The following antibodies were used: Mouse Anti-IIIβ-tubulin (MMS-435P, Covance), Rabbit mAb Anti-P-Akt (Ser473; #4060, Cell Signaling), Goat Anti-Akt (C-20; sc-1618, Santa Cruz), Donkey anti-Mouse IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (A-21202, Thermo Fisher), Swine Anti-Rabbit Immunoglobulins/HRP (P0217, Dako), Rabbit Anti-Goat Immunoglobulins/HRP (P0449, Dako).

The following drugs and reagents were used: Poly-D-Lysine (P7280, Sigma), rat tail collagen Type I, Rat Tail (354236, Corning), Nystatin dihydrate (N4014, Sigma), DMSO (D5879, Sigma), Methyl-β-cyclodextrin (C4555, Sigma), Phalloidin—TRITC (P1951, Sigma), NG-Monomethyl-L-arginine, monoacetate salt (L-NMMA; ab120137, Abcam), diamino-fluorescein Diacetate (DAF-FM DA; D-23844, Molecular Probes), CellTracker™ RedCMTPX Dye (C34552, Thermo Fisher), Complete Protease Inhibitor Cocktail Tablets (11697498001, Roche), MK-2206-2HCl (A10003, AdooQ Bioscience).

Hippocampal and forebrain primary cell cultures and explants were obtained from E16-E17 (embryonic day 16–17) mice embryos. Pregnant CD1 dams were sacrificed by cervical dislocation and the fetuses were collected and decapitated. Brain tissues were maintained during the dissection procedure constantly submerged in ice-cold 0.3% glucose-phosphate-buffered saline (PBS) solution. For primary cell cultures, hippocampi or forebrains were isolated and trypsinized for 6 min at 37°C. Trypsin was neutralized with FBS, the tissues were incubated with DNase I for 10 min at 37°C, and then they were mechanically dissociated by gentle trituration. The neurons were centrifuged at 800 rpm for 5 min, resuspended and plated in culture glasses pre-coated with 0.5 mg/ml poly-D-lysine. The composition of the neuronal culture medium was Neurobasal (w/o L-glutamine, w/ Phenol Red; 21103-049, GIBCO), 1% penicillin/streptomycin (15140-122, GIBCO), 1% Glutamine (25030-024, GIBCO) and 2% B27 (17504-044, GIBCO). Explants were obtained from dissected hippocampi, plated in 15.6 mm dishes pre-coated culture glasses with 0.5 mg/ml poly-D-lysine (P7280, Sigma) and 0.03 mg/ml collagen (354236, Corning) with neuronal culture medium. Explants were cultured for 3 DIV (3 days in vitro) in the experiments of axon extension, or were cultured for 7 or 14 DIV in the experiments of axon regeneration. After 7 or 14 days, axotomy was performed using a hypodermic needle (302200, BD Microlance) to cut the axons close to the explant body (Finn et al., 2000). Axotomized explants were collected using a pipette and moved to a new dish, where they will be immersed in a collagen matrix. When collagen was coagulated, culture media was applied (Lumsden and Davies, 1986) and explants were kept in culture for 3 more days with the corresponding treatments.

For acute treatment experiments on dissociated forebrain and hippocampal neurons to measure growth cone size, filopodia density, Akt phosphorylation, and NO formation, cells were incubated with Nystatin at the following concentrations: 2.5 μM, 10 μM or 25 μM during 30 min. DMSO was used as vehicle control condition, applied to match the same volume of Nystatin used.

For chronic treatment experiments on hippocampal explants and dissociated neurons to measure axonal extension and regeneration, the samples were incubated with 2.5 μM Nystatin or DMSO for 3 days. Nystatin was added to the culture medium immediately after plating dissected neurons (experiments of axon extension) or after axotomy (experiments of axon regeneration).

