Milrinone Ameliorates the Neuroinflammation and Memory Function of Alzheimer’s Disease in an APP/PS1 Mouse Model

Introduction

Alzheimer’s disease (AD) is a complex neurodegenerative disorder, which commonly results in hypomnesis and cognitive deficits.¹ There are approximately 50 million people suffering from AD in 2017 all over the world, and this number is constantly increasing in the next 30 years.² The histopathologic characterizations of AD include a series of interactive processes such as neuronal amyloid-β (Aβ) plaque formation, microglial activation, tau protein hyperphosphorylation, and neuroinflammation.^3,4 Aβ deposits are considered to enrich the activation of microglial cells.⁵ The activated microglial cell is a crucial neuroinflammation factor to accelerate NLRP3 inflammasomes formation⁶ and promotes the secretion of inflammatory cytokines such as IL-1β and IL-6, which in turn increases the level of Aβ plaque.^7,8 Meanwhile, tau hyperphosphorylation especially the sites threonine 231 (T231) and serine 396 (S396) phosphorylated prematurely is an important hallmark for the occurrence of AD.^9,10 Currently, only few drugs including Tacrine, Donepezil, Galantamine, Memantine, and Rivastigmine are confirmed to be used for treatment with AD.^11–15 However, these therapies are generally used to control symptoms rather than alter the course of AD, or even lead to some severe side effects such as hepatotoxicity, gastrointestinal fatigue, and muscle cramps.^11–15 Therefore, exploring clinical drugs based on inhibition of AD pathological features is particularly important for improving AD.

Milrinone is a kind of bipyridine with vasodilator properties, which is generally used in the therapies for cardiopulmonary diseases.^16,17 For instance, Qasim and Jain revealed that milrinone can ameliorate persistent pulmonary hypertension in newborn through the great improvement for cardiac function and the reduction for pulmonary vascular resistance.¹⁶ Tang et al performed a meta-analysis for the function of milrinone on patients with acute heart failure (AHF) or acute myocardial infarction (AMI); they uncovered that milrinone may be an effectively clinical drug for the treatment of AHF and AMI.¹⁷ More importantly, growing evidences uncovered the clinical use of milrinone in neurological and neurocritical care patients, including subarachnoid hemorrhage (SAH) and stroke.^18,19 For example, a clinical study of the application of milrinone on SAH patients in Montreal Neurological Hospital has indicated that intravenous milrinone infusion and the maintenance of homeostasis is simple to use and requires less intensive monitoring and resources than the standard hypertension, hypervolemia, and hemodilution (triple-H) therapy for SAH.¹⁸ Labeyrie et al found that intravenous milrinone can effectively decrease the incidence of stroke.¹⁹ In addition, the neuroprotective role of milrinone on nerve or brain injury-related diseases has been also confirmed in animal models.^20–22 Arac et al reported that milrinone represses neurocytes apoptosis and inflammation in spinal cord injury (SCI) rats, thereby making valuable contributions on the function recovery of spinal cord tissues.²⁰ Choi et al made the comparisons among the effects of milrinone, sodium nitroprusside, and nitroglycerin on patients with cerebral perfusion injury, and found that milrinone treatment is more helpful for the cognitive function recovery relative to the other two treatments.²¹ Saklani et al further explored the protective effect of milrinone on cerebral ischemia-reperfusion injury (CI/RI) mice, and indicated that milrinone protect against memory loss after CI/RI through regulation of calcium level.²² As a novel drug treatment with brain diseases, however, there are relatively rare studies on the effects of milrinone for AD therapies.

In this study, the function of milrinone on the malignant behaviors of AD and its detailed action mechanism on AD pathogenesis were investigated. Our results may provide an underlying drug to attenuate AD and not just control its clinical symptoms.

