Mitochondrial Uncouplers Confer Protection by Activating AMP-Activated Protein Kinase to Inhibit Neuroinflammation Following Intracerebral Hemorrhage

Abstract

Intracerebral hemorrhage (ICH) is a disease with high disability and mortality rates. Currently, the efficacy of therapies available for ICH is limited. Microglia-mediated neuroinflammation substantially exacerbates brain damage following ICH. Here, we investigated whether mitochondrial uncouplers conferred protection by suppressing neuroinflammation following ICH. To mimic ICH-induced neuroinflammation in vitro, we treated microglia with red blood cell (RBC) lysate. RBC lysate enhanced the expression of pro-inflammatory cytokines in microglia. A clinically used uncoupler, niclosamide (Nic), reduced the RBC lysate-induced expression of pro-inflammatory cytokines in microglia. Moreover, Nic ameliorated brain edema, decreased neuroinflammation, and improved neurological deficits in a well-established mouse model of ICH. Like niclosamide, the structurally unrelated uncoupler carbonyl cyanide p-triflouromethoxyphenylhydrazone (FCCP) reduced brain edema, decreased neuroinflammation, and improved neurological deficits following ICH. It has been reported that mitochondrial uncouplers activate AMP-activated protein kinase (AMPK). Mechanistically, Nic enhanced AMPK activation following ICH, and AMPK knockdown abolished the beneficial effects of Nic following ICH. In conclusion, mitochondrial uncouplers conferred protection by activating AMPK to inhibit microglial neuroinflammation following ICH.

INTRODUCTION

Intracerebral hemorrhage (ICH) is a disease associated with high morbidity and mortality rates and accounts for approximately 20% of all strokes.¹⁾ Currently, the efficacy of therapies available for ICH is limited. Following ICH, primary damage is caused by the accumulation of blood in the hemorrhagic lesions, which occurs within minutes to hours ICH onset. Immunocytes, including brain-resident microglia and peripherally derived macrophages, are vital for hematoma clearance.²⁾ However, immunocytes also augment secondary damage by releasing immune-active molecules such as inflammatory cytokines. Among these cells, microglia are some of the first immunocytes to respond to ICH. Following ICH, blood components such as red cells and heme activate microglia.^3,4) In turn, activated microglia produce and secrete various deleterious molecules such as inflammatory cytokines, chemokines, reactive superoxide, and prostaglandins, leading to inflammatory damage.^5–9) Accordingly, inhibition of microglial activation has been reported to reduce brain injury and edema following ICH. For instance, minocycline and the microglia/macrophage inhibitory factor tuftsin fragment 1–3 inhibit microglial activation and attenuate brain injury following ICH.^10,11) Moreover, mesenchymal stem cells and regulatory T cells also ameliorate ICH-induced brain injury by inhibiting microglial activation.^12,13) Thus, uncovering new agents that inhibit microglia-mediated neuroinflammation is important for the development of new treatments for ICH.

Mitochondria play a central role in metabolism. Accumulating evidence suggests that changes in mitochondrial activity substantially impact microglial function.¹⁴⁾ Mitochondrial uncouplers promote proton leak across the mitochondrial inner membrane, dissipating the mitochondrial membrane potential that drives mitochondrial ATP synthesis. Thus, mitochondrial uncouplers lead to a futile cycle of substrate oxidation which does not generate ATP, causing a decrease in intracellular ATP levels which promotes AMP-activated protein kinase (AMPK) activation.¹⁵⁾ Mitochondrial uncoupling is therefore considered a promising strategy for treating diseases such as obesity, diabetes, and neuroinflammatory diseases like optic neuritis.^15,16) However, the effects of mitochondrial uncouplers on the outcomes of ICH have not been investigated. Notably, we recently showed that AMPK activation was the major mechanism underlying the inhibition of microglia-mediated neuroinflammation.^17,18) Since microglia-mediated neuroinflammation essentially contributes to brain damage following ICH, here we investigated whether mitochondrial uncouplers confer protection by activating AMPK to inhibit neuroinflammation following ICH.

MATERIALS AND METHODS

Cell Cultures and in Vitro Model of ICH-Induced Neuroinflammation in Microglia

Murine BV2 microglia were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal bovine serum and 1% penicillin and streptomycin at 37°C and 5% CO₂. BV2 microglia were used 2–3 passages after thawing. To mimic ICH-induced neuroinflammation in cultured microglia, we treated BV2 microglia with red blood cell (RBC) lysate as previously reported.^19,20) In brief, 1 mL of blood drawn from the mouse heart was centrifuged at 300 × g for 5 min, and the plasma and buffy coat were discarded. Packed RBCs were then frozen in liquid nitrogen for 5 min and thawed at 37°C for 10 min. This freeze-thaw cycle was repeated three times.¹⁹⁾ Following this, 10 µL of RBC lysate was used to treat microglia cultured on 24-well plates. An equal volume of saline (10 µL) was added to control cells.

Measurement of ATP Levels and ADP/ATP Ratio

An ADP/ATP ratio assay kit (Sigma-Aldrich, St. Louis, MO, U.S.A.) was used to measure cellular ATP and ADP levels. The cells were seeded on a 96-well plate at a density of 1 × 10⁴ cells/well, and were then treated with 25 µM niclosamide (Nic) or vehicle for 30 min. Then, ATP levels and the ADP/ATP ratio were measured and calculated according to the manufacturer’s manual. Results of ATP were expressed as the percentages of ATP levels to that of control cells treated with vehicle.

Measurement of the Mitochondrial Membrane Potential

The fluorescent probe tetramethylrhodamine ethyl ester (TMRM) (Sigma-Aldrich) was used to assess the mitochondrial membrane potential in BV2 microglia. The cells were seeded on 24-well or 96-well plates and cultured overnight. The next day, the cells were treated with 25 µM Nic or vehicle for 30 min. Following this, they were washed with phosphate buffered saline (PBS) and stained with 100 nM TMRM for 10 min. For cells cultured in 24-well plates, fluorescent images were captured using a confocal microscope (Zeiss LSM700, Germany) and constant parameters were used to acquire all of the images. For cells cultured in 96-well plates, fluorescence intensity was assessed using a microplate reader (TECAN, Männedorf, Switzerland) according to the manufacturer’s instructions.

