A Natural Benzofuran from the Patagonic Aleurodiscus vitellinus Fungus has Potent Neuroprotective Properties on a Cellular Model of Amyloid-β Peptide Toxicity

Abstract

Alzheimer’s disease (AD) is characterized by amyloid plaques that form due to an increase in amyloid-β peptide (Aβ) aggregation. One strategy in the search of new treatments for AD focuses on compounds that decrease Aβ accumulation. Compounds containing a benzofuran ring have been described to play an important role in decreasing Aβ-induced toxicity; however, only synthetic benzofurans have been tested thus far. The aim of the present study was to examine the in vitro neuroprotective properties of fomannoxin (Fx), a natural benzofuran isolated from cultures of the Andean-Patagonian fungi Aleurodiscus vitellinus, and evaluate its effect on Aβ peptide. We tested the effect of Fx at a wide concentration range (10^–11–10^–4 M) in PC-12 cells, and found the compound did not alter cellular viability. Fx also showed a concentration-dependent effect on the Aβ-induced toxicity in PC12 cells, showing viability above 100% at 10^–6 M. We then measured the effect of Fx (10^–7–10^–5 M) on the frequency of cytosolic Ca²⁺ transients in rat hippocampal neurons at both acute and chronic (24 h) times. Acute incubation with Fx increased the frequency of cytosolic Ca²⁺ transients to values around 200%, whereas chronic incubation with Fx increased the frequency of Ca²⁺ transients. Finally, the Aβ-induced decrease in intracellular Ca²⁺ transients was prevented when Fx (10^–6 M) was co-incubated with Aβ (5×10^–6 M). The results suggest a potent neuroprotective effect of this naturally occurring benzofuran against Aβ peptide toxicity that could be mediated by an interference with it binding to plasma membrane, and lead Fx as new chemical entity to develop pharmacological tools against Aβ peptide neurotoxicity.

Keywords

Alzheimer’s disease Aβ peptide inhibitor benzofuran fomannoxin

INTRODUCTION

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by the accumulation of hyperphosphorylated tau protein and extracellular amyloid-β peptide (Aβ) [1, 2], which is the most prevalent cause of dementia [3]. Scientists are constantly turning their attention to nature, attempting to discover new molecules or entities that may show related biological activity [4]. In this sense, molecules such as polyphenols [5 –9], alkaloids [10 –14], and other compounds [15 –18] have shown interesting neuroactive properties that suggest they may have a relevant application on central nervous system pathologies, like AD.

In recent decades, diverse structures with interesting biological activities have been isolated from Basidiomycetes fungi. For example, Basidiomycetes of Russulales order produce benzofuran compounds [19 –23], which have been distinguished as the major group of biologically active heterocyclic compounds and recognized as pharmacologically relevant molecules [24]. Several studies indicate that compounds containing a benzofuran nucleus possess a broad range of biological activities, including antibacterial, antifungal, antioxidant, antitumoral, anti-inflammatory, anticonvulsant, and anti VIH [25 –28].

Interestingly, benzofurans have been found to be inhibitors of the agents involved in AD. Although several factors are involved in the development of AD, studies performed over decades have come to a consensus that the generation of toxic Aβ is important in the development of AD pathogenesis [29].

Currently, efforts to combat AD are focused on treatments that involve its symptomatology, and research has focused on the development of small molecules and peptides that modify Aβ aggregation [30]. Since 1999, benzofuran compounds have played an important role as inhibitors of Aβ fibril formation [31 –34]; however, the search for benzofuran compounds inhibiting Aβ peptides have been limited to chemical synthesis. On the other hand, secondary metabolites play a significant role in the development of new therapies for neurodegenerative diseases [35], which makes them an important source for highly selective candidate molecules.

Considering Russulales fungi are known to produce benzofuran compounds and the close correlation between the activity of benzofurans as inhibitors of fibril formation and their ability to bind to Aβ [32], it seems reasonable to test the bioactivity of a natural benzofuran compound from Russulale fungi against Aβ peptide. Chemical and antioxidant properties of Aleurodiscus vitellinus (Russulales, Basidiomycota), a native edible mushroom growing widely in the Andean-patagonian forests, have recently been shown to have a higher content of trehalose and monounsaturated oleic acid [36] demonstrating their nutritional and medicinal potential. Therefore, we chose to examine this species as a source of bioactive natural products and proceeded to isolate from mycelial cultures the most prevalent benzofuran compound, fomannoxin (Fx), which was previously reported as a toxin involved in the pathogenicity of the root rotting fungus Heterobasidion annosum [20]. Despite this information, Fx has never been tested as a molecule with pharmacological activity; thus, our study evaluated whether Fx acts as a neuroprotective compound in a cellular model of Aβ peptide toxicity.

MATERIALS AND METHODS

Fungal material

Samples of Alleurodiscus vitellinus (RM314) were collected during the 2012 autumn season from the Magallanes National Reserve located in Punta Arenas, XII region, Chile (53° 8′ 46.12′′ S, 71° 0′ 11.4′′ W) and stored in the Laboratory of Chemistry of Natural Products at the University of Concepción. Morphologic and microscopic identification of the collected specimens were performed according to the literature [37]. Mycelia were isolated from the fruiting body of the fungi, where the representative strain was cultivated in a solid YMG in vitro medium (20 g glucose, 10 g malt extract, 5 g yeast extract, and 10 g/L agar/distilled water). Cultures were incubated in a growth camera at 20°C. The strain was deposited and stored at 4°C in the fungal ceparium of the Laboratory of Chemistry of Natural Products at the University of Concepción

Culture of A. vitellinus RM314 mycelium

Mycelial cultures at 10 days of in vitro (DIV) growth were transferred into 0.5 L Erlenmeyer flasks containing 0.2 L of liquid YMG medium which were kept 10 days in an orbital shaker Rzr-2051 – Heidolph with constant stirring at 120 rpm. Submerged fermentation was carried out in 5 L Erlenmeyer flasks containing 2.8 L liquid YMG medium and then cultivated under the same conditions for 21 days.

