Chemical Class–Driven Polyphenolic Profiles Shape In Vitro Regenerative Activity of Four Medicinal Plants Relevant to Burn Wound Healing

Abstract

Background: Burn wound repair is driven by oxidative balance and keratinocyte regeneration. Polyphenol-rich botanicals are considered promising due to combined antioxidant and pro-regenerative properties. This study compares four ethnopharmacologically relevant species—Boswellia serrata (BS), Sambucus nigra (SN), Ocimum basilicum (OB), and Galium verum (GV)—to determine how their polyphenolic class profiles relate to in vitro regenerative activity. Methods: Ethanolic (E—99.5%) and hydroalcoholic (H—70%) extracts were profiled by LC–MS, total polyphenol content (TPC), and DPPH assays. Biological effects were assessed in HaCaT keratinocytes using Alamar Blue (24/48 h) and scratch wound closure (24 h), and results were correlated with chemical profiles. Results: The H extract of OB (OB-H) and of GV (GV-H) had the highest TPC (62.6 and 63.9 mg GAE/g) and lowest DPPH IC₅₀ (18.7 and 17.1 μg/mL), aligning with the strongest biological responses—HaCaT viability up to 169.1% and wound closure up to 414%. SN extracts, dominated by rutin, promoted moderate migration with preserved viability, whereas BS produced modest viability gains. Conclusions: Polyphenolic composition—particularly the dominance of phenolic acids—correlates strongly with in vitro regenerative responses in HaCaT keratinocytes. O. basilicum and G. verum hydroalcoholic extracts displayed the most favorable profiles.

1. Introduction

Burn wound repair is a dynamic and multistage process involving hemostasis, inflammation, cell proliferation, and tissue remodeling, each coordinated by specific molecular and cellular interactions. While these phases ensure restoration of skin integrity, their dysregulation often leads to delayed healing, chronic wounds, or pathological scarring [1,2,3]. Thermal injury rapidly activates resident skin and immune cells, triggering the release of pro-inflammatory mediators such as cytokines, prostaglandins, and leukotrienes. This vascular reaction promotes endothelial activation, neutrophil infiltration, and oxidative stress through NF-κB and MAPK signaling pathways [4,5,6]. Although these events are essential for pathogen clearance, excessive or prolonged inflammation can intensify tissue damage and disrupt the transition toward the proliferative phase [4,5,6]. Their relatively low cost, accessibility, and favorable safety profiles further increase their relevance compared with synthetic drugs [7,8,9]. Among the species traditionally used in southern Romania, four stand out due to their ethnopharmacological importance and reported biological activities. Boswellia serrata (BS) is rich in boswellic acids, triterpenoids known for strong anti-inflammatory activity via 5-lipoxygenase inhibition [10]. Sambucus nigra (SN) flowers contain rutin and related flavonoids, associated with antioxidant and collagenase-inhibitory effects [11]. Ocimum basilicum (OB) is a phenolic acid species, particularly abundant in rosmarinic acid, that has shown keratinocyte-stimulating and antioxidant properties [12,13,14,15]. Galium verum (GV) contains rosmarinic acid and glycosylated flavonoids such as isoquercitrin and quercitrin. The alcoholic extract of GV was reported not to exert any cytotoxic activity against healthy keratinocytes [16], whereas the decoction preparation was shown to protect the mucosal keratinocytes against toxic DNA damage [17]. Medicinal plants have historically been employed in wound care owing to their antioxidant, anti-inflammatory, antimicrobial, and tissue-supportive properties [18].

Despite evidence from individual studies describing antioxidant or anti-inflammatory potential for these plants, no comparative analysis has examined how differences in their polyphenolic composition relate to keratinocyte responses in vitro. This study addresses this gap by characterizing and comparing the polyphenolic profiles, antioxidant capacity, and in vitro regenerative effects of two different extracts from BS, SN, OB, and GV.

2. Results

2.1. Total Polyphenolic Content (TPC)

In the present study, aqueous ethanol was considered the most appropriate solvent for extraction of polyphenols from different plant extracts.

