The chemicals and reagents used included Dulbecco’s modified Eagle medium (DMEM) (Invitrogen), phosphate-buffered saline (PBS) (Dominique Dutscher), fetal bovine serum (FBS) (Corning), horse serum (HS) (BioConcept), penicillin/streptomycin (BioConcept), GlutaMAX (Thermo Fisher Scientific), Accutase (Innovative), dimethylsulfoxide (DMSO) (Sigma-Aldrich), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma-Aldrich), ATPlite 1step Luminescence Assay Kit (Perkin Elmer), trypan blue (Thermo Fisher Scientific), Hank’s balanced salt solution (HBSS) (Sigma-Aldrich), tetramethylrhodamine-methylester (TMRM) (Sigma Aldrich), MitoSOX (Invitrogen), D-glucose (Sigma-Aldrich), pyruvate (Sigma-Aldrich), L-glutamine (Sigma-Aldrich), CellTracker Blue CMAC (Invitrogen), gelatin (Merck), paraformaldehyde (PFA) (Sigma-Aldrich), oligomycin (Sigma-Aldrich), rotenone (Sigma-Aldrich), antimycin (Sigma-Aldrich), XF Calibrant Seahorse (Agilent Technologies), XF DMEM Seahorse (Agilent Technologies), MitoTracker Red CMXROS (Invitrogen), MitoTracker Green FM (Invitrogen), RNeasy Mini Kit (25) (Qiagen), GoScript™ Reverse Transcription Mix, Oligo (dT) (Promega), GoTaq^® Probe qPCR Master Mix (Promega), primers for RT-qPCR (Microsynth AG), neurobasal medium NB (Sigma-Aldrich), retinoic acid RA (Sigma-Aldrich), B27 (Gibco Invitrogen), nerve growth factor (NGF) (Lubio), rat tail collagen I (Corning), anti-βIII-tubuline (Sigma-Aldrich), and Alexa Fluor 488-conjugated secondary antibody (Abcam).

Ginkgo biloba extract EGb 761^® (batch PSC0148/WS1133/EXCh.053) was produced and supplied by Dr. Willmar Schwabe GmbH & Co. KG, Karlsruhe, Germany. The PACs fraction (batch PSC0148/WS1133/EXCh.053/B/Rö16-235-C) was isolated from EGb 761^® by multiple column chromatography, as described in the literature (Kulić Z. et al., 2022), and has a content of 90.81% Ginkgo PACs and 9.19% residual solvent (water).

Human SH-SY5Y neuroblastoma cells were maintained at 37°C in a humidified incubator with 5% CO₂ in DMEM supplemented with 10% (v/v) heat-inactivated FCS, 5% horse serum, and 1% penicillin–streptomycin–GlutaMAX (PSG). Cells were passaged biweekly and plated 1 day prior to treatment, when 80% confluency was reached.

For experiments, cells were plated with 6–12 replicates into clear-bottom 96-well cell culture plates at a density of 1 × 10⁴ cells/well, unless otherwise specified.

For differentiation, 96-well plates were coated with collagen type I (rat tail BD Bioscience) at 0.05 mg/mL for 3 h, followed by cell seeding.

Seahorse XF Cell Culture miniplates were coated with 0.1% gelatin for 30 min.

No coating was applied for other experimental conditions.

The selection of doses was based on the literature, where common concentrations of EGb 761^® or other Ginkgo extracts are typically reported to range between 1 and 100 μg/mL (Lejri et al., 2019; Rhein et al., 2010). EGb 761^® contains approximately 7% PACs. To compare the effects of PACs to the corresponding concentrations of EGb 761^®, PACs concentrations were selected accordingly. EGb 761^® (1, 10, and 100 μg/mL) or PACs (0.1, 1, and 10 μg/mL) treatments were applied 24 h post-seeding. Treatment duration was 24 h. Control specimens received 10 µL/well DMEM supplemented with 1% PSG. DMSO 0.1% was used as the vehicle control.

Treatment solutions were prepared from the dried extract and isolated fraction in DMSO and serum-free DMEM supplemented with 1% PSG.

SH-SY5Y neuroblastoma cells were seeded into black 96-well cell culture plates at 5 × 10³ cells/well in high-glucose DMEM (+10% FCS and 1% PSG).