Aleurodiscus vitellinus RM314 extract

Culture broth was separated by filtration from mycelia and subsequently extracted three times with equal volumes of hexane. Total hexanic extract was evaporated until dry at 35°C under reduced pressure (Rotary evaporator Büchi Model). The extract was then examined by TLC (Silica gel 60 F₂₅₆, Merck) using chloroform as the eluent, where preliminary presence of Fx (Rf = 0.52) [38] was observed under UV light (254 and 366 nm) and revealed with sulphuric acid (10%).

Compound isolation

Hexane total extract (100 mg) was dissolved in dichloromethane: methanol (1 : 3, v/v) and fractionated by Sephadex LH-20 chromatography (column 30×1.5 cm) using dichloromethane: methanol (1 : 3, v/v) as the eluent [39].

In order to detect and isolate the pure Fx compound, fractions were analyzed by HPLC. The HPLC system consisted of YL9111S HPLC equipped with a Diode Array Detector YL9160 PDA (Younglin Instrument Co. Ltd., Korea). Separation was performed on a KROMASIL^® 100-5 C18 column (250×10 mm) at room temperature using a mobile phase consisting of water (TFA 0.01%) with acetonitrile by an isocratic method of 40 : 60, respectively. Flow rate of the mobile phase was set at 5 mL/min for 15 min. Detection by DAD was performed by scanning from 254 to 310 nm. YL-Clarity version 4.0.3.797 Software was used for recording and processing chromatographic data. Identification of Fx was conducted by comparison of UV-spectra according to a previous study [38].

Also, fractions were analyzed by GC-MS on a Hewlett Packard gas chromatograph 5890A, coupled with Hewlett Packard 5975 mass spectrometer system equipped with a HP5- MS (30 m long, 0.25 mm id, and 0.25 μm film thickness). Temperature was programmed from 100°C to 275°C at a rate of 10° C/min with a 10-min hold. Helium was used as a carrier gas with a constant flow at 1.4 mL/min. Ionization voltage was 70 eV. Compound identification was performed compared to the obtained spectra reported in the literature.

Nuclear magnetic resonance

¹H and ¹³C NMR spectra were acquired on a Varian Unity 450 MHz spectrometer, equipped with a 5 mm magic-angle-spinning probe head optimized for use with liquid samples. Microtubes with 40 μL of sample volume were used. The structure was elucidated on the basis of spectroscopic analysis and chemical evidence.

Fomannoxin compound for biological assays

Dry Fx (100 mM) was resuspended in DMSO and used as a stock solution. For the following assays, extracts were subjected to 8 serial dilutions in DPBS (10^–9–10^–2 M), with the aim of reaching the concentrations of the experimental conditions (10^–11–10^–4 M).

Hippocampal cultures

Primary 18 days embryonic rat hippocampal neurons were cultured and plated at 300,000 cells/mL onto plates pre-treated with poly-L-lysine. Cultures were maintained in Minimal Essential Media (MEM) supplemented with horse serum (10%), DNAse (4 μg/mL) and L-glutamine (2 mM) during the first 24 h. Subsequently, the culture medium was replaced with MEM supplemented with horse serum (2%), fetal bovine serum (2%), and N3 (0.5%, a mixture of nutrient supplements). Neurons were maintained in a thermo-regulated incubator at 37°C with 5% CO₂. Neurons were used after 9 DIV incubation.

PC-12 cells

PC-12 cells were maintained in a DMEM culture medium supplemented with fetal bovine serum (5%), horse serum (5%), penicillin and streptomycin (1%). Cells were maintained in a thermo-regulated incubator at 37°C with 5% CO₂ and used after reaching ∼80% confluence.

Aggregation of Aβ_1 - 40 peptide and obtention of oligomers

Aβ_1 - 40 peptide (Peptide, Bogart, GA, USA) was reconstituted in DMSO at a concentration of 2.3 mM. An aliquot of 2 μL was taken from this stock and dissolved in sterile distilled water until reaching a concentration of 80 μM. This solution was subjected to vertical stirring (500 rpm) with a magnetic agitator for 2 h at room temperature. The peptide was used at a concentration of 0.5 μM.

Cell viability assay

The different concentrations of Fx were incubated for 24 h in 96-well microplates of PC-12 cells. To measure cellular viability, cellular cultures were incubated with MTT (1 mg/mL) during 30 min. Later, the precipitated MTT was dissolved using isopropanol for 15 min. The absorbance was measured in a microplate reader NOVOstar (BGM, Germany) at two wavelengths: 560 and 620 nm, and the difference was quantified using NovoStar Software for the different experimental conditions.

