Polyphenolic Profile and Antioxidant Capacity of Coffee Silverskin Extracts: Insights from HPLC and GC-MS Analyses and Protective Effect in Schwann-like Cells

Abstract

Coffee silverskin (CS) is an abundant leftover of the coffee roasting process known to contain significant concentrations of bioactive molecules, including polyphenols and flavonoids, with established antioxidant properties and potential applications in nutraceutical and functional-food formulations. This study systematically optimized extraction conditions to maximize the recovery of phenolics and antioxidants from CS by evaluating the effects of solvent type, temperature, and sonication time. Ethanol extraction at 20 °C for 30 min yielded the most enriched polyphenolic fraction, with the highest total phenolic content (TPC), total flavonoid content (TFC), and antioxidant activity across 2,2-diphenyl-1-picrylhydrazyl (DPPH), trolox equivalent antioxidant capacity (TEAC), and ferric reducing antioxidant power (FRAP) assays. Comprehensive chemical characterization via high-performance liquid chromatography (HPLC) and gas chromatography–mass spectrometry (GC-MS) identified key phenolics (chlorogenic acid, hyperoside, quercetin), alongside abundant caffeine, tocopherols, and phytosterols. The biological relevance of the optimized extract was assessed for the first time in RT4-D6P2T Schwann-like cells, demonstrating significant cytoprotective and antioxidant effects against H₂O₂- and lipopolysaccharide-induced oxidative stress, thereby establishing a previously unreported neuroprotective potential.

1. Introduction

Coffee is one of the most consumed beverages in the world and as such, the coffee industry generates over 10 million tons of coffee waste worldwide [1]. Among the various leftovers, coffee silverskin (CS) is the thin epidermal layer that detaches from coffee beans during roasting. It amounts to roughly 0.75–0.85% of the roasted bean weight (~7.5 kg per ton of coffee). Currently, most CS is discarded (sent to compost or landfill), which creates disposal costs and environmental burdens. Nevertheless, it shares some bioactive compounds with roasted beans but exhibits a distinct phytochemical profile due to its exposure to heat and concentration effects during roasting. CS often retains moderate levels of caffeine (ranging from 0.8 to 7.3 mg/g dry weight) compared to roasted beans. Meanwhile, CS is significantly enriched in dietary fiber (~18% cellulose, 13% hemicellulose) along with ~19% protein and 8% ash (minerals), as well as phenolic compounds, particularly CGAs and their derivatives; one study reported CS rich in 3-, 5-, and 4,5-di-caffeoylquinic acids, present both in free and bound forms [2]. In addition, the roasting process drives Maillard reactions, leading to formation of volatiles like pyrroles and pyridines in CS, which impart unique sensory and antioxidant characteristics. The roasting process also generates melanoidins and concentrates plant polyphenols and alkaloids in the silverskin [3]. These compounds exhibit strong antioxidant and health-promoting activities (e.g., neuroprotective, cardioprotective, antimicrobial, and anti-inflammatory effects) [4,5]. This composition suggests multiple valorization routes: CS can be exploited as a source of fiber, fuels (via combustion or biogas), and other valuable compounds [6,7]. In a circular-economy context, CS is viewed as an underutilized resource, and sustainable management strategies are being sought to recover its components rather than discarding it [8,9,10]. However, the use of CS as a functional food ingredient faces important challenges, including its classification as a Novel Food under EU Regulation 2015/2283 [11], which requires formal safety assessment and authorization, and the environmental pressures associated with its disposal, which underscore the need for sustainable valorization strategies within the circular-economy framework.

Plant polyphenols are well recognized in nutrition and food science for their health benefits, and epidemiological studies have linked long-term consumption of polyphenol-rich diets to lower incidence of chronic diseases (cardiovascular disease, diabetes, neurodegenerative disorders, etc.) [12,13]. In parallel, food scientists value polyphenols for their functional properties: they contribute to color, flavor, and oxidative stability of foods, and they can serve as natural alternatives to synthetic preservatives [14,15]. Thus, recovering antioxidants from agro-industrial by-products like CS represents an opportunity in developing health-promoting and clean-label food ingredients.

Several research groups have characterized CS extracts, evidencing their high total phenolic content (TPC) and strong radical-scavenging activity. For example, Narita and Inouye reported that subcritical water extracts of CS had very high antioxidant activity (H-ORAC ~2629 μmol TE/g, DPPH ~379 μmol TE/g) that correlated linearly with extract phenolic content [16]. Guglielmetti and colleagues have found that adding CS extracts to bread significantly elevates measured TPC and improves TEAC/DPPH/FRAP values [17]. These results indicate that CS contains chlorogenic acids and related phenolics at levels comparable to green coffee or brew, justifying its strong antioxidative profile. However, reported antioxidant effects vary widely across studies due to methodological differences. Moreover, most studies rely solely on bulk colorimetric assays (e.g., Folin–Ciocalteu TPC, DPPH, FRAP, etc.), while fewer reports use mass spectrometry coupled with chromatographic techniques to identify specific CS polyphenols.

Oxidative stress is a key pathogenic process in numerous neurodegenerative and demyelinating diseases affecting both the central and peripheral nervous systems (PNSs) [18,19]. Schwann cells, which are primarily responsible for axonal myelination in the peripheral nervous system, are particularly vulnerable to the accumulation of reactive oxygen species (ROS) [20,21,22], resulting in a redox imbalance that alters their neurotrophic and myelinating functions, promotes mitochondrial dysfunction, and triggers inflammatory and apoptotic signaling pathways. Under physiological conditions, Schwann cells maintain redox homeostasis via endogenous antioxidant mechanisms, including the involvement of enzymes such as superoxide dismutase (SOD), catalase, and glutathione peroxidase (GPx) [23,24,25]. However, in response to toxic or inflammatory insults, such as those related to ischemia, neuroinflammation, or the action of environmental toxins, this balance can be disrupted, leading to myelin degeneration and neuronal dysfunction [26,27].

Although CS has been recognized as a valuable source of bioactive compounds, significant gaps remain regarding how different extraction solvents and conditions influence its detailed phytochemical composition and antioxidant potential. In this study, we aimed to systematically compare different conditions to obtain CS extracts and evaluate their antioxidant capacity. Specifically, we used solvent extraction followed by chromatographic analysis to identify and quantify the major phenolic compounds in the CS. We also measured antioxidant activity using multiple assays (DPPH, TEAC, FRAP) to provide robust data on its radical-scavenging potential. Ultimately, the antioxidant and cytoprotective effects of the CS extracts were evaluated by their ability to modulate the oxidative stress response induced by hydrogen peroxide and lipopolysaccharide (LPS) stimuli in RT4-D6P2T rat Schwann cells, which could lead to potential future nutraceuticals able to mitigate oxidative stress associated with pathophysiological conditions of the PNS.

