17S-docosatriene (Chen et al. 2009; Serhan et al. 2006), OGD/REOX-induced changes in Ca²⁺_ER were largely inhibited, similar to DHA (Figure 4 B). Taken together, these results suggest that DHA oxidation products are important in maintaining astrocyte Ca²⁺_ER homeostasis under ischemic conditions.

Mechanisms underlying release of Ca²⁺_ER stores in astrocytes following in vitro ischemia

Ca²⁺ is released from ER primarily through the ryanodine receptor (RyR) and IP₃ receptor Ca²⁺ release channels (Henzi and MacDermott 1992). We examined whether the Ca²⁺_ER release results from IP₃R and/or RyR activation between 90 and 120 min REOX. As shown in Figure 4 C, inhibition of IP₃R with a non-selective inhibitor 2-APB (100 µM) reduced the OGD/REOX-mediated [Ca²⁺]_ER release by 60% (p < 0.05). These data imply that activation of IP₃R plays a partial role in the Ca²⁺_ER release. To assess the role of RyR in release of Ca²⁺_ER stores, we examined the effects of ryanodine on depletion of Ca²⁺_ER stores. When RyR was inhibited by 100 µM ryanodine, there was 34 % reduction of the OGD/REOX-mediated release of Ca²⁺_ER stores at 120 min REOX (p < 0.05). Moreover, in the presence of both 2-APB and ryanodine, the effects were additive and Ca²⁺_ER release was absent. In fact, [Ca²⁺]_ER was significantly higher than the peak values in the non-treated astrocytes (p < 0.05, data no shown). These findings suggest that both IP₃R and RyR are involved in Ca²⁺_ER release in astrocytes following in vitro ischemia.

To specifically investigate effects of DHA on ER Ca²⁺ efflux, cells were exposed to DHA (10 µM) at 80 min REOX, a time just prior to Ca²⁺_ER release. [Ca²⁺]_ER was subsequently monitored at 120,140 or160 min REOX. As shown in Figure 4 D, the delayed addition of DHA at 80 min REOX effectively blocked ER Ca²⁺ efflux. Similarly, blocking P2Y1 receptor activity with MRS2179 at 80 min REOX also prevented the ER Ca²⁺ efflux (Figure 4 D). These findings further suggest that DHA indeed prevents ER Ca²⁺ efflux in astrocytes under ischemic conditions, which involves IP₃ mediated release pathways.

Effect of P2Y receptor activation in astrocytes under ischemic conditions

Figure 5 A shows that P2Y1 receptor agonist 2Me-S-ADP triggered a transient Ca²⁺_cyt response in normoxic astrocytes. However, P2Y1 receptor activity was transiently inhibited following 2 h OGD (~ 60%) and partially recovered at 120 min REOX (Figure 5 B). These data suggest that P2Y1 receptor activation may contribute to delayed Ca²⁺_ER release and Ca²⁺_cyt rise. This speculation was supported by the finding that inhibition of P2Y1 receptor activity with MRS2179 significantly attenuated Ca²⁺_cyt dysregulation during 120 min REOX (Figure 5 C). In addition, inhibition of mGluR5 receptor by MPEP did not further inhibit the delayed Ca²⁺_cyt rise in mouse astrocytes in the presence of MRS2179 (Figure 5 D). These results suggest that REOX triggered Ca²⁺_cyt increase is largely mediated by the activation of P2Y1 receptors coupled to IP3 mediated Ca²⁺_ER release.

Figure 5. P2Y1 receptor-mediated Ca²⁺ transients after OGD/REOX.

