1Toll-like receptor 4 activation enhances Orai1-mediated calcium signal promoting cytokine production in spinal astrocytes

Abstract

Toll-like receptor 4 (TLR4) has been implicated in pathological conditions including chronic pain. Activation of astrocytic TLRs leads to the synthesis of pro-inflammatory cytokines like interleukin 6 (IL-6) and tumor necrosis factor-ɑ (TNF-α), which can cause pathological inflammation and tissue damage in the central nervous system. However, the mechanisms of TLR4-mediated cytokine releases from astrocytes are incomplete understood. Our previous study has shown that Orai1, a key component of calcium release activated calcium channels (CRACs), mediates Ca²⁺ entry in astrocytes. How Orai1 contributes to TLR4 signaling remains unclear. Here we show that Orai1 deficiency drastically attenuated lipopolysaccharides (LPS)-induced TNF-α and IL-6 production in astrocytes. Acute LPS treatment did not induce Ca²⁺ response and had no effect on thapsigargin (Ca²⁺-ATPase inhibitor)-induced store-dependent Ca²⁺ entry. Inhibition or knockdown of Orai1 showed no reduction in LPS-induced p-ERK1/2, p-c-Jun N-terminal kinase, or p-p38 MAPK activation. Interestingly, Orai1 protein level was significantly increased after LPS exposure, which was blocked by inhibition of NF-κB activity. LPS significantly increased basal Ca²⁺ level and SOCE after exposure to astrocytes. Moreover, elevating extracellular Ca²⁺ concentration increased cytosolic Ca²⁺ level, which was almost eliminated in Orai1 KO astrocytes. Our study reports novel findings that Orai1 acts as a Ca²⁺ leak channel regulating the basal Ca²⁺ level and enhancing cytokine production in astrocytes under the inflammatory condition. These findings highlight an important role of Orai1 in astrocytic TRL4 function and may suggest that Orai1 could be a potential therapeutic target for neuroinflammatory disorders including chronic pain.

1. Introduction

Astrocytes play a key role in maintaining ion and pH homeostasis, promoting gliotransmitter synthesis/release, removal of neurotransmitters within the central nervous system (CNS) [1–4]. Astrocytes possess numerous fine processes surrounding neurons and blood vessels providing brain glucose supply and antioxidant defense [5, 6]. However, during injury or disease, astrocytes become reactive, producing proinflammatory mediators and releasing reactive oxygen species [7, 8]. Reactive astrocytes have been implicated in many central nervous system (CNS) disorders, such as epilepsy, Alzheimer’s Disease, Parkinson’s disease, and multiple sclerosis [9–12]. Spinal astrocytes have been recognized as an important player in chronic pain and itch [13–15]. Peripheral inflammation or injury can trigger astrocyte activation (reactive astrogliosis), which is thought to be involved in central mechanisms underlying the maintenance and exaggeration of chronic pain [16, 17].

Toll-like receptor 4 (TLR4) is a member of the TLR family proteins that play a crucial role in the initiation of innate immune response. TLR4 is one of the most studied among the family members and widely expressed on many different cell types in the human body. TLR4 is expressed on astrocytes and is involved in a variety of inflammatory conditions associated with excessive cytokine production [18–20]. TLR4 signaling is involved in several CNS diseases, such as AD progression and Parkinson’s disease [18, 21, 22]. Accumulating evidence indicates that TLR4 contributes to the development of pain hypersensitivity in several animal models of pain [20, 23–25]. TLR4 signaling also enhances histamine-mediated itch sensation and is essential for the development of chronic itch [15, 26]. It is well established that activation of astrocytic TLR4 leads to the downstream stimulation of the MAPK and NF-κB signaling pathways that trigger the production of inflammatory mediators and cytokines, however, the molecular entities in these pathways remain poorly defined.

Intracellular Ca²⁺ is vital for a wide range of cellular processes in astrocytes and mediates remarkable cellular functions, such as gliotransmission and cytokine production [27, 28], which are involved in the process of pathological CNS diseases [29–31]. Astrocytic Ca²⁺ signals can be derived from intracellular Ca²⁺ release by intracellular organelles and extracellular Ca²⁺ influx through various Ca²⁺ permeable channels on the plasma membrane. Astrocytes express numerous Ca²⁺ permeable channels such as L-type voltage-gated Ca²⁺ channels and transient receptor potential (TRP) channels, regulating Ca²⁺ homeostasis [28, 32]. Ca²⁺ also acts as a critical molecular coordinator of neuron-glia communication [33, 34]. Given the important role of Ca²⁺ in physiological and pathological functions of astrocytes, the precise profile of Ca²⁺ signaling in astrocytes remains to be defined.

Store-operated Ca²⁺ channels (SOCs) are the major Ca²⁺ entry pathway in non-excitatory cells. Calcium release activated calcium channels (CRACs, known as Orai1/2/3) have emerged as important Ca²⁺ signals in many cell types. CRACs can be activated by Ca²⁺ release from the endoplasmic reticulum (ER) [35, 36]. It is well known that SOC entry is required for immune cell activation and that loss of SOC entry leads to various diseases including severe combined immunodeficiency, thrombocytopenia, tubular aggregate myopathy [37, 38]. SOCs have emerged as the leading Ca²⁺ pathway in non-excitatory and excitatory cells [39, 40]. Orai1 is a primary component of CRACs and its role in immune responses and autoimmunity is well-documented. Orai1-mediated Ca²⁺ entry plays a key role in regulating NF-κB activation and nuclear localization [41, 42]. Previous studies have shown that SOCs are expressed in hippocampal astrocytes and contribute to spontaneous astrocytic Ca²⁺ transients and play a major role in gliotransmission [1, 43]. We have reported that Orai1 is a key player in astrocytic SOC entry [44]. Activation of SOCs by thapsigargin (a Ca²⁺-ATPase inhibitor) triggers cytokine production, and inhibition CRACs or knockdown of Orai1 markedly attenuates lipopolysaccharides (LPS)-induced cytokine production in spinal astrocytes [44]. However, how Orai1 is involved in this process is unknown.