2. Materials and Methods

2.1. Chemicals

Analytical standards used in this study were of certified purity. Gallic acid (#91215), ascorbic acid (#A92902), quercetin (#PHR1488), p-coumaric acid (#95777), (±)-α-tocopherol (#PHR1031), chlorogenic acid (#PHR2202), kaempferol (#1354900), trans-caffeic acid (#51868), and caffeine (#56396) were purchased from Sigma-Aldrich (Milan, Italy) and used as received. Analytical standards of neochlorogenic acid (#N1155) and rutin hydrate (#R0035) were purchased from TCI (Tokyo, Japan) and used as received. Analytical standard of (+)-catechin (#BISN0111) was purchased from Apollo Scientific (Manchester, UK) and used as received. HPLC-grade acetonitrile (#K981) and water (#K978) were purchased from VWR chemicals (Milan, Italy). Sodium nitrite (#11483970), sodium carbonate anhydrous (#10548070), hydrochloric acid (#12666846), and HPLC-grade formic acid (#30280925) were obtained by J.T. Baker Chemicals (Deventer, Holland). 2,2-Diphenyl-1-picrylhydrazyl (DPPH, #D9132), potassium persulfate (K₂S₂O₈, #379824), sodium hydroxide (#S5881), 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS, C₁₈H₂₄N₆O₆S₄, #A1888), ammonium acetate (#238074), copper(II) chloride dihydrate (CuCl₂·2H₂O, #307483), sodium acetate trihydrate (CH₃COONa·3H₂O, #117418), 2,4,6-tris(2-pyridyl)-1,3,5-triazine (TPTZ, C₁₈H₁₂N₆, #T1253), iron(III) chloride hexahydrate (FeCl₃·6H₂O, #44944), aluminum (III) chloride (AlCl₃, #563919), potassium peroxydisulfate (#60489), 96% ethanol (#100971), methanol (#179337), and acetic acid (#695092) were purchased from Sigma-Aldrich (Milan, Italy). Solvents were used without further purification.

2.2. Raw Materials

Dried samples of ground coffee silverskin, produced by roasting Arabica coffee beans, were kindly provided by Quarta Caffè S.R.L. (Lecce, Italy). Beans were roasted under industrial conditions (approx. 200–230 °C for 12–15 min), and the silverskin layer was mechanically separated from the beans immediately after roasting and cooling. The collected CS was sieved to remove impurities, dried to constant weight, and ground to a fine powder (particle size < 1 mm) before packaging. The raw material was stored in a sealed bag under inert atmosphere (N₂), at room temperature in a cool and dry place, away from direct light exposure.

2.3. Extraction Procedure

Polyphenols extraction procedure was based on methods reported by Zengin et al. with slight modifications [28], testing different extraction parameters, such as solvent (water, ethanol, and a 1:1 (v/v) hydroalcoholic solution), time (30 and 60 min), and temperature (20 and 40 °C). Briefly, the ground coffee silverskin (4 g) was suspended in either H₂O, EtOH, or 50% EtOH/H₂O (50 mL). The suspension was sonicated at a frequency of 40 KHz with a LABSONIC LBS2 ultrasonic bath (FALC, Treviglio (BG), Italy) at either 20 or 40 °C and for either 30 or 60 min. Throughout each extraction, the bath temperature was monitored using the built-in digital thermometer, and the system automatically compensated for heat generated by sonication to maintain the selected temperature. Following sonication, the mixtures were filtered under vacuum and the solvent evaporated using a Rotavapor R-200 (BÜCHI Labortechnik AG, Flawil, Switzerland) operated at a bath temperature of 40 °C under reduced pressure (150–200 mbar). The residues were finally lyophilized (Freeze Dryer BFBT-101-A, Biolab, Toronto, ON, Canada) and dissolved in methanol to obtain a 1 mg/mL stock solution of raw extract. The yield was calculated as the mass of the extract obtained relative to the initial mass of the whole plant and is expressed as a percentage (%).

2.4. Total Phenolic Content (TPC) and Total Flavonoid Content (TFC) Determination

TPC was determined according to Singleton et al. [29]. Briefly, a water-diluted 1:10 Folin–Ciocalteu reagent (0.2 N, 250 μL) was added to 50 μL of the extract (1 mg/mL). After 5 min of incubation in the dark, 700 μL of 7.5% sodium carbonate was added. The resulting mixture was then incubated for an additional 2 h in the dark at room temperature, and the absorbance was subsequently measured at a wavelength (λ) of 765 nm. Gallic acid (25–100 μg/mL) served as the standard for constructing the calibration curve (r² = 0.994). The TPC was expressed as mg of gallic acid equivalents per gram of dry sample (mg GAE/g).

The total content of flavonoids was determined by the method described by Yan et al. [30]. 0.5 M NaNO₂ (75 μL) and 0.5 M AlCl₃ (75 μL) were added to 250 μL of the extract (1 mg/mL). The mixture was incubated in the dark at ambient temperature for 5 min. Subsequently, 1 M NaOH (1 mL) was added, and the absorbance was measured at a wavelength of 510 nm. Rutin (0.5–0.1 mg/mL) was used as the standard for constructing the calibration curve (r² = 0.995). The TFC was expressed as mg of rutin equivalents per gram of dry sample (mg RE/g).

2.5. DPPH Assay

The scavenging activity of the extracts was evaluated spectrophotometrically according to Nzekoue et al. [31]. DPPH was diluted with methanol to obtain a working solution with an initial absorbance of about 0.70 a.u. at a wavelength (λ) of 517 nm. For the assay, 100 µL of the extract was mixed with 900 µL of the prepared DPPH solution. After 30 min of incubation in the dark at room temperature, the DPPH absorbance was measured at 517 nm. Ascorbic acid (5.0–50.0 µg/mL) was used as the standard for constructing the calibration curve (r² = 0.998). The DPPH activity was expressed as mg of ascorbic acid equivalents per gram of dry sample (mg AAE/g).

2.6. Trolox Equivalent Antioxidant Capacity (TEAC) Assay

The TEAC assay was employed to evaluate the samples’ ability to scavenge 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic) acid radical cations (ABTS^•+) [32]. The radical cation solution was generated by the reaction between a 7 mM ABTS stock solution and a 2.45 mM K₂S₂O₈ solution after 16–18 h of incubation in the dark. The resulting radical solution, which remained stable for up to 2 days, was diluted with distilled water to obtain a working solution with an absorbance of 0.70 ± 0.02 at a wavelength (λ) of 734 nm. For the assay, the diluted ABTS^•+ solution (950 µL) was added to 50 µL of each sample. The absorbance reading at 734 nm was taken after 10 min of incubation at room temperature and in the dark. Trolox (3.9–125.0 µg/mL) was used as the standard for constructing the calibration curve (r² = 0.999). The ABTS activity was expressed as mg of Trolox equivalents per gram of dry sample (mg TE/g).