Microfluorimetry of cytosolic Ca²⁺

Hippocampal neurons were incubated with the fluorescent probe Fluo-4AM (5 μM) (Invitrogen, Carlsbad, CA, USA) in DMEM for 20 min at 37°C. Subsequently, a wash was performed with DPBS for 20 min with the purpose of removing excess Fluo-4AM. Afterwards, 2 washes were carried out with normal external solution (NES). Intracellular Ca²⁺ transients were recorded using an inverted fluorescent Nikon TE-2000 microscope (Tokyo, Japan) coupled to an EM-CCD iXon+ camera of 16 bit (Andor, Belfast, North Irland). Each image was recorded every 1 s with a time exposure of 200 ms. The results were analyzed using the software ImagingWorkbench 6.0 (INDEC Biosystems, Santa Clara, CA, USA).

Immunofluorescence

Hippocampal neurons of 10 DIV incubation were treated for 24 h with Aβ_1 - 40 (5×10^–7 M) and Fx (10^–6 and 10^–5 M). After, samples were washed with PBS 1X and fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. Then, non-specific blocking was done with 10% horse serum and permeability with 0.1% Triton X-100 for 5 min. To detect the Synaptic Vesicle glycoprotein 2 (SV2), we used a specific SV2 antibody (1 : 200, mouse, Hybridoma Bank), and to identify neuronal structure we used Microtubule Associated Protein 2 (MAP2) antibody (1 : 100, Rabbit, Santa Cruz) incubated for 1 h at room temperature. Subsequently, secondary antibodies Cy3 (1 : 200, anti-mouse, Jackson Immuno Research Laboratories) and Alexa Fluor (1 : 200, anti-rabbit, Abcam) were used to detect SV2 and MAP2, respectively, and were incubated for 45 min at room temperature. Cells were mounted with Dako Fluorescent Mounting Medium (Dako, USA) while the spatial localization of Aβ was determined using a fluorescent Aβ form (Aβ1-40 TAMRA-labeled, Anaspec) and correlated with MAP2. Images were obtained in the CMA Bio-Bio (Center for Advanced Microscopy, U. de Concepcion), using Epifluorescence microscopy (Nikon Eclipse Ti-U) and the software Nis Elements 4.20.01 (Nikon). Microphotographs were analyzed using the software MacBiophotonics ImageJ.

Thioflavin T binding assay

Aβ aggregation was performed on a 96-well plate in the presence or absence of FX (1–10 μM) in PBS buffer with 20 μM of Thioflavin T (ThT, Sigma-Aldrich). The aggregation process was followed by fluorescence measurements of the Thioflavin T-Aβ complex (ex: 440 nm, em: 485 nm) every 3 min for 4 h. The plate was kept at room temperature with an orbital agitation of 500 rpm.

Statistical analysis

Data represented in the figures are expressed as mean±SEM and as percentages with respect to the control without treatment. Statistical analysis was performed using one way ANOVA. The following results were considered to be statistically significant: ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001 versus control; ^#p < 0.05, ^# #p < 0.01, ^# # #p < 0.001 versus Aβ. All analyses were performed using the software GraphPad Prism (GraphPad Prism, CA, USA).

RESULTS

Chemistry and analytical characterization

Fomannoxin (Fx) benzofuran was isolated and detected by HPLC-DAD showing a Fx peak at 7.81 min (Fig. 1A), indicating a high purity of Fx, that was near 95.5%. Additionally, to ensure molecular identity, we used two analytical approaches: first the UV spectra analyses (Fig. 1B) where we observed maximum absorbance at three wavelengths: 202, 230, and 297 nm; and second, GC-MS to confirm the identity and purity of Fx. Benzofuran was detected at 11.13 min (Fig. 2A), indicating a molecular mass of 188 Da coinciding with the secondary fungal metabolite (2 S)-2-Isopropenyl-2,3-dihydro-1-benzofuran-5-carbaldehyde with a molecular formula of C₁₂H₁₂O₂ [38].

Fig.1

A) HPLC-DAD chromatogram of fomannoxin detected at 7.81 min (Fx). B) UV Spectra of fomannoxin compound.

Fig.2

A) Chromatogram obtained by GC-MS. B) Mass spectrum of Fomannoxin ((2S)-2 Isopropenyl-2,3-dihydro-1-benzofuran-5-carbaldehyd).

In order to complete Fx characterization, nuclear magnetic resonance was performed (Fig. 2B). The ¹H and ¹³C NMR spectra were in full agreement with previously published parameters for benzofurans and Fx [22, 38]. ¹H NMR (CDCl₃, 450 MHz) δ 1.78 (s, 3H, 1’ -CH3), 3.1 (dd, IH, J = 15.9, 7.8 Hz, Ha), 3.4 (dd, IH, J = 15.9, 9.6 Hz, Hb), 4.97 (bs, 1H, Hc), 5.1 (bs, IH, Hd), 5.3 (t, 1H, J = 8.8 Hz, 2-H), 6.9 (d, 1 H, J = 8.1 Hz, 7-H), 7.7 (d, 1H, J = 8.2 Hz, 6-H), 7.74 (s, 1H, H-4), 9.85 (s, 1H, -CHO); ¹³C NMR(CDCl₃, 450 MHz) δ 17.05(C3’), 33.71(C3), 87.19(C2), 109.5(C7), 112,8(C2’), 125.8(C4), 128.4(C-3a),130.7 (C-5), 133.2(C-6), 143.3(C-1’), 165.5 (C7a), 190.6(C-1”). Taken together, we can confirm that the biological effects observed on the neurotoxicity models can be attributable to Fx.