In this study, we show that Orai1 deficiency drastically decreased LPS-induced cytokine production. Acute activation of TLR4 by LPS neither induced calcium release nor calcium entry. LPS application did not modulate Orai1 function directly. Inhibition or knockdown of Orai1 did not alter in LPS-induced MAPK activation. Interestingly, we found that Orai1 protein level was significantly increased after LPS exposure for a few hours, which was blocked by inhibition of NF-κB activity. LPS significantly increased basal Ca²⁺ level and SOCE after 3–4 hours of exposure. Moreover, elevating extracellular Ca²⁺ concentration increased cytosolic Ca²⁺ level, which was almost eliminated in Orai1 KO astrocytes. Taken together, these findings identify Orai1 as a Ca²⁺ leak channel, contributing astrocytic Ca²⁺ homeostasis and TLR4-mediated cytokine production.

2. Results

2.1. Orai1 deficiency impairs thapsigargin (TG)- and LPS-induced cytokine production in spinal astrocytes

Ca²⁺ is an important regulator of many cellular functions in a variety of cell types. Non-selective Ca²⁺ permeable channels-mediated Ca²⁺ entry is crucial for TLR4-mediated NFκB activation and cytokine production in macrophages and epithelial cells [32, 45–47]. However, the role of store-dependent Ca²⁺ entry in TLR4-mediated functions has not been established. It has been reported that SOC entry is dispensable for LPS-induced cytokine production in microphages [48]. We have shown that inhibition or knockdown of Orai1 drastically reduces SOCE and cytokine production [44]. Since pharmacological approaches may not specifically inhibit Orai1, here we used an Orai1 mutant mouse line, which had been previously characterized and backcrossed to CD-1 for at least 8 generations [49–51], to confirm these observations. Spinal astrocytes were cultured from Orai1 deficient (KO) and its littermate wild type (WT) mice. Western blotting results revealed that Orai1 protein was absent, while other CRAC family members were intact in Orai1 KO astrocytes compared to WT astrocytes (Fig. 1A). We performed live cell Ca²⁺ imaging in spinal cord astrocytes. Astrocytes were pretreated with the Ca²⁺-free Tyrode’s solution. Consistent with our previous study [44], 1μM TG (Ca²⁺-ATPase inhibitor), transiently induced Ca²⁺ release from the ER, which was followed by a sustained increase when 2 mM Ca²⁺ was added. TG-induced Ca²⁺ entry was drastically reduced, while Ca²⁺ release was not affected in Orai1 KO astrocytes (Fig. 1B, C). To determine whether reduction of SOCE observed in Orai1^−/− astrocytes is due to loss of Orai1 expression, we transfected Orai1^−/− astrocytes with wild type Orai1-CFP. Astrocytes transfected with Orai1-CFP showed increased basal Ca²⁺ level and SOCE compared to astrocytes transfected with GFP plasmid (Supplementary Figure 1), suggesting that Orai1 is responsible for astrocytic SOCE. As expected, TG-induced production of TNF-α and IL-6 was eliminated in Orai1 KO group (Fig. 1D, E). These findings demonstrate that Orai1 is necessary for TG-induced SOCE and cytokine secretion.

Figure 1. Deficiency of Orai1 abolishes thapsigargin (TG)-induced cytokine production in spinal astrocytes.

(A) Representative images of western blots probed with antibodies against Orai1, Orai2, Orai3 and actin, respectively. Spinal cord samples were collected from adult wild type (WT) and Orai1 knockout (KO) mice. (B) Representative Ca²⁺ imaging recordings of TG-induced Ca²⁺ release and SOCE in WT and Orai1 KO astrocytes. (C) Summary of the result of TG-induced SOCE in WT and Orai1 KO astrocytes, n=13–18. (D, E) TG-induced IL-6 (D) and TNF-ɑ (E) production in WT and Orai1 KO astrocytes, n=9. (F-I) LPS-induced cytokine production in Orai1 KO and WT astrocytes: IL-6 (F) and TNF-ɑ (G) production after 24 h treatment; IL-6 (H) and TNF-ɑ (I) production after 6 h treatment, n = 5–8 in each group. Values represent mean ± SEM; *** P < 0.001 compared with WT by Student’s t test (C) or compared with control by one-way ANOVA followed by the Tukey’s multiple comparison test (D-I). ### P < 0.001 compared with WT by one-way ANOVA followed by the Tukey’s multiple comparison test F-I).

