The main players of the inflammatory response in the central nervous system (CNS) are microglia and astrocytes, which, once activated to a proinflammatory profile, begin to release different inflammatory mediators, such as cytokines, chemokines, and reactive oxygen and nitrogen species [1
]. This sustained neuroinflammatory stimulus may contribute to neuronal death [2
] and, therefore, to the pathogenesis of neurodegenerative diseases, including Multiple Sclerosis (MS), Alzheimer’s Disease (AD), and Parkinson’s Disease (PD) [3
]. Drugs that display anti-inflammatory and antioxidant activity have considerable general potential for new therapies for these diseases.
As surveillant cells, microglia present a highly plastic phenotype [4
]. Once a stressor is recognized, microglia rapidly shift to an activated state, which leads to changes in their morphology and molecular profile [5
]. Activated microglia can be recognized as having an M1 phenotype (classically activated or proinflammatory) or M2 phenotype (alternatively activated or anti-inflammatory). Classically activated microglia (M1) can be characterized by the expression of MHC class II molecules and proinflammatory molecules, such as TNF-α, interleukin (IL)-β, IL-6, IL-18, nitric oxide, CCL2, and CCL5 [6
]. Alternatively, activated microglia (M2) help to recover tissue homeostasis and express anti-inflammatory cytokines, such as arginase, IL-10, and TGF-β [7
Flavonoids are natural products derived from plants that have been shown to exert potent anti-inflammatory, antioxidant, and immunomodulatory effects. Many studies suggest a correlation between flavonoid intake in the diet and a reduction in levels of dementia, in addition to its benefits to memory and the learning process [8
]. Flavonoids modulate inflammatory responses involved in neurodegenerative diseases through the reduction of the expression of proinflammatory cytokines, including IL-6, TNF-α, IL-1β, and COX-2 [10
]. In the CNS, the anti-inflammatory effects of flavonoids have been related to the control of astrogliosis and microgliosis [12
]. Therefore, flavonoids are regarded as potential therapeutic agents for controlling inflammatory processes involved in neurodegenerative diseases [14
In this study, we have investigated the neuroprotective and immunomodulatory effects of the phytoestrogen agathisflavone (FAB), which has known activity upon activation of estrogen receptors (ERs). FAB is a biflavonoid and, in this study, was extracted from Poincianella pyramidalis
(Tul.), an endemic plant common in northeastern Brazil. We used an in vitro model of neuroinflammation, in which cocultures of neurons and glial cells were exposed to lipopolysaccharide (LPS) and IL-1β; LPS is a component of the Gram-negative bacteria cell membrane known to activate microglia and promote secondary neuronal damage [15
], whereas IL-1β is an important cytokine responsible for the activation of the M1 proinflammatory profile in microglia [16
]. The results demonstrated that FAB displayed a significant cytoprotective effect on neurons and an immunomodulatory effect on microglia, but these effects were not mediated via estrogen receptors in this model.
2.1. Neuron–Glial Cell Cocultures
Cell cultures were obtained from cerebral hemispheres from Wistar rats. The animals were provided by the Department of Physiology of the Institute of Health Sciences of the Federal University of Bahia (Salvador, BA, Brazil). Primary cultures of glial cells were obtained from Wistar rats as described in our previous work [12
]. Cerebral hemispheres from postnatal Wistar rats at the age of one to two days old were isolated aseptically and the meninges were mechanically removed. The cortex was dissociated mechanically and suspended in DMEM HAM F12 medium (Gibco®
, Life Technologies, Burlington, ONT, Canada, 12500-062), supplemented with 2 mM l
-glutamine, 0.011 g/L pyruvate, 10% FBS, 3.6 g/L Hepes, 33 mM glucose (Cultilab, Campinas-SP, Brazil), 100 IU/mL penicillin G, and 100 µg/mL streptomycin, and cultured in 100 mm Ø plates in a humidified atmosphere with 5% CO2
at 37 °C. Culture medium was changed every 2 days and cells were cultured for 15 days. Cells were then washed 3× with phosphate-buffered saline (PBS), detached with trypsin (Trypsin EDTA, Sigma Aldrich, Saint Luis, MO, USA, 9002077), plated at a density of 1 × 105
, and maintained in culture for 72 h. After incubation, neurons obtained from cerebral hemispheres of 14–16-day-old Wistar rat embryos, using the same method described above for glial isolation, were suspended in supplemented DMEM/HAM F12 (Gibco®
, Life Technologies, Carlsbad, CA-USA, 12500-062), and seeded at half the number of glial cells (5 × 104
) onto the astroglial monolayer. Cells were incubated in a humidified atmosphere with 5% CO2
at 37 °C for 8 days in vitro (DIV), when treatments were performed.