Differentiation was induced with neurobasal medium, containing 1% PSG, 2% B27, and retinoic acid (RA, 10 µM), which was added for 3 days (Lejri et al., 2019). Post-treatment with EGb 761^® and PACs, cells were fixated with 2% paraformaldehyde. DMSO 0.1% and NGF (50 ng/mL) served as negative and positive controls, respectively. Media were refreshed every second day.

To account for differences in cell number, readouts were normalized to cell count using CellTracker Blue staining. CellTracker Blue CMAC (7-amino-4-chloromethylcoumarin) was loaded at 5 µM and incubated for 1 h at 37°C at 5% CO₂ (Szabo et al., 2024). CMAC was co-incubated with other dyes, following the respective protocols specific to each dye during combined staining procedures. Fluorescence was measured at 353 nm (excitation)/466 nm (emission) using the Cytation 3 reader.

Two days prior to the assay, the cells were plated with 6–12 replicates into a black 96-well cell culture plate at a density of 10,000 cells per well. The following day, the treatment protocol was followed. On the day of the assay, the cells were loaded with a final concentration of 5 µM of MitoSOX and then incubated for 2 h at 37°C and 5% CO₂. After washing twice with HBSS, fluorescence was detected using the Cytation 3 cell imaging multi-mode reader at 531 nm (excitation)/595 nm (emission) (Szabo et al., 2024).

Mitochondrial membrane potential (MMP) was assessed using TMRM at 0.4 μM for 30 min in the dark (Szabo et al., 2024). The cells were plated with 6–12 replicates into a black 96-well cell culture plate at a density of 10,000 cells per well. The following day, the treatment protocol was followed. Two days after plating, the cells were treated with the dye at a final concentration of 0.4 µM, and afterward they were incubated for 30 min under agitation at room temperature. Before measuring the fluorescence at 531 nm (excitation)/595 nm (emission) using the Cytation 3 cell imaging multi-mode reader (BioTek), the cells were washed twice with HBSS.

The total ATP was measured using the ATPlite 1step bioluminescence kit (Perkin Elmer) as per the manufacturer’s protocol, with luminescence detected by the multiplate reader Cytation 3 (BioTek). Two days prior to the assay, the cells were plated with 6–12 replicates into a black 96-well cell culture plate at a density of 10,000 cells per well. The following day, the treatment protocol was followed. On the day of the assay, after the preparation for the ATP standard curve, 100 µL of the ATP substrate solution was added to each well. After incubation in the dark under agitation at room temperature for 5 min, luminescence was measured using the Cytation 3 cell imaging multi-mode plate reader (BioTek) (Szabo et al., 2024).

Forty-eight hours prior to the assay, the cells were plated with 6–12 replicates into a 96-well cell culture plate at a density of 10,000 cells per well. The following day, the treatment protocol was followed. On the day of the assay, 10 µL of MTT solution (5 mg/mL MTT (3-(4,5-dimethylthyazol-2-yl)-2,5-diphenyltetrazolium bromide)) were added to every well and incubated for 3 h at 37°C and 5% CO₂. During the incubation time, metabolically active cells transformed MTT to formazan crystals. After adding DMSO to dissolve the formazan crystals, the absorbance was measured at 550 nm using the Cytation 3 cell imaging multi-mode plate reader (BioTek).

Cells were plated with 6–12 replicates into a black 96-well cell culture plate at a density of 10,000 cells per well 48 h before the assay. The following day, the treatment protocol was followed. On the day of the assay, the cells were incubated for 1 h at 37°C and 5% CO₂ with MitoTracker Green FM (100 nM) and then washed with HBSS. Fluorescence was detected using the Cytation 3 cell imaging multi-mode reader (BioTek) at 490 nm (excitation)/516 nm (emission).

SH-SY5Y cells were plated into 6-well cell culture plates at 10⁵ cells/well (Szabo et al., 2023). EGb 761^® (10 μg/mL) and PACs (1 μg/mL) treatment duration was 24 h. After PBS washing and cell lysis, total RNA was isolated using the RNeasy Mini Kit. RNA concentration and purity were quantified and assessed using NanoDrop. RNA was reverse transcribed to cDNA using the GoScript™ Reverse Transcription Mix, Oligo (dT). qRT-PCR was performed using the GoTaq^® qPCR Master Mix (ref. dye CXR, 300 nM/primers 900 nM) with the StepOnePlus^TM Real-Time PCR System. The relative fold gene expression was quantified using the 2^−ΔΔCT method with GAPDH as an endogenous reference.