To determine whether Orai1 deficiency affects TLR4-mediated cytokine production, we used Lipopolysaccharide (LPS), a potent TLR4 agonist. LPS dose-dependently increased production of IL-6 and TNF-α after 24 hours of treatment (Supplementary Figure 2A, B), which was completely blocked by a specific TLR4 antagonist TAK-242 [52] (Supplementary Figure 2C, D). To confirm LPS-induced cytokine production is mediated by TLR4, astrocytes were transfected with control siRNA or TLR4 siRNA. Similarly, TLR4 siRNA significantly reduced IL-6 and TNF-α production (Supplementary Figure 2E, F). The TLR4 reduction was evaluated by RT-PCR (Supplementary Figure 2G). Importantly, LPS-induced production of IL-6 and TNF-α was markedly attenuated in Orai1 KO astrocytes (Fig. 1F, G). To determine whether Orai1 is involved in early cytokine production, we examined LPS-induced cytokine production after 6 hours of treatment and the similar result was observed (Fig. 1H, I). These results indicate that Orai1 plays a key role in TLR4-mediated cytokine production.

2.2. Acute activation of TLR4 does not modulate Orai1 functions

It has been shown that LPS-induced activation of TLR4 causes Ca²⁺ release from ER in endothelial cells and Ca²⁺ flux in murine macrophages [53, 54]. We wondered whether astrocytic Orai1 is involved in the LPS-induced Ca²⁺ response. Ca²⁺ imaging was performed in astrocytes. Interestingly, bath application of LPS neither elicited Ca²⁺ release in the absence of Ca²⁺, nor changed basal Ca²⁺ level in the presence of 2 mM Ca²⁺ (Fig. 2A, B), while 2-Methylthioadenosine (purinoceptor P2Y agonist) application caused Ca²⁺ release from intracellular organelles (Fig. 2A), and ATP 100 μM induced robust Ca²⁺ response in these cells (Fig. 2B, C). We then tested if LPS enhances Orai1 function. Astrocytes were pretreated with vehicle (PBS) or 10 ng/mL LPS for 10 min. TG-induced Ca²⁺ release and entry in LPS-treated cells were comparable to those in PBS-treated cells (Fig. 2D–F). These results suggest that acute activation of TLR4 is not able to directly trigger Ca²⁺ response or modulate Orai1 function.

Figure 2. Acute LPS treatment neither induce Ca²⁺ response nor increase SOCE.

(A) Effect of LPS 10 ng/ml and 1 μM 2-Methylthioadenosine (P2 purinoceptor agonist) on cytosolic Ca²⁺ level under the 0 Ca²⁺ condition. (B) Effect of LPS and ATP on cytosolic Ca²⁺ level under the 2 mM Ca²⁺ condition. (C) Summary of the effect of 10 ng/ml LPS on basal Ca²⁺ level. (D) Representative Ca²⁺ imaging recordings of TG-induced Ca²⁺ responses. (E, F) Summary of the effect of LPS on TG-induced Ca²⁺ release (E) and Ca²⁺ influx (F). Values represent mean ± SEM; n = 13–22.

2.3. Inhibition or deficiency of Orai1 does not affect LPS-induced initial mitogen-activated protein kinase (MAPK) activation.

It is well known that MAPK activation is required for the production of proinflammatory cytokines in TLR signaling [55, 56]. To determine whether Orai1 regulates cytokines production via the MAPK pathway, we examined LPS-induced MAPK activation in astrocytes. The activation of ERK, p38, and JNK were assessed by measuring the relative expression of p-ERK, p-p38, and p-JNK. Astrocytes were treated with LPS from 5 minutes to 1 hour. ERK and p38 activation occurred within 5 min and maintained for at least 1 hour, but JNK activation started at later time points (Supplementary Figure 3). Inhibition of MEK (upstream of ERK) by PD58059, p38 by SB203580, or JNK by SP600125 reduced LPS-induced cytokine production in a concentration-dependent manner (Supplementary Figure 4A–F), respectively. These effects were not due to changes of cell viability (Supplementary Figure 4G). To determine whether store-operated Ca²⁺ entry contributes to TLR4-mediated MAPK activation, we pretreated astrocytes with vehicle (0.1% DMSO) or a CRAC channel blocker YM-58483 (10 μM) for 30 min. Surprisingly, ERK, p38 and JNK activation was not affected by YM-58483 (Fig. 3A–C). To further determine whether Orai1 contributes to TLR4-mediated MAPK activation, astrocytes were exposed to 10 ng/mL LPS for 20 min. Consistently, LPS-induced ERK and p38 activation in Orai1 KO cells were similar to those in WT astrocytes (Fig. 3D–F). These data suggest that Orai1 is downstream of TLR4-MAPK signaling.

Figure 3. Inhibition or deficiency of Orai1 does not affect LPS-induced mitogen-activated protein kinase (MAPK) activation in astrocytes.

(A-C) Effect of YM-58483 on LPS-induced ERK activation (A), p38 activation (B) and JNK activation (C). The p-ERK, p-p38 and p-JNK levels were quantified as a ratio to total ERK, p38 and JNK, and were normalized to levels of their PBS control. (D) Representative images of western blots probed with antibodies against p-ERK, ERK, p-p38 and p38, respectively. Samples were collected from cultured WT or Orai1 KO astrocytes treated with PBS or LPS. (E, F) LPS-induced ERK (E) and p38 (F) activation in astrocytes cultured from Orai1 KO and WT littermates, n = 3–4 samples (each group). Values represent mean ± SEM; *P < 0.05 compared with PBS control by one-way ANOVA followed by the Tukey’s multiple comparison test.