RNA Interference

Small interfering RNA (siRNA) targeting murine AMPK α1/α2 (5′-GAG AAG CAG AAG CAC GAC GTT-3′), which has been used in our previous publication,¹⁷⁾ was synthesized by Genepharma (Shanghai, China) and used in the study. Nonsense siRNA (Ctrl-siRNA) which does not target any mammalian DNA sequence was used as the negative control. AMPK siRNA (0.5 µg/mL) or Ctrl-siRNA was transfected into BV2 microglia using Entranster™-R (Engreen Biosystem Co., Ltd., Beijing, China) according to the manufacturer’s instructions. At 48 h after transfection, cells were either harvested to assess the knockdown efficiency via Western blot or used for further experiments.

Western Blot

Tissue and cell samples were harvested and lysed in RIPA lysis buffer (Beyotime, Shanghai, China) supplemented with a protease inhibitor cocktail and phosphatase inhibitor cocktail on ice for 30 min. The cell lysates were then centrifuged and the supernatants were collected. Protein concentration was measured using a BCA kit (Thermo Scientific, Waltham, MA, U.S.A.). Following this, 40 µg proteins per sample were mixed with 5 × loading buffer and boiled for 5 min at 100°C. Then, the proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes. The PVDF membranes were blocked in PBS containing 5% non-fat milk and 0.1% (v/v) Tween-20 for 2 h. The membranes were then incubated with primary antibodies (anti-AMPK, anti-p-AMPK, anti-S6, and anti-p-S6) followed by appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies. Protein bands were visualized with enhanced chemiluminescent Western blotting reagents (SuperSignal West Pico, Pierce, Rockford, IL, U.S.A.) and images were captured using the Chemiluminescence Imaging System with Image Lab software (Bio-Rad, CA, U.S.A.). The optical density of the protein bands was semi-quantified by ImageJ software (NIH), and the results were expressed as the ratio of the target protein expression to β-actin or total protein expression.

Enzyme-Linked Immunosorbent Assay (ELISA)

Interleukin (IL)-1β, IL-6, and tumor necrosis factor-alpha (TNF-α) concentrations in the culture medium and mouse striatal tissue samples were measured using commercial ELISA kits (Invitrogen, San Diego, CA, U.S.A.). BV2 microglia were treated as indicated. The culture medium was collected for ELISA, and the cells were harvested to measure protein concentration with a BCA kit. Brain tissues were harvested from the ipsilateral and contralateral striata at 3 d post-ICH. Protein concentration was determined using a BCA kit. IL-1β, IL-6, and TNF-α protein levels were measured by ELISA according to the manufacturer’s instructions. Optical density was measured at 450 nm using an Infinite M1000 Pro Reader (TECAN). The concentrations of IL-1β, IL-6, and TNF-α were calculated from standard curves and normalized to total protein concentration.

Animal Research

All animal procedures were approved by the Animal Care and Use Committee of Soochow University. The collagenase-induced ICH model was adapted to mice as previously described with minor modifications.²¹⁾ In brief, the mice were placed onto a stereotaxic frame (Model 500, Kopf Instruments, Tujunga, CA, U.S.A.). A 26s-gauge needle was stereotaxically inserted into the left striatum (coordinates: 0.5 mm anterior to the bregma, 2 mm lateral to the midline, and 3.5 mm below the skull) through a cranial burr hole. Collagenase VII-S (0.03 U in 1 µL saline; Sigma-Aldrich) was infused at a rate of 0.1 µL/minute using a microinfusion pump (Harvard Apparatus Inc., South Natick, MA, U.S.A.). For the sham surgery, 1 µL of saline was infused into the striatum. After injection, the needle was kept in place for an additional 10 min to prevent leakage of the collagenase solution. After the needle was removed, the skin incision was closed with a suture. Mice were randomly allocated into groups and received a daily intraperitoneal injection of vehicle (10% dimethyl sulfoxide (DMSO) in corn oil) or Nic (25 mg/kg/d) for 3 d, the first of which was administered at 3 h post-ICH induction. For other experiments, mice were randomized to receive an intrastriatal injection of vehicle (DMSO, 2 µL) or carbonyl cyanide p-triflouromethoxyphenylhydrazone (FCCP, 50 mM, 2 µL) at 3 h post-ICH induction. No animal died before assessment and no animal was excluded from the analysis.

Intrastriatal Microinjection of Lentiviral Vectors

Lentiviruses were injected into the left striatum as previously reported.²²⁾ In brief, mice were placed onto a stereotaxic frame. Lentiviruses expressing non-targeted control short hairpin RNA (Ctrl-shRNA) or AMPK-shRNA (GeneChem, Shanghai, China) were injected into the left striatum at two separate locations at a rate of 0.5 µL/minute with a 30-gauge needle (1.5 µL of 1 × 10⁸ TU/mL lentivirus per injection). Two injection sites were used—one 0.5 mm anterior to the bregma, 2.0 mm lateral to the midline, and 3.5 mm below the skull, and the other 1 mm anterior to the bregma, 1.5 mm lateral to the midline, and 3.2 mm below the skull. At 2 weeks after injection, the mice were either harvested to assess the in vivo knockdown efficiency in the striatum or subjected to ICH induction and further experiments.