2.2. Agents and Treatments
Agathisflavone (FAB) was extracted from Poincianella pyramidalis
(Tul.) as previously described [17
], stored at 100 mM in dimethyl sulfoxide (DMSO, Sigma Aldrich, Saint Luis, MO, USA, 472301), and kept out of light at −20 °C until use. For the experiments, FAB was diluted in culture medium to make a final concentration of 0.1 or 1 µM in the neuron–glial cocultures; control cultures were treated with DMSO, the vehicle of dilution of FAB. To induce neuroinflammation, at 26 DIV, cocultures were treated for 24 h with LPS (1 µg/mL; Sigma Aldrich, Saint Luis, MO, USA, L2880) or IL-1β (10 ng/mL; R&D Systems, Minneapolis, MN, USA, 501-RL-010); then, the medium was removed and replaced with medium containing just agathisflavone (0.1 or 1 µM) or vehicle, and cultures were analyzed after 24 h. To assess whether the effects of agathisflavone were mediated through ERs, neuron–glial cocultures were treated with the selective ER-α antagonist MPP dihydrochloride at 10 nM (1,3-bis(4-hydroxyphenyl)-4-methyl-5-[4-(2-piperidinylethoxy)phenol]-1H-pyrazole dihydrochloride; Sigma Aldrich, Saint Luis, MO, USA, M7068) or the selective ER-β antagonist PHTPP at 1 μM (4-[2-phenyl-5,7-bis(trifluoromethyl) pyrazolo[1,5-a]pyrimidin-3-yl]phenol; Tocris, Bristol, UK, #2662); control cultures were treated with DMSO vehicle.
For immunocytochemistry, cells were washed with PBS three times and fixed with 4% paraformaldehyde for 15 min at room temperature (RT). Cultures were washed three times with PBS, incubated with 0.3% Triton X-100 in PBS (Sigma Aldrich, Saint Luis, MO, USA, 9002-93-1) for 5 min, and blocked by incubation with PBS containing 5% bovine serum albumin (BSA) (Sigma Aldrich, Saint Luis, MO, USA, A9418)) for 1 h. After blocking, samples were incubated with primary antibodies diluted in PBS containing 1% of BSA overnight. Cells were washed with PBS three times. Then, secondary antibodies were added to cells and incubated for 2 h. The cells were washed with PBS three more times and incubated with 5.0 μg/mL 4,6-diamidino-2-phenylindole (DAPI, Invitrogen - Molecular Probes, Eugene, OR-USA) for nuclear staining. Staining was visualized on a fluorescence microscope (Leica, Wetzlar-Germany, DFC7000). Images were captured with a 20× or 40× objective. The following primary antibodies were used at the indicated dilutions: anti-Tubulin β3 (mouse, 1:500; BioLegend, San Diego, CA, USA, 801201), anti-glial fibrillary acidic protein (GFAP) (rabbit, 1:300; DAKO, Glostrup-Denmark, Z0334), anti-Iba-1 (ionized calcium-binding adaptor molecule 1, rabbit, 1:200; Wako, Richmond, VA, USA, 019-19741), anti-CD68 (rat, 1:100; Abcam, ab53444), anti-active caspase-3 (rabbit, 1:300; Chemicon, ab3623), anti-neurofilament (1:400; Abcam, Cambridge, UK, AB24574), and anti-NF-κB-P50 (mouse, 1:200; Santa Cruz Biotechnology, Santa Cruz, CA, USA, SC8414). The following secondary antibodies were used at the indicated dilutions: Alexa Fluor 488-conjugated goat anti-mouse IgG (1:500; Molecular Probes, A11001), Alexa Fluor 594-conjugated goat anti-rabbit IgG (1:500; Invitrogen, Molecular Probes, Eugene, OR, USA, A11037), Alexa Fluor 555-conjugated goat anti-rat IgG (1:500; Invitrogen - Molecular Probes, Eugene, OR, USA, A21434), Alexa Fluor 488-conjugated goat anti-rabbit IgG (1:500; Invitrogen, Molecular Probes, Eugene, OR, USA, A11008), and Alexa Fluor 594-conjugated goat anti-mouse IgG (1:500, Invitrogen - Molecular Probes, Eugene, OR, USA). The quantification was performed by analyzing the total number of positive cells (per marker), divided by the total number of nuclei (DAPI positive) × 100.