A. 2Me-S-ADP (10 µM)-induced Ca²⁺_cyt transient in normoxic or OGD/REOX-treated cells. B. Summary data of IA under the Ca²⁺_cyt transient in response to 2Me-S-ADP. Data are mean ± SEM, n = 3. * p < 0.05 vs. Normoxia, # p < 0.05 vs. OGD. C. Changes of [Ca²⁺]_cyt in astrocytes was determined with fura-2 under normoxic conditions and with OGD/REOX. In some studies, the P2Y1 antagonist MRS2179 (20 µM) was present during 0–180 min h REOX. Data are mean ± SEM, n = 3. * p < 0.05 vs. normoxia, # p <0.05 vs. OGD/REOX. D. The mGluR5 antagonist MPEP (50 µM) did not provide further inhibition of OGD/REOX mediated rise in [Ca²⁺]_cyt. Data are means ± SEM, n = 3. * p < 0.05 vs. normoxia, # p < 0.05 vs. OGD/REOX

Effect of DHA on astrocyte ER stress

Disruption in ER Ca²⁺ homeostasis can lead to unfolded protein response and development of ER stress. In our previous study, we have shown that disruption in ER Ca²⁺ homeostasis increases the expression of phosphorylated eIF2α (p-eIF2α) and ER stress following OGD/REOX (Liu et al. 2010). In the present study, we first investigated effects of DHA on expression of p-eIF2α. As shown in Figure 6 A, expression of p-eIF2α was increased during 1–3 h REOX (~ 3.4 folds). Treatment of astrocytes with DHA during 0–3 h REOX significantly attenuated p-eIF2α protein expression (p < 0.05). Moreover, 1–3 h REOX led to significant elevation of ATF-4 protein in astrocytes (Figure 6 B). DHA also blocked the up-regulation of ATF-4 protein. These results collectively indicate that DHA not only blocks dysregulation of Ca²⁺_ER but also attenuates ER stress in astrocytes following OGD/REOX.

Figure 6. DHA inhibits the OGD/REOX-mediated up regulation of ER stress proteins and cell death.

A. Representative immunoblot showing regulation of p-eIF2α protein in normoxic control, 2 h OGD, 1–3 h REOX. Data are mean ± SEM. n=4. * p < 0.05 vs. Con; # p < 0.05 vs. corresponding 1–3 h REOX. B. Representative immunoblot of ATF4 protein. DHA (10 µM) was added during 1–3 h REOX. The same blot was probed with anti α-tubulin antibody as a loading control. Data are mean ± SEM. n=4. * p < 0.05 vs. Con; # p < 0.05 vs. corresponding 1–3 h REOX. C. Representative tracings of fura-2 fluorescence (Ca²⁺-insensitive wavelength, 340 nm) during REOX following 2 h of OGD from 10 cells. The abrupt loss of fluorescence (arrow) indicates compromise of plasma membrane integrity and cell death. D. Summary data of cell survival under normoxic conditions and at the end of 180 min REOX in OGD and OGD + DHA astrocytes. Data are mean ± SEM, n=3–4. * p < 0.05 vs. Con; # p < 0.05 vs. OGD/REOX.

We further investigated whether suppression of ER stress via DHA would have an impact on cell survival. We calculated cell survival by accessing the fura-2 dye retention in astrocytes during 180 min REOX. A sudden loss of the fluorescent dye reflects loss of plasma membrane integrity and cell death (Kintner et al. 2007). As shown in Figure 6 C, under normoxic conditions, most astrocytes retained the fluorescence dye during 3 h of perfusion and exhibited little cell damage (< 5%). 2 h OGD and 3 h REOX reduced cell survival to 73 ± 3% (Figure 6 C, D). Cells died between 140 and 180 min of REOX (shown as an abrupt loss of the dye, Figure 6 C), which coincided with the increases in cytosolic Ca²⁺ and loss of ER Ca²⁺ (see Figure 4). Interestingly, when 10 µM DHA was present throughout REOX, cell survival was maintained at 92 ± 2% (p < 0.05, Figure 6 C). Taken together, these data suggest that DHA suppressed ER stress as well as astrocyte death following in vitro ischemia.