Assessment of Brain Edema Following ICH

At 3 d post-ICH induction, the brains of the mice were harvested and the left and right striata were dissected and collected. The wet weight of the striatal tissues was measured, and they were then stored in an oven at 95°C for 24 h. Following this, the dry weight of the tissues was measured. Brain water content was calculated to indicate edema according to the following formula: ((wet weight-dry weight)/wet weight) × 100%.^23,24)

Measurement of Neurological Deficits Following ICH

Neurological deficits were blindly assessed using a composite neurological score test and the vibrissae-elicited forelimb placement test.^25,26) For neurological score assessment, the mice were subjected to gait, body symmetry, climbing, circling behavior, front limb symmetry, compulsory circling, and whisker response tests. Each test was graded 0–4, and a higher point indicated more severe neurological deficit. For assessment of the gait, mice were given points based on their performance on an open bench top: 0 was given to the mice that displayed normal gait; 1 point was given to the mice that displayed stiff or inflexible gait; 2 points were given to the mice that limped; 3 points were given to the mice that trembled, drifted or fell; 4 points were given to the mice that did not walk at all. For assessment of body symmetry, mice were given points also based on their performance on an open bench top: 0 was given to the mice that displayed normal body symmetry; 1 point was given to the mice that displayed slight asymmetry; 2 points were given to the mice that displayed moderate asymmetry; 3 points were given to the mice that displayed prominent asymmetry; 4 points were given to the mice that displayed extreme asymmetry. For assessment of climbing performance, mice were given points based on their climbing performance on a 45° angle slope: 0 was given to the mice that climbed normally; 1 point was given to the mice that climbed with limb weakness; 2 points were given to the mice that held on the slope and did not slip or climb; 3 points were given to the mice that slid down the slope and displayed unsuccessful effort to prevent fall; 4 points were given to the mice that slid immediately and did not display any effort to prevent fall. For assessment of circling behavior, mice were given points based on their circling behavior on an open bench top: 0 was given to the mice that did not circle; 1 point was given to the mice that made one-sided turns predominantly; 2 points were given to the mice that circled to one side occasionally; 3 points were given to the mice that circled to one side constantly; 4 points were given to the mice that pivoted, swayed or did not move at all. For assessment of front limb symmetry, mice were given points based on their performance when they were suspended by their tails: 0 was given to the mice that displayed normal symmetry; 1 point was given to the mice that displayed slight asymmetry; 2 points were given to the mice that displayed marked asymmetry; 3 points were given to the mice that displayed prominent asymmetry; 4 points were given to the mice that displayed no body or limb movement. For assessment of the behavior of compulsory circling, mice were suspended by tails and their front limbs were placed on a bench: 0 was given to the mice that did not circle; 1 point was given to the mice that displayed tendency to turn one-side; 2 points were given to the mice that circled to one side; 3 points were given to the mice that pivoted to one side sluggishly; 4 points were given to the mice that did not advance. For assessment of the whisker response, mice were given points based on their responses when their vibrissae on each side were slightly touched from the rear: 0 was given to the mice that reacted normally to the stimulus on both sides; 1 point was given to the mice that displayed slight asymmetry response to the stimulus; 2 points were given to the mice that displayed prominent asymmetry response to the stimulus; 3 points were given to the mice whose ipsilateral responses were absent; 4 points were given to the mice whose bilateral proprioceptive responses were absent. Finally, these points were summed to produce a total neuroscore for each mouse.

For the forelimb placement test, the vibrissae were stimulated by gently brushing them on the edge of a countertop. Normal mice quickly placed the forelimb ipsilateral to the stimulated vibrissae onto the countertop. Following ICH, placement of the forelimb contralateral to the injured hemisphere in response to vibrissae stimulation was impaired. This impairment was dependent on the severity of the injury. Each mouse was tested 10 times. The number of trials in which a mouse appropriately placed its forelimb on the countertop was calculated as a percentage of the total number of trials.

Statistical Analysis

Statistical analysis was performed with SPSS Statistics 17.0. One-way ANOVA was used for multiple comparisons, and two-tailed Student’s t-tests were used for pairwise comparisons. Data obtained from the behavioral tests were analyzed by two-way ANOVA. p < 0.05 was considered statistically significant.

RESULTS

Nic is a clinically used anthelmintic drug approved by the U.S. Food and Drug Administration (FDA). This drug acts as a mitochondrial uncoupler, and displays therapeutic effects beyond its anthelmintic effects in a variety of disease models.^15,27) To mimic ICH-induced neuroinflammation in cultured microglia, we treated BV2 microglia with RBC lysate as previously reported.^19,20) We found that Nic dose-dependently reduced RBC lysate-induced neuroinflammation in microglia, as indicated by the fact that Nic reduced the RBC lysate-induced expression of the pro-inflammatory mediators IL-1β, TNF-α, and IL-6 (Figs. 1a–c). Mitochondrial uncouplers reduce intracellular ATP levels and mitochondrial membrane potential. As expected, Nic decreased intracellular ATP levels, increased the ADP/ATP ratio, and reduced mitochondrial membrane potential in BV2 microglia (Figs. 1d–g).

Fig. 1. The Mitochondrial Uncoupler Nic Inhibits Microglia-Mediated Neuroinflammation in an in Vitro Model of ICH

(a–c) Protein levels of IL-1β (a), IL-6 (b), and TNF-α (c) in the culture medium of BV2 microglia treated with RBC lysate (Lysate) plus vehicle or Nic for 24 h (n = 3). *: p < 0.05. Ctrl: control cells treated with vehicle. (d–g) Treatment with 25 µM Nic reduced intracellular ATP levels (d, n = 3), enhanced the ADP/ATP ratio (e, n = 3), and reduced mitochondrial membrane potential (f: representative images of TMRM staining from three independent replicates, g: quantification of TMRM red fluorescence at 575 nm) in BV2 microglia at 30 min after treatment. **: p < 0.01. IL, interleukin; TMRM, tetramethylrhodamine methyl ester; TNF-α, tumor necrosis factor-alpha. (Color figure can be accessed in the online version.)