Biological assays

Evaluation of intrinsic Fx toxicity and its effects on Aβ-induced toxicity in PC12 cells

In order to evaluate the eventual toxic properties of Fx, PC-12 cells were incubated for 24 h with a wide range of Fx concentrations (10^–11 to 10^–4 M), and cell viability was evaluated. A mitochondrial uncoupler FCCP (10^–5 M) was used as a positive control of toxicity. Results from MTT experiments demonstrated that Fx did not reduce the cellular viability of PC12 cells with respect to control, showing minimum and maximum viability values in the range of 106±2% (10^–4 M) and 126±4% (10^–5 M), respectively (Fig. 3A), suggesting that Fx does not have toxic effect on PC12 cells. Some statistical increases in viability could be attributed to a normal increase in cell numbers despite the use of standardized cell densities at the beginning of the experiment.

Fig.3

Effect of fomannoxin on PC12 cellular viability. A) Chronic treatment of 24 h with increasing concentrations of fomannoxin (10^–11 to 10^–4 M) to evaluate cytotoxicity using the MTT technique. B) Effects of fomannoxin on the reduction of cellular viability induced by Aβ oligomer peptides (0.5 μM) was prevented by a concentration-dependent effect with different concentrations of fomannoxin. FCCP (10^–5 M) was used as a positive control for toxicity. Values are expressed as percent of control without treatment (n = 3; n = 9).

According to previous bibliographic evidences, that suggest a preventive potential role on the Aβ toxicity, exerted by molecules benzofuran-like (see Introduction), we used the same experimental approach described in Fig. 3A, to measure the potential effects of Fx on PC12 cell viability in cells treated chronically with Aβ (0.5 μM, 24 h) in the same range of concentrations previously evaluated. PC-12 cells exposed to Aβ demonstrated a reduced viability of about 40±2% (Fig. 3B) with respect to control conditions (untreated control), while co-incubation with Fx induced a recovery of viability by preventing the cytotoxicity of Aβ in a concentration-dependent manner and reaching values similar to control when 10^–5 M was used (109±1%); however, this preventive effect was observed from low concentrations that was able to prevent an important part of Aβ toxicity (Fig. 3B). It is relevant to note that FCCP showed the same toxic behavior in both experimental approaches, validating the sensitivity of this methodology.

Preservation of synaptic function in hippocampal neurons by Fx

We and others have demonstrated that synaptic failure is one of the key events associated with early Aβ neurotoxicity [5 , 41] and this network disconnection represents an important step in neuronal death. The functional hippocampal neuronal network is characterized by spontaneous electrophysiological activity accompanied by a transient increase in cytosolic Ca²⁺, showed as spikes that are important tools for the evaluation of synaptic connectivity.

With the purpose of evaluating the effect of Fx on neural network activity, we first performed acute treatments of Fx at different concentrations (10^–7–10^–5 M). Using microfluorimetry and Fluo4-AM dye, we observed that Fx significantly increased the frequency of cytosolic Ca²⁺ transients with respect to control (Fig. 4A), obtaining values of 327±9% (10^–7 M), 204±5% (10^–6 M) and 190±5% (10^–5 M) (Fig. 4B) without affecting neuronal viability (data not shown). These results are interesting because far from demonstrating toxicity, the effect of Fx appears to be more similar to activation observed under high frequency stimulation that induces long term potentiation (LTP) in hippocampal slices. Additionally, lower Fx concentrations (10^–7 M) are shown to have high potency to induce an increase in Ca²⁺ transients frequency, suggesting an enhanced synaptic activity under Fx presence (Fig. 4B). This effect cannot to be observed in PC-12 cells, because this cell line is unable to make synaptic connections like neurons, but this enhanced Ca²⁺ activity could be related to the increase observed in PC-12 cell viability (Fig. 3A). Under chronic conditions, 24 h incubation of neurons with the same Fx concentration range (10^–7–10^–5 M, Fig. 5A), the network exhibited a different range of effects on the frequency of Ca²⁺ transients; the most striking effect was observed when hippocampal neurons were treated with Fx 10^–6 M, there was an increase in Ca²⁺ transients to 147±5% as compared to control indicating a beneficial effect, whereas a lower Fx 10^–7 M decreased Ca²⁺ transients to about 19±2% with respect the control. Fx at 10^–5 M, on the other hand, abolished any synaptic activity or increment in the frequency of Ca²⁺ transients in hippocampal neurons, suggesting a specific neuronal toxicity at high Fx concentrations (Fig. 5B). Therefore, Fx 10^–6 M was selected as the concentration to test neuroprotective properties on in vitro neuronal toxicity in an Aβ model.

Fig.4

A) Fomannoxin increases frequency of cytosolic Ca²⁺ transients in rat hippocampal neurons with acute treatment. Original traces of cytosolic Ca²⁺ transients showing neuronal network activity. B) Quantification of the events shown in A, after acute treatment with different concentrations of fomannoxin (10^–7, 10^–6, and 10^–5 M) (n = 3; n > 40).

Fig.5

Effect of fomannoxin on the frequency of cytosolic Ca²⁺ transients in rat hippocampal neurons at chronic times. A) Original traces of cytosolic Ca²⁺ transients showing neuronal network activity. B) Quantification of the events shown in (A) after 24 h of treatment with fomannoxin (10^–7, 10^–6, and 10^–5 M) (n = 3; n > 40).