2.7. Ferric Reducing Antioxidant Power (FRAP) Assay

The FRAP assay was carried out according to Fu et al. [33] with minor modifications. The FRAP reagent was prepared daily by mixing 300 mM acetate buffer (pH 3.6), 10 mM TPTZ (dissolved in 40 mM HCl), and 20 mM FeCl₃·6H₂O in a volume ratio of 10:1:1. The resulting mixture was warmed for 10 min at 37 °C. For the assay, 50 µL of each sample were added to 950 µL of the FRAP working solution. Six minutes later, the absorbance was measured at 593 nm. Ascorbic acid (5.0–50.0 µg/mL) was used as the standard for constructing the calibration curve (r² = 0.994). Results were expressed as mg of ascorbic acid equivalents per gram of dry sample (mg AAE/g).

2.8. HPLC Analysis

The HPLC analysis was carried out using an Agilent 1220 Infinity HPLC (Agilent Technologies, Palo Alto, CA, USA) equipped with an Eclipse Plus C18 (particle size 5 μm; 4.6 × 250 mm, Agilent) stationary phase column and interfaced with a diode array detector (model G1315B DAD system; Agilent). The procedure followed published methods with slight modifications [34,35,36]. The mobile phase consisted of a mixture of water (solvent A) and acetonitrile (solvent B), both containing 0.1% formic acid. The flow rate was maintained at 1.0 mL/min in gradient elution mode. The gradient program was executed as follows: 0 to 2 min, 20% of solvent B; 2 to 10 min, linear increase from 20% to 80% of solvent B; 14 to 24 min, linear increase from 80% to 100% of solvent B; 24 to 34 min, 100% of solvent B; 34 to 35 min, linear decrease from 100% to 20% of solvent B; 35 to 40 min, 20% of solvent B (re-equilibration step). The injection volume was set at 20 μL, the column temperature at 30 °C, and the UV-Vis detection wavelength at 280 nm. Compound identification was performed by comparing retention times and UV-Vis spectra with those of authenticated reference standards (see Section 2.1). Quantification was carried out using external calibration curves (r² ≥ 0.998) prepared from pure standards at known concentrations (0.5, 1.0, 5.0, 10.0, and 15.0 mg/mL). To facilitate the detection of minor constituents, concentrated stock solutions of the extract were utilized. Results are expressed as mg of compound per mg of extract.

2.9. Gas Chromatography

The GC-MS analysis was performed using a splitless injection mode on a Thermo Scientific TRACE GC coupled to a Thermo Scientific PolarisQ external ionization ion trap mass spectrometer (Thermo Fisher Scientific, Bellefonte, PA, USA), operating in electron ionization (EI) mode at 70 eV, with a scan range of m/z 50–650, ion source temperature set at 200 °C, transfer line temperature at 250 °C, and injector temperature at 250 °C. Chromatographic separation was achieved using a TraceGOLD TG-5SilMS (30 m × 0.25 mm × 0.25 µm) column (Thermo Fisher Scientific, Bellefonte, PA, USA). Prior to injection, aliquots of the samples were filtered using a 0.2 µm pore size filter. Identification of analytes was performed through comparison of the obtained mass spectra with entries in the NIST/EPA/NIH Mass Spectral Library. Quantification was conducted by calculating relative peak areas from the total ion chromatogram (TIC). The analysis of semi-volatile organic compounds (SVOCs) adhered to the standard US EPA Method 8270 methodology [37].

2.10. Cell Culture Maintenance and Treatments

The RT4-D6P2T rat schwannoma cell line (ATCC cod. CRL-2768™) was maintained in Dulbecco’s Modified Eagle’s Medium (DMEM, 4500 mg/L glucose, EuroClone, Milan, Italy) supplemented with 10% Fetal Bovine Serum (FBS), 100 U/mL penicillin, 100 µg/mL streptomycin, and 2 mM glutamine, according to standard procedures [38,39]. Cells were incubated at 37 °C in a humidified atmosphere containing 5% CO₂. Cell cultures were subsequently exposed to different treatments with CS extract, caffeine (Sigma Aldrich, code: C0750, St. Louis, MO, USA), and lipopolysaccharide (LPS, Invitrogen, code: 00-4976-93, Carlsbad, CA, USA). Stock solutions of the CS extract and caffeine were prepared in a methanol/DMSO (90:10, v:v) mixture to ensure complete solubilization of both hydrophilic and partially lipophilic compounds. Prior to cell treatment, these stock solutions were diluted into the culture medium to ensure the final DMSO concentration never exceeded 0.1% (v/v). Control experiments using the methanol/DMSO (90:10, v/v) solution were performed to confirm that the vehicle did not influence cell viability.

2.11. Cell Viability Assay

Cells were seeded in a 96-well plate at a density of 5 × 10³/well, with each well containing 100 μL of complete medium. Cells were then allowed to adhere to the plate overnight. Cell viability was determined using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay, performed according to the manufacturer’s instructions (Sigma Aldrich, St. Louis, MO, USA). After the viable cells reduced the MTT to purple formazan crystals, the crystals were dissolved using dimethyl sulfoxide (DMSO). The absorbance of the dissolved formazan was measured at 570 nm using a microplate reader (Multiskan™ FC Microplate Photometer (Thermo Fisher Scientific, Waltham, MA, USA).

2.12. DCFDA Assay

The production of ROS was estimated using the 2′,7′-dichlorofluorescein diacetate (DCFDA) assay (Sigma) [40]. Cells were seeded in Corning^® Falcon^® Microplate (Sigma-Aldrich, code: CLS353376) at a density of 0.3 × 10⁵ cells/well. Following the specified treatments, the medium was replaced with fresh medium containing 50 µM of DCFDA and incubated for 30 min. Cells were then washed with PBS and analyzed using a Cytation 5 Imaging plate reader (Biotek Instruments Inc., Agilent, Winooski, VT, USA) at excitation/emission wavelengths (Ex/Em) of 485/535 nm. The fluorescence generated is directly proportional to the amount of DCFDA oxidized to DCF.

2.13. Statistical Analysis

All experiments (extraction procedures, analytical measurements, and cellular experiments) were performed in triplicate, and the results are presented as the mean ± standard deviation (SD). Statistical analysis was performed using Prism v. 8.0.1 software (GraphPad Software Inc., La Jolla, CA, USA). Multiple comparisons were conducted using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. A p-value of <0.05 was accepted as the level of statistical significance. The following statistical representations were used in the text: * p < 0.05, ** p < 0.005, *** p < 0.0005; **** p-value < 0.0001.