To investigate whether Nic inhibited post-ICH neuroinflammation in vivo, we induced ICH in mice by injecting collagenase into the left striatum. ICH induced neuroinflammation in the ipsilateral striatum, as indicated by the enhanced expression of the pro-inflammatory mediators IL-1β, TNF-α, and IL-6 in the left striatum at 3 d post-ICH (Figs. 2a–c). Nic, injected via the intraperitoneal route, suppressed post-ICH neuroinflammation, as demonstrated by the fact that Nic reduced the expression of the pro-inflammatory mediators IL-1β, TNF-α, and IL-6 in the ipsilateral striatum at 3 d post-ICH (Figs. 2a–c). To further show that Nic conferred therapeutic effects in mice with experimentally induced ICH, we assessed ICH-induced brain edema by measuring the water content of the striatum. Consistent with our finding that Nic suppressed ICH-induced neuroinflammation, Nic significantly suppressed ICH-induced brain edema at 3 d post-ICH (Fig. 2d). Moreover, we assessed neurological deficits with the neuroscore and the vibrissae-elicited forelimb placement test. Before ICH induction, the mice did not exhibit any obvious neurological deficits. Moreover, the degree of neurological impairment did not differ among the groups immediately before the injection of Nic (3 h after ICH induction), suggesting that a similar degree of neurological damage was induced in all of the mice. Notably, Nic significantly improved neurological deficits at 3 d post-ICH (Figs. 2e, f), which correlated with our findings that Nic inhibited neuroinflammation and brain edema at 3 d post-ICH.

Fig. 2. The Mitochondrial Uncoupler Nic Inhibits Neuroinflammation and Displays Therapeutic Effects in a Mouse Model of ICH

(a–c) Protein levels of IL-1β (a, n = 3), IL-6 (b, n = 3), and TNF-α (c, n = 3) in the hemorrhagic, ipsilateral (Ipsil) and contralateral (Contra) striata at 3 d post-ICH. Nic or vehicle was injected intraperitoneally. **: p < 0.01. (d) Brain edema was assessed by measuring the water content in the ipsilateral (Ipsil) and contralateral (Contra) striata at 3 d post-ICH (n = 6). **: p < 0.01. (e, f) Neurological deficits of Nic-treated mice and vehicle-treated mice (Veh). Sham: Sham-operated mice. Neurological deficits were assessed with two behavioral tests before ICH induction (Baseline), immediately before Nic injection (3 h after ICH induction, 0 d), 1 d, and 3 d after ICH. The neuroscore test (e, n = 6) comprised a battery of motor tests, and a higher total score indicated more severe neurological deficit. For the forelimb placement test, the number of trials in which a mouse appropriately placed its forelimb on the countertop was calculated as a percentage of the total number of trials. *: p < 0.05.

FCCP is a classical mitochondrial uncoupler which has been widely used in research to investigate the effects of mitochondrial uncouplers. Importantly, FCCP is structurally unrelated to Nic. FCCP was injected into the ipsilateral striatum following ICH. Like Nic, FCCP ameliorated brain edema (Fig. 3a), improved neurological deficits (Figs. 3b, c), and inhibited the ICH-induced expression of the pro-inflammatory mediators IL-1β, TNF-α, and IL-6 in the ipsilateral striatum at 3 d post-ICH (Figs. 3d–f). Thus, by using two structurally unrelated mitochondrial uncouplers, we showed that mitochondrial uncouplers inhibited neuroinflammation and conferred protection against ICH.

Fig. 3. The Classical Mitochondrial Uncoupler FCCP Displays Therapeutic Effects and Inhibits Neuroinflammation in the Mouse ICH Model

(a) Brain edema was assessed by measuring the water content in the ipsilateral (Ipsil) and contralateral (Contra) striata at 3 d post-ICH (n = 6). *: p < 0.05; **: p < 0.01. FCCP or vehicle was injected into the ipsilateral striatum following ICH. (b, c) Neurological deficits of FCCP-treated mice and vehicle-treated mice (Veh). Neurological deficits assessed with the neuroscore (b, n = 6) and the forelimb placement test (c, n = 6) before ICH induction (Baseline), immediately before FCCP or vehicle injection (3 h after ICH induction, 0 d), 1 d, and 3 d after ICH. Sham: Sham-operated mice. *: p < 0.05; **: p < 0.01. (d–f) Protein levels of IL-1β (e, n = 3), IL-6 (f, n = 3), and TNF-α (g, n = 3) in the ipsilateral (Ipsil) and contralateral (Contra) striata at 3 d post-ICH. **: p < 0.01.

To investigate the mechanism by which mitochondrial uncouplers inhibit neuroinflammation following ICH, we focused on Nic since Nic is a drug used clinically Mitochondrial uncouplers decrease intracellular ATP levels, which activates AMPK.¹⁵⁾ Moreover, Nic has been shown to exert anti-diabetic effects in mice via AMPK activation.¹⁵⁾ Notably, we recently showed that AMPK activation is the major mechanism underlying the inhibition of microglia-mediated neuroinflammation.^17,18) Thus, we hypothesized that Nic would inhibit microglia-mediated post-ICH neuroinflammation by activating AMPK. In support of this hypothesis, Nic significantly enhanced AMPK activation (as indicated by enhanced phosphorylation at threonine (Thr)-172) in BV2 microglia treated with RBC lysate (Figs. 4a, b). More relevantly, siRNA-mediated AMPK knockdown in BV2 microglia (Figs. 4c, d) abolished the inhibitory effects of Nic on RBC lysate-induced expression of pro-inflammatory mediators (Figs. 4e–g). Collectively, the in vitro results suggested that Nic suppressed microglia-mediated neuroinflammation by activating AMPK.

Fig. 4. The Mitochondrial Uncoupler Nic Inhibits Microglia-Mediated Neuroinflammation by Activating AMPK in the in Vitro ICH Model

(a, b) Representative images and quantification (n = 3) of AMPK activation (p-AMPK: AMPK phosphorylation,) in BV2 microglia treated with RBC lysate and Nic (n = 3) for 2 h. **: p < 0.01. (c, d) Representative images and quantification (n = 3) of AMPK expression in BV2 microglia following treatment with AMPK-siRNA or Ctrl-siRNA for 2 d. NS: not significant; *: p < 0.05. (e–g) AMPK knockdown abolished the inhibitory effect of Nic on RBC lysate-induced neuroinflammation in BV2 microglia. BV2 microglia were transfected with Ctrl-siRNA or AMPK-siRNA. At 2 d after transfection, cells were treated with RBC lysate (Lysate) plus vehicle or Nic for 24 h. The protein levels of the pro-inflammatory mediators IL-1β (e, n = 3), IL-6 (f, n = 3), and TNF-α (g, n = 3) in the culture medium were measured by ELISA. *: p < 0.05. (Color figure can be accessed in the online version.)