Materials and Methods

AD Mouse Model and Drug Treatment

All animal experiments in this study were in strict accordance with the protocols stated in the Guide for the Care and Use of Laboratory Animals and approved by the ethical committee of The Third Affiliated Hospital of Qiqihar Medical University. Female APP/PS1 mice (30 weeks) and age-matched female wild-type C57BL6/L mice (control group) were procured from Jackson Laboratory. The mice were allowed to adapt to the laboratory environment for 2 weeks before testing. Afterwards, the APP/PS1 mice (32 weeks) were further assigned into two groups randomly: the APP/PS1 and APP/PS1 + M groups (n = 6). The mice in APP/PS1 + M group were injected with milrinone (i.p., 0.5 mg/kg),²⁰ while the mice in the control and APP/PS1 groups were intraperitoneally injected with the equal saline. All mice were treated once a day for a month.

Passive Avoidance Test

Passive avoidance test (PAT) was performed according to the previous study.²³ The experimental apparatus used in this study was purchased from Taimeng Tech (Chengdu, China), which was divided into two independent compartments (one for light, and one for dark) connected together via an automatic guillotine door. This test included a training test and a formal test (one day after the training). For training test, the mice of the above groups were initially placed into the light compartment to habituate for 180 sec, and then the guillotine door was opened to allow the mice to enter the dark compartment, in which a footshock (0.5 mA, 2 sec) was delivered. Ten sec later, the mice were removed from the dark compartment. Similar procedures were performed in the formal test except for footshock. The frequencies and latencies for the mice entering from the light compartment to the dark one were recorded with a maximum of 300 sec.

Morris Water Maze Test

As the previous study mentioned,²⁴ the Morris water maze test (MWMT) was conducted. Briefly, all the mice were initially placed into a water pool (temperature: 23 ± 1°C, diameter: 1.5 m, height: 0.6 m) for 1 min to adapt to the environment. Subsequently, a hidden platform (immersed 1 cm below the water) was set in the middle of the platform. The mice were placed into the pool to find the platform within 60 sec. Each mouse was tested three times a day at intervals of at least 15 min. The time required to find the hidden platform was recorded as the escape latency. To assess the memory ability, the invisible platform was removed and the mice were allowed to swim freely for 60 sec. The numbers of crossing the original position of the platform were recorded. All the data were obtained using a video tracking system (Taimeng Tech).

Brain Tissues Collection

After the above behavior tests, all the mice were anaesthetized by pentobarbital sodium (i.p., 50 mg/kg) and sacrificed by decapitation. The hippocampus and cortex tissues were collected and immediately frozen at −80°C for further analysis.

Cell Culture, Grouping, and Treatments

Mouse microglial cell line BV-2 (immortalized cell) was procured from Cobioer biotech (Nanjing, China) and cultured in Dulbecco’s Modified Eagle’s medium (DMEM) with 5% fetal bovine serum and 1% streptomycin/penicillin at 37°C with 5% CO₂. Milrinone was dissolved in 0.3% carboxyl methyl cellulose sodium (CMC-Na), and Aβ_1–42 was dissolved in sterile phosphate buffered saline (PBS). Afterwards, the cells were initially pretreated with different concentrations of milrinone (0, 10, 25, 50 μmol/L) for 1 h, followed by the stimulation with Aβ_1–42 (10 μmol/L) and lipopolysaccharide (LPS) (1 μg/mL). The cells treated with equal PBS were used as controls. Approximately 1 day later, the levels of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6 were assessed by enzyme-linked immunosorbent assay (ELISA).

ELISA Analysis

According to the manufacturer’s protocol, the levels of the inflammatory factors (TNF-α, IL-1β, and IL-6) in mouse hippocampus, cortex, and BV-2 cells were measured using specific ELISA kits (MSK Biotech, Ltd., Wuhan, China), while the levels of superoxide dismutase (SOD), malondialdehyde (MDA) and glutathione (GSH) in mouse hippocampus, and the levels of Aβ₄₀ and Aβ₄₂ in mouse hippocampus and cortex tissues were measured using the corresponding commercial assay kits (MSK Biotech).