The TPC of the tested ethanolic (E) and hydroalcoholic (H) extracts ranged from 31.3 to 63.9 mg of gallic acid equivalents (mg GAE)/g of dry extract (

Table 1. Antioxidant activities and total phenolic content of the extracts. Values are expressed as mean ± standard deviation (SD) of duplicate experiments.

Plant Extract	Extraction Solvent	* IC50 (µg/mL)	* Total Phenolic Content (mg GAE/g Extract)
Boswellia serrata	EtOH 99.5%	3244.0	46.0
	EtOH 70%	1907.0	47.6
Sambucus nigra flowers	EtOH 99.5%	46.1	35.8
	EtOH 70%	27.3	35.5
Ocimum basilicum	EtOH 99.5%	43.5	50.8
	EtOH 70%	18.7	62.6
Galium verum	EtOH 99.5%	45.7	31.3
	EtOH 70%	17.1	63.9
Ascorbic acid	EtOH 99.5%		3.3

* IC₅₀, concentration needed to reduce the absorbance of the DPPH by 50%.; GAE, gallic acid equivalent.

). The H extract of GV (GV-H) and the H extract of OB (OB-H) displayed the highest content of 63.9 mg GAE/g and 62.6 mg GAE/g respectively, while the E extract of GV (GV-E) showed the lowest TPC of 31.3 mg GAE/g.

2.2. Polyphenolic Composition by Chemical Class

The LC–MS analysis of E and H extracts from BS (BS-E and BS-H), from SN (SN-E and SN-H), OB and GV was previously reported in full detail by Rosca et al., 2025 [19].

Class dominance was assigned based on the summed quantitative contribution of compounds belonging to each class (phenolic acids vs. flavonoids) within the targeted LC-MS panel and visualized here as a qualitative dominance map to facilitate comparison across extracts.

Here, the data were reorganized by chemical class and used for comparative qualitative assessment of the predominant polyphenolic profiles (Figure 1). Based on this qualitative dominance mapping, BS-E and BS-H exhibited a mixed polyphenolic profile, with comparable contributions of flavonoids (F) and phenolic acids (PA). In contrast, both ethanolic and hydroalcoholic extracts of Sambucus nigra displayed a flavonoid-dominant profile (F > PA), consistent with the prevalence of flavonoid-type constituents reported previously. Ocimum basilicum (OB-E and OB-H) and Galium verum (GV-E and GV-H) extracts were classified as phenolic acid-dominant (PA > F), reflecting the predominance of phenolic acid-type compounds within their polyphenolic profiles.

2.3. Determination of Antioxidant Activity

The H extracts showed stronger radical-scavenging activity than E extracts (Table 1). The highest antioxidant capacity was recorded for GV-H (IC₅₀ = 17.1 μg/mL) and OB-H, while BS-E had the lowest (IC₅₀ = 3243.98 μg/mL). Dose–response curves for the DPPH assay are provided in the Supplementary Materials (Figure S1).

2.4. Cell Viability

Following a 24 h treatment period, the effect of BS-E, BS-H, SN-E, SN-H, OB-E, OB-H, GV-E and GV-H on human immortalized keratinocytes (HaCaT) cell viability was evaluated. At 24 h, HaCaT cell viability significantly increased after treatment with BS-H (113.3 ± 14.5), SN-H (128.8 ± 15.2), OB-E (127.7 ± 13.0), OB-H (143.2 ± 19.8) and GV-H (123.0 ± 7.8) at 1000 μg/mL. At 48 h, treatment increased the proliferative effect of compounds BS-H, SN-H, OB-E, OB-H, GV-E and GV-H on HaCaT cells. For example, cell viability of HaCaT cells treated for 48 h with OB-H was 169.1 ± 9.3 vs. control (100%) (Figure 2).

At 48 h, the stimulatory effects on viability became more pronounced. BS-H increased viability to 138.2 ± 11.3%, SN-H to 144.6 ± 8.8%, OB-E to 161.0 ± 12.8%, and OB-H to the highest value recorded, 169.1 ± 9.3%. Both GV-E and GV-H also promoted proliferation, reaching 146.5 ± 11.6% and 166.4 ± 9.3%, respectively. In contrast, BS-E and SN-E had lower effects on viability at both time points (Figure 3).