2.4. Bromodeoxyuridine Cell Proliferation Assay
Proliferation was evaluated using bromodeoxyuridine (BrdU) (Sigma Aldrich, Saint Luis, MO, USA). BrdU (10 µM) was added to the wells at the start of each treatment. Cells were fixed and DNA was denatured by treatment with denaturing solution (2 N HCl) for 20 min at room temperature. Mouse anti-BrdU monoclonal antibody (1:200, Sigma Aldrich, Saint Luis, MO, USA, B8434) diluted in PBS was pipetted into the wells and allowed to incubate for 1 h. Unbound antibody was washed away and cells were incubated with Alexa Fluor 594 antibody specific for mouse IgG (1:500, Invitrogen - Molecular Probes, Eugene, OR, USA) diluted in PBS-T for 1 h under slow agitation at room temperature. After incubation, the cell nuclei were stained with DAPI (5 µg/mL) for 10 min at room temperature. All reagents were used in accordance with the manufacturer’s instructions. Experiments were performed in triplicate. Thereafter, cells were analyzed using a fluorescence microscope (Leica, Wetzlar-Germany, DFC7000). Quantification was analyzed with ImageJ 1.33u (Bethesda, MD, USA).
2.5. Fluoro-Jade B (FJ-B) Staining
FJ-B staining was used to investigate neuronal death. Neuron–glial cocultures were cultivated in 96-well black plates (1.5 × 104 cells/cm2) and were treated with DMSO, LPS (1 µg/mL), or IL-1β in the corresponding wells for 24 h. Then, the medium was removed and replaced with medium containing agathisflavone 0.1 and 1 µM or IL-1 receptor antagonist (raIL-1, Sigma Aldrich, Saint Luis-MO, USA, 1 μg/mL) and kept for more than 24 h. After treatment, the cultures were fixed in ethanol at 4 °C for 10 min. Cultures were washed three times with PBS and then incubated with 0.3% Triton X-100 in PBS (Sigma Aldrich, Saint Luis, MO, USA) for 10 min. After washing in PBS three times, cells were incubated with 0.001% Fluoro-Jade B in PBS for 30 min at RT, under agitation and protection from light. After incubation, the cells were washed three times with PBS, incubated for 5 min at RT in the dark with 5 µg/mL DAPI for nuclear staining, and then washed three times with PBS. Analyses were performed on a spectrophotometer (Varioskan™ Flash Multimode Reader, Thermo Plate, Thermo Fisher Scientific, Inc., Vantaa, Finland), and the fluorescence intensity of each sample was measured at 480 nm for Fluoro-Jade B and 350 nm for DAPI. The values of absorbance of Fluoro-Jade B of each well were normalized to the DAPI absorbance in the same well.
2.6. RNA Isolation and cDNA Synthesis
Total RNA was isolated from primary cultures of rat microglia with Trizol® reagent according to the manufacturer’s specifications. Afterwards, 1 × 104 cells/cm2 were seeded in 60 mm plates and then treated for 24 h with 1 µg/mL LPS or IL-1β, 0.1 and 1 µM of FAB, or the combination of LPS and FAB. The samples were stored at −80 °C until the time of the analysis. The concentration and purity of RNA were determined by spectrophotometric analysis using a nanospectrum Kasvi (KASVI, Sao Jose dos Pinhais, PR, Brazil, K23-0002). DNA contaminants were removed by treating the RNA samples with DNase using the Ambion DNA-free kit (cat# AM1906, Invitrogen™, Life Technologies™, Carlsbad, CA, USA). For cDNA synthesis, SuperScript® VILO™MasterMix (cat# MAN0004286, Invitrogen™, Life Technologies, Carlsbad, CA, USA) was used in a 20 µL reaction with a concentration of 2.5 µg of total RNA.