2.4. Activation of TLR4 increases Orai1 expression via NF-κB signaling

We then asked whether LPS treatment alters Orai1 expression. Western blotting was performed in cultured astrocytes treated with PBS or 10 ng/mL LPS for 6 hours (the time point for measuring cytokine production). LPS treatment significantly increased Orai1 and STIM1 expression after 6 hours (Fig. 4A). To determine the time course of LPS-induced Orai1 upregulation, astrocytes were treated with LPS 10 ng/mL at different time points from 30 min to 6 hours. Orai1 upregulation occurred one hour after LPS treatment with the peak Orai1 level at 4-hour time point. (Fig. 4B). To determine whether Orai1 upregulation increases Ca²⁺ signal, we performed calcium imaging recordings in astrocytes treated with PBS or 10 ng/mL LPS for 3–4 hours. Basal Ca²⁺ level was significantly higher in LPS-treated cells compared to PBS-treated cells (Fig. 4C, D). Moreover, in LPS-treated astrocytes, TG-induced SOCE was also significantly increased (Fig. 4C, E), but TG-induced Ca²⁺ release was not altered (Fig. 4C). These results suggest that upregulation of Orai1 protein contributes to Orai1-mediated Ca²⁺ signal.

Figure 4. Activation of TLR4 increases expression and function of Orai1 and STIM1 in astrocytes.

(A) Effect of LPS treatment for 6 h on Orai1 and STIM1 protein expression, n = 3. (B) Time course of LPS-induced Orai1 upregulation, n = 2. (C) Representative recordings of TG-induced Ca²⁺ release and Ca²⁺ entry after PBS or LPS treatment (3–4 hours). (D) Effect of LPS treatment on basal Ca²⁺ level, n = 72–201. (E) Effect of LPS treatment on TG-induced SOCE, n = 72–102. (F) Effect of inhibition of IK/NF-κB with 10 μM IKK-16 (IKK) on LPS-induced NF-κB activation, n = 2. (G) Effect of IKK-16 on LPS-induced Orai1 upregulation, n = 5. Values represent mean ± SEM; ** P < 0.05, ** P < 0.01, *** P < 0.01 compared with PBS by Student’s t test.

NF-κB is one of the key transcription factors in the inflammatory response. It is well known that LPS stimulation triggers post-translational modifications in the p65 subunit of NF-κB (NF-κB p65) prior to nuclear translocation [57, 58]. To determine whether Orai1 upregulation is mediated by NF-κB activity, we used IKK-16 (a selective IκB kinase (IKK) inhibitor), which inhibits NF-κB activation [59, 60]. Astrocytes were pretreated with IKK-16 at 5 or 10 μM for 30 min. IKK-16 at a higher concentration (10 μM) abolished LPS-induced phosphorylation of NF-κB (p-NF-κB) (Fig. 4F). Using this concentration, we found that Orai1 upregulation was eliminated by IKK-16 (Fig. 4G). This result indicates that Orai1 upregulation was mediated by NF-κB signaling.

2.5. Orai1 contributes to basal Ca²⁺ homeostasis

To determine whether Orai1 plays a role in Ca²⁺ homeostasis in astrocytes, we measured resting Ca²⁺ levels in cultured astrocytes using live-cell Ca²⁺ imaging. Under 2 mM Ca²⁺ condition, YM-58483 and 2-APB (another CRAC channel blocker) dose-dependently decreased basal Ca²⁺ level in astrocytes (Fig. 5A, B). To confirm that these effects are mediated by the CRAC channel family, we used control siRNA or targeting siRNAs against Orai1, Orai2, and Orai3, respectively. Only Orai1 siRNA reduced basal Ca²⁺ level (Fig. 5C). Their knockdown efficiency was evaluated in our previous study [44], and confirmed in the present study by Western blot analysis (Supplementary Figure 5A). To further confirm the Orai1 siRNA result, we cultured astrocytes from Orai1 KO and WT neonatal mice. Consistently, astrocytes from Orai1 deficiency mice showed significant reduction of basal Ca²⁺ level (Fig. 5D). To determine whether Orai1’s modulation is store (STIM)-dependent, we examined the effect of the STIM inhibitor ML-9 on basal Ca²⁺ level. Twenty-five μM ML-9 (an effective concentration for SOCE) treatment did not reduce basal Ca²⁺ level (Fig. 5E), while increasing ML-9 concentration to 50 μM resulted in slight Ca²⁺ response. To further rule out the possibility of STIM protein’s involvement, we also used the knockdown approach and found that transfection of STIM1 siRNA or STIM2 siRNA into astrocytes did not alter basal Ca²⁺ level (Fig. 5F). The reduction of STIM1 and STIM2 protein expression was confirmed in STIM1 siRNA- or STIM2 siRNA-treated astrocytes (Supplementary Figure 5B). These results indicate that Orai1 mediates a store-independent Ca²⁺ signal regulating basal Ca²⁺ homeostasis.

Figure 5. Orai1 contributes to basal Ca²⁺ homeostasis.

(A) Effect of YM-58483 on basal Ca²⁺ level, n = 14–22. (B) Effect of 2-APB on basal Ca²⁺ level, n = 14–22. (C) Effects of specific siRNAs against Orai members on the basal Ca²⁺ concentration measured in astrocytes transfected with control siRNA, Orai1 siRNA, Orai2 siRNA or Orai3 siRNA, n = 15–28. (D) The basal Ca²⁺ concentrations measured in astrocytes cultured from WT and Orai1 KO mice, n = 21–25. (E) Effect of ML-9 on basal Ca²⁺ level, n = 13. (F) Effect of STIM1 siRNA and STIM2 siRNA on basal Ca²⁺ concentration measured in astrocytes transfected with control siRNA, STIM1 siRNA or STIM2 siRNA, n = 34–45. Values represent mean ± SEM, ***P < 0.001 compared with control by one-way ANOVA followed by the Tukey’s multiple comparison test or compared with WT by Student’s t test.