The Seahorse XF HS Mini Analyzer (Agilent Technologies) was used to measure the oxygen consumption rate (OCR) and extracellular acidification rate (ECAR), with SH-SY5Y cells plated at 1.5 × 10⁴ cells/well (Szabo et al., 2024). The XF Mito Stress Test protocol was followed as per the manufacturer’s specifications. Mitochondrial agents were injected at the following concentrations: oligomycin (2 μM) and a combination of antimycin A (1 μM) and rotenone (1 μM). The assay medium consisted of the Seahorse XF DMEM (pH of 7.4) with additional 18 mM D-glucose, 4 mm pyruvate, and 2 mM L-glutamine.

Cell nuclei were stained with DAPI (3 μM, 10 min of incubation), and neurites were immunolabeled with anti-βIII-tubulin (5 μg/mL, incubation overnight) and Alexa Fluor 488-conjugated secondary antibody (5 μg/mL, 1 h of incubation) (Lejri et al., 2019).

Images were captured using the Cytation 3 cell imaging multi-mode reader (×20 objective) and analyzed with ImageJ (neurophology plugin). The following parameters of neuroplasticity were evaluated: the number of contact and branching points (endpoint and attachment point), total neurite length, soma count, and neurite count. In total, approximately 10,000 cells and 57,000–101,000 neurites were evaluated. Images were analyzed to quantify neurite extension between cells, which were visualized using anti-βIII-tubulin/Alexa488 staining for neurites and DAPI for nuclei.

Data are presented as the mean ± SEM, normalized first to the cell count with CellTracker Blue and afterward to the control condition (CTRL = 100%). Statistical analysis was performed using the GraphPad Prism software. When comparing more than two experimental groups, one-way ANOVA and post hoc Dunnett’s multiple comparisons test versus control were applied. For direct comparisons between two groups, an unpaired Student’s t-test was used. P-values <0.05 were considered statistically significant.

Both EGb 761^® and PACs exhibited antioxidant and ROS scavenging properties as treatment with both compounds led to a decrease in mitochondrial superoxide in the neuroblastoma cell line SH-SY5Y. Lowest concentrations of mitochondrial superoxide were measured in the treatment group of EGb 761^® 100 μg/mL (−9.7% compared to CTRL, p < 0.0001) and PAC 0.1 μg/mL (−9.9% compared to CTRL, p < 0.0001) conditions (Figure 1A). ROS have shown direct effects on the electron transport chain, OXPHOS, and ATP production. Therefore, we decided to investigate the MMP and total ATP levels in the following assays. EGb 761^® (10 μg/mL, p = 0.0179) and PAC (1 μg/mL, p = 0.0416) concentrations positively impacted the total ATP production in neuroblastoma cells (Figure 1C). Other concentrations did not alter total ATP levels compared to the control. To further validate the beneficial effects of both compounds on ATP production, measuring the MMP was crucial, as it plays a critical role in driving oxidative phosphorylation. Both PACs and EGb 761^® increased MMP, but PACs showed a more pronounced effect than EGb 761^®. Specifically, PACs treatment elevated the mitochondrial membrane potential up to 42% in comparison to the control condition, with both 1 μg/mL and 10 μg/mL exhibiting similar results (for both p < 0.0001), suggesting a ceiling effect. In contrast, treatment with EGb 761^® revealed its maximal effect at a concentration of 10 μg/mL (p < 0.0001), resulting in a 30% increase. The highest concentration led to only a 22% elevation (p < 0.0001), which is slightly above the 16% (p < 0.0001) observed at the lowest concentration (Figure 1B).

Increases in ATP and MMP were paralleled by elevated metabolic activity. The metabolic activity was significantly increased up to 31% for the EGb 761^® 100 μg/mL-treated cells (p < 0.0001) and up to 21% for the PAC 10 μg/mL-treated cells (p < 0.0001) (Figure 2A). Lower concentrations also demonstrated substantial effects on EGb 761^® at 10 and 1 μg/mL of EGb 761^® increased activity by 26% and 22%, respectively (both p < 0.0001), whereas PACs at 1 and 0.1 μg/mL resulted in a 16% increase, with both concentrations to 16% (p = 0.0006 and p = 0.0009, respectively). Ameliorated metabolic activity and ATP levels come with increased energy demands; therefore, investigating effects on mitochondrial mass was the next logical step.