Overall, a clear pattern was observed in which H extracts tended to produce higher viability values than their ethanolic counterparts, with OB-H and GV-H producing the most substantial increases.

2.5. In Vitro Evaluation of Proliferative and Migratory Activities

The wound closure analysis after 24 h treatment revealed that all extracts, except BS-E, significantly enhanced the migratory capacity of HaCaT keratinocytes compared to the untreated control (118 ± 9% closure) (Figure 4 and Figure 5). The most pronounced effect was observed for OB-H, achieving 414 ± 15% closure, followed by OB-E (333 ± 22%). GV-H reached 303 ± 17% and GV-E 246 ± 13%. SN-H induced 296 ± 18% closure, while SN-E reached 208 ± 25%. BS-H showed a comparable effect to SN-E (208 ± 25%), whereas BS-E remained close to control levels. Overall, H extracts demonstrated a stronger stimulatory effect on keratinocyte migration than ethanolic counterparts, with OB-H consistently outperforming all other treatments. A quantitative comparison of wound closure at 24 h for the highest tested concentration (1000 µg/mL) is shown in Figure 6.

Images were acquired immediately after scratching (0 h) and after 24 h of incubation with the indicated plant extract conditions (Control and BS-E, BS-H, SN-E, SN-H, OB-E, OB-H, GV-E and GV-H) (Figure 4). The red line marks the measured wound width at each time point. Scale bar = 100 μm.

Phase-contrast micrographs taken at 0 h and 24 h illustrate the wound gap used for quantitative measurements (red line). Scale bar = 100 μm (Figure 5).

3. Discussion

The TPC values obtained for GV-H and OB-H were among the highest reported for these species, and higher than some literature values, suggesting that the extraction method and solvent polarity strongly influence polyphenol recovery. Similar trends have been reported by Faur et al. 2020 [20], where aqueous ethanol extraction yielded higher TPC than absolute ethanol for similar plant materials.

In parallel with their higher TPC, GV-H and OB-H exhibited pronounced antioxidant and regenerative responses, whereas SN extracts showed moderate effects and BS extracts limited activity. The lower IC₅₀ values observed for hydroalcoholic extracts are likely related to enhanced recovery of polar phenolic compounds in more polar solvents [20]. These observations can be discussed in the context of literature-reported antioxidant and inflammation-modulating pathways associated with the major phytochemicals (Figure 7), without implying direct mechanistic validation in the present study [21].

3.1. Correlation Between Polyphenolic Composition and In Vitro Activity

LC–MS profiling of the E and H extracts highlighted distinct patterns in polyphenolic composition, which showed parallel trends with their in vitro antioxidant and regenerative performance (Figure 1). Extracts classified as phenolic acid-dominant (PA > F), including OB-H and GV-H [20,22], exhibited the strongest antioxidant activity (lowest IC₅₀ values) and the highest keratinocyte proliferation and migration rates [16,20]. A flavonoid-dominant profile (F > PA) characterized SN extracts which showed moderate wound closure enhancement with stable cell viability [12,23]. BS extracts displayed a more balanced composition of phenolic acids and flavonoids (F ≈ PA), notably gallic acid and quercitrin, and induced modest proliferative effects, aligning with their primarily anti-inflammatory profile as reported by Siddiqui MZ et al. rather than direct regenerative stimulation [24]. The lower concentrations used in antioxidant assays (<100 μg/mL) compared to those applied in cell-based assays (up to 1000 μg/mL) reflect methodological differences, as DPPH detects radical-scavenging at low doses [21], whereas cellular responses require higher extract concentrations to elicit measurable effects.

Overall, the relative dominance of phenolic acids was associated with superior antioxidant capacity and regenerative outcomes in vitro, indicating that chemical class profiles may help prioritize extracts for further investigation [25].

3.2. Boswellia serrata

Compared to the other plants analyzed, BS extracts displayed the weakest antioxidant and regenerative activity, consistent with their mixed polyphenolic profile (F ≈ PA).

This may partly reflect assay specificity, as DPPH captures direct radical-scavenging capacity, whereas boswellic acids are mainly associated with inflammation-modulatory effects rather than strong radical-scavenging under DPPH conditions.