2.6. Orai1 functions as a leak Ca²⁺ channel

We then asked whether Orai1 functions as a leak Ca²⁺ channel in astrocytes. To answer this question, we applied modified Tyrode’s solutions that contained different Ca²⁺ concentrations to astrocytes. Astrocytes were perfused with 0.2 mM Ca²⁺ solution for 2 minutes. Changing the extracellular Ca²⁺ concentration to 2 mM immediately elevated intracellular Ca²⁺ level. Subsequently increasing Ca²⁺ concentration to 10 mM resulted in further Ca²⁺ entry in astrocytes (Fig. 6A). 2-APB (30 μM) or YM-58483 (10 μM) reduced Ca²⁺ level under 0.2 mM Ca²⁺ condition, almost abolished 2 mM- or 10 mM-Ca²⁺-induced Ca²⁺ influx (Fig. 6A, B). We also observed that Ca²⁺ level under the 0.2 mM Ca²⁺ condition was significantly lower in Orai1 siRNA-transfected astrocytes compared to control siRNA-transfected astrocytes (Fig. 6C). Raising Ca²⁺ concentration-induced Ca²⁺ influx was almost eliminated in Orai1 siRNA-transfected astrocytes (Fig. 6C). This knockdown effect was further confirmed in Orai1 KO astrocytes (Fig. 6D). To determine whether STIM proteins are involved in Orai1-mediated Ca²⁺ entry, we knocked down STIM1 or STIM2 respectively. Elevating extracellular Ca²⁺ concentration-induced Ca²⁺ influx was intact in STIM1 or STIM2 knocked down astrocytes (Supplementary Figure 6A, B). These findings suggest that Orai1 acts as the leak Ca²⁺ channel and plays a crucial role in maintaining astrocytic Ca²⁺ homeostasis.

Figure 6. Orai1 functions as a leak Ca²⁺ channel.

(A) Effect of 2-APB on elevation of extracellular Ca²⁺-induced Ca²⁺ entry, n = 18. (B) Effect of YM-58483 on elevation of extracellular Ca²⁺-induced Ca²⁺ entry, n = 20. (C) Effect of Orai1 siRNAs on elevation of extracellular Ca²⁺-induced Ca²⁺ entry in astrocytes transfected with control siRNA or Orai1 siRNA, n = 13–25. (D) Effect of Orai1 deficiency on elevation of extracellular Ca²⁺-induced Ca²⁺ entry in astrocytes cultured from WT and Orai1 KO mice, n = 19–20. Values represent mean ± SEM, ***P < 0.001, compared with vehicle, control siRNA or WT by one-way ANOVA.

3. Discussion

TLR4 signaling plays a crucial role in neuroinflammation and has been linked to the pathogenesis of several CNS diseases. Activation of TLR4 leading to an increase in the production of inflammatory cytokines was well characterized, however, our understanding of the underlying mechanisms remains incomplete. It is known that cytosolic Ca²⁺ is essential for TLR4-mediated downstream events in macrophages and epithelial cells [32, 45–47]. Whether and how Orai1-mediated Ca²⁺ entry in astrocytic TLR4 signaling has not been documented. In the present study, we identified Orai1 as a Ca²⁺ leak channel regulating the basal Ca²⁺ level and a downstream effector of TLR4 signaling in astrocytes. Our data reveal a functional link between TLR4 and the CRAC channel Orai1. LPS stimulates TLR4 to induce orai1 upregulation, consequently, increases cytosolic Ca²⁺ level, enhancing Ca²⁺-mediated cellular processes.

It has been shown that LPS exposure induces Ca²⁺ release from ER stores and Ca²⁺ influx in endothelial cells and macrophages [53, 54]. We tested if LPS has a similar effect in astrocytes. Our data showed that acute treatment of LPS neither triggered Ca²⁺ release nor induced Ca²⁺ influx in astrocytes, while positive controls 2-Methylthioadenosine and ATP had effects in the same cells. These results suggest that acute treatment of LPS does not activate Ca²⁺ permeable channels and regulate basal Ca²⁺ homeostasis in spinal astrocytes. CRAC channels are primarily activated by the ER Ca²⁺ sensor STIM proteins when ER Ca²⁺ level is decreased by Gq protein-coupled receptor (GPCR) activation under physiological and pathological conditions. TLR4 does not belongs to the GPCR family. Therefore, acute LPS treatment did not trigger IP3-mediated Ca²⁺ release. The inconsistency may be attributed to the different cell types and the magnitude of stimulation used. We applied a relative low concentration of LPS. At this concentration, TLR4 antagonist completely block LPS-induced cytokine production. Our Ca²⁺ imaging recordings also revealed that acute LPS treatment had no effect on TG-induced Ca²⁺ release and Ca²⁺ influx. Clearly, acute treatment of LPS did not modulate SOCE.

It is well documented that LPS exposure induces robust MAPK activation in many cell types including astrocytes [55]. We also observed activation of ERK, p38 and JNK after LPS exposure. Our ELISA data showed that inhibition of MEK/ERK significantly attenuated LPS-induced cytokine production, which is consistent with previous studies [61–63]. We have shown that activation of Orai1 by TG increased ERK activation in dorsal horn neurons [64]. We expect that Orai1 is an upstream of ERK signaling. Surprisingly, inhibition or deficiency of Orai1 did not change ERK activation. Similarly, Inhibition of p38 or JNK drastically decreased cytokine production, while inhibition of Orai1 did not affect LPS-induced p38 and JNK activation. These results suggest that Orai1 is downstream of TLR4-MAPK.