DISCUSSION

DHA attenuates P2Y1 receptor-mediated Ca²⁺_cyt transients

Astrocytes play a key role in the synthesis and supply of DHA within the brain (Moore 2001). DHA is released readily from the astroglial membranes under basal and stimulated conditions, and supplied to the neurons (Moore 2001). However, the effects of DHA on astrocyte function are not well studied. The functions of astrocytes are regulated by changes of intracellular Ca²⁺ concentration and much of the Ca²⁺ signal may come from the ER Ca²⁺ stores (Takano et al. 2009). DHA plays an important role in the regulation of intracellular Ca²⁺ in astrocytes. In rat astrocytes, DHA inhibits the SOCE, reduces the amplitude of Ca²⁺ response to agonists of G protein coupled receptors, and suppresses intracellular Ca²⁺ oscillations (Sergeeva et al. 2005).

Interestingly, in our current study, we found that the P2Y1 receptor-mediated Ca²⁺_cyt increase was blocked by DHA in a dose-dependent manner. 10 µM DHA blocked nearly 90% of the P2Y1 receptor-mediated Ca²⁺_cyt increase. It is known that activation of P2Y1 receptors leads to phospholipase C activation and IP₃ generation, which triggers Ca²⁺ release from ER stores in astrocytes (James and Butt 2002). Our data show that DHA decreased the exogenous IP₃-mediated Ca²⁺_ER release rate and reduced the amount of Ca²⁺_ER release by ~ 40%. These data support our speculation that DHA can directly inhibit IP₃ receptor-mediated Ca²⁺_ER release.

In contrast to the finding of Sergeeva et al. that DHA inhibits SOCE in rat astrocytes (Sergeeva et al. 2005), in our study, DHA failed to inhibit SOCE following depletion of Ca²⁺_ER stores in mouse cortical astrocytes. Our findings suggest that DHA selectively modulates IP₃ receptor-mediated Ca²⁺_ER release but not the SOCE in mouse astrocytes.

Effects of DHA in regulating ER Ca²⁺ and ER stress

DHA is the most abundant essential omega-3 fatty acid in the brain and has been shown to be neuroprotective in numerous in vivo and in vitro experiments. For instance, DHA protects neuronal cells against serum starvation-mediated apoptosis by up-regulation of Akt and extracellular-signal-regulated kinase (ERK)(Akbar et al. 2005). Administration of DHA to rats stimulated the ERK and Bcl-2 mediated anti-apoptotic cascade in cerebral ischemia (Akbar et al. 2005; Pan et al. 2009). DHA treatment also suppresses lipopolysaccharide/interferon-γ stimulated release of interleukin-6 in cultured cortical glial cells (Pan et al. 2009). However, the function of DHA in regulation of astrocyte Ca²⁺ homeostasis under ischemic conditions is unknown.

We found that DHA blocked the ER Ca²⁺ release in ischemic astrocytes. It is well documented that IP₃- and caffeine/ryanodine-sensitive stores are the two major ER Ca²⁺ stores in astrocytes (Golovina and Blaustein 2000). Activation of these two receptors and increases in Ca²⁺_cyt have been detected in astrocytes in vivo under ischemic conditions (Ding et al. 2009). We show in this study that inhibition of IP₃R is more effective in attenuating Ca²⁺_ER store release than RyR inhibition following OGD/REOX. Blocking both IP₃R and RyR had an additive effect and completely prevented Ca²⁺_ER release. These results suggest that both IP₃R and RyR are involved in Ca²⁺_ER release in astrocytes following in vitro ischemia. We concluded that DHA can attenuate Ca²⁺_ER release via inhibiting IP₃R. In addition to affecting Ca²⁺_ER release, exposing DHA to astrocytes also attenuated Ca²⁺_ER overload following OGD. We speculate that this, in part, results from DHA-mediated inhibition of ER Ca²⁺-ATPase. This view is supported by a study that cardiac sarcoplasmic reticulum Ca²⁺-ATPase activity was reduced by ~30% in rats fed with a DHA enriched diet (Taffet et al. 1993). The underlying mechanisms could involve changes of the composition and/or structure of the phospholipid membranes of ER, which can affect Ca²⁺-ATPase protein turnover and activity (Taffet et al. 1993).