Western Blotting Analysis

RIPA buffer containing protease inhibitors was used to extract proteins from mouse brain tissues and BV-2 cells. Protein concentrations were then determined using a BCA Protein Assay Kit (Abcam, Cambridge, UK). Protein samples were separated via 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and the resolved proteins were transferred onto polyvinylidene fluoride (PVDF) membranes. Membranes were blocked with 5% bovine serum albumin at room temperature. After blocking, membranes were incubated overnight at 4°C with primary antibodies against TLR4 (1:1000; Abcam), MyD88 (1:1000; Abcam), NF-κB (p65) (1:1000; Abcam), phospho-NF-κB (p-p65) (1:1000; Cell Signaling), NLRP3 (1:1000; Abcam), ASC (1:1000; Affinity Biosciences), caspase 1 (1:1000; Affinity Biosciences), IL-1β (1:1000; Abcam), IL-18 (1:1000; Abcam), Iba-1 (1:1000; Abcam), p-Tau (S396) (1:1000; Abcam), p-Tau (T231) (1:1000; Abcam), and GAPDH (1:1000; Abcam). Thereafter, they were washed three times with Tris-buffered saline Tween-20. Subsequently, an HRP-conjugated IgG secondary antibody (1:5000; Santa Cruz, Waltham, MA, USA) was added and membranes were incubated at room temperature for 1 h. GAPDH was used as the internal reference. An enhanced chemiluminescence detection kit (Thermo Fisher Scientific) was used to detect the bands, which were then quantified using Gel-Pro Analyzer software (version 4.0; Media Cybernetics, Silver Spring, MD, USA).

Statistical Analysis

SPSS 20.0 software (Chicago, USA) was used in analysis. One-way ANOVA, followed by Tukey’s multiple comparisons test was used to assess experimental data. Data were presented as means ± SD. P-value less than 0.05 indicated a statistically significant difference.

Results

Milrinone Ameliorates the Memory Functions of APP/PS1 Mice

To explore the therapeutic efficacy of milrinone on AD, relevant behaviour tests were performed on APP/PS1 mice. As presented in Figure 1A and B, we found that APP/PS1 mice had a relatively higher frequency to enter the dark compartment and a shorter latency before entering the dark compartment than the mice of the WT group (P < 0.01). However, these situations were all reversed in the milrinone treatment group (P < 0.05). Subsequently, the results of MWMT demonstrated that there seemed no significant differences among the three aforementioned groups on day 1 and 2. From day 3 to day 4, the escape latency of APP/PS1 mice was longer compared to that of the WT mice, while milrinone treatment remarkably shortened the escape latency (Figure 1C, P < 0.01). Further probe experiments revealed that decreased frequency of crossing over the platform position was observed in APP/PS1 mice by contrast to the WT mice (Figure 1D, P < 0.01). On the contrary, it was increased in the treatment group (P < 0.05).

Figure 1 Milrinone ameliorates the memory functions of APP/PS1 mice. (A) The number of APP/PS1 mice entering the dark compartment during the passive avoidance test (PAT). (B) The latency of APP/PS1 mice in the PAT. (C) The escape latency of APP/PS1 mice in the Morris water maze test (MWMT). (D) The number of crossings through the original position of the platform in MWMT. **P < 0.01 vs the WT group; #P < 0.05, ##P < 0.01 vs the APP/PS1 group.

Milrinone Inhibits Aβ Plaque Formation, Microglia Activation, and Tau Hyperphosphorylation in APP/PS1 Mice

The formation of Aβ plaque is identified as an important hallmark of AD. Meanwhile, both microglia activation and tau hyperphosphorylation are also accompanied with the progression of AD. Therefore, the effects of milrinone on the above biological processes were further investigated. As illustrated in Figure 2A–D, we found that the levels of Aβ₄₀ and Aβ₄₂ in cortex and hippocampus of APP/PS1 mice were expected increased (P < 0.01). In contrast, the mice in the milrinone group had relatively low levels of Aβ plaque (P < 0.05). In addition, the level of Iba-1 (microglia marker) was also elevated in APP/PS1 mice (Figure 2E and F, P < 0.01), whereas it was repressed in the milrinone treatment group (P < 0.01). Similar patterns were observed in the protein levels of p-Tau (S396) and p-Tau (T231) (Figure 2G, P < 0.01).