3. Results

3.1. Analysis of the Extraction Conditions

In order to optimize the conditions for extracting biomolecules from the CS, three variables were investigated, i.e., solvent polarity, extraction temperature, and sonication time. More specifically, water and ethanol were used as solvents, deliberately avoiding highly organic solvents (such as hexane, ethyl acetate, and acetone) due to their higher cost and lower biocompatibility. A 1:1 mixture of water and ethanol was also tested to evaluate conditions of intermediate polarity and lipophilicity. Two conditions each were also checked for optimizing the extraction temperature, i.e., room temperature (20 °C) and a mildly elevated temperature (40 °C). The higher temperature was chosen to improve the solubility of the biomolecules while minimizing the risk of thermal degradation. Finally, both a shorter (30 min) and a longer (60 min) sonication time were tested to determine if extended sonication would allow for the extraction of a higher amount of bioactive compounds.

To identify the optimal extraction procedure, we assessed both the extraction yield (reported relative to the initial raw material quantity) and, more importantly, the type of biomolecules extracted (quantified as the total polyphenolic content). The results are summarized in Table S1.

As clearly visible in the scatter plot in Figure 1a, the extraction study revealed a clear solvent-dependent trade-off between total mass yield and polyphenol concentration. Conversely, time and temperature were less influential, as demonstrated by the tight clustering of the data points related to the same extraction solvent.

Aqueous extraction produced the highest yields (between 13% and 19%), while pure ethanol yielded the lowest yields (5% maximum). Conversely, pure ethanol resulted in the highest TPC (between 27 and 38 mg GAE/g extract), and the 50% ethanol/water mixture provided intermediate yields and TPC values. Ethanol thus allowed for the most selective extraction of polyphenols; water extracted large amounts of hydrophilic, non-phenolic material, whereas the hydroalcoholic mixtures provided intermediate results between polyphenolic content and mass transfer, and no significant differences (p > 0.05) were observed varying extraction time and temperature.

Temperature and time effects in our dataset indicate a significant (p < 0.05)decreased TPC at 40 °C relative to 20 °C for ethanol extracts, alongside a reduction in efficiency for aqueous extraction at the higher temperature/longer time condition (40 °C, 60 min) (Figure 1b).

To gain a better insight into the composition of the extract, the TFC was also determined, providing a complementary measure to the TPC (Table S1). The data revealed that, despite flavonoids comprising the majority of the TPC extracted in all conditions, a broader chemical diversity of polyphenols is present. Ethanol extracts exhibited the highest TPC and TFC values, confirming ethanol’s strong affinity for both flavonoid and non-flavonoid phenolics. At 20 °C, TPC reached about 35–38 mg GAE/g, while TFC ranged between 202 and 227 mg RE/g, indicating that although TPC is higher in concentration, flavonoids make up a substantial portion of the extractable phenolics under mild conditions.

For aqueous extracts, both TPC and TFC decreased compared to ethanol, but the reduction is more dramatic for TFC. Water’s polarity favors the extraction of hydrophilic phenolics, especially chlorogenic acids, while many flavonoids are less water-soluble. As a result, TPC in water remained moderate (~11 mg GAE/g at 20 °C), whereas TFC dropped to around 21 mg RE/g. This highlights the limited ability of water to dissolve flavonoid-rich fractions. The 50% ethanol/water mixture showed intermediate behavior, combining water’s efficiency for hydrophilic polyphenols with ethanol’s affinity for flavonoids. TPC values averaged at 16–18 mg GAE/g, while TFC averaged at about 50 mg RE/g. These proportions indicate that mixed solvents extract a more balanced phenolic profile, though they are still less effective than pure ethanol for flavonoids.

3.2. Antioxidant Assays

Considering that high TPC and TFC might not be directly imputable to the highest antioxidant effect, a series of antioxidant assays were also performed on the different extracts to confirm the desired ability to scavenge radicals.

Table 1. Antioxidant assays on the CS extracts at different extraction conditions of solvent, temperature, and time.

	Temp (°C)	Time (min)	Ethanol	H2O	50% Ethanol/H2O
DPPH (mg AAE/g)	20	30	156.50 ± 2.52	8.68 ± 0.16	21.09 ± 2.38
	20	60	149.18 ± 2.59	6.98 ± 0.13	20.18 ± 0.57
	40	30	116.95 ± 18.96	5.00 ± 0.39	19.03 ± 1.34
	40	60	138.90 ± 2.82	6.53 ± 0.45	22.73 ± 0.13
TEAC (mg TE/g)	20	30	206.91 ± 15.79	39.61 ± 1.44	51.17 ± 0.86
	20	60	275.48 ± 12.02	51.51 ± 1.26	80.97 ± 1.00
	40	30	208.54 ± 7.58	34.98 ± 1.61	75.46 ± 1.11
	40	60	180.79 ± 22.48	38.38 ± 24.30	50.01 ± 2.67
FRAP (mg AAE/g)	20	30	79.86 ± 6.15	8.40 ± 0.28	17.82 ± 0.56
	20	60	100.64 ± 1.25	9.59 ± 0.92	21.19 ± 0.57
	40	30	91.63 ± 4.45	8.37 ± 0.06	18.52 ± 0.43
	40	60	87.78 ± 0.15	3.98 ± 0.05	17.20 ± 0.08

presents the antioxidant capacity of the CS extracts, as determined by DPPH, TEAC, and FRAP assays, obtained under varying extraction conditions of solvent, temperature, and duration.

3.3. Chromatographic Characterization

3.3.1. HPLC Analysis

The high-performance liquid chromatography (HPLC) chromatogram of the CS ethanolic extract (Figure S1) revealed a complex profile rich in phenolic and antioxidant compounds, reflecting the chemical diversity characteristic of this coffee-processing by-product. The chromatogram shows multiple well-defined peaks between 2 and 38 min of retention time, with peaks corresponding to a variety of bioactive molecules, ranging from alkaloids, flavonoids, and vitamins, as indicated in

Table 2. Summary of the bioactive molecules identified by HPLC analysis of the CS ethanolic extract.

Retention Time (min)	Compound	Chemical Formula	Class	Amount * (μg/mg of Extract)
2.9	Cyanidin	C15H10O6	Flavonoid	1.30 ± 0.08
3.1	Quinine	C20H24N2O2	Alkaloid	n.d.
4.1	Gallic acid	C7H6O5	Phenolic acid	0.05 ± 0.01
4.5	Neochlorogenic acid	C8H10N4O2	Phenolic acid	0.15 ± 0.01
5.6	Caffeine	C16H18O9	Alkaloid	44.82 ± 1.49
6.6	Caffeic acid	C9H8O4	Phenolic acid	0.08 ± 0.01
6.7	Catechin	C15H14O6	Flavonoid	n.d.
6.9	Chlorogenic acid	C16H18O9	Phenolic acid	0.37 ± 0.02
7.3	Rutin	C27H30O16	Flavonoid	0.82 ± 0.04
7.8	Hyperoside	C21H20O12	Flavonoid	3.17 ± 0.10
8.8	p-Coumaric acid	C9H8O3	Phenolic acid	0.06 ± 0.01
9.0	trans-Ferulic acid	C10H10O4	Phenolic acid	0.94 ± 0.07
10.4	Cinnamic acid	C9H8O2	Phenolic acid	0.25 ± 0.02
10.8	Quercetin	C15H10O7	Flavonoid	2.38 ± 0.17
11.1	Kaempferol	C15H10O6	Flavonoid	1.14 ± 0.07
13.3	Isogentisin	C14H10O5	Flavonoid	n.d.
31.9	α-Tocopherol	C29H50O2	Vitamin	22.60 ± 1.53

* n.d.: not detected.