These findings support previous evidence by Tsai et al. (2022), who showed that α-boswellic acid reduced epidermal thickening and inflammatory cytokine production in atopic dermatitis models but did not enhance keratinocyte proliferation [26]. Similarly, Ranzato et al. (2017) demonstrated that boswellic acids modulate MMP-9 activity in an IL-1α–dependent context, exerting dose-dependent anti-inflammatory effects rather than directly stimulating cell migration [27]. Comparable anti-inflammatory and modulatory effects of pentacyclic triterpenes and triterpenoids have been reported in other in vitro systems, supporting their role as inflammation-modulating rather than strongly pro-proliferative agents [28,29].

Our results contribute to the existing literature by showing that BS extracts maintain keratinocyte cytocompatibility but display limited pro-regenerative activity in our in vitro assays. This functional profile is consistent with a predominantly modulatory, non-proliferative effect, suggesting that such extracts may complement polyphenol-rich, more proliferative plant extracts such as OB or GV when considered for topical formulation strategies.

3.3. Ocimum basilicum

Among the studied plants, OB-H demonstrated the strongest regenerative activity on keratinocytes. This observation is consistent with Faur et al. (2020), who showed a dose-dependent proliferative effect of OB-H on HaCaT cells, attributed largely to rosmarinic acid and flavonoid glycosides [20]. Antonescu et al. (2021) further confirmed the regenerative effect in a co-formulation with Trifolium pratense, where basil extract enhanced wound closure in both keratinocytes and fibroblasts [30].

Our study extends these observations by showing that OB extracts, classified as phenolic acid-dominant (PA > F), preserve cell viability and enhance keratinocyte migration and wound closure in an in vitro scratch assay. These effects are consistent with literature reports describing the antioxidant and inflammation-modulating properties of rosmarinic acid, including interactions with NF-κB–related pathways. Overall, these findings support the potential incorporation of OB extracts into topical formulations designed to promote wound repair.

3.4. Galium verum

Galium verum extracts, particularly GV-H, demonstrated notable regenerative activity on keratinocytes, comparable to that observed for Ocimum basilicum. These results are in line with Semenescu et al. (2023), who demonstrated a dose-dependent stimulatory effect of GV fractions on keratinocyte viability, particularly in ethyl acetate extracts [31]. The phenolic acid-dominant profile of GV extracts (PA > F), together with the presence of constituents such as rosmarinic acid, isoquercitrin and quercitrin, is consistent with their reported antioxidant and pro-regenerative properties [32].

Our findings expand current knowledge by showing that GV extracts perform strongly in an in vitro keratinocyte model, indicating their potential to support cell migration. These effects are consistent with literature reporting antioxidant and cytoprotective properties of their major constituents, although such mechanisms were not assessed in this study. Although less studied than OB, the comparable activity observed here positions GV as a promising alternative or complementary candidate for topical formulations aimed at supporting wound repair.

3.5. Sambucus nigra

SN flower extracts demonstrated moderate regenerative activity, maintaining keratinocyte viability and inducing measurable though less pronounced effects on cell migration compared to basil and GV. These observations are consistent with Mota et al. (2020), who described strong antioxidant and collagenase-inhibitory properties of elderflower extracts, mainly attributed to flavonoids such as rutin and isoquercitrin [14], and with Palomino et al. (2021), who confirmed their cytocompatibility with human keratinocytes [23].

Our findings add new evidence by demonstrating that SN extracts are also active in the wound model, where their activity appears to be associated primarily with extracellular matrix stabilization rather than direct proliferative stimulation. While less potent than basil or bedstraw in promoting cell migration, SN may contribute to wound healing through supportive roles, limiting matrix degradation and maintaining tissue integrity during repair.

3.6. Limitations of the Study

This study focused on the polyphenolic fraction of the selected plant extracts, as determined by LC–MS analysis. Other classes of bioactive compounds naturally present in these species—such as terpenoids, essential oils, or polysaccharides—were not profiled and may also contribute to the observed biological effects.

Moreover, the LC-MS analysis was based on a predefined, targeted panel of polyphenolic standards, and compounds present below the limits of detection or lacking available reference standards could not be quantitatively assessed.