Primary hippocampal neurons were obtained as described above. After 3 DIV, neurons were pre-incubated with 100 μM L-NMMA or control medium for 1 h at 37°C. Immediately after, the pre-incubation medium was removed and neuronal NO was labeled by incubating with 5 μM DAF-FM supplemented with 100 μM L-NMMA or control medium for 30 additional minutes at 37°C. After three washes with Neurobasal, neurons were further incubated with 10 μM CellTracker™ RedCMTPX Dye in addition to 2.5 μM Nystatin or 0.5 μM MβCD supplemented with 100 μM L-NMMA or control medium for 30 min at 37°C. CellTracker was used as a marker of the surface of the neurons.

Hippocampal explants were cultured with 100 μM L-NMMA or control medium for 2 h immediately before axotomy was performed. Explants were then returned to their culture medium for three additional days in the presence of 2.5 μM Nystatin or DMSO, supplemented with or without L-NMMA.

Primary hippocampal neurons were obtained as described above. After 3 DIV, neurons were pre-incubated with 2 μM MK-2206-2HCl or control medium for 4 h at 37°C. Immediately after, the pre-incubation medium was removed and neurons were incubated with 2.5, 10, 25 μM Nystatin or control medium supplemented with or without 2 μM MK-2206-2HCl for 30 min at 37°C.

Hippocampal dissociated cultures were fixed with 4% PFA in PBS for 10 min at room temperature (RT), permeabilized with 0.1% PBS-Triton for 10 min. To detect the actin cytoskeleton, neurons were stained with a solution of 1 μg/ml phalloidin-TRITC in PBS for 30 min, rinsed with PBS and mounted in Mowiol. To detect P-Akt, Akt or tubulin, neurons were incubated with a blocking solution, 10% normal horse serum (NHS) in TBS, for 2 h and with primary antibody diluted in blocking solution for 2 h. Then, neurons were washed and incubated with secondary antibody in blocking solution for 1 h rinsed with TPBS and mounted in Mowiol. To detect cholesterol, neurons were fixed with a solution of 0.12 mM sucrose in 4% PFA in PBS for 15 min at RT. Then, neurons were stained with a freshly prepared solution of 0.05mg/ml filipin in PBS for 90 min, rinsed with PBS, fixed again with 0.12 mM sucrose in 4% PFA for 20 min and mounted in Mowiol (Gu et al., 1997; Feng et al., 2003).

Explants were fixed with a solution of 4% PFA in PBS or 30 min at RT. Then, the explants were rinsed with PBS and permeabilized with a solution of 0.5% Triton X-100 in PBS for 30 min. Explants were then incubated with blocking solution, NHS 10% in PBS, for 2 h. After blocking, explants were incubated overnight at 4°C with the primary antibody diluted in blocking solution. Explants were then washed three times with PBS and incubated with the respective secondary antibodies diluted in blocking solution for 2 h at RT. Finally, explants were washed three times in PBS and mounted in Mowiol.

Forebrain neurons were dissected and cultured in 35 mm diameter dishes during 3 DIV. On the third day, neurons were treated with control medium (containing DMSO) or Nystatin medium (at different concentrations) for 30 min. After each respective treatment, neurons were placed on ice and lysed with ice-cold lysis buffer supplemented with a protease inhibitor cocktail (11697498001, Roche), 10 mM NaF, 1 mM Na₃VO₄ and 10 mM Na₂H₂P₂O₇. Cell lysates were diluted with loading buffer and boiled for 5 min. Samples were separated by electrophoresis using an 8% polyacrylamide gel. Proteins were transferred to nitrocellulose membranes (10600002, GE Healthcare Life Sciences) and incubated with different primary antibody and secondary antibodies. Bands were quantified using GelPro Analyzer software (version 3.1, Media Cybernetics).

Images from explants, growth cones and filopodia were acquired using an epifluorescence microscope (Eclipse Nikon E1000) under a 5× and 10× objective (for explants) or a 60× oil-immersion objective (for growth cones and filopodia). A confocal microscope (Leica TCS SP5) was used to acquire a z-stack of images (every 0.5 μm) from DAF-FM and P-Akt intensity with 63x oil-immersion objective.