.

Among the 17 compounds identified, caffeine was the most abundant, at 44.82 ± 1.49 µg/mg extract, followed by α-tocopherol (22.60 ± 1.53 µg/mg) and hyperoside (3.17 ± 0.10 µg/mg). These molecules together account for almost 60% of the total chromatographic area, indicating that both alkaloids and lipid-soluble antioxidants contribute significantly to the bioactivity of silverskin extracts. Other flavonoids, such as quercetin (2.38 ± 0.17 µg/mg) and kaempferol (1.14 ± 0.07 µg/mg), were also present in significant amount.

Polyphenolic acids, such as ferulic, cinnamic, gallic, caffeic, chlorogenic, neochlorogenic, and coumaric acids (about 0.94, 0.25, 0.05, 0.08, 0.37, 0.15, and 0.06 µg/mg, respectively), were present in moderate concentrations but accounted for several significant peaks in the chromatogram.

The retention time distribution of compounds reflects a clear polarity gradient. Early eluting peaks (2–5 min) correspond mainly to hydrophilic organic acids (cyanidin, quinine, gallic acid), while mid-range retention (5–15 min) includes polyphenolic acids and flavonoids of intermediate polarity. Late peaks (>25 min) represent non-polar compounds such as tocopherols. The most intense peak, appearing at 5.6 min (~43% of total area), corresponds to caffeine, which is consistent with its high concentration and strong UV absorbance maximum at 273 nm. Meanwhile, the peaks at 28.0 min and 31.9 min (each about 14% of total area) suggest compounds of lower polarity, such as α-tocopherol and co-eluting flavonoid derivatives, both contributing significantly to the antioxidant potential measured by FRAP and DPPH assays.

3.3.2. GC-MS Analysis

The chromatogram shows two major peaks corresponding to caffeine (retention time 10.59 min, 43.86% relative area) and β-sitosterol (retention time 27.39 min, 33.57% relative area), indicating their predominance among the volatile and semi-volatile constituents (Figure S2 and

Table 3. Summary of the bioactive molecules identified by GC-MS analysis of the CS ethanolic extract.

Retention Time (min)	Compound	Molecular Weight (u.m.a.)	Relative Area (%)
5.03	Benzothiazole	135.186	1.46
5.46	Benzophenone	182.218	0.59
7.79	2-Methyl-1,4-benzenedicarboxaldehyde	148.159	0.36
8.06	2,4-Di-tert-butylphenol	206.324	0.60
9.56	Ethyl 4-ethoxybenzoate	194.227	2.04
9.64	Methyl 2-((6-oxo-5-phenyl-2,3,6,6a-tetrahydro-1,3-methanopentalen-3a(1H)-yl)oxy)acrylate	310.349	1.19
9.76	3,5-Di-tert-butyl-4-hydroxybenzaldehyde	234.334	0.23
10.59	Caffeine	194.191	43.86
10.91	7,9-Di-tert-butil-1-oxaspiro[4.5]deca-6,9-diene-2,8-dione	276.173	1.05
11.29	Diphenyl sulfone	218.040	4.73
13.04	(3-Heptyl)undecyl adipate	398.340	0.89
25.27	α-Tocopherol	430.706	0.85
26.46	Campesterol	400.680	2.35
26.72	Stigmasterol	412.691	1.63
27.39	β-Sitosterol	414.707	33.57
28.05	Anhydrovariabilin	282.290	0.23
28.47	α-Tocopheryl acetate	472.743	0.44

). Alongside these, several minor peaks correspond to phenolic antioxidants, vitamins, aromatic aldehydes, and other sterols, confirming the biochemical complexity of the silverskin extract.

The presence of α-tocopherol (vitamin E) at 0.85% and α-tocopheryl acetate at 0.44% indicates that CS contains lipid-soluble antioxidants capable of protecting the extract from oxidative degradation. Of particular interest is the identification of sterol compounds, campesterol (2.45%), stigmasterol (1.63%), and β-sitosterol (3.37%), which collectively account for nearly 7.5% of the total volatile profile.

Among the less abundant but chemically relevant components were diphenyl sulfone (4.73%) and diisobutyl azelate (1.55%), which may originate from the thermal oxidation of fatty acids or the degradation of complex lipids. Although these are not naturally occurring antioxidants, their presence reflects the complex chemistry of roasted biomass and possible solvent extraction of degradation intermediates. Similarly, the detection of 7,9-di-tert-butyl-1-oxaspiro[4.5]deca-6,9-diene-2,8-dione (1.05%) corresponds to known pyrolysis products of phenolic antioxidants and stabilizers, which may also exhibit residual radical-scavenging activity. The detection of aromatic aldehydes such as 2-methyl-1,4-benzenedicarboxaldehyde and methyl-2-((6-oxo-5-phenyl-2,3,6,6a-tetrahydro-1,3-methanopentalen-3a(1H)-yl)oxy)acrylate suggests the occurrence of complex Maillard and pyrolytic reactions during coffee bean roasting. These compounds are frequently reported in roasted coffee volatiles and contribute to both aroma and antioxidant behavior [41]. Though present in small amounts (<1.5%), their biological significance lies in their radical-trapping activity and ability to modulate oxidative pathways in lipid systems. A relevant set of phenolic derivatives and benzene-based compounds was also identified, including 2,4-di-tert-butylphenol (0.60%), ethyl 4-ethoxybenzoate (2.04%), benzothiazole (1.46%), and benzophenone (0.59%).

In addition to all these compounds, traces of additional both saturated and unsaturated hydrocarbons with varying chain lengths (from C9 to C20) were also detected.

3.4. Cellular Assays

3.4.1. Effects of the CS Extract and H₂O₂ on Cell Viability and ROS Production in a Schwann-like Cell Line

Following the chemical characterization of the ethanolic CS extract, we investigated the relative biological activity using an in vitro model based on the RT4-D6P2T Schwann-like cells. To find optimal conditions of oxidative stress conditions to be used in the subsequent studies, RT4-D6P2T cells were exposed to increasing concentrations of H₂O₂ to induce intracellular ROS production. The protective effects of the extract were evaluated by measuring parameters such as cell viability/metabolic activity, by means of MTT test, and ROS levels, by means of DCFDA assay.