The extraction procedure employed a single hydroalcoholic solvent composition (70% ethanol), selected based on its common use in the literature rather than on a systematic solvent optimization within the present study [20,33]. Consequently, the influence of different ethanol–water ratios on extraction yield, polyphenolic recovery, or biological activity was not evaluated.

Extraction yields (% w/w) were not determined, which limits direct comparison with other studies and may affect reproducibility assessments.

In addition, no assessment of batch-to-batch phytochemical consistency was performed, which would be necessary to confirm reproducibility of the chemical profile over time and across different plant material sources. Future analytical work should integrate comprehensive phytochemical profiling and reproducibility testing to capture the full spectrum of active constituents and ensure consistency for potential therapeutic applications.

Antioxidant capacity was assessed only by the DPPH assay, which primarily reflects direct radical scavenging and does not capture other antioxidant mechanisms; complementary assays (e.g., ABTS or FRAP) were not performed within the scope of this study.

Furthermore, the in vitro keratinocyte model used in this study does not replicate the complex multicellular interactions that occur in actual burn wounds, where fibroblasts, endothelial cells, and immune cells dynamically contribute to inflammation, angiogenesis, and tissue remodeling.

Finally, we did not evaluate the local metabolism or cutaneous bioavailability of the identified compounds. For example, rosmarinic acid can undergo hydrolysis to caffeic and ferulic acids within the skin, which may alter its antioxidant and regenerative activity. Flavonoids such as rutin and isoquercitrin are known to have limited penetration through the stratum corneum, potentially restricting their in vivo efficacy, while lipophilic boswellic acids may penetrate more deeply into dermal layers. Without experimental data on cutaneous permeation and metabolic transformation, it remains uncertain which phytochemicals—or their metabolites—are ultimately responsible for the therapeutic effects observed in vivo.

Topical efficacy is strongly formulation dependent and relies on achieving adequate cutaneous delivery of active constituents. Recent work highlights that advanced transdermal delivery approaches—supported increasingly by AI-based predictive modeling of skin penetration and release kinetics—can inform the selection of carriers, permeation enhancers, and microneedle/stimuli-responsive platforms to improve cutaneous delivery and enable more controlled local exposure profiles [34].

Implementing such an integrated validation framework in a single experimental cycle generally requires substantial resources, specialized infrastructure, and dedicated funding; as a result, individual studies may prioritize selected endpoints, while more extensive, multi-assay and translational programs can be addressed in subsequent research.

4. Materials and Methods

4.1. Plant Material and Extraction

The plant materials and extraction protocol were previously described in detail by Rosca et al. [35] and are only summarized here. Plant material consisted of commercially available, processed products obtained from certified manufacturers, with documented supplier information and batch/lot numbers ensuring traceability and botanical identity as provided by the producers.

Specifically, commercially available tea preparations were purchased from an online pharmacy (planteea.ro, SC Brotăcel SRL, Bucharest, Romania), including Ocimum basilicum-aerial parts; Lot 1958, SC Stefmar Producție SRL, Râmnicu Vâlcea, Romania; Galium verum-aerial parts; Lot 84941, Dacia Plant, Bod, Romania; Sambucus nigra-flowers; Lot 85,053 Dacia Plant, Bod, Romania. In addition, incense resin granules (Boswellia serrata-gum resin; 100% tămâie; Lot 5819, LIFE Bio, SC Bionovativ SRL, Podu Olt, Romania) were used [35].

Botanical identification was based on supplier documentation and morphological characteristics prior to extraction.

The plant materials were extracted using either ethanol 99.5% or a 70:30 (v/v) ethanol–water mixture under controlled temperature and time conditions. The resulting extracts were coded as BS-E, SN-E, OB-E, GV-E and BS-H, SN-H, OB-H, GV-H.

4.2. Determination of TPC by Folin–Ciocalteu Method

Quantitative determination of TPC of the plant extracts studied was performed according to the method described by Coelho et al. 2018 [33].