Figure 2 Milrinone inhibits the formation of Aβ plaque, activation of microglia, and the hyperphosphorylation of tau in APP/PS1 mice. (A) The level of Aβ40 in APP/PS1 mice cortex was measured by ELISA analysis. (B) The level of Aβ42 in APP/PS1 mice cortex was measured by ELISA analysis. (C) The level of Aβ40 in APP/PS1 mice hippocampus was measured by ELISA analysis. (D) The level of Aβ42 in APP/PS1 mice hippocampus was measured by ELISA analysis. (E) The protein level of Iba-1 in APP/PS1 mice cortex was determined by Western blot assay. (F) The protein level of Iba-1 in APP/PS1 mice hippocampus was determined by Western blot assay. (G) The protein levels of p-Tau (Ser396) and p-Tau (Thr231) in APP/PS1 mice brain tissues were determined by Western blot assay. **P < 0.01 vs the WT group; #P < 0.05, ##P < 0.01 vs the APP/PS1 group.

Milrinone Attenuates the Oxidative Stress in APP/PS1 Mice

To further explore the effect of milrinone on oxidative stress, the levels of MDA, SOD, and GSH-Px were measured by Western blotting. We found that MDA level was increased, and both SOD and GSH-Px were decreased in hippocampus of APP/PS1 mice (Figure 3A–C, P < 0.01). Unsurprisingly, these situations were reversed in APP/PS1 mice treatment with milrinone (P < 0.05).

Figure 3 Milrinone attenuates the oxidative stress in APP/PS1 mice. (A) The level of MDA in APP/PS1 mice brain tissues was measured by ELISA analysis. (B) The level of SOD in APP/PS1 mice brain tissues was measured by ELISA analysis. (C) The level of GSH-Px in APP/PS1 mice brain tissues was measured by ELISA analysis. **P < 0.01 vs the WT group; #P < 0.05, ##P < 0.01 vs the APP/PS1 group.

Milrinone Represses the Inflammatory Responses in APP/PS1 Mice

Inflammatory responses occur in the development of AD. We then assessed the function of milrinone on AD inflammation. As shown in Figure 4A–F, high levels of IL-1β, IL-6, and TNF-α were found in cortex and hippocampus tissues of APP/PS1 mice (P < 0.01), whereas milrinone repressed the release of these inflammatory cytokines from the damaged brain tissues (P < 0.05).

Figure 4 Milrinone represses the inflammatory responses in APP/PS1 mice. (A) The level of IL-1β in APP/PS1 mice cortex was measured by ELISA analysis. (B) The level of IL-6 in APP/PS1 mice cortex was measured by ELISA analysis. (C) The level of TNF-α in APP/PS1 mice cortex was measured by ELISA analysis. (D) The level of IL-1β in APP/PS1 mice hippocampus was measured by ELISA analysis. (E) The level of IL-6 in APP/PS1 mice hippocampus was measured by ELISA analysis. (F) The level of TNF-α in APP/PS1 mice hippocampus was measured by ELISA analysis. **P < 0.01 vs the WT group; #P < 0.05, ##P < 0.01 vs the APP/PS1 group.

Milrinone Diminishes the Secretion of Inflammatory Cytokines and Inhibits the Formation of NLRP3 Inflammasomes in BV-2 Cells

As presented in Figure 5A–C, different concentrations of milrinone (0, 10, 25, 50 μmol/L) were assigned to explore its effect on inflammation in BV-2 cells. We found that LPS/Aβ significantly promoted inflammation levels of BV-2 cells (P < 0.01). Different milrinone concentrations reduced the secretion of inflammatory cytokines to a certain extent (P < 0.05). Meanwhile, we further demonstrated that there were no significant differences between the LPS/Aβ and LPS/Aβ + M10 group, as well as between the LPS/Aβ + M25 and LPS/Aβ + M50 group in inflammation level. Therefore, M25 was used to perform the subsequent trails. NLRP3 inflammasomes are commonly strongly correlated with neuroinflammation.^25,26 The levels of NLRP3 inflammasomes-related proteins (ASC, NLRP3, active caspase 1, IL-1β, and IL-18) were assessed in LPS/Aβ-induced BV-2 cells. We found that the levels of the aforementioned proteins were all elevated in the LPS/Aβ group (Figure 5D–I, P < 0.01). In the LPS/Aβ + M25 group, however, the levels of ASC, NLRP3, active caspase 1, IL-1β, and IL-18 were suppressed compared to those of the LPS/Aβ group (P < 0.05).