Our results showed that Orai1 was upregulated in astrocytes after LPS exposure for a few hours, which is consistent with a previous study in mast cells [65]. The time course study revealed that Orai1 upregulation started at the 1-hour time point and lasted at least 6 hours. We tested if the upregulation of Orai1 enhances Orai1-mediated Ca²⁺ signal. The Ca²⁺ imaging data demonstrated that basal Ca²⁺ level and TG-induced SOCE were significantly increased after LPS 4-hour exposure. It is well known that NF-κB signaling is required for induction of a large number of inflammatory genes. Previous reports have shown that NF-κB activation occurred at 30 min after LPS exposure [66, 67], which is earlier than the time we observed for Orai1 upregulation. We thereby examined the effect of a specific IKK/NF-κB inhibitor (IKK-16) on LPS-induced Orai1 upregulation. Pretreated astrocytes with 10 μM IKK-16 for 30 min, completely inhibited NF-κB activation and Orai1 upregulation, suggesting that Orai1 is a downstream effector of TLR4-NF-κB signaling.

We and others have demonstrated that Orai1 is an important Ca²⁺ entry pathway in astrocytes [1, 44]. Indeed, Orai1 KO astrocytes showed a marked reduction (60%) of LPS-induced cytokine production. However, this cytokine secretion was not eliminated in Orai1 KO astrocytes. Definitely, other signaling pathways may also play roles in the cytokine production. It is known that astrocytes express numerous calcium-permeable ion channels and Na⁺/Ca²⁺ exchanger that mediated Ca²⁺ entry pathways [28, 68–73]. A previous report has demonstrated that TRPM2, a Ca²⁺-permeable nonselective cation channel, is involved in LPS-induced cytokine production [74]. A recent study also shows that Pannexin 1 (Panx 1), functions as a Ca²⁺-permeable channel[75], plays a role in LPS-induced cytokine production [76]. Some Ca²⁺-independent pathways may also contribute to LPS-induced cytokine production in astrocytes.

Previous reports have demonstrated that deletion of Orai2 increases Orai1-mediated SOCE in astrocytes and T cells [77–79]. However, our data from previous and present studies showed that knockdown of Orai2 did not significantly change SOCE or basal Ca²⁺ level in spinal cord neurons and astrocytes, which are consistent with a previous study that loss of Orai2 diminishes SOCE in neurons [80]. The possible reason for the different findings might be attributed to differences between the knockout and knockdown approaches. Knockdown of Orai2 protein may have been insufficient to determine the clear role of Orai2 in astrocytes. This has been observed in the recent study that knockdown of Orai2 does not increase SOCE while deletion of Orai2 increases SOCE [77].

Our Ca²⁺ imaging data showed that inhibition of Orai1 drastically decreased basal Ca²⁺ level. This effect was confirmed by Orai1 deficiency. Orai1 is generally believed to be activated in a store-dependent manner. Interestingly, Inhibition of STIM proteins by ML-9 had no effect on basal Ca²⁺ level. Consistently, knockdown of STIM1 or STIM2 did not change cytosolic Ca²⁺ concentration. These findings suggest that Orai1 is constituently activated and independent of STIM proteins under the resting state. We further refined our understanding of the functional role of Orai1 in astrocytic Ca²⁺ signal by determining whether Orai1 is a leak channel in astrocytes. Increasing extracellular Ca²⁺ concentrations from 0.2 mM to 2 mM then 10 mM, we observed a large Ca²⁺ influx, which was almost eliminated by inhibition of Orai1 with 2-APB or YM-58483. The similar results were seen in Orai1 knockdown or Orai1 deficient astrocytes, but not in STIM1 siRNA- or STIM2 siRNA-transfected astrocytes. These findings provide strong evidence that Orai1 functions as the leak Ca²⁺ channel contributing to basal Ca²⁺ homeostasis.

Taken together, we demonstrated that Orai1 plays a key role in TLR4-mediated cytokine production and is a downstream effector of TLR4-NF-κB signaling. We also identified Orai1 as a leak Ca²⁺ channel maintaining astrocytic Ca²⁺ homeostasis. Our findings revealed a novel link between TLR4 and Orai1. Collectively, the present data indicated that Orai1 is a crucial Ca²⁺ signal in spinal astrocytes, and may be implicated in Ca²⁺-dependent pathological events including neuroinflammation.

4. Methods

4.1. Ethics

Pregnant wild type CD-1 or Orai1 mutant mice were produced in our animal facility and housed in standard cages in 12-h light/dark cycle. Both male and female neonatal (post neonatal 2–4 days) mice were used. All experiments were conducted in accordance with the guidelines of the National Institutes of Health and the Committee for Research and Ethical Issues of IASP and were approved by the Rutgers New Jersey Medical School Animal Care and Use Committee.

