Nonanimal Euglena gracilis‐Derived Extracellular Vesicles Enhance Skin‐Regenerative Wound Healing

This study proposes using microalgae‐containing carbohydrate bioactives, an Euglena gracilis‐derived extracellular microvesicle (EMVEG) system, for enhanced skin regeneration. The critical deformation ratio, 1.67, during cell extrusion enables the authors to tune the particle size of the EMVEG at ≈1 µm, thus satisfying the encapsulation yield of β‐1,3‐glucan and the cellular delivery performance. In vitro 5‐bromo‐2'‐deoxyuridine and cell scratch assays reveal that the EMVEG promotes the proliferation and migration of skin cells, thereby increasing both collagen synthesis and the expressions of proliferation‐associated proteins. An ex vivo wound‐healing test using both artificial and porcine skin reveals that similar to that seen using β‐1,3‐glucan, the EMVEG can substantially increase the cell population, expressing the proliferation‐related protein, termed proliferating cell nuclear antigen. These results demonstrate that the EMVEG system shows considerable potential in the field of skin regeneration. This technique is expected to design new types of extracellular vesicles that are applicable for skin regeneration in the pharmaceutical and cosmetic industries.

first, we fabricated the EMV EG via cell extrusion while regulating the vesicle size to optimize the β-glucan content therein. Next, we characterized β-glucan in the EMV EG via an aniline blue assay [10] and optimized the particle size of the EMV EG to micron scales because vesicles of such length scales can contain a considerable amount of β-glucan while retaining their thermal diffusivity and vesicular structure. Thereafter, we experimentally verified the in vitro cell proliferation and migration of EMV EG in human keratinocytes and analyzed ex vivo tissue viability in reconstructed human epidermis and porcine skin to establish our EMV EG as a useful skin regeneration agent.

Design of Cell Extrusion for Fabrication of EMV EG
The cell walls of most microalgae have high mechanical strength. Therefore, EV-mimetic vesicles cannot easily be realized using conventional external stresses. Compare with microalgae, EG is enveloped only by a plasma membrane composed of lipids and proteins quite similar to those of other eukaryotic cells. [11] Therefore, external stress can easily break its cell membrane to spontaneously produce EG-derived vesicles. This concept enables us to establish an EG-derived vesicle system using a cell extrusion process, as illustrated in Figure 1. Confocal laser scanning microscopy (CLSM) observation reveals that EG is surrounded by a lipid-based membrane (red), and β-glucan (blue) occupies the inner phase of EG by ≈70% by volume (see the fluorescence microscope image in Figure 1). When EG passes through polycarbonate (PC) porous filters, its cell membrane is elongated, deformed, and ruptured owing to the pressure difference generated by the EG-containing fluid passing through such small pores. [12] The broken lipid membrane fragments would self-assemble to form the EMV EG presumably encapsulating active ingredients, including β-glucans, proteins, and minerals. In particular, β-glucan, a primary constituent of EG, shows excellent efficacy in regenerative tissue engineering by enhancing cell proliferation and migration.

Controllable Production of EMV EG based on Critical Deformation Ratio
During the extrusion process, the EG particles passing through the pore channel are deformed and fractured by the applied pressure, which produces the membrane fragments that eventually assemble to form EMV EG . [8c,13] This process is decisively affected by the pore size under constant pressure conditions. In principle, the Blake-Kozeny equation explained that as particles pass through a filter, the clogging of the filter pores is determined by the critical deformation ratio (CDR), defined by CDR = D pore /D avg , where D pore is the pore diameter of a filter, and D avg is the particle diameter of a vesicle. The theoretically known CDR is ≈1.67. [14] If CDR < 1.67, EG particles are deformed and consequently their cell membranes are fragmented in the PC pore channel (Figure 2a). Current studies employed the five-step sequential extrusion from 12 down to 0.2-µm pore size of PC filters. When the pore size was reduced stepwise to 5 µm, EG was deformed and fragmented along the PC pore channel, resulting in an ≈2-µm-sized EMV EG . Further extrusion through a 2-µm pore-sized PC filter led to the production of an ≈1-µm-sized EMV EG . The particle size of the EMV EG generated in the CDR ≈1.67 was linearly proportional to the pore size of the PC filters and also coincided with the calculated CDR pattern (Figure 2b). We determined the protein concentration in the EMV EG by using a bicinchoninic acid (BCA) assay. The protein concentration in the EMV EG almost remained unchanged with the variation of its particle size. As proteins commonly have hydrodynamic particle sizes in the nanoscales, they can be captured regardless of the EMV EG particle size. However, the β-glucan concentration rather increased in logarithmic proportion to the EMV EG particle size. Notably, as β-glucan is encapsulated in the form of sub-microfibrillar fragments from paramylon with a dimension approximately within the range of www.advmatinterfaces.de 2-8 µm, [14] the encapsulation yield increased as the size of the EMV increased.