Figure 5 Milrinone diminishes the secretion of inflammatory cytokines and inhibits the formation of NLRP3 inflammasomes. (A) The level of IL-1β in LPS/Aβ-treated BV-2 cells with different concentrations of milrinone (0, 10, 25, 50 μmol/L) was measured by ELISA analysis. (B) The level of IL-6 in LPS/Aβ-treated BV-2 cells with different concentrations of milrinone (0, 10, 25, 50 μmol/L) was measured by ELISA analysis. (C) The level of TNF-α in LPS/Aβ-treated BV-2 cells with different concentrations of milrinone (0, 10, 25, 50 μmol/L) was measured by ELISA analysis. (D) The Western blot assay images for the levels of ASC, NLRP3, active caspase 1, IL-1β, and IL-18 in LPS/Aβ-treated BV-2 cells. (E) The protein level of ASC in LPS/Aβ-treated BV-2 cells was measured by Western blot assay. (F) The protein level of NLRP3 in LPS/Aβ-treated BV-2 cells was measured by Western blot assay. (G) The protein level of active caspase 1 in LPS/Aβ-treated BV-2 cells was measured by Western blot assay. (H) The protein level of IL-1β in LPS/Aβ-treated BV-2 cells was measured by Western blot assay. (I) The protein level of IL-18 in LPS/Aβ-treated BV-2 cells was measured by Western blot assay. **P < 0.01 vs the control group; #P < 0.05, ##P < 0.01, && P < 0.01 vs the LPS/Aβ group. ns: not significant.

Milrinone Inactivates the TLR4/MyD88/NF-κB Signaling Pathway in BV-2 Cells

Finally, the interaction between milrinone and TLR4/MyD88/NF-κB signaling pathway was investigated due to the important role of this pathway in microglial mediated inflammation of brain injury. As shown in Figure 6A–D, we discovered that the levels of TLR4, MyD88, and p-p65/p65 were upregulated in the LPS/Aβ group (P < 0.01), whereas milrinone reversed the increased levels of TLR4, MyD88, and p-p65/p65 caused by LPS/Aβ inducement (P < 0.01).

Figure 6 Milrinone inactivates the TLR4/MyD88/NF-κB signaling pathway in BV-2 cells. (A) The Western blot assay images for the levels of TLR4, MyD88, p65, and p-p65 in LPS/Aβ-treated BV-2 cells. (B) The protein level of TLR4 in LPS/Aβ-treated BV-2 cells was measured by Western blot assay. (C) The protein level of MyD88 in BV-2 cells was measured by Western blot assay. (D) The protein level of p-p65/p65 in BV-2 cells was measured by Western blot assay. **P < 0.01 vs the control group. ##P < 0.01 vs the LPS/Aβ group.

Discussion

Alzheimer’s disease mainly occurs in the aged population and its important characterizations are memory loss and cognitive decline.¹ The interactions among Aβ plaque, microglial activation, tau protein hyperphosphorylation, and neuroinflammation were the main histopathologic features that affect the progression of AD.^3,4 This study focused on the regulatory mechanisms of milrinone on Aβ deposits, microglial activation, tau phosphorylation, oxidative stress, and inflammation, which demonstrated that milrinone could ameliorate memory deficits and cognitive decline caused by AD.