We further investigated whether Nic suppressed post-ICH neuroinflammation by activating AMPK in vivo. Consistent with the in vitro results, Nic treatment significantly enhanced AMPK activation in the ipsilateral, hemorrhagic striatum following ICH (Figs. 5a, b). It is well-established that AMPK activation inhibits the activity of mammalian target of rapamycin (mTOR).²⁸⁾ Since recent studies showed that mTOR inhibition is an important mechanism by which macrophage-mediated inflammation is suppressed,²⁹⁾ we examined whether Nic inhibited mTOR activity following ICH. As expected, Nic reduced the phosphorylation of ribosomal protein S6 (Figs. 5c, d), the substrate of mTOR, suggesting that Nic inhibited mTOR activity in the ipsilateral striatum following ICH.²⁹⁾ These results further suggested that the inhibition of post-ICH neuroinflammation by Nic may be mediated via the activation of AMPK. To further show that AMPK activation was required for the inhibitory effect of Nic on ICH-induced neuroinflammation in vivo, we injected a lentivirus expressing AMPK-shRNA into the left striatum. Compared to a control lentivirus (Ctrl-shRNA), the lentivirus expressing AMPK shRNA (AMPK-shRNA) significantly decreased endogenous AMPK expression in the striatum ipsilateral to lentiviral injection at 14 d after injection (Figs. 5e, f). According to the results, the mice were then subjected to ICH surgery at 14 d after lentiviral injection. Consistent with the in vitro results, the inhibitory effect of Nic on the ICH-induced expression of pro-inflammatory cytokines was abolished by AMPK-shRNA (Figs. 5g–i).

Fig. 5. The Mitochondrial Uncoupler Nic Inhibits ICH-Induced Neuroinflammation via AMPK Activation in Vivo

(a, b) Representative images and quantification (n = 3) of Western blot analysis of AMPK activation (p-AMPK) in the striata of mice treated with niclosamide Nic for 3 d following ICH. *: p < 0.05. Ipsil: ipsilateral, hemorrhagic side; Contra: contralateral side. Nic or vehicle was injected intraperitoneally. (c, d) Representative images and quantification (n = 3) of ribosomal protein S6 phosphorylation in the mouse striatum at 3 d post-ICH. **: p < 0.01. (e, f) Representative images and quantification (n = 3) of AMPK expression in the striatum ipsilateral to injection (Ipsil) at 14 d following the injection of lentiviruses expressing AMPK-shRNA or nonsense shRNA (Ctrl-shRNA) into the left striatum. Contra: contralateral, non-injected side. Ipsil-C: the striatum ipsilateral to the control lentivirus injection; Ipsil-AMPK: the striatum ipsilateral to the AMPK-shRNA-expressing lentivirus injection. **: p < 0.01. (g–i) The inhibitory effect of Nic on ICH-induced neuroinflammation was abolished by AMPK-shRNA. Lentiviruses expressing Ctrl-shRNA or AMPK-shRNA were injected into the left striatum. At 2 weeks after injection, ICH was induced by injecting collagenase into the left striatum. Vehicle (Veh) or Nic was administered daily for 3 d, starting at 3 h post-ICH. The protein levels of IL-1β (g, n = 3), IL-6 (h, n = 3), and TNF-α (i, n = 3) in the ipsilateral (Ipsil) and contralateral (Contra) striata were assessed by ELISA at 3 d post-ICH. NS: not significant; *: p < 0.05; **: p < 0.01.

shRNA-mediated AMPK knockdown also abolished the therapeutic effects of Nic on brain edema (Fig. 6a). Moreover, we assessed neurological deficits with the neuroscore and the forelimb placement test. Before ICH induction, the mice did not exhibit any obvious neurological deficits, suggesting that lentiviral injection and AMPK knockdown did not affect the baseline behavioral functions of the mice. The degree of neurological impairment did not differ among the groups immediately before the injection of Nic (3 h after ICH induction, 0 d), suggesting that a similar degree of neurological damage was induced in these mice. Notably, Nic significantly improved neurological deficits in the mice injected with the lentivirus expressing Ctrl-shRNA, but not in the mice injected with the lentivirus expressing AMPK-shRNA at 3 d post-ICH (Figs. 6b, c). These results suggested that AMPK knockdown abolished the therapeutic effects of Nic on neurological deficits, which correlated with our findings that AMPK knockdown abolished the effects of Nic on neuroinflammation and brain edema at 3 d post-ICH. Taken together, we showed that the mitochondrial uncoupler Nic conferred protection following ICH by activating AMPK to inhibit neuroinflammation.

Fig. 6. AMPK Is Required for the Therapeutic Effects of Nic in a Mouse ICH Model

(a) Brain edema was assessed by measuring the water content in the hemorrhagic, ipsilateral (Ipsil) and contralateral (Contra) striata at 3 d post-ICH (n = 6). NS: not significant; **: p < 0.01. Nic or vehicle was injected intraperitoneally. (b, c) Neurological deficits assessed with the neuroscore (b, n = 6) and the forelimb placement test (c, n = 6) before ICH induction (Baseline), immediately before Nic injection (3 h after ICH induction), 1 d, and 3 d after ICH induction. *: p < 0.05; **: p < 0.01. Ctrl-shRNA: control lentivirus; AMPK-shRNA: lentivirus expressing AMPK-shRNA. Lentiviruses expressing Ctrl-shRNA or AMPK-shRNA were injected into the left striatum 14 d before ICH induction. Ipsil: ipsilateral; Contra: contralateral; Veh: vehicle. (Color figure can be accessed in the online version.)

DISCUSSION

In this study, using the classic mitochondrial uncoupler FCCP and the clinically used mitochondrial uncoupler Nic, we presented evidence that mitochondrial uncouplers conferred protection following ICH by activating AMPK to inhibit neuroinflammation.