Confirmation of Vesicular Structure and Enhanced Cellular Uptake of EMV EG
Considering effective cell uptake, dispersibility in suspension, and the encapsulation yield of β-glucan, we adjusted the cell extrusion process to produce the EMV EG with ≈1 µm of average particle size and 20 µg mL −1 of β-glucan encapsulation yield (Figure 3a,b and Figure S2, Supporting Information). Transmission electron microscopy (TEM) observation revealed that EMV EG membrane buckling occurred irrelative to the pore size of the PC filters, indicating the presence of a thin solid layer, a lipid-base cell membrane, at the periphery of the EMV EG ( Figure S3, Supporting Information). [15] To further characterize the morphological features of the EMV EG , we stained their lipid membranes with Texas red-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine (DHPE) and β-glucan with aniline blue, respectively, and then conducted CLSM observations. β-Glucan was stably collected inside the EMV EG . Interestingly, we consistently observed that lipid substances were also present with nonuniform concentration differences in the EMV. In the EG membrane, intramembranous proteins are intercalated between the lipid bilayer, and Articulin 80&86 anchors them directly underneath. [11,16] Articulin 80&86 structurally strengthens the cell membrane as they are repeatedly interlocked and linked similar to a molecular spring. As some small lipid bilayer fragments generated during the cell extrusion will still be captured by Articulin 80&86, they seem to prefer to be entrapped inside rather than self-associate to form the shell layer of the EMV EG . As a control, the neat β-glucan, derived from paramylon, was also extruded and used in the form of sub-microfibrillar fragments. Typically, both the neat β-glucan control and EMV EG showed no toxicity in HaCaT cells ( Figure 3d). Furthermore, the cellular uptake and the CLSM observations of the EMV EG against the HaCaT cells revealed a significant increase in the cellular uptake with an increase in the concentration of the EMV EG (Figure 3e,f). This experimentally proves that the EMV EG can enhance the intracellular uptake of β-glucan.

In Vitro Proliferation and Migration of Skin Cells Enhanced by EMV EG
β-Glucan is a well-known bioactive that promotes the proliferation and migration of skin cells. [17] We investigated the implication of our EMV EG enclosing β-glucan on skin cell activations. 5-Bromo-2'-deoxyuridine (BrdU) is a common reagent that can detect and quantify proliferating cells. [18] Therefore, we compared the portion of BrdU-positive HaCaT cells treated with the neat β-glucan and EMV EG using immunocytochemical (ICC) analysis. We observed that the EMV EG -treated cells contained more BrdU-positive cells than those controlled by 44% and β-glucan-treated cells by 27% after 24-h incubation (Figure 4a,b and Figure S5, Supporting Information). Furthermore, we assessed the migration capability of the EMV EG using an in vitro scratch assay. [19] We scratched the HaCaT monolayer and then treated the cells with different concentrations of β-glucan and EMV EG . A decrease in the scratched area with time was observed ( Figure 4c and Figure S6, Supporting Information). Compared with the case of the control and treatment of β-glucan only after 24-h incubation, the treatment of the EMV EG resulted in more migration of the HaCaT cells. This phenomenon was more distinguishable after 48 h ( Figure 4d). These results indicated that the proposed EMV EG facilitated the proliferation and migration of keratinocytes because of their more effective delivery of encapsulated β-glucan than the same amount of neat β-glucan.