2.7. Quantitative PCR (qPCR)
Quantitative real-time PCR was performed using Taqman® Gene Expression Assays (Applied Biosystems, CA, USA) containing two primers to amplify the sequence of interest, a specific Taqman® MGB probe, and TaqMan Universal Master Mix II with UNG (catalogue# 4440038 Invitrogen, Life Technologies™, Carlsbad, CA, USA). The assays corresponding to the genes quantified in this study were IL1B (Rn00580432_m1), TNF Loc1036 (Rn01525859_m1), Nos2 (Rn00561646_m1), IL6 (Rn01410330_m1), CCL2 (Rn00580555_m1), CCL5 (Rn00579590_m1_m1), IL10 (Rn01483988_m1), ARG (Rn00691090_m1), and TGFB (Rn00572010_m1). Real-time PCR was performed using the Quant Studio 7 Flex™ Real-Time PCR System (Applied Biosystems™ by Life technology, Carlsbad, CA, USA). The thermocycling conditions were performed according to the manufacturer’s specifications. The actin beta (Actb) (Rn00667869_m1) and Hypoxanthine Phosphoribosyl Transferase 1 (HPRT1) (Rn01527840_m1) targets were used as reference genes (endogenous controls) for normalization of gene expression data. Data were analyzed using the 2−ΔΔCt method. The results represent the average of 3 independent experiments.
2.8. NO Production
Nitric oxide (NO) production was assessed as sodium nitrite (NaNO2−) accumulation in the culture medium using a colorimetric test based on the Griess reagent (Wang et al., 2002). Samples (50 µL) were collected after LPS damage and/or 24 h treatment with 0.1 and 1 µM of FAB. Equal volumes of culture medium and Griess reagent (1% sulfanilamide, 0.1% N-(1-naphthyl) ethylenediamine dihydrochloride, and 2% phosphoric acid Sigma Aldrich, Saint Luis, MO, USA) were mixed. The mixture was incubated for 10 min at room temperature and then the absorbance at 550 nm was measured using a microplate reader (Varioskan™ Flash Multimode Reader, Thermo Plate, Thermo Fisher Scientific, Inc., Vantaa-Finland). The concentrations of nitrite in the samples were determined based on a sodium nitrite standard curve (1.26–100 mmol/L NaNO2). Three independent experiments were performed.
2.9. Statistical Analyses
Statistical analyses were performed using GraphPad Prism 5. We first analyzed whether the values came from a Gaussian distribution. Kruskal–Wallis followed by Dunn’s multiple comparison test was performed for non-normal samples; for normal samples, we performed one- or two-way analysis of variance (ANOVA) followed by Tukey’s or Bonferroni’s post-tests, respectively. Confidence intervals were defined at a 95% confidence level (p < 0.05 was considered to be statistically significant). Fold change was calculated by dividing the average (mean) value of the experimental group by that of the control group. In all figures, error bars represent SEM of at least 3 independent experiments.
2.10. Ethics Approval
All experiments were performed in accordance with the local Ethical Committee for Animal Experimentation of the Health Sciences Institute (Protocol No. 027/2012).
Flavonoids are natural plant-derived compounds that have attracted considerable attention as potential treatments for neurodegenerative diseases because of their antioxidant [29
] and immunomodulatory [31
] activities. They are polyphenolic compounds synthesized in response to stress and found in fruits, seeds, grains, vegetables, flowers, wines and teas [32
]. Biflavonoids are a specific class of flavonoids composed of a combination of C-C or C-O-C bonded flavonoid dimers [33
] and have shown greater pharmaceutical efficiency than their respective monomers [34
]. Here, using an established LPS and IL-1β model of neuroinflammation [35
], we demonstrated that the biflavonoid agathisflavone (FAB) had potent neuroprotective and immunomodulatory effects in neuron–glial cocultures. A key action of FAB is to inhibit the proinflammatory function of microglia and direct them towards an anti-inflammatory M2-like phenotype. The results support FAB as a promising therapy for protecting against neurodegeneration and promoting tissue repair.