Mitochondria play a central role in metabolism, and changes in mitochondrial activity substantially impact microglial function.¹⁴⁾ Mitochondrial uncouplers dissipate the proton gradient across the mitochondrial inner membrane which drives mitochondrial ATP synthesis. Therefore, mitochondrial uncouplers result in a futile cycle of substrate oxidation without generating ATP. Exogenous uncoupling agents have been shown to be protective against ischemic brain injuries. For instance, the uncoupling agent 2,4-dinitrophenol (DNP) reduces the infarct damage following cerebral ischemia via a mechanism involving a reduction in mitochondrial reactive oxygen species generation and calcium uptake.³⁰⁾ Moreover, deletion of endogenous uncoupling protein 2 (UCP2) in mice increases infarct volumes, suppresses the expression of antioxidant, cell-cycle, and DNA repair genes, and significantly enhances the expression of inflammatory cytokines after cerebral ischemia.³¹⁾ On the other hand, overexpression of UCP2 inhibits the production of inflammatory cytokines following cerebral ischemia.³²⁾ However, whether mitochondrial uncouplers protect against ICH-induced brain injury remains unclear.

Evidence shows that inhibiting microglial activation protects the brain following ICH and represents a promising therapeutic strategy for treating ICH. During ICH, blood rapidly accumulates in the brain parenchyma, causing primary damage. Secondary damage following ICH, for the most part, results from inflammation-related injuries.^2,33) Microglia are some of the first inflammatory cells to respond to ICH. They are activated by blood components following ICH,³⁴⁾ and inhibiting microglial activation decreases brain injury and edema following ICH.^20,33) Agents that have been shown to inhibit microglial activation following ICH include the antibiotic minocycline,¹¹⁾ the microglia/macrophage inhibitory factor tuftsin fragment 1–3,¹⁰⁾ curcumin,³⁵⁾ sinomenine,³⁶⁾ and the prostaglandin E2 receptor agonist misoprostol.³⁷⁾ These compounds attenuate brain injury and improve neurological function following ICH. Moreover, infusion of mesenchymal stem cells¹²⁾ and regulatory T cells¹³⁾ also reduces ICH-induced brain injury by inhibiting microglial activation. However, the effects of mitochondrial uncouplers on microglia-mediated neuroinflammation following ICH have not yet been investigated.

In the study, we examined whether exogenous mitochondrial uncouplers inhibited post-ICH neuroinflammation. The mitochondrial uncoupler Nic suppressed RBC lysate-induced inflammation in microglia in vitro and ICH-induced neuroinflammation in the mouse brain in vivo. Moreover, the structurally unrelated uncoupler FCCP, which is widely used in research, also inhibited neuroinflammation in mice with experimentally induced ICH. Consequently, both Nic and FCCP ameliorated brain edema and improved neurological deficits following ICH, suggesting that mitochondrial uncouplers exerted therapeutic effects following ICH. Nic is a clinically used mild uncoupler. Systematic administration of Nic displays therapeutic effects in disease models.¹⁵⁾ To increase the translatability of the study, we injected Nic intraperitoneally. Unlike niclosamide, FCCP is a potent mitochondrial uncoupler and displays high toxicity. To prevent intolerable toxicity that might be induced by systematic administration of FCCP, we injected FCCP directly into the striatum. It is worth noting that intraperitoneally injected Nic likely affected microglia as well as peripheral immune cells, such as macrophages that infiltrated into the brain following ICH, while intrastriatally injected FCCP mainly affected microglia. This may partly explain why the stronger uncoupler FCCP displayed the similar effect as that conferred by the mild uncoupler Nic following ICH. Moreover, the dose of FCCP was also different from that of Nic.

Although mitochondrial uncouplers could potentially be used to treat various diseases including stroke, their clinical application is limited by the toxicity of existing mitochondrial uncouplers.¹⁵⁾ For instance, DNP, the best known chemical mitochondrial uncoupler, was approved by the FDA to treat obesity, and numerous studies have shown that this drug has beneficial effects in various disease models. However, DNP causes hyperthermia and was withdrawn from the market due to its narrow therapeutic index.¹⁵⁾ DNP also displays a quite steep toxicity profile in small animals.³⁸⁾ Therefore, the development of safe mitochondrial uncouplers is necessary. Nic is a clinically approved anthelmintic drug used to treat tapeworm infections.¹⁵⁾ Nic acts by uncoupling the mitochondria of parasitic worms.^39,40) Nic displays an excellent safety profile, and long-term treatment with high doses of Nic salt does not have adverse effects in animals. In this study, we showed that Nic exerted therapeutic effects following ICH. FCCP also displayed therapeutic effects in this mouse model of ICH. Thus, our study indicates that safe mitochondrial uncouplers could potentially be used to treat hemorrhagic stroke. In the study, we examined the mechanism underlying the inhibition of post-ICH neuroinflammation by the mitochondrial uncoupler Nic. Our previous studies showed that AMPK activation was a major mechanism underlying the suppression of microglial activation.^17,18) Notably, AMPK activation is a downstream effect of the decrease in intracellular ATP levels induced by mitochondrial uncouplers.¹⁵⁾ As expected, we observed that Nic enhanced AMPK activation in microglia treated with RBC lysate. We observed a trend that ICH enhanced AMPK activation both in the ipsilateral and contralateral striatum following ICH (Fig. 5a), possibly because ICH exerted remote effects on AMPK activation in the contralateral side. Notably, intraperitoneally injected Nic further enhanced AMPK activation only in ipsilateral striatum. Thus, it was likely that only the ipsilateral striatum was permeable to Nic due to the breakdown of blood–brain barrier following ICH. It is well-established that AMPK activation inhibits the activity of mTOR.²⁸⁾ Since mTOR inhibition is an important mechanism by which macrophage-mediated inflammation is suppressed,²⁹⁾ we examined whether Nic induced mTOR inhibition following ICH. As expected, Nic treatment inhibited mTOR in the ipsilateral striatum, further suggesting that AMPK activation may be critical to the inhibition of post-ICH neuroinflammation by Nic. More relevantly, we showed that AMPK knockdown abolished the inhibition of neuroinflammation by Nic in RBS lysate-treated BV2 microglia and in the mouse brain following ICH. Consequently, AMPK knockdown abolished the therapeutic effects of Nic on brain edema and neurological deficits following ICH.