The cytotoxic effect of H₂O₂ at concentrations of 1, 2, and 2.5 mM, was assessed after 1 h and 4 h of treatment. A significant reduction in cell viability was observed, as early as 1 h after treatment, at the lowest concentration tested of 1 mM, compared to the untreated control (Figure 2a). Based on this result, subsequent experiments were designed to investigate the oxidative stress induced by 500 µM and 1 mM H₂O₂ administered over 1 h at time steps of 15 min. A mild but statistically insignificant increase in ROS levels was detected with 500 μM H₂O₂, whereas 1 mM H₂O₂ induced rapid and marked oxidative response that was quite constant over the whole experiment time (Figure 2b).

The cytotoxic effect of the CS extract was assessed following 24 h of treatment at concentrations of 10, 25, 50, and 100 μg/mL. No significant changes in cell viability were observed with concentrations lower than 100 μg/mL (Figure 3a). Then, 10 μg/mL and 50 μg/mL concentrations were selected for subsequent experiments to evaluate the ROS generated over 1 h by the extract. No significant increase in ROS production was observed with 10 μg/mL CS extract, whereas it was detected with 50 μg/mL (Figure 3b). Based on these results, 10 μg/mL of CS extract was selected for further experiments.

3.4.2. CS Extract Exerts Antioxidant Properties on a Schwann-like Cell Line

ROS production was evaluated at 30, 45, 60, and 75 min following treatment of RT4-D6P2T cells with 10 μg/mL of CS extract, 1 mM H₂O₂, and their combination (CS extract + H₂O₂). The oxidative stress response was expressed as a percentage relative to the H₂O₂-treated condition (set as 100%). Overall, treatment with the extract alone resulted in lower ROS levels compared to the untreated control, suggesting an enhanced antioxidant effect that might ameliorate cellular readiness to respond to H₂O₂ exposure. In fact, co-treatment with CS extract and H₂O₂ led to a marked reduction in ROS levels relative to H₂O₂ alone, indicating a robust protective effect of the extract against oxidative stress. This trend was confirmed at all time points analyzed (30, 45, 60, and 75 min), hinting that the extract exerts a sustained protective effect against H₂O₂-induced oxidative stress (Figure 4).

3.4.3. Caffeine Contribution to the Antioxidant Activity of the Extract

At this stage, we aimed to evaluate whether the observed effects of the CS extract were primarily due to its most abundant compound, caffeine, or to the combined action of the full phytochemical profile as reflected in the chromatographic fingerprint.

ROS levels were evaluated at 30, 45, 60, and 75 min following the treatment of RT4-D6P2T cells with 10 μg/mL of CS extract, 2.34 µM caffeine, 1 mM H₂O₂, and their combination (CS extract + H₂O₂; caffeine + H₂O₂). The 2.34 µM concentration of pure caffeine was chosen because it corresponds to approximately 100 times the concentration of caffeine administered within the 10 μg/mL extract (based on the HPLC analysis showing a concentration of ~45 μg/mg of dry sample). This elevated concentration was used to ensure the observation of a significant antioxidant effect for the pure compound. The oxidative stress response was expressed as a percentage relative to the H₂O₂-treated condition (set as 100%). Overall, the antioxidant effects provided by the CS extract in this cell model were found to be comparable to those of the 100-times concentrated pure caffeine, suggesting a synergistic effect involving caffeine and the other bioactive molecules present in the extract (Figure 5).

3.4.4. CS Extract Counteracts Lipopolysaccharide (LPS)-Induced ROS Production

LPS, a potent endotoxin derived from the outer membrane of Gram-negative bacteria, is known to trigger a significant inflammatory response, leading to an increased production of ROS and subsequent oxidative stress [42,43,44]. To further evaluate the antioxidant potential of the CS extract, we treated immortalized rat Schwann cell lines with LPS (0.5 µg/mL) for 30, 45, and 60 min to induce ROS production and assessed whether co-treatment with CS could mitigate this effect. We chose this LPS concentration based on published data; for instance, Li and colleagues established that concentrations of LPS below 10 µg/mL were appropriate to induce inflammation, without negatively affecting cell viability in RSC96 rat Schwann cells [45]. Intracellular ROS levels were quantified using the DCFDA probe. We observed that cells co-treated with 10 μg/mL CS exhibited a statistically significant decrease in LPS-induced ROS production compared to cells treated with LPS alone (p < 0.0001) (Figure 6).

The antioxidant activity of CS extracts varied significantly depending on the extraction solvent, temperature, and duration. Ethanolic extracts exhibited the highest antioxidant capacity, particularly under moderate extraction conditions (20 °C). For instance, the DPPH radical scavenging capacity reached 156.50 mg AAE/g at 30 min, while TEAC and FRAP values were also notably high (275.48 mg TE/g and 100.64 mg AAE/g, respectively), albeit after longer extraction times (60 min).

Conversely, water extracts displayed significantly lower antioxidant values across all three tests (e.g., approximately 8 mg AAE/g for DPPH, between 40 and 50 mg TE/g for TEAC, and about 9 mg AAE/g for FRAP, all at 20 °C), despite yielding a higher total mass extract. The hydroalcoholic solvent system provided intermediate scavenging values, although these were much closer to the results obtained by aqueous solvent than to those of pure ethanol. For instance, at 20 °C, TEAC activity peaked at approximately 81 mg TE/g, FRAP about 21 mg AAE/g, and DPPH about 9 mg AAE/g. Therefore, the mild condition of 20 °C and 30 min was selected as optimal for balancing compound recovery and chemical integrity. These conditions were subsequently used for all chromatographic and biological characterization studies.

4. Discussion

The initial phase of the research focused on the optimization of the solvent used during the ultrasound-assisted extraction, maximizing the yield and efficacy of the final extract. The procedure employed for the extraction reflects modern green chemistry principles, offering enhanced extraction kinetics while minimizing solvent usage and energy consumption compared to conventional methods.

The investigation into solvent polarity (water, ethanol, and 50% ethanol in water) and temperature (20 °C and 40 °C) directly addresses the physiochemical nature of the target molecules. As the major phenolic constituents in CS are known to be derivatives of chlorogenic acids (CGAs), which exhibit amphiphilic properties, the observation that the ethanol extract likely yielded the highest content of phenols (TPC) and flavonoids (TFC) is chemically sound. This optimal ethanolic solution provides a balanced polarity environment, effectively disrupting the matrix and solubilizing a broader spectrum of both lipophilic (e.g., certain flavonoids and volatile compounds) and hydrophilic (e.g., highly glycosylated phenolics) compounds. Water alone and its combination with ethanol may have limited efficacy due to the non-polar domains of the CS matrix. These findings are consistent with the broader literature on CS. For example, Costa et al. found that a 50:50 hydroalcoholic solvent was a practical compromise to maximize antioxidant recovery in conventional solid–liquid extractions [46]. Likewise, several recent valorization studies report that hydroalcoholic and aqueous ultrasonic procedures permit high total recovery while remaining compatible with food applications [47].