Each experiment contains a mixed culture of neurons isolated from more than three embryos. All experiments were repeated three independent times (independent dissections). or the analysis of growth cone area and filopodia density, actin staining through phalloidin was used to identify growth cones and axonal filopodia. Filopodia were manually counted as actin-enriched protrusions formed in discrete segments along the axon. Proximal (<50 μm) and distant (>50 μm) axonal regions were randomly selected for quantifications. The growth cone compartment was outlined based on a differential staining for actin in the growth cone concerning the axon compartment. An intensity threshold mask was created using ImageJ (Schneider et al., 2012) and the growth cone perimeter was selected using the wand tool (similar results were obtained by manual selection of growth cone perimeter). In the DAF-FM experiments, the intensity was measured (mean gray value × area) inside the cell body (maximum of Z projections) using ImageJ. The measurements were normalized to the signal intensity obtained in control conditions, to avoid basal fluorescence. Between 20–30 images were acquired for each condition. In P-Akt images, the intensity was quantified inside growth cones and in the cell body, and was normalized to the control condition. In the explant growth and regeneration experiments, the signal from dapi staining (nuclei) was subtracted from each explant and considered as the beginning of the axons. In 3 DIV and 7 DIV explants, the length of eight lines drawn from the beginning of the axons until their most distal part was measured. In 14 DIV explants, the 10 longest axons were selected and their length from the beginning of the axons until their tip was measured. For each explant, 8 (in 7 DIV) or 10 (in 14 DIV) axon measurements were obtained and averaged.

All the data shown in the graphs represent the mean ± SEM. The number of neurons and explants used in each experiment is specified in the corresponding figure legend. Data was analyzed using GraphPad software. Normal distribution was evaluated by applying the D’Agostino and Pearson normality test. Two-tailed, unpaired Student’s t-test was used to compare two conditions, and one-way ANOVA with Tukey post hoc test was used when the experiment had more than two conditions. Significance is considered when *p-value < 0.05, **p < 0.01, ***p < 0.001.

Growth cones play a key role in axon regeneration and membrane cholesterol levels can modify growth cone dynamics. Nystatin is a polyene antifungal agent widely used in experimental research to alter cholesterol levels and modify the function of lipid rafts. We recently proposed a reduction of cholesterol levels as a possible strategy to promote axon regrowth after sciatic nerve lesion (Roselló-Busquets et al., 2019). However, the potential applicability of Nystatin remains unexplored. To answer this question, we first analyzed the effect of different concentrations of Nystatin on the growth cone area in hippocampal neurons (Figure 1). We cultured mice hippocampal neurons for 3 days in vitro (3 DIV) and treated them acutely with different widely used concentrations of Nystatin, from high doses that can remove membrane cholesterol (25 μM) to low doses that do not affect cholesterol levels (2.5 μM; Johnson et al., 1998; Koide et al., 2009; Kim et al., 2013). All three concentrations tested, 2.5 μM (Figures 1A,D), 10 μM (Figures 1B,E) and 25 μM (Figures 1C,F) increased growth cone area (Figure 1G).

The results from Figure 1 suggest a cholesterol-independent effect of Nystatin in controlling growth cone dynamics. To find out the mechanism through which Nystatin is increasing growth cone size, we studied whether Akt phosphorylation is affected under our three Nystatin concentrations tested. Western Blot analysis of P-Akt/Akt levels from primary cultured forebrain neurons treated with 2.5 μM, 10 μM and 25 μM of Nystatin (Figure 2A) revealed no significant differences, but only a tendency in increasing Akt phosphorylation under the highest dose tested (Figure 2B). However, the evaluation of specific P-Akt levels in growth cones (Figure 2C) and cell bodies by immunocytochemistry assay showed that only the highest doses of Nystatin (10 μM and 25 μM) significantly increase Akt phosphorylation locally inside growth cones (Figure 2D).

To analyze whether Nystatin Akt phosphorylation is required for the observed effects of Nystatin on growth cones, we used the Akt inhibitor MK-2206. Whereas lowest concentration of Nystatin (2.5 μM) increases growth cone area independently of Akt inhibition (Figures 3A,D), the effect induced by highest doses (10 μM and 25 μM) depends on Akt activity (Figures 3B,C,E,F). These results are consistent with the phosphorylation of Akt in growth cones. Immunocytochemistry analysis of P-Akt/Akt levels within growth cones revealed that those concentrations of Nystatin that increased Akt phosphorylation also affected growth cone dynamics in an Akt-dependent manner (Figures 3G–I).