There are some limitations of our study. We provided evidence that mitochondrial uncouplers inhibited microglial neuroinflammation following ICH. However, whether mitochondrial uncouplers also act through astrocytes to inhibit neuroinflammation following ICH remains further investigation. For instance, the uncoupler Nic has been reported to inhibit nuclear factor-kappaB (NF-κB),⁴¹⁾ and inhibiting astrocytic NF-κB has been shown to be an important mechanism underlying the suppression of neuroinflammation following brain injury.⁴²⁾ Moreover, the uncoupler Nic is a potent inhibitor of signal transducer and activator of transcription (STAT)3.⁴¹⁾ Interestingly, activation of AMPK inhibits inflammation by suppressing of SATA3.⁴³⁾ Thus, it remains to be investigated whether the anti-inflammatory effects of niclosamide are mediated by the suppression of STAT3.

In conclusion, our results suggested that the exogenous mitochondrial uncouplers conferred protection against ICH by activating AMPK to inhibit neuroinflammation.

Acknowledgments

This work was supported by following Grants: National Natural Science Foundation of China (81971119, 81671310, 81571124), Priority Academic Program Development of the Jiangsu Higher Education Institutions, Suzhou Clinical Research Center of Neurological Disease (Szzx201503) and the Jiangsu key laboratory Grant (BM2013003).