According to our previous results, Aβ at chronic times (24 h) induced a synaptic silencing reflected as a reduction in the frequency of intracellular Ca²⁺ transients and electrophysiological silencing of synaptic activity [7 , 41]. We confirm, that in our hands the hippocampal neurons incubated with the Aβ peptide (0.5 μM, 24 h) showed a significant reduction in intracellular Ca²⁺ transients to about 47±4% with respect to control (Fig. 6A). This decrease was prevented when Fx (10^–6 M) was co-incubated with Aβ, demonstrating a protective effect on the neurons with values near control condition (98±2%) (Fig. 6B), suggesting that Fx could prevent Aβ toxicity at low concentrations. It is worth mentioning that the benzofuran alone, as well in Fig. 5, showed significantly increased frequency of cytosolic Ca²⁺ transients.

Fig.6

Fomannoxin prevents reduction in the frequency of intracellular Ca²⁺ transients induced by oligomers of Aβ_1 - 40 peptide in rat hippocampal neurons. A) Original traces of cytosolic Ca²⁺ transients showing neuronal network activity. B) Quantification of the events shown in (A) after 24 h of treatment with Aβ_1 - 40 (0.5 μM) and fomannoxin (10^–6 M) (n = 3; n > 40).

Using the same experimental protocol to evaluate Ca²⁺ signals, we evaluated if Fx could prevent the cytotoxic effects of Aβ on neural synaptic organization in hippocampal cell cultures. The neurons were labeled with antibodies to SV-2, a key presynaptic protein, and MAP-2, a structural cytoskeletal protein, and changes in SV-2 inmunoreactivity were evaluated via fluorescence microscopy. We found that Fx (10^–6 and 10^–5 M) prevented the decrease in SV-2 puncta induced by chronic incubation with Aβ (95% and 104%, respectively), reinforcing the idea that Fx can protect neurons and network against synaptic failure induced by toxic and degenerative stimulus like the Aβ peptide (Fig. 7).

Fig.7

Fomannoxin prevents synaptic network alterations in hippocampal neurons. The panels show fluorescence images of hippocampal neurons under the following experimental conditions: 1) control; 2) Fx (1 μM, 24 h); 3) Fx (10 μM, 24 h); 4) Aβ (0.5 μM, 24 h); 5) Fx (1 μM) plus Aβ (0.5 μM); and 6) Fx (10 μM) plus Aβ (0.5 μM) during 24 h. Fluorescent SV-2 signal is in red and MAP2 is in blue. (n = 3; n > 6).

In order to search the potential action mechanism of Fx, we use fluorescent Aβ (Aβ FAM), to evaluate the Aβ binding to the plasma membrane (Fig. 8A), where Aβ were distributed widely over the neuronal shapes stained with specific MAP-2 antibody. Surprisingly, the presence of Fx (10^–5 M), potently reduced Aβ binding to neurons, suggesting that the neuroprotective properties of Fx could be related with its capacity to block the peptide binding to the plasma membrane (Fig. 8). Another possibility for the neuroprotective effects of Fx could be related to its capacity to interfere with Aβ aggregation. To test this possibility, we used thioflavin assay to check the aggregation kinetic of Aβ peptide alone and to compare this process with the presence of Fx (10^–6–10^–5, Fig. 9); the aggregation Aβ kinetics were unaltered in the presence of different concentrations of Fx, which reinforces the idea that Fx avoid the binding or insertion of Aβ peptide on the plasma membrane, but this preventive action is not related with a change in peptide aggregations status.

Fig.8

Fomannoxin inhibit the binding of fluorescent Aβ to hippocampal neurons. The upper panels show hippocampal neurons treated with Aβ-FAM (1 μM, 24 h, in green) and stained with MAP2 (in white) and SV2 (in red). The lower panels show the same experimental conditions, in presence of Fx (10 μM, 24 h) co-incubated with Aβ-FAM (n = 3; n > 6).

Fig.9

Effect of fomanoxin in Aβ aggregation. Aβ aggregation (80 μM) was measured by Thioflavin T (ThT 20 μM) fluorescence every 3 min for 350 min at 37°C, in absence of fomanoxin (red) or in presence of 1 μM (green) or 10 μM (blue). As observed in the graph, neither concentration has an effect in the aggregation properties of Aβ, suggesting that the protective properties are mediated by other mechanisms. The graph also shows the effect of fomanoxin over the background fluorescence of ThT (n = 5).

DISCUSSION

Benzofuran derivatives represent an interesting class of heterocyclic compounds based on oxygen that have a wide distribution in nature and can be recognized by their broad range of biological activities. Recent studies have demonstrated the neuroprotective effect of compounds having benzofuran rings. The first study of benzofurans as inhibitors of Aβ fibril formation tested synthetic compounds at micromolar concentrations in cultures of IMR32 human neuroblastoma cells, showing the inhibition mechanism affected binding to specific sites on Aβ, particularly regarding benzofurans with substitutions at the second and third positions [31]. Since then, diverse research groups have synthesized benzofurans with substitutions at these positions and evaluated the inhibition of Aβ fibril formation [3 , 38]. In our study, Fx prevented the Abeta -induced toxicity in PC-12 cells, and increased neural network activity in hippocampal cells. These results align with previous reports since the Fx used in our study has an isopropenyl radical at its second carbon (Fig. 2). Furthermore, Fx exerted some positive effects on neuronal activity suggesting that this benzofuran has a direct effect on neuronal function which needs to be further characterized in future experiments; however, the present study showed Fx has the ability to block binding to the plasma membrane, instead of interfering with the aggregation process. In our experimental conditions, this effect appears to be more relevant than reported effects that have indicated synthetic benzofuran inhibits aggregation of Aβ, and this could be related with the concentrations used in the literature, where the most potent synthetic benzofuran that can inhibit the aggregation of Aβ peptide, is known to require three times higher concentrations (about 30 μM) than were used in this work [32]. Then, we cannot discard that Fx could have both effects at higher concentrations, but we contend that the blocking of binding at lower concentrations represents more interesting and selective properties. Thus, Fx could be an important pharmacological tool to help understand the cytotoxic mechanism of action of aggregating proteins like Aβ and alpha-synuclein that are considered to be responsible for neuron death and neurodegenerative pathologies, particularly Aβ in AD.