Similarly, the moderate elevation of temperature to 40 °C served to evaluate compound solubility and diffusion kinetics without risking the thermal degradation of sensitive phenolic molecules, such as certain flavanols or CGA isomers. An apparent decline in extractable free phenolics under slightly harsher conditions was indeed observed and is plausible because higher temperatures can both increase extraction kinetics and promote (i) thermal degradation of free chlorogenic acids and other hydroxycinnamates, as well as (ii) polymerization or binding to melanoidins produced during roasting, which reduces Folin–Ciocalteu-reactive free phenolics, as previously demonstrated by Narita & Inouye [16]. Conversely, some high-temperature techniques (e.g., subcritical water and pressurized liquid extraction) can effectively solubilize bound or high-molecular-weight antioxidant species, resulting in high ORAC/DPPH responses; however, these techniques frequently yield different molecular profiles [16,48].

With increasing temperature (40 °C), TPC decreased more sharply than TFC in ethanol extracts, suggesting thermal sensitivity of certain non-flavonoid polyphenols such as chlorogenic acids, which are known to degrade at elevated temperatures. Flavonoids, in contrast, appeared more stable under moderate heat, maintaining relatively consistent concentrations. This behavior may reflect structural differences: many flavonoids possess more rigid carbon ring frameworks that resist thermal decomposition better than esterified phenolic acids.

The strong correlation expected between the determined TPC/TFC values and the observed antioxidant capacities across the three complementary assays (DPPH, TEAC, and FRAP) confirms the primary role of polyphenols in the extract’s biological function. These assays collectively assess both radical scavenging and reducing abilities, allowing a comprehensive comparison of solvent efficiency and thermal effects. Ethanol proved again to be the most effective solvent in extracting phenolic compounds with high antioxidant potential, followed by the 50% ethanol/water mixture and, lastly, water. This pattern was consistent across all assays, i.e., DPPH, TEAC, and FRAP, and aligns with previous reports highlighting the superior extraction efficiency of mid-polar solvents for polyphenols from coffee by-products [49]. The predominance of ethanol can be attributed to its intermediate polarity, which effectively solubilizes chlorogenic acids, caffeic acid derivatives, and other polyphenolic constituents characteristic of coffee residues [50].

The discrepancy between high yield and low antioxidant capacity of the water extracts reflects the lower solubility of key polyphenols in purely aqueous media, where non-polar or weakly polar phenolic compounds are less efficiently extracted. The limited antioxidant activity of water extracts is consistent with prior findings for coffee silverskin and other coffee by-products [51]. While the hydroalcoholic mixture supports the well-documented synergistic effect of facilitating the extraction of both hydrophilic and lipophilic phenolic compounds, the resulting antioxidant effect was notably lower compared to pure ethanol [52].

While increasing the temperature from 20 °C to 40 °C slightly enhanced the overall yield, the TPC, TFC, and the antioxidant capacity were all lower. This decline is likely due to the potential degradation of thermolabile antioxidants upon prolonged exposure to higher temperatures [53]. On the other hand, the influence of extraction time appeared modest in magnitude; although longer times sometimes yielded marginally better results, the increase was not significant enough to warrant the extended duration.

The DPPH (radical scavenging, hydrogen atom transfer) and FRAP (reducing power, single electron transfer) assays interrogate different mechanistic pathways of antioxidant action. A high rating in all three assays suggests that the ethanol CS extract contains a complex mixture of compounds capable of acting through multiple mechanisms, providing superior, broad-spectrum protection compared to single-mechanism antioxidants, and it was thus chosen for the subsequent chemical and cellular characterization.

The HPLC profile of the CS ethanolic extract demonstrated a rich phenolic composition. From a functional perspective, the quantitative data suggest that caffeine and tocopherol are the major contributors to total antioxidant capacity in this extract, although polyphenolic compounds such as hyperoside, quercetin, and chlorogenic acids likely play synergistic roles. Caffeine’s predominance is expected given its chemical stability and high solubility in ethanolic media. Its extraction behavior mirrors that observed in roasted coffee beans, where caffeine typically represents 1% to 2% of dry matter and is among the most efficiently recovered compounds during mild ethanol-assisted extraction [54]. α-Tocopherol (vitamin E) represents another key bioactive molecule. Its relatively late elution is consistent with its hydrophobic nature, suggesting its co-elution with non-polar lipid-soluble components. The presence of tocopherol in the silverskin matrix is consistent with reports by del Castillo et al., who found significant quantities of lipid-soluble antioxidants in roasted coffee by-products [55].

Flavonoids were also prominent in the chromatographic profile. Hyperoside (quercetin-3-O-galactoside, 3.17 ± 0.10 µg/mg), quercetin (2.38 ± 0.17 µg/mg), and kaempferol (1.14 ± 0.07 µg/mg) were the most abundant. These compounds exhibit strong free radical scavenging activity due to their catechol moieties and conjugated double bonds, which facilitate electron transfer and metal chelation [56]. The detection of hyperoside is particularly relevant, as it has been previously identified in coffee pulp and husk but rarely quantified in silverskin.

Among the phenolic acids, chlorogenic acid (0.37 ± 0.02 µg/mg) and neochlorogenic acid (0.15 ± 0.01 µg/mg), both hydroxycinnamate esters of quinic acid, are well-known for their potent antioxidant and metal-chelating properties [57]. The relatively low abundance observed here compared with raw coffee beans is likely due to thermal degradation during roasting, as silverskin is exposed to high temperatures (>200 °C) in the roasting drum. Nonetheless, their persistence indicates partial thermal stability and their contribution to the overall antioxidant potential of the extract. The co-detection of caffeic acid (0.08 ± 0.01 µg/mg) further supports this, as it may result from partial hydrolysis of chlorogenic acid esters. Additionally, the presence of rutin (0.82 ± 0.04 µg/mg) and p-coumaric, ferulic, and cinnamic acids (each <1 µg/mg) enriches the antioxidant profile through complementary redox mechanisms. Hydroxycinnamic acids such as trans-ferulic acid are known for their UV-absorbing and radical-stabilizing properties, suggesting potential applications of silverskin extracts in cosmetic and food preservation formulations [58]. The detection of isogentisin, a xanthone derivative, although below the quantification limit in this analysis, further supports the presence of minor antioxidant components typical of roasted coffee matrices.

Several studies have emphasized the synergistic antioxidant mechanisms between phenolics and alkaloids in coffee by-products, wherein caffeine stabilizes free radicals and regenerates phenolic antioxidants [59]. The concurrent presence of multiple antioxidant classes—phenolic acids, flavonoids, xanthones, and lipophilic vitamins—may explain the robust antioxidant performance observed across all assays (DPPH, TEAC, FRAP) in the CS extracts. The relatively low concentrations of certain compounds (e.g., gallic acid and chlorogenic acid) compared to other coffee residues may reflect matrix effects and roasting-related transformations. During roasting, Maillard reaction products and melanoidins can bind phenolic acids, reducing their extractability but not necessarily their antioxidant potential, as these complexes may still exhibit radical-scavenging properties [60,61]. Therefore, the quantified compounds represent only part of the total antioxidant matrix.