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Qureshi AI, Mendelow AD, Hanley DF. Intracerebral haemorrhage. Lancet, 373, 1632–1644 (2009).
• 2) Donnan GA, Hankey GJ, Davis SM. Intracerebral haemorrhage: A need for more data and new research directions. Lancet Neurol., 9, 133–134 (2010).
• 3) Garton T, Keep RF, Wilkinson DA, Strahle JM, Hua Y, Garton HJ, Xi G. Intraventricular hemorrhage: the role of blood components in secondary injury and hydrocephalus. Transl. Stroke Res., 7, 447–451 (2016).
• 4) Xiong XY, Yang QW. Rethinking the roles of inflammation in the intracerebral hemorrhage. Transl. Stroke Res., 6, 339–341 (2015).
• 5) Lan X, Han X, Li Q, Yang QW, Wang J. Modulators of microglial activation and polarization after intracerebral haemorrhage. Nat. Rev. Neurol., 13, 420–433 (2017).
• 6) Mracsko E, Veltkamp R. Neuroinflammation after intracerebral hemorrhage. Front. Cell. Neurosci., 8, 388 (2014).
• 7) Wang J, Dore S. Inflammation after intracerebral hemorrhage. J. Cereb. Blood Flow Metab., 27, 894–908 (2007).
• 8) Wen YD, Zhang HL, Qin ZH. Inflammatory mechanism in ischemic neuronal injury. Neurosci. Bull., 22, 171–182 (2006).
• 9) Zhang Z, Zhang Z, Lu H, Yang Q, Wu H, Wang J. Microglial polarization and inflammatory mediators after intracerebral hemorrhage. Mol. Neurobiol., 54, 1874–1886 (2017).
• 10) Wang J, Rogove AD, Tsirka AE, Tsirka SE. Protective role of tuftsin fragment 1–3 in an animal model of intracerebral hemorrhage. Ann. Neurol., 54, 655–664 (2003).
• 11) Zhao F, Hua Y, He Y, Keep RF, Xi G. Minocycline-induced attenuation of iron overload and brain injury after experimental intracerebral hemorrhage. Stroke, 42, 3587–3593 (2011).
• 12) Bao XJ, Liu FY, Lu S, Han Q, Feng M, Wei JJ, Li GL, Zhao RC, Wang RZ. Transplantation of flk-1+ human bone marrow-derived mesenchymal stem cells promotes behavioral recovery and anti-inflammatory and angiogenesis effects in an intracerebral hemorrhage rat model. Int. J. Mol. Med., 31, 1087–1096 (2013).
• 13) Yang Z, Yu A, Liu Y, Shen H, Lin C, Lin L, Wang S, Yuan B. Regulatory t cells inhibit microglia activation and protect against inflammatory injury in intracerebral hemorrhage. Int. Immunopharmacol., 22, 522–525 (2014).
• 14) De Simone R, Ajmone-Cat MA, Pandolfi M, Bernardo A, De Nuccio C, Minghetti L, Visentin S. The mitochondrial uncoupling protein-2 is a master regulator of both m1 and m2 microglial responses. J. Neurochem., 135, 147–156 (2015).
• 15) Tao H, Zhang Y, Zeng X, Shulman GI, Jin S. Niclosamide ethanolamine-induced mild mitochondrial uncoupling improves diabetic symptoms in mice. Nat. Med., 20, 1263–1269 (2014).
• 16) Khan RS, Dine K, Geisler JG, Shindler KS. Mitochondrial uncoupler prodrug of 2,4-dinitrophenol, mp 201°C, prevents neuronal damage and preserves vision in experimental optic neuritis. Oxid. Med. Cell. Longev., 2017, 7180632 (2017).
• 17) Zhou X, Cao Y, Ao G, Hu L, Liu H, Wu J, Wang X, Jin M, Zheng S, Zhen X, Alkayed NJ, Jia J, Cheng J. Camkkbeta-dependent activation of amp-activated protein kinase is critical to suppressive effects of hydrogen sulfide on neuroinflammation. Antioxid. Redox Signal., 21, 1741–1758 (2014).
• 18) Zhang M, Wu X, Xu Y, He M, Yang J, Li J, Li Y, Ao G, Cheng J, Jia J. The cystathionine beta-synthase/hydrogen sulfide pathway contributes to microglia-mediated neuroinflammation following cerebral ischemia. Brain Behav. Immun., 66, 332–346 (2017).
• 19) Bhasin RR, Xi G, Hua Y, Keep RF, Hoff JT. Experimental intracerebral hemorrhage: Effect of lysed erythrocytes on brain edema and blood-brain barrier permeability. Acta Neurochir. Suppl., 81, 249–251 (2002).
• 20) Wang YC, Zhou Y, Fang H, Lin S, Wang PF, Xiong RP, Chen J, Xiong XY, Lv FL, Liang QL, Yang QW. Toll-like receptor 2/4 heterodimer mediates inflammatory injury in intracerebral hemorrhage. Ann. Neurol., 75, 876–889 (2014).
• 21) Wu H, Wu T, Xu X, Wang J, Wang J. Iron toxicity in mice with collagenase-induced intracerebral hemorrhage. J. Cereb. Blood Flow Metab., 31, 1243–1250 (2011).
• 22) Cheng J, Uchida M, Zhang W, Grafe MR, Herson PS, Hurn PD. Role of salt-induced kinase 1 in androgen neuroprotection against cerebral ischemia. J. Cereb. Blood Flow Metab., 31, 339–350 (2011).
• 23) Hu HM, Li B, Wang XD, Guo YS, Hui H, Zhang HP, Wang B, Huang DG, Hao DJ. Fluoxetine is neuroprotective in early brain injury via its anti-inflammatory and anti-apoptotic effects in a rat experimental subarachnoid hemorrhage model. Neurosci. Bull., 34, 951–962 (2018).
• 24) Liu H, Wang Y, Xiao Y, Hua Z, Cheng J, Jia J. Hydrogen sulfide attenuates tissue plasminogen activator-induced cerebral hemorrhage following experimental stroke. Transl. Stroke Res., 7, 209–219 (2016).
• 25) Liu RY, Wang JJ, Qiu X, Wu JM. Acute hyperglycemia together with hematoma of high-glucose blood exacerbates neurological injury in a rat model of intracerebral hemorrhage. Neurosci. Bull., 30, 90–98 (2014).
• 26) Ni W, Mao S, Xi G, Keep RF, Hua Y. Role of erythrocyte cd47 in intracerebral hematoma clearance. Stroke, 47, 505–511 (2016).
• 27) Chen W, Mook RA Jr, Premont RT, Wang J. Niclosamide: Beyond an antihelminthic drug. Cell. Signal., 41, 89–96 (2018).
• 28) Laplante M, Sabatini DM. Mtor signaling in growth control and disease. Cell, 149, 274–293 (2012).
• 29) Ip WKE, Hoshi N, Shouval DS, Snapper S, Medzhitov R. Anti-inflammatory effect of il-10 mediated by metabolic reprogramming of macrophages. Science, 356, 513–519 (2017).
• 30) Korde AS, Pettigrew LC, Craddock SD, Maragos WF. The mitochondrial uncoupler 2,4-dinitrophenol attenuates tissue damage and improves mitochondrial homeostasis following transient focal cerebral ischemia. J. Neurochem., 94, 1676–1684 (2005).
• 31) Haines BA, Mehta SL, Pratt SM, Warden CH, Li PA. Deletion of mitochondrial uncoupling protein-2 increases ischemic brain damage after transient focal ischemia by altering gene expression patterns and enhancing inflammatory cytokines. J. Cereb. Blood Flow Metab., 30, 1825–1833 (2010).
• 32) Haines B, Li PA. Overexpression of mitochondrial uncoupling protein 2 inhibits inflammatory cytokines and activates cell survival factors after cerebral ischemia. PLOS ONE, 7, e31739 (2012).
• 33) Gao Z, Wang J, Thiex R, Rogove AD, Heppner FL, Tsirka SE. Microglial activation and intracerebral hemorrhage. Acta Neurochir. Suppl., 105, 51–53 (2008).
• 34) Wu J, Yang S, Xi G, Song S, Fu G, Keep RF, Hua Y. Microglial activation and brain injury after intracerebral hemorrhage. Acta Neurochir. Suppl., 105, 59–65 (2008).
• 35) Yang Z, Zhao T, Zou Y, Zhang JH, Feng H. Curcumin inhibits microglia inflammation and confers neuroprotection in intracerebral hemorrhage. Immunol. Lett., 160, 89–95 (2014).
• 36) Yang Z, Liu Y, Yuan F, Li Z, Huang S, Shen H, Yuan B. Sinomenine inhibits microglia activation and attenuates brain injury in intracerebral hemorrhage. Mol. Immunol., 60, 109–114 (2014).
• 37) Wu H, Wu T, Hua W, Dong X, Gao Y, Zhao X, Chen W, Cao W, Yang Q, Qi J, Zhou J, Wang J. Pge2 receptor agonist misoprostol protects brain against intracerebral hemorrhage in mice. Neurobiol. Aging, 36, 1439–1450 (2015).
• 38) Harper JA, Dickinson K, Brand MD. Mitochondrial uncoupling as a target for drug development for the treatment of obesity. Obes. Rev., 2, 255–265 (2001).
• 39) Sheth UK. Mechanisms of anthelmintic action. Prog. Drug Res., 19, 147–157 (1975).
• 40) Weinbach EC, Garbus J. Mechanism of action of reagents that uncouple oxidative phosphorylation. Nature, 221, 1016–1018 (1969).
• 41) Sekulovski N, Whorton AE, Tanaka T, Hirota Y, Shi M, MacLean JA II, de Mola JRL, Groesch K, Diaz-Sylvester P, Wilson T, Hayashi K. Niclosamide suppresses macrophage induced inflammation in endometriosis. Biol. Reprod., 102, 1011–1019 (2020).
• 42) Colombo E, Farina C. Astrocytes: Key Regulators of neuroinflammation. Trends Immunol., 37, 608–620 (2016).
• 43) Gong H, Tai H, Huang N, Xiao P, Mo C, Wang X, Han X, Zhou J, Chen H, Tang X, Zhao T, Xu W, Gong C, Zhang G, Yang Y, Wang S, Xiao H. Nrf2-SHP cascade-mediated STAT3 inactivation contributes to AMPK-driven protection against endotoxic inflammation. Front. Immunol., 11, 414 (2020).