4.2. Spinal astrocyte culture

Primary cultures of spinal cord astrocytes were prepared from CD-1 and Orai1 mutant (CD-1 background) neonatal mice as we described previously [44]. Briefly, following the induction of hypothermia on ice, neonatal mice were decapitated. The spinal cord was carefully removed after a laminectomy. After removing meninges, the spinal cord strips were incubated for 30 min at 37 °C in Hank’s balanced salt solution (HBSS, Invitrogen, Carlsbad, CA) (in mM: 137 NaCl, 5.4 KCl, 0.4 KH₂PO₄, 1 CaCl₂, 0.5 MgCl₂, 0.4 MgSO₄, 4.2 NaHCO₃, 0.3 Na₂HPO₄, and 5.6 glucose) containing papain (15 U/ml; Worthington Biochemical, Lakewood, NJ). The strips were mechanically dissociated by gently triturating using a pipette. The cells were grown in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen) in a 75 cm² flask containing 10% fetal bovine serum with 5% CO₂ at 37°C. When the cells reached about 70–80% confluence (usually 7–8 days), the flask was placed on an orbital shaker at 320 rpm for 2–3 hours to detach cells mainly microglia and precursor cells sitting on top of the astrocyte monolayer. The attached cells were trypsinized and re-plated to six-well plates for Western blotting analysis, 96-well plates for ELISA, and 12-mm coverslips for Ca²⁺ imaging in MEM culture media. The purity of astrocytes was determined by immunostaining GFAP (the percentage of positive cells was 92%).

4.3. Transfection

Astrocytes were collected from T75 flasks after 7–8 days in culture. Transfection was performed as described in our previous study [44]. Briefly, astrocytes were electroporated using a mouse Nucleofector kit according to the manufacturer’s instructions (Lonza Group Ltd, Basel, Switzerland). For siRNA transfection, astrocytes were transfected with 10 μg of TLR4 siRNA, STIM1 siRNA, STIM2 siRNA, Orai1 siRNA (Life Technologies, Grand Island, NY), Orai2 siRNA (Dharmacon), Orai3 siRNA (Dharmacon), or Scramble siRNA (Life Technologies for STIM1, STIM2 and Orai1; Dharmacon for Orai2 and Orai3) per 1 × 10⁶ cells. For transfection of Orai1-CFP (inserted in the pIRESneo plasmid, a generous gift from Dr. Gill, Penn State College of Medicine, PA), astrocytes were transfected with Orai1-CFP (2 μg) per 1 × 10⁶ cells. Transfected cells were seeded on 12-mm glass coverslips or 6-well plates. After 16 h, transfection medium was removed and astrocytes were fed with fresh culture medium. Calcium imaging and Western blot analysis were performed 48–72 h post transfection.

4.4. Western blot analysis

Astrocytes in six-well plates were quickly rinsed with phosphate buffer saline (PBS) before being lysed in an ice-cold radio-immunoprecipitation assay (RIPA) buffer containing 150 mM NaCl, 50 mM Tris-HCl, 0.2 mM ethylenediaminetetraacetic acid (EDTA), 2% sodium dodecyl sulphate (SDS), 1% Triton X-100, 1 % deoxycholate, 0.1 mM phenylmethanesulfonyl fluoride (PMSF), and protease cocktails inhibitor (Thermo Fisher Scientific). The supernatants were collected and utilized to determine total protein concentration. Proteins were resolved using SDS-PAGE, and then transferred to polyvinylidene difluoride membrane (PVDF; GE Healthcare, Piscataway, NY). The membrane was probed with primary antibodies, including rabbit anti-Orai1 (1:500, ProSci), rabbit anti-STIM1 (1:8000, Cell Signaling Technology), rabbit anti-STIM2 (1:8000, ProSci), anti-ERK (1: 10,000; Cell Signaling Technology), anti-p-ERK (1:1000; Cell Signaling Technology), anti-p-p38 (1:1000, Cell Signal), anti-p38 (1:5000, Cell Signal), anti-JNK (1:5000, Cell Signal) anti-p-JNK (1:1000, Cell Signal), or mouse anti-Actin (1:20,000; Sigma-Aldrich), for overnight at 4°C. The blots were rinsed and incubated with either HRP-conjugated secondary antibody (1:10 000, Cell Signal) followed by reaction with enhanced chemiluminescence (ECL) reagents, or IR Dye goat anti-rabbit/mouse secondary antibodies (1:10,000; LI-COR). Image J (NIH) or Odyssey Image Studio Software (LI-COR) were used to quantify the bands.

4.5. ELISA

Astrocytes were plated in 96-well plates (2×10⁴ cells per well) overnight. For testing the effect of drugs on LPS-induced cytokine production, astrocytes were pretreated with different concentrations of drugs for 30 min before application of LPS. After 6 or 24 hours of LPS treatments, astrocytes were centrifuged at 1000 rpm for 5 min, and IL-6 and TNF-α in the supernatants were measured by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s instructions (R&D Systems, Minneapolis, MN, USA).

4.6. MTT cell viability assay

The cell viability was determined by an assay measuring dehydrogenase activity (Trevigen, Gaithersburg, MD) according to the manufacturer’s instructions as we described previously [44]. After collecting supernatants for ELISA, 100 μL of MEM was immediately added to each well that contained astrocytes. Ten microliters of MTT reagent were added to each well and incubated for 2 h at 37 °C. Once the purple precipitate was formed and clearly visible intracellularly under the microscope, MTT detergent was applied to dissolve the formazan crystals. The absorbance was measured directly from 96-well plates at 570 nm using a microplate reader (Spectramax Plus, Molecular Devices, CA).