The most abundant compound identified by GC-Ms was caffeine, accounting for nearly 44% of the total volatile composition, in perfect agreement with the HPLC data [49]. The analysis also revealed compounds likely formed during the roasting process, such as melanoidins, which also possess significant antioxidant capacity, and potentially furan derivatives, which may contribute to the characteristic aroma profile. In addition, tocopherols, which play a key role in scavenging lipid radicals and stabilizing unsaturated fatty acids [62], were also detected in the ethanolic extract. This finding corroborates earlier reports by Campos-Vega et al., who observed tocopherols among the lipid-soluble antioxidants in coffee residues and silverskin extracts [54].

Phytosterols are known for their cholesterol-lowering effects and anti-inflammatory properties [63]. Their detection in CS indicates the potential nutritional and pharmacological value of this by-product, extending its use beyond antioxidant supplementation to functional food formulation. The co-existence of tocopherols and phytosterols is noteworthy, as both act synergistically in lipid protection systems, enhancing the stability and bioefficacy of extracts [64].

Phenolic derivatives typically arise from the degradation of lignin, chlorogenic acids, and other polyphenols during roasting [42]. Many of these phenolic derivatives, particularly 2,4-di-tert-butylphenol, are known for their strong antioxidant and antimicrobial activity [43]. The presence of such compounds indicates that the alcoholic extract retains both the natural phenolic antioxidants and the thermally derived secondary products that may contribute to overall radical scavenging activity, emphasizing the potential valorization of CS as an ingredient in nutraceutical, cosmetic, and food formulations.

When compared to the HPLC analysis of non-volatile compounds in the same extract, a complementary pattern emerges. While HPLC detected polar phenolic acids and flavonoids, GC-MS revealed more lipophilic and thermally stable molecules, e.g., tocopherols, sterols, and phenolic derivatives. This duality underscores the chemical heterogeneity of silverskin and supports the notion that its antioxidant potential arises from both hydrophilic and lipophilic molecules. Such findings agree with those of Bresciani et al., who reported that CS contains both classes of antioxidants, contributing to sustained radical inhibition across different polarity systems [65].

The investigation into the cellular effects of the optimal CS extract shifts the study from chemical characterization to practical biological relevance. Several scientific studies have tested the biological activity and neuroprotective effect of CS extracts on neural and microglia models [66,67]. To the best of our knowledge, there are no published data addressing the biological effects on an in vitro model of Schwann cells. For these reasons, we decided to conduct a preliminary exploration in this sense. The MTT assay result, indicating “minimal cytotoxicity” and high cell viability, is paramount for the potential commercial application of the CS extract in food, cosmetic, or therapeutic contexts.

Oxidative stress was then assessed at short time points (5, 15, 30, and 60 min) following exposure to the pro-oxidant agent due to biological and experimental considerations: ROS generation is a rapid and transient event, typically occurring within minutes or hours after the oxidative insult. Oxidative species are characterized by short half-lives and are quickly scavenged by intracellular antioxidant systems (e.g., glutathione, catalase, superoxide dismutase). Consequently, the primary cellular events triggered by oxidative stress—including DNA damage, mitochondrial dysfunction, and activation of redox-sensitive pathways (e.g., Nrf2, MAPK, caspases)—take place within the early phases of exposure. Moreover, short-term evaluation helps to avoid confounding effects associated with non-specific toxicity observed in prolonged exposures. The use of short exposure times also reflects the early events that characterize acute oxidative stress in neurodegenerative or demyelinating diseases, where Schwann cells are rapidly affected by toxic, inflammatory, or metabolic insults.

The fact that the extract reduced the DCFDA fluorescence signal following an oxidative insult (induced by either H₂O₂ or LPS administration) confirms that the bioactive compounds are successfully internalized by the glial cells and remain functional in the complex intracellular environment. These results are consistent with those published by Nzekoue and colleagues, who demonstrated for the first time that CS extract exhibits neuroprotective activity against H₂O₂-induced oxidative damage, with methanol and ethanol/water (70:30) extracts effectively preserving cell viability and counteracting oxidative stress [31]. To determine whether the observed antioxidant activity was primarily attributable to the major component of the CS extract, comparative experiments were conducted using pure caffeine. The results demonstrated that a 100-fold higher concentration of caffeine than that found in the extract was required to achieve a comparable antioxidant effect. This finding highlights a significant synergistic contribution from other phytochemical constituents, which exert potent effects despite their relatively low abundance. These results are consistent with recent literature supporting the protective role of caffeine in the context of neurodegenerative diseases, both in neural cell and animal models, especially if combined with other antioxidant molecules [59,68,69,70].

5. Conclusions

This study evaluated different conditions, i.e., solvent, temperature, and time, for extracting bioactive molecules from CS residues from coffee production. Temperature and time were shown to only slightly affect the polyphenolic content. On the other hand, solvents influenced significantly the type of biomolecules extracted. To summarize, although the hydroalcoholic extraction (~50% ethanol) under mild conditions gave the highest yield, pure ethanol was the best solvent to obtain the most enriched polyphenolic fraction (high TPC per g extract). Considering that our future aim is to exploit that industrial waste for potential nutraceutical purposes, the ethanolic extract was then used, and to maintain the highest quality of extracted compounds, milder conditions (20 °C and 30 min) were kept.

The findings of this study robustly establish CS, a principal leftover of the coffee roasting industry, as a viable and potent source of bioactive compounds, such as polyphenols, sterols, and vitamins. The overarching rationale for this investigation—the valorization of industrial waste into functional ingredients—is reinforced by the compelling chemical and biological data generated. The optimized procedure yielded an extract with potent, multi-mechanistic antioxidant activity, whose effects were directly supported by the chemical fingerprint (HPLC and GC-MS) and validated by superior safety and efficacy in a Schwann-like cellular model. However, for industrial scale-up, modern green technologies (e.g., PLE, SWE or supercritical CO₂ with co-solvent) could be used to shorten times and improve yields, but they will change the phytochemical speciation and must be optimized to obtain the desired amount of polyphenols, vitamins, sterols, and alkaloids [47,48], and this will be the objective of a future study.

Future work must also focus on the translational aspects. This includes conducting in vivo studies to confirm bioavailability and protective effects in animal models of oxidative stress-related diseases, defining the specific mechanisms of action (e.g., Nrf2 pathway modulation), and developing stable, high-quality formulations suitable for the food and nutraceutical market. Nevertheless, the findings here solidify CS’s position not merely as a waste product, but as a rich, untapped resource for bio-based pharmaceutical and nutritional ingredients.