The method was based on mixing 30 µL of extract solution with 150 µL of Folin–Ciocalteau reagent (2M) in a Thermo Scientific Nunc 96-well plate. After 5 min at room temperature, 120 µL of sodium carbonate solution (0.25 mg/L) was added in each well. The microplate containing the reaction mixtures was incubated in the dark for 30 min at 40 °C, and the absorbance was recorded against a blank sample at 665 nm using a Varioskan LUX Multimode Microplate Reader from Thermo Fisher Scientific (Waltham, MA, USA). Gallic acid was used as standard for the quantification of the total phenolic content. The calibration curve was obtained by plotting the absorbance at 665 nm of different concentrations of standard solutions in ethanol (12.5–600 mg/L). All tests were carried out in duplicate. The total phenolic content was expressed as milligrams of GAE per gram of dry extract (mg GAE/g dry extract). A schematic overview of the Folin–Ciocalteu assay workflow is provided in the Supplementary Materials (Figure S2).

4.3. Chemicals

The list of chemicals, suppliers, and purity grades used in this study was identical to that reported by Rosca et al. [19,35], except for any specific additional reagents indicated in the corresponding experimental sections.

4.4. Phytochemical Profiling (HPLC-MS)

The extraction protocol and LC–MS analytical method applied in this study were previously developed and validated by Ghiulai et al. 2020 and Vlase et al. 2014 [36,37]. Details of the workflow application to the investigated extracts were previously published in Rosca et al. [19] and are only summarized here.

Separation was performed on a C18 reversed-phase column with gradient elution of acidified water and acetonitrile, and detection in negative ionization mode. Identification and quantification were based on authentic standards of major polyphenols.

The method targets polyphenolic compounds and enables screening and quantification of a predefined panel of phenolic acids and flavonoids using authentic external standards for calibration. Non-polyphenolic constituents (e.g., triterpenoids reported for Boswellia spp.) were outside the scope of the analytical method and were not assessed.

Limits of detection (LOD), limits of quantification (LOQ), and recovery parameters were established during the original method validation and are reported in the original publication [36,37].

4.5. Antioxidant Activity

The antioxidant capacity of the analyzed plant extracts was measured using the method from Brand-Williams et al. 1995 [21] with some modifications. The estimation of the free radical reduction capacity using a solution of DPPH was adapted to evaluate the antioxidant activity of the different plant extracts. A total of 2.85 mL of DPPH solution (0.063 µM) in ethanol was added to 0.15 mL of antioxidant solutions at different concentrations (4.3–78 μg/mL for OB, 3.6–44 μg/mL for SN, 3.4–39 μg/mL for GV, 30–190 μg/mL for BS in absolute ethanol or 70% ethanol; 0.4–5 μg/mL for standard ascorbic acid in water). The mixtures were homogenized and incubated in the dark for 60 min at room temperature. The absorbance of the samples was recorded at 517 nm on a Cary 60 UV-VIS, Agilent Technologies, using absolute ethanol or 70% ethanol as a blank. The control solution was prepared by replacing 0.15 mL sample (ethanolic extract) with absolute or 70% ethanol. By using the same procedure, positive controls of ascorbic acid in ethanol were also completed. All tests were carried out in duplicate. The antioxidant concentration necessary to decrease the initial DPPH concentration by 50% inhibition (IC₅₀) was calculated using the graph obtained by plotting the percentage of the inhibition capacity against the concentration of antioxidant extracts. The percentage of remaining DPPH and inhibition were calculated as follows:

Inhibition (%) = [(A₀ − Aₛ)/A₀] × 100

where A_s—absorbance of the samples at 517 nm after 60 min incubation in the dark, and A₀—absorbance of the blank with solvent instead of antioxidant at the same incubation time.

The antioxidant capacity of the extracts was expressed as IC₅₀ value and compared with that of ascorbic acid.