Our data demonstrates that Fx is more potent than synthetic compounds tested, and could be associated to groups present in the structure at the second and third carbon, substitutions which have never been studied [31 –33]. Therefore, we propose that the Fx benzofuran with substitutions at the second and third position that was isolated, purified and chemically characterized by our laboratory is the most potent compound studied thus far; and futures works will be done to discriminate the precise mechanism by which Fx blocks the Aβ binding to plasma membrane.

Our data shows that Fx at 10^–6 M co-incubated with Aβ recuperated the Ca⁺² transient signals to about 44±4% in hippocampal cells with respect to the Aβ condition. However, we also demonstrated that the use of Fx alone in hippocampal cells significantly increased frequency of cytosolic Ca²⁺ transients in both acute and chronic treatments, suggesting that the neuroprotective effect against Aβ may be due not only blockage of their binding to plasma membrane, but also to intrinsic effects in the neuron that could be associated to glutamate receptor, especially on NMDA receptors. These effects need to be studied more in-depth, since it has been reported, for example, that benzofuran compounds also act as inhibitors of acetylcholinesterase (AChE) [42, 43], and it is well known that AChE could promote Aβ fibril and plaque formation [43] since AChE is an important factor in the aggregation and deposition of Aβ. Thus, we also hypothesize that the intrinsic neuroprotective effect of Fx against Aβ could be due to reduction in AChE levels in hippocampal cells or on the modulation of NMDA receptors. Studies of structure-activity relationships reported for this type of compound suggest that the furan ring, the presence of ester groups, and the length of the hydrocarbon side chain are key structural features contributing to inhibitory potency of benzofuran derivatives against AChE or eventually on allosteric modulation of glutamate receptors.

Conclusion

We have isolated and identified the benzofuran fomannoxin from Aleurodiscus vitellinus that demonstrates a significant neuroprotective effect against Aβ toxicity by a mechanism that blocks Aβ binding, and promotes an acute enhancement in synaptic activity and reinforces the network function (LTP-like phenomena). Additionally, we cannot discard the possibility that fomannoxin (at high concentration) modulates the aggregation processes of the Aβ peptide and/or other aggregating proteins. The structural characteristics that contribute to the effects of Fx should be examined in future studies in order to develop new chemical entities as therapeutic tools to inhibit the toxic effects of Aβ peptide or other proteins.

Footnotes

ACKNOWLEDGMENTS

We thank Ms. Laurie Aguayo for editing the manuscript and Ms. Ixia Cid for technical assistance. This work was supported by FONDECYT 1151028 (CP); FONDECYT 1161078 (JF) and 1130747 (JF); CONICYT-MINCYT PCCI 130036 (MG).

Authors’ disclosures available online ().

References

Hardy

, Selkoe

(2002) The amyloid hypothesis of Alzheimer’s disease: Progress and problems on the road to therapeutics. Science 297, 353–356.

Mazanetz

, Fischer

(2007) Untangling tau hyperphosphorylation in drug design for neurodegenerative diseases. Nat Rev Drug Discov 6, 464–479.

van der Flier

, Scheltens

(2005) Epidemiology and risk factors of dementia. J Neurol Neurosurg Psychiatry 76, v2–v7.

Aguayo

, Guzman

, Perez

, Aguayo

, Silva

, Becerra

, Fuentealba

(2006) Historical and current perspectives of neuroactive compounds derived from Latin America. Mini Rev Med Chem 6, 997–1008.

Fuentealba

, Dibarrart

, Saez-Orellana

, Fuentes-Fuentes

, Oyanedel

, Guzmán

, Perez

, Becerra

, Aguayo

(2012) Synaptic silencing and plasma membrane dyshomeostasis induced by amyloid-β peptide are prevented by Aristotelia chilensis enriched extract. J Alzheimers Dis 31, 10.

Garcia

, Nanni

, Figueira

, Ivanov

, McDougall

, Stewart

, Ferreira

, Pinto

, Silva

RFM

, Brites

, Santos

(2017) Bioaccessible (poly)phenol metabolites from raspberry protect neural cells from oxidative stress and attenuate microglia activation. Food Chem 215, 274–283.

Han

Y-S

, Zheng

W-H

, Bastianetto

, Chabot

J-G

, Quirion

(2004) Neuroprotective effects of resveratrol against β-amyloid-induced neurotoxicity in rat hippocampalneurons: Involvement of protein kinase C. Br J Pharmacol 141, 997–1005.

Rivière

, Richard

, Vitrac

, Mérillon

J-M

, Valls

, Monti

J-P

(2008) New polyphenols active on β-amyloid aggregation. Bioorg Med Chem Lett 8, 828–831.