4.7. Ca²⁺ imaging

The intracellular Ca²⁺ level was measured using fura-2-based microfluorimetry and imaging analysis (Nicolai et al. 2010). Astrocytes were loaded with 4 μM fura-2AM (Life Technologies) in HBSS for 30 min at RT, and then washed for another 20 min in normal bath solution (containing 5.6 mM glucose, 5 mM KCl, 140 mM NaCl, 1 mM MgCl₂, 2 mM CaCl₂, and 10 mM Hepes). Then cells on the coverslip were mounted in a small laminar-flow perfusion chamber (Model RC-25, Warner Instruments, Hamden, CT, USA) and continuously perfused with Tyrode’s solution (5–7 ml/min). Images were acquired using an Olympus inverted microscope equipped with a CCD camera (Hamamatsu ORCA-03G, Tokyo, Japan) at 3 s intervals (20–22 °C). The fluorescence ratio was calculated using the excitation and emission intensities at 340 and 380 nm with background subtraction. Data were analyzed using the software MetaFluor 7.7.9 (Molecular Devices, Sunnyvale, CA, USA). Free intracellular Ca²⁺ concentration was calculated by the formula [Ca²⁺]_I = Kd*β*(R–R_min)/ (R_max–R), where β=(I_{380 max})/(I_{380 min}). R_min, R_max, and β were calculated using in situ calibration [81]. From each coverslip, only one recording was made.

4.8. Real-Time (RT) PCR

Real-Time PCR was performed as described in our previous study [44]. Total RNA was extracted from cultured astrocytes using TRIzol Reagent (Ambion life technologies). The RNA content was analyzed by nan drop UV-Vis Spectrophotometer at 260 nm (Thermo Scientific) for cDNA preparation. Total RNA was reverse transcribed into cDNA for each sample using the superscript first strand cDNA synthesis kit (Invitrogen) according to the manufacturer’s protocol. Specific primers for mouse TLR4 (Forward:5′-CACTGTTCTTCTCCTGCCTGAC-3′, Reverse:5′TGGTTGAAGA AGGAATGTCATC-3’) and Tublin (Forwad: 5’GTGCATCTCCATCCATTGTTG3’, Reverse: 5’GTGGGTTCCAGGTCTACGAA3’) were purchased from IDT (North Carolina, USA). The CFX Touch real time PCR System (Bio-Red) was used to amplify genes with the following amplification conditions: 5 min of initial denaturation at 95 °C, then 40 cycles of 95 °C for 30 s, 57 °C for 30 s, and 72 °C for 1.5 min. The relative gene expression was determined by the 2(−Delta Delta C_T, 2^−ΔΔCT) method. The expression of TLR4 was normalized with Tublin (housekeeping gene).

4.9. Drugs

PD98059 (2-(2-Amino-3-methoxyphenyl)-4H-1-benzopyran-4-one), YM98483 (N-[4-[3,5-Bis (trifluoromethyl)-1H-pyrazol-1-yl]phenyl]-4-methyl-1,2,3-thiadiazole-5-carboxamide), SB203580 (4-[5-(4- Fluorophenyl)-2-[4-(methylsulfonyl)phenyl]-1H-imidazol-4-yl]pyridine), ML-9 (1-(5-Chloronaphthalene- 1-sulfonyl)-1H-hexahydro-1,4-diazepine), SP600125 (Anthra[1-9-cd]pyrazol-6(2H)-one), IKK 16 (N-(4-Pyrrolidin-1-yl-piperidin-1-yl)-[4-(4-benzo[b]thiophen-2-yl-pyrimidin-2-ylamino)phenyl]carboxamide dihydrochloride) and TAK-242 ((R)-Ethyl 6-(N-(2-chloro-4-fluorophenyl)sulfamoyl)cyclohex-1- enecarboxylate) were purchased from Tocris (Minneapolis, MN). Thapsigargin ((3S,3aR,4S,6S,6AR, 7S,8S,9bS)-6-(Acetyloxy)-2,3,3a,4,5,6,6a,7,8,9b-decahydro-3,3a-dihydroxy-3,6,9-trimethyl-8-[[(2Z)-2-methyl-1-oxo-2-butenyl]oxy]-2-oxo-4-(1-oxobutoxy)azuleno[4,5-b]furan-7-yloctanoate), lipopolysaccharide (LPS), lipopolysaccharide (LPS) and 2-APB (2-Aminoethoxydiphenyl borane) were purchased from Sigma (St Louis, MO, USA). They were dissolved in Milli-Q water or dimethyl sulfoxide (DMSO) as stock solutions and further diluted to working concentrations.

4.10. Data analysis

Data are expressed as original traces or as mean ± SEM. All statistical analyses were carried out using the statistical software Origin 8.1. Treatment effects were statistically assessed using one-way ANOVA followed by the Tukey’s multiple comparison test. When the comparisons were limited to two means, paired or two-sample Student’s t-tests were performed. Error probabilities of P < 0.05 were considered statistical significant.

Abbreviations

TLR4

Toll-like receptor

SOCs

store-operated calcium channels

STIM1

stromal interaction molecule 1

STIM2

stromal interaction molecule 2

ER

endoplasmic reticulum

TG

thapsigargin

LPS

lipopolysaccharide

ER

endoplasmic reticulum

FCS

fetal calf serum

GFAP

glial fibrillary acidic protein

NF-κb

Nuclear factor kappa B

T NFα

Tumor necrosis factor alpha

IL6

Interleukin-6

DMEM

Dulbecco’s modified Eagle medium

CNS

central nervous system

PBS

Phosphate buffered saline

ELISA

enzymelinked immunosorbent assay

YM

YM-58483

PD

PD98059

SB

SB203580

SP

SP600125

HBSS

Hanks’ balanced salt solution

FBS

Fetal bovine serum

ANOVA

Analysis of variance

Availability of data and materials

All material and datasets used in this manuscript will be made available to researchers on request.