4.6. Cell Culture

The HaCaTcells were acquired from CLS Cell Lines Service GmbH (Eppelheim, Germany). The cells were received as frozen samples and were stored in liquid nitrogen. HaCaT cells were cultured in Dulbecco’s modified Eagle Medium (DMEM) high glucose supplemented with 10% fetal bovine serum (FBS) and 1% antibiotic mixture of penicillin and streptomycin (Sigma-Aldrich, Munich, Germany). They were maintained in a humidified incubator with 5% CO₂ at 37 °C. After reaching 80–90% confluence, the cells were treated with the tested extracts (BS-E, BS-H, SN-E, SN-H, OB-E, OB-H, GV-E and GV-H) at 5 increasing concentrations (0.36, 36, 360, 760 and 1000 μg/mL). The number of the HaCat cells was determined in the presence of Trypan blue using an automated cell counting device (Thermo Fisher Scientific, Inc., Waltham, MA, USA).

4.7. Cell Viability Assessment (Alamar Blue Assay)

The Alamar Blue staining method (see Figure S3) was used to determine the cell viability of HaCaT cells 24 h and 48 h post stimulation, with the tested compounds at increasing concentrations (0.36, 36, 360, 760 and 1000 μg/mL) [38]. The cells (1 × 10⁴ cells/well) were seeded onto 96-well plates and incubated at 37 °C and 5% CO₂ until reaching 80–90% confluence. The used medium was removed with an aspiration station and replaced with a fresh medium containing the tested compounds. After 24 h and 48 h, respectively, the cells were stained with 20 μL/well of Alamar Blue reagent and incubated again for 3 h at 37 °C. The absorbance was measured at two wavelengths, 570 nm and 600 nm, using a xMark™ Microplate Spectrophotometer, Bio-Rad (Hercules, CA, USA). The Alamar Blue assay workflow is illustrated in the Supplementary Materials (Figure S3).

4.8. Scratch Wound Healing Assay

The scratch test was performed to assess the wound healing potential of the tested compounds on HaCaT cells. Prior to the scratch assay, HaCaT cells were cultured in low-serum medium (0.5% FBS) to suppress proliferation, ensuring that wound closure primarily reflected cell migration [19]. A number of 2 × 10⁵ cells/well were seeded onto 12-well plates. After reaching 80–90% confluence, the old medium was removed, and the attached cells were scratched along the diameter of the well using a sterile pipette tip. Afterwards, the cells were washed with phosphate-buffered saline PBS (Thermo Fisher Scientific, Boston, MA, USA) and stimulated with the tested compounds at five increasing concentrations (0.36, 36, 360, 760 and 1000 μg/mL). To determine the healing potential of the tested compounds compared to control in early stages, the cells were captured on images at 0 h, 24 h and 48 h using the Olympus IX73 inverted microscope, 20× objective, using the CellSens V3.2 software (Olympus, Tokyo, Japan) [39]. The percentage of wound closure was calculated relative to the initial wound area, and results are presented as mean ± SD of three independent experiments.

4.9. Statistical Analysis

TPC and antioxidant activity (DPPH) assays were performed in duplicate, whereas cell-based assays were conducted in three independent experiments performed in triplicate.

Data are presented as mean ± standard deviation (SD). The statistical tests were carried out using one-way ANOVA followed by Dunnett’s post-test (GraphPad Prism version 6.0.0, GraphPad Software, San Diego, CA, USA). Due to the limited number of independent biological replicates, formal testing of data normality was not performed. The differences between the groups were considered statistically significant if p < 0.05, as follows: * p < 0.05, ** p < 0.01, and *** p < 0.001.

5. Conclusions

In this study, the extracts from OB and GV, characterized by higher levels of phenolic acids, showed the strongest pro-regenerative profiles in both viability and migration assays. Specifically, both OB extracts (OB-E and OB-H) and GV-H exhibited a significant stimulation of HaCaT proliferation and migration, making them the most promising candidates for further in vivo validation. These extracts surpassed the effect of other plant extracts, such as BS-H and SN-H; SN-H, characterized by a higher flavonoid content, supported a moderate cytoprotection, while BS, rich in boswellic acids, presented a pattern compatible with modulatory, non-proliferative activity.

The specialized literature that has studied the plants included in our study individually regarding the antioxidant and anti-inflammatory activities regarding the main constituents mentioned provides a useful framework for interpreting the functional profiles observed in our in vitro assays. The strong proliferative effect and regenerative potential of OB-H and GV-H, combined with their antioxidant properties, positions them as strong candidates for further in vivo testing for therapeutic applications in wound healing and skin regeneration.