The increase in mitochondrial mass in human neuroblastoma cells seemed to be cohesive with the results from previous experiments. PACs had a slightly better effect on mitochondrial mass than EGb 761^®, with mitochondrial mass increases of 15%, 21%, and 24% at 0.1, 1, and 10 μg/mL (all p < 0.0001), respectively. EGb 761^® treatment led to elevations of 8% (p = 0.0266), 11% (p = 0.0005), and 13% (p = 0.0002) at 1, 10, and 100 μg/mL, respectively (Figure 2B).

PPARGC1-α is coding for a transcriptional coactivator regulating gene involved in energy metabolism, linking external stimuli and the regulation of mitochondrial biogenesis. Based on the optimal conditions for metabolic activity and mitochondrial mass—EGb 761^® at 10 μg/mL and PACs at 1 μg/mL—gene expression was assessed accordingly. EGb 761^® and PAC treatment of the SH-SY5Y cell line increased the protein expression of PPARGC1-α after 24 h (Figure 2C), suggesting upregulated mitochondrial biogenesis. EGb 761^® at 10 μg/mL elevated PPARGC1-α expression by 60% (p = 0.0019), whereas PACs only elevated it by 27% (p = 0.0068).

Based on previous observations, we evaluated the impact of EGb 761^® and PACs on the bioenergetic profile of human neuroblastoma cells by performing the Seahorse XF Mito Stress Test. We observed a statistically notable increase in the OCR (EGb 761^® 10 μg/mL, p = 0.0017, PACs 1 μg/mL, p = 0.0041) (Figure 3A) and ECAR (EGb 761^® 10 μg/mL p = 0.0003, PACs 1 μg/mL, p = 0.0001) (Figure 3B) in both treatment groups. The amelioration of the extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) as indicators of glycolysis and OXPHOS, respectively, indicate a metabolic shift in the cells toward a more energetic and metabolic state in comparison to the untreated control (Figure 3C). This is further illustrated by the energy phenotype map, where both EGb 761^® and PACs treatments shift cells from a quiescent phenotype into the energetic quadrant. Notably, PACs induced a slightly stronger enhancement in oxidative capacity (OCR) relative to ECAR, suggesting a more pronounced stimulation of mitochondrial function. Overall, both treatments promote a similar bioenergetic reprogramming in human neuroblastoma cells, favoring a more metabolically active phenotype.

Neuroplasticity is a process with high energy demand. As treatment with EGb 761^® and PACs can increase the energetic state of cells, it was of interest to study whether we can detect an increase in outgrowth measures. Imaging neurite outgrowth in human neuroblastoma SH-SY5Y cells (×20, Supplementary Figure S4) using the Cytation 3 cell imaging multi-mode reader allows for both visualization and quantification of neurite formation and their projections between cells (Figure 4). After 3 days of treatment with EGb 761^®, concentrations of 10 μg/mL (+195%, p < 0.001) and 100 μg/mL (+170%, p < 0.001) were found to be the most effective, boosting the neurite count to nearly the same extent as NGF (+223%, p < 0.001) compared to that in untreated cells (CTRL). An amount of 1 μg/mL of EGb 761^® increased the neurite count by 124% (p = 0.002). Similarly, PACs treatment showed comparable effects and enhanced neurite count by 111% (p = 0.008), 203% (p < 0.001), and 253% (p < 0.001) for the 0.1, 1, and 10 μg/mL concentrations, respectively. In addition, treatment with 10 and 100 μg/mL of EGb 761^® significantly enhanced neurite length by up to 298% (p < 0.001) and 205% (p < 0.001), respectively. With an increase of 153% (p = 0.001), 293% (p < 0.001), and 360% (p < 0.001) for the 0.1, 1, and 10 μg/mL PACs treatments, respectively, PACs seem to have a stronger effect on neurite length. PACs treatment also significantly increased the number of attachment points, with 262% and 282% increases at 1 and 10 μg/mL, respectively (p < 0.001), exceeding the effects of both NGF (+254%, p < 0.001) and EGb 761^® (165%, 232%, and 199% increases at 1, 10, and 100 μg/mL, respectively; all p < 0.001). Similar trends were observed for endpoints, where PACs at 10 μg/mL increased counts by 292% (p < 0.001), surpassing both NGF (+261%) and EGb 761^® (up to 212%, p < 0.001).