Dal-Pan

, Dudonné

, Bourassa

, Bourdoulous

, Tremblay

, Desjardins

, Calon

(2017) Cognitive-enhancing effects of a polyphenols-rich extract from fruits without changes in neuropathology in an animal model of Alzheimer’s disease. J Alzheimers Dis 55, 20.

10.

Araya

, Ramírez

, Figueroa-Aroca

, Sotes

, Pérez

, Becerra

, Saez-Orellana

, Guzmán

, Aguayo

, Fuentealba

(2014) Modulation of neuronal nicotinic receptor by quinolizidine alkaloids causes neuroprotection on a cellular Alzheimer model. J Alzheimers Dis 42, 12.

11.

Chlebek

, De Simone

, Hošt’álková

, Opletal

, Pérez

, Havlíková

, Cahlíková

, Andrisano

(2016) Application of BACE1 immobilized enzyme reactor for the characterization of multifunctional alkaloids from Corydalis cava (Fumariaceae) asAlzheimer’s disease targets. Fitoterapia 109, 241–247.

12.

Wan Othman

WNN

, Liew

, Khaw

, Murugaiyah

, Litaudon

, Awang

(2016) Cholinesterase inhibitory activity of isoquinoline alkaloids from three Cryptocarya species (Lauraceae). Bioorg Med Chem 24, 4464–4469.

13.

Winblad

, Giacobini

, Frölich

, Friedhoff

, Bruinsma

, Becker

, Greig

(2010) Phenserine efficacy in Alzheimer’s disease. J Alzheimers Dis 22, 7.

14.

Zhang

, Oh

, Kim

, Houghton

, Whang

(2013) Epimediphine, a novel alkaloid from Epimedium koreanum inhibits acetylcholinesterase. Nat Prod Res 27, 1067–1074.

15.

Fuentealba

, Muñoz

, Yévenes

, Moraga-Cid

, Pérez

, Guzmán

, Rigo

, Aguayo

(2011) Potentiation and inhibition of glycine receptors by tutin. Neuropharmacology 60, 453–459.

16.

Hoi

, Ho

, Baum

, Chow

AHL

(2010) Neuroprotective effect of honokiol and magnolol, compounds from Magnolia officinalis, on beta-amyloid-induced toxicity in PC12 cells. Phytother Res 24, 1538–1542.

17.

Lemes

LFN

, de Andrade Ramos

, de Oliveira

, da Silva

FMR

, de Castro Couto

, da Silva Boni

, Guimarães

MJR

, Souza

INO

, Bartolini

, Andrisano

, do Nascimento Nogueira

, Silveira

, Brand

, Soukup

, Korábečný

, Romeiro

, Castro

, Bolognesi

, Romeiro

LAS

(2016) Cardanol-derived AChE inhibitors: Towards the development of dual binding derivatives for Alzheimer’s disease. Eur J Med Chem 108, 687–700.

18.

Somani

, Kulkarni

, Shinde

, Shelke

, Laddha

, Sathaye

(2015) In vitro acetylcholinesterase inhibition by psoralen usingmolecular docking and enzymatic studies. J Pharm Bioallied Sci 7, 32–36.

19.

Arnone

, Assante

, Nasini

, Vajna de Pava

(1992) Isolation and structure elucidation of dihydrobenzofurans isolated from Laurilia taxodii|. Gazz Chim Ital 122, 3.

20.

Donnelly

DMX

, Fukuda

, Kouno

, Martin

, O’Reilly

(1988) Dihydrobenzofurans from Heterobasidion annosum. Phytochemistry 27, 2709–2713.

21.

Hansson

, Menkis

, Olson

, Stenlid

, Broberg

, Karlsson

(2012) Biosynthesis of fomannoxin in the root rotting pathogen Heterobasidion occidentale. Phytochemistry 84, 31–39.

22.

Hansson

, Wubshet

, Olson

, Karlsson

, Staerk

, Broberg

(2014) Secondary metabolite comparison of the species within the Heterobasidion annosum s.l. complex. Phytochemistry 108, 243–251.

23.

Hong-Jun

, Ze-Jun

, Chuan-Xing

, Ji-Kai

(2007) Dihydrobenzofurans from culture broth of the basidiomycete Bondarzewia berkeyeli. Acta Botanica Yunnanica 29, 3.

24.

Nevagi

, Dighe

(2015) Biological and medicinal significance of benzofuran. Eur J Med Chem 97, 561–581.

25.

Dawood

, Abdel-Gawad

, Rageb

, Ellithey

, Mohamed

(2006) Synthesis, anticonvulsant, and anti-inflammatory evaluation of some new benzotriazole and benzofuran-based heterocycles. Bioorg Med Chem 14, 3672–3680.

26.

Masubuchi

, Ebiike

, Kawasaki

K-i

, Sogabe

, Morikami

, Shiratori

, Tsujii

, Fujii

, Sakata

, Hayase

, Shindoh

, Aoki

, Ohtsuka

, Shimma

(2003) Synthesis and biological activities of benzofuran antifungal agents targeting fungal N-myristoyltransferase. Bioorg Med Chem 11, 4463–4478.

27.

Randhavane

, Karale

(2009) Synthesis and biological screening of some fluorinated dibenzofuran containing 3-chlorochromones and benzothiazepines. J Heterocycl Chem 46, 732–736.

28.