Current work using an in vivo model demonstrated that albino Swiss mice treated with high doses of agathisflavone did not show toxicity in any analyzed parameter (hematological, biochemical, histopathological, behavioral, as well as physiological) and presented LD50 larger than 2000 mg/kg, which is indicative of low toxicity [36
]. A key finding of this study is that FAB is not toxic in both concentrations tested and has an important neuroprotective effect, in support of studies demonstrating neuroprotective effects of other biflavonoids, including amentoflavone, ginkgetin, and isoginkgetin, in oxidative-stress-induced and amyloid-β-peptide-induced cell death [37
]. In the present study, LPS and IL-1β were shown to result in neuronal disruption and increased neuronal cell death, as measured by expression of caspase-3, which plays a pivotal role in apoptosis [38
]. Notably, FAB decreased the number of caspase-3+
neurons and increased the overall number of neurons (β-Tubulin III+
cells) and BrdU+
cells, indicating that in addition to being neuroprotective, FAB has a neurogenic effect, as we have shown previously for FAB in murine pluripotent stem cells [39
The neuroprotective effect of FAB is undoubtedly related to its inhibition of LPS- or IL-1β-induced microglial activation. It is known that microglial activation in the CNS is heterogeneous and often categorized into M1 and M2 phenotypes, which respectively have either cytotoxic or neuroprotective actions. Notably, dynamic changes in M1/M2 phenotypes have been associated with multiple neurodegenerative diseases, including AD, MS, stroke, and traumatic injury [40
]. In general, M1 microglia predominate at the site of injury, and persistently activated M1 responses lead to neuronal loss. In contrast, the M2 phenotype is anti-inflammatory and is associated with the repair process. Hence, controlling the switch from M1 to M2 phenotype has considerable therapeutic benefit. LPS and IL-1β are potent inducers of the proinflammatory M1 microglial phenotype, characterized by the expression of the CD68 and NF-κB markers, as well as microglial proliferation (identified by BrdU incorporation) [41
]. Our results demonstrate that, compared with the cells treated with LPS or IL-1β, FAB improved microglial morphology, reduced microglial proliferation (BrDU+
cells), and decreased the M1 markers CD68 and NF-κB. In addition, we demonstrated that LPS directly induced production of NO in the cocultures and that this was significantly reduced by FAB. At high concentrations, NO reversibly inhibits mitochondrial respiration by competing with O2
in cytochrome c oxidase and is neurotoxic [6
]. Blockade of NO production by FAB is a likely mechanism by which it protects against LPS-induced neuronal death.
An important finding is that FAB reduced the LPS- and IL-1β-induced expression of proinflammatory molecules, such as IL-1β, TNF, IL-6, CCL2, and CCL5, and increased the expression of regulatory molecule IL-10, protecting cortical neurons from inflammation. Comparable results were found in another study published by our group using the same approach (coculture of neurons and glia), which showed that after excitotoxicity induced by glutamate, FAB was able to reduce cell death induced by glutamate through reducing the expression of proinflammatory cytokines, including TNF, IL-1β, and IL6, and increasing the expression of IL10 and arginase 1 [12
]. Astrocytes are also involved in the regulation of immune responses in neurodegenerative diseases [47
]. Astrocytes respond to most CNS insults by reactive astrogliosis, characterized by cellular hypertrophy and increased GFAP expression, often associated with cell proliferation [48
]. Under continuous stimulation, reactive astrocytes can potentiate neuroinflammation, as well as compromise homeostasis and synaptic function [48
]. Here, we demonstrated that LPS and IL-1β altered astrocyte morphology, characteristic of astrocytic reactivity, and this was controlled by FAB.
LPS is recognized by TLR4, which is highly present in the plasma membrane of microglia and astrocytes [49
], which activates several signal transduction pathways and, at the end, causes NF-κB activation and results in raised transcription of genes encoding proinflammatory cytokines, especially the IL-1 family of cytokines [50
], chemokines, and inducible enzymes that lead to neuroinflammation and consequent neuronal death. Once activated, NF-κB may regulate the expression of immune and inflammatory genes [51
]. Interestingly, FAB treatment caused an overall decrease in NF-κB and this was even more pronounced in microglia. It is known that in neurons, NF-κB promotes survival and plasticity, contrary to its effects in glial cells, which play an important role in promoting inflammation and leading to neuronal damage [52
Although previous studies [12
] have shown that FAB acts via estrogen signaling to improve the neuroprotective properties of microglia, the immunomodulatory effect of FAB on NF-κB expression in the neuroinflammation model induced by IL-1β, however, was not mediated by estrogen receptors, agonists of which have proved to play a pivotal role in modulating microglial inflammatory response [53
]. In addition, we demonstrated that neurodegeneration seen through Fluoro-jade B in LPS and IL-1β treatments was similar, as well as the neuroprotective effect of FAB, which suggests that FAB reduces neuronal death by a mechanism that seems to be related to the inhibition or control of the IL-1β signaling pathway.