ER is a multifunctional organelle involved in folding and processing of proteins, intracellular Ca²⁺ homeostasis, and cell death signal activation (Baumann and Walz 2001). Disturbance of Ca²⁺ homeostasis and accumulation of unfolded proteins in the ER cause ER stress and activates UPR (Kaufman 1999). UPR results in the phosphorylation and activation of PERK (PKR-like ER kinase), which triggers a cascade of events including the phosphorylation of eIF2α, inhibition of protein synthesis, and activation of apoptotic signals (Harding et al. 2002; DeGracia and Hu 2007; Paschen 2003). We detected a rapid up regulation of p-eIF2α and ATF-4 proteins during 1–3 h REOX, which coincided with the ER Ca²⁺ overload and subsequent depletion. Interestingly, DHA not only blocked the ER Ca²⁺ overload and release, it also reduced p-eIF2α and ATF-4 expression and cell death. Taken together, our data strongly suggest that DHA may increase cellular tolerance to stress by prevention of Ca²⁺_ER overload, depletion of Ca²⁺ stores, and ER stress.

Contribution of bioactive DHA metabolites

DHA’s effects on ER Ca²⁺ dysregulation can result from its direct action or through its metabolites. DHA can be transformed by a 15-lipidoxygenases-like enzyme to NPD1 (Lukiw et al. 2005). In the presence of the 5 and 15 LOXs potent inhibitor ebselen, the effect of DHA on ER Ca²⁺ loading and depletion was blocked. These results clearly indicated involvement of DHA metabolites. A possible candidate is NPD1. This speculation was confirmed by the use of DiHDoHE, an isomer of NPD1. DiHDoHE blocked ER Ca²⁺ release in ischemic astrocytes, mimicking the DHA effects. NPD1 is neuroprotective in models of experimental stroke by down-regulating brain ischemia reperfusion-induced leukocyte infiltration, proinflammatory signaling, and infarct size (Bazan, NG et al., 2009). Our study suggests that NPD1 could reduce ischemic brain damage via decreasing astrocyte ER Ca²⁺ dysregulation and ER stress.

In summary, we report here that DHA and its oxidized product NPD1 blocked purinergic receptor P2Y1-mediated Ca²⁺ release from ER Ca²⁺ stores in part via directly inhibiting IP₃R. DHA also attenuated the delayed rise in Ca²⁺_cyt, ER stress, and astrocyte death following in vitro ischemia. Taken together, our findings suggest that the DHA-mediated protective effects in astrocytes may in part be attributable to its inhibition of astrocyte Ca²⁺ dysregulation and ER stress following in vitro ischemia.

Abbreviations

dBcAMP

dibutyryl cyclic AMP

[Ca²⁺]_cyt

calcium concentration cytoplasmic

[Ca²⁺]_ER

calcium concentration endoplasmic reticulum

DIV

days in culture

ER

endoplasmatic reticulum

FBS

fetal bovine serum

Fura-2 AM

Fura-2 acetoxymethyl ester

HBSS

Hanks balanced salt solution

IP₃R

inositol 1,4,5-triphosphate receptor

mGluR5

Group I metabotropic glutamate receptor 5

OGD/REOX

oxygen glucose deprivation and reoxygenation

p-eIF2α

phosphorylated eukaryotic initiation factor 2 alpha

RyR₂

ryanodine receptor type 2

DHA

Docosahexaenoic acid (22:6n-3)

2Me-S-ADP

methylthioadenosine- 5'- O- diphosphate

MPEP

6-Methyl-2-(phenylethynyl) pyridine hydrochloride

IP₃

D-myo-inositol 1,4,5-triphosphate

Mag-fura 2-AM

mag fura-2 acetoxymethyl ester

Fura 2-AM

fura-2 acetoxymethyl ester

2-APB

2-Aminoethoxydiphenylborane

Ebselen

2-phenyl-1,2-benzisoselenazol-d(2H)-one

DiHDoHE

10(S),17(S)-dihydroxy-4Z,7Z,11E,13Z,15E,19Z-docosahexaenoic acid10(S),17(S)-DiHDoHE

ATF-4

activating transcription factor 4

UPR

unfolded protein response

SOCE

store-operated Ca²⁺ entry