Rida

, El-Hawash

SAM

, Fahmy

HTY

, Hazzaa

, El-Meligy

MMM

(2006) Synthesis of novel benzofuran and related benzimidazole derivatives for evaluation of in vitro anti-HIV-1, anticancer and antimicrobial activities. Arch Pharm Res 29, 826–833.

29.

Sengupta

, Nilson

, Kayed

(2016) The role of amyloid-β oligomers in toxicity, propagation, and immunotherapy. EBioMedicine 6, 42–49.

30.

Turner

, Chastain

, Park

, Moss

, Servoss

(2017) Modulating amyloid-β aggregation: The effects of peptoid side chain placement and chirality. Bioorg Med Chem 25, 20–26.

31.

Allsop

, Gibson

, Martin

, Moore

, Turnbull

, Twyman

(2001) 3-p-toluoyl-2-[4′-(3-diethylaminopropoxy)-phenyl]-benzofuran and 2-[4′-(3-diethylaminopropoxy)-phenyl]-benzofuran do not act as surfactants or micelles when inhibiting the aggregation of β-amyloid peptide. Bioorg Med Chem Lett 11, 255–257.

32.

Howlett

, Perry

, Godfrey

, Swatton

, Jennings

, Spitzfaden

, Wadsworth

, Wood

, Markwell

(1999) Inhibition of fibril formation in β-amyloid peptide by a novel series of benzofurans. Biochem J 340, 283–289.

33.

Rizzo

, Rivière

, Piazzi

, Bisi

, Gobbi

, Bartolini

, Andrisano

, Morroni

, Tarozzi

, Monti

J-P

, Rampa

(2008) Benzofuran-based hybrid compounds for the inhibition of cholinesterase activity, β amyloid aggregation, and Aβ neurotoxicity. J Med Chem 51, 2883–2886.

34.

Sashidhara

, Modukuri

, Jadiya

, Dodda

, Kumar

, Sridhar

, Kumar

, Haque

, Siddiqi

, Nazir

(2014) Benzofuran–chalcone hybrids as potential multifunctional agents against Alzheimer’s disease: Synthesis and in vivo studies with transgenic Caenorhabditis elegans. ChemMedChem 9, 2671–2684.

35.

Joyner

, Cichewicz

(2011) Bringing natural products into the fold - exploring the therapeutic lead potential of secondary metabolites for the treatment of protein-misfolding-related neurodegenerative diseases. Nat Prod Rep 28, 26–47.

36.

Toledo

, Barroetaveña

, Fernandes

, Barros

, Ferreira

(2016) Chemical and antioxidant properties of wild edible mushrooms from native Nothofagus spp. forest, Argentina. Molecules 21, 1201.

37.

Gorjón

, Greslebin

, Rajchenberg

(2013) The genus Aleurodiscus s.l. (Stereaceae, Russulales) in the Patagonian Andes. Mycol Prog 12, 91–108.

38.

Heslin

, Stuart

, Murchú

PÓ

, Donnelly

DMX

(1983) Fomannoxin, a phytotoxic metabolite of Fomes annosus: In vitro production, host toxicity and isolation from naturally infected Sitka spruce heartwood. Eur J Forest Pathol 13, 11–23.

39.

Horlacher

, Nachtigall

, Schulz

, Süssmuth

, Hampp

, Fiedler

H-P

, Schrey

(2013) Biotransformation of the fungal phytotoxin fomannoxin by soil streptomycetes. J Chem Ecol 39, 931–941.

40.

Sepulveda

, Parodi

, Peoples

, Opazo

, Aguayo

(2010) Synaptotoxicity of Alzheimer beta amyloid can be explained by its membrane perforating property. PLoS One 5, e11820.

41.

Parodi

, Sepulveda

, Roa

, Opazo

, Inestrosa

, Aguayo

(2010) Beta-amyloid causes depletion of synaptic vesicles leading to neurotransmission failure. J Biol Chem 285, 2506–2514.

42.

Liu

, Dumontet

, Simonin

, Iorga

, Guerineau

, Litaudon

, Nguyen

, Gueritte

(2011) Benzofurans from Styrax agrestis as acetylcholinesterase inhibitor: structure-activity relationships and molecular modeling studies. J Nat Prod 74(10), 2081–2088.

43.

Rees

, Hammond

, Soreq

, Youkin

, Brimijoin

(2003) Acetylcholinesterase promotes beta-amyloid plaques in cerebral cortex. Neurobiol Aging 24(6), 777–787.

A Natural Benzofuran from the Patagonic Aleurodiscus vitellinus Fungus has Potent Neuroprotective Properties on a Cellular Model of Amyloid-β Peptide Toxicity

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Fungal material

Culture of A. vitellinus RM314 mycelium

Aleurodiscus vitellinus RM314 extract

Compound isolation

Nuclear magnetic resonance

Fomannoxin compound for biological assays

Hippocampal cultures

PC-12 cells

Aggregation of Aβ1 - 40 peptide and obtention of oligomers

Cell viability assay

Microfluorimetry of cytosolic Ca2+

Immunofluorescence

Thioflavin T binding assay

Statistical analysis

RESULTS

Chemistry and analytical characterization

Biological assays

Evaluation of intrinsic Fx toxicity and its effects on Aβ-induced toxicity in PC12 cells

Preservation of synaptic function in hippocampal neurons by Fx

DISCUSSION

Conclusion

Footnotes

ACKNOWLEDGMENTS

References

Aggregation of Aβ_1 - 40 peptide and obtention of oligomers

Microfluorimetry of cytosolic Ca²⁺