We then wanted to evaluate whether chronic treatments of Nystatin promote axonal growth in hippocampal neurons. To avoid possible toxicity effects of longer exposure to Nystatin, we used the lowest concentration tested (2.5 μM) which was sufficient to promote a significant increase in growth cone size (Figure 4). Hippocampal explants were cultured inside a collagen matrix with control media (Figure 4A) or with 2.5 μM Nystatin-containing media (Figure 4B) for 3 days. The length of explant-protruding axons was analyzed, revealing that chronic exposure to a low concentration of Nystatin promotes axon growth (Figure 4C). We then used primary cell cultures of hippocampal neurons (Figures 4D,E) to study the chronic effect of 2.5 μM Nystatin on axonal growth and branching. After 3 days of incubation with Nystatin, we found increases on both axon length and axonal branching (Figure 4F).

Our results suggest a cholesterol and Akt phosphorylation independent effect on growth cone dynamics by low concentrations of Nystatin in acute (min) and chronic (days) treatments. NO is a gaseous second messenger that participates in actin cytoskeleton remodeling, filopodia formation (Welshhans and Rehder, 2005) and growth cone guidance (Tojima et al., 2009). To evaluate whether the observed effects of low concentrations of Nystatin on growth cones depend on NO production, we incubated hippocampal neurons with a NOS inhibitor (L-NMMA) and detected the formation of NO using Diamino-fluorescein Diacetate (DAF-FM DA), a cell-permeable reagent used to quantify low concentrations of NO in solution. DAF-FM DA remains non-fluorescent until it is hydrolyzed to DAF-FM by intracellular esterases, allowing its reaction with NO to form a fluorescent benzotriazole. DAF fluorescent intensity was quantified, revealing increased levels of NO upon low-dose Nystatin treatment (Figures 5A–C). Importantly, the treatment of neurons with the highest concentrations of Nystatin (10 μM and 25 μM) did not affect NO production (Figures 5D,E). The effect of Nystatin in the growth cone area (Figure 6A) and filopodia density (Figure 6B) was prevented by NOS inhibition (Figures 6C,D), suggesting NO production as an alternative mechanism used by Nystatin to modulate axon dynamics at low concentrations. Consistent with the lack of action on NO production, the effect of highest doses of Nystatin (10 μM and 25 μM) on growth cone area and filopodia density is independent of NOS inhibition (Figures 6E–H). Methyl-beta-cyclodextrin (MβCD) is a compound that, similar to Nystatin, extracts cholesterol from cell membranes, increasing growth cone area and filopodia number (Roselló-Busquets et al., 2019). NO measurement using DAF-FM showed that MβCD treatment did not affect NO production (Supplementary Figure S1). These results suggest that the growth cone area and filopodia number can be modulated by decreasing cholesterol levels or by regulating NO production using low concentrations of Nystatin.

Previous results showed that Nystatin increases axon regeneration post-axotomy in primary cell cultures of hippocampal neurons (Roselló-Busquets et al., 2019). The ability of axons to regenerate is lost in adult CNS neurons (He and Jin, 2016; Curcio and Bradke, 2018; Fawcett, 2019). Primary CNS neuronal cultures lose their regenerative capacities during their in vitro differentiation (del Rio and Soriano, 2010). To study whether low concentrations of Nystatin promote axon regeneration post-axotomy and whether this effect is maintained in differentiated neurons, we performed axotomy experiments with hippocampal explants cultured during 7 and 14 DIV (Figures 7A,B). Then, axon explants were mechanically ablated and further regrowth for 3 DIV in the presence of 2.5 μM Nystatin supplemented with the NOS inhibitor L-NMMA (Figures 7C–J). Quantification of axon length revealed that Nystatin promotes regrowth of axons after axotomy regardless of the differentiation state of the neuronal cultures, and in a process that requires NO production (Figures 7C–K).