APP/PS1 transgenic mouse was generally used for AD researches.^27,28 In this study, the analyses for PAT and MWMT demonstrated that milrinone treatment ameliorated cognitive deficits in learning and memory function in APP/PS1 mice. We speculated milrinone may be an effective drug for AD therapy. The enrichment of Aβ deposits is the initial changes in brain tissues after AD and aggravates the toxicity to central neural system.^29,30 Numerous studies further confirmed that inhibition of Aβ deposit levels in brain tissues is beneficial to the cognitive restoration of AD.^27,28,31,32 In our study, relatively high levels of Aβ₄₀ and Aβ₄₂ were found in both cortex and hippocampus of APP/PS1 mice, whereas they were repressed by injection of milrinone, suggesting that milrinone suppresses the accumulation of Aβ deposits in APP/PS1 mice.

Previous study has revealed that the accumulated Aβ triggered the activation of microglia, in which process is accompanied with the release of inflammatory cytokines.³³ Therefore, inflammation reactions in APP/PS1 mice with or without milrinone treatment were initially studied. We discovered that the levels of IL-1β, IL-6, and TNF-α in LPS/Aβ-treated BV-2 cells and APP/PS1 mice brain tissues (cortex and hippocampus) were elevated. By contrast, milrinone restrained the inflammation responses. Similarly, a recent study on milrinone for SCI therapy has indicated that the inflammation caused by SCI can be attenuated by milrinone treatment.²⁰ The results implied that milrinone may function as an anti-inflammatory role in AD pathogenesis. It has been confirmed that NLRP3 inflammasome serves as a vital factor in the progression of neuroinflammation.^34–36 We further speculated NLRP3 inflammasomes-related proteins (ASC, NLRP3, active caspase 1, IL-1β, and IL-18) may be also inhibited by milrinone. The experimental data that the levels of ASC, NLRP3, active caspase 1, IL-1β, and IL-18 in LPS/Aβ-treated BV-2 cells were suppressed by milrinone inducement validated this hypothesis. In addition, numerous studies demonstrated that Aβ deposits can enter the mitochondria to induce oxidative stress and result in toxicity to the neurocytes.^37,38 Interestingly, in this study, we found milrinone reversed the toxic effects of oxidative stress on AD mice, which process may attenuate the oxidative stress damages to neurons. Furthermore, the sites T231 and S396 phosphorylated earlier are considered as important hallmarks for the occurrence of AD.^9,10 In the current study, the levels of p-Tau (S396) and p-Tau (T231) were unsurprisingly elevated in the brain tissues of AD mice, while were restrained by milrinone. Meanwhile, we also uncovered that the protein level of Iba-1, a marker of microglial activation, was inhibited by milrinone, pointing out that milrinone is an inhibitor for the activation of microglia. All the above results suggested that milrinon may ameliorate AD via repression of neuroinflammation, oxidative stress, and tau hyperphosphorylation as well as regulation of Aβ burden and microgliosis.

The activation of TLR4/MyD88/NF-κB signalling pathway is closely correlated with several inflammation diseases, such as coronary microembolization,³⁹ acute lung injury,⁴⁰ and knee osteoarthritis.⁴¹ In the course of AD, the occurrence of neuroinflammation has been confirmed in our above conclusions. Therefore, we speculated TLR4/MyD88/NF-κB signalling may also involve in the regulation of neuroinflammation in AD. As shown in Figure 6, relevant protein levels were measured. We found that in cellular model, AD may activate the TLR4/MyD88/NF-κB signalling. Meanwhile, milrinone treatment expectedly repressed the expression of this pathway. Some previous studies on AD inflammation suggested that the inhibition of TLR4/MyD88/NF-κB signalling is crucial for the reduction of neuroinflammation in AD,^42,43 which validated our experimental data that milrinone inhibited neuroinflammation through suppression of TLR4/MyD88/NF-κB signalling pathway.

Conclusion

In a word, our in vivo experimental data revealed the neuroprotective role of milrinone on AD pathogenesis via the regulation of AB plaque enrichment, microglial activation, tau protein hyperphosphorylation, oxidative stress, and inflammation reactions. At the same time, the in vitro model results uncovered the inhibitory effect of milrinone on the inflammatory-related TLR4/MyD88/NF-κB signalling pathway. Our findings may provide a promising therapeutic drug for the amelioration of AD.