Exhibition of Remarkable Skin-Regenerative Wound Healing
To investigate cell migration and proliferation capabilities during skin regeneration, we treated CCD-986sk as fibroblasts and HaCaT as keratinocytes with EMV EG and quantified the levels of collagen and keratin production. After treating neat β-glucan and EMV EG for 72 h, the EMV EG effectively stimulated fibroblasts to produce more collagen than did neat β-glucan (Figure 5a). The fibroblasts treated with EMV EG showed a 20% increase in collagen production than those treated with neat β-glucan ( Figure 5b). Furthermore, the expressions of proliferation-associated proteins, such as cytokeratin 10 (K10), cytokeratin 14 (K14), Slug, and Src, in . c) Schematic of EMV EG encapsulating protein and lipid layer fragments and its confocal microscopy images of β-glucan and lipids labeled aniline blue (blue colored) and Texas red-DHPE (red colored), respectively. The scale bars are 1 µm. d) Viability of HaCaT cells treated with β-glucan and EMV EG with varying the volume fraction diluted by Dulbecco's modified Eagle's medium (DMEM) (φ). e) Quantitative cellular uptake (n = 3 in each group) and f) confocal microscopy visualization of EMV EG in HaCaT cells after incubation for 24 h. The lipids of EMV EG are labeled with Texas red-DHPE (red colored) and 4,6-diamidino-2-phenylindole (DAPI) (blue colored) for these observations. The scale bars are 20 µm.
www.advmatinterfaces.de keratinocytes were determined using Western blot analysis [20] after treating the HaCaT cells with neat β-glucan and EMV EG (Figure 5c and Figure S7, Supporting Information). Notably, compared with those treated with neat β-glucan, the expressions of K10 and K14 treated with EMV EG for 24 h were increased by ≈40%. Similarly, after treating with EMV EG , the expressions of Src and Slug treated for 24 h were enhanced by ≈45%. This result is comparable with the cases of treating with neat β-glucan ( Figure 5d). Src, a critical regulatory protein, upregulates the focal adhesion kinase (FAK) pathway, promoting the migration of keratinocytes and Slug, a crucial transcript factor, activates keratinocytes to spread and migrate during skin regeneration. Such increases in protein expressions indicate that the EMV EG significantly contributes to the wound-healing process through the FAK pathway and Slug transcription. [5c,21] Finally, we verified that such impressively enhanced in vitro efficacies of EMV EG on skin cells are also expressed in the actual skin. To evaluate ex vivo wound-healing performance, the neat β-glucan and EMV EG were topically treated to Keraskin and porcine skin, respectively. [22] The tissue viability of Keraskin indicated that compared with the control and neat β-glucan, the EMV EG showed better viability (Figure 6a,e). To quantify the cell population expressing the proliferation-related protein, a proliferating cell nuclear antigen (PCNA) was selectively stained with hematoxylin and eosin (H&E) (Figure 6b). As indicated by the PCNA staining, compared with the control and neat β-glucan, the EMV EG increased cell proliferation by ≈20% (Figure 6e). Additionally, the viabilities of the porcine skin also indicated that both neat β-glucan and EMV EG with hydrogen peroxide (HPO) were viable treatments (Figure 6c,f). The quantification of the stained PCNA following the topical treatment of porcine skins demonstrated that compared with the samples treated with only HPO, those treated with the neat β-glucan and EMV EG contained more PCNA-positive cells (Figure 6d,f). Furthermore, in the ex vivo tests, the EMV EG treatment reliably exhibited better wound-healing performance than the neat β-glucan treatment. www.advmatinterfaces.de

Conclusion
This study identified and developed a new type of nonanimalderived extracellular vesicle system, EMV EG , that exhibits skin regeneration performance by using a combination of experimental methodology and in vitro and ex vivo assays. By employing a modified cell extrusion process, we showed that pressure-mediated EG rupturing and consequent reassembly of cell membrane fragments led to the facile production of the EMV EG . Depending on the particle size corresponding to the CDR and the encapsulated amounts of β-glucans, we adjusted the particle size of the EMV EG to ≈1 µm while maintaining a typical vesicular morphology. We then verified that the EMV EG exhibited excellent in vitro and ex vivo skin regeneration performances, which were interpreted as enhancing the expressions of related proteins through a favorable cellular influx of β-glucan. This technique is expected to be applied to other cells, thereby enabling the design of new types of EVs that are applicable for skin treatments and care in the pharmaceutical and cosmetic industries.

Fabrication of EMV EG by Cell Extrusion:
To obtain EG, first, the culture medium was removed by centrifugation at 1500 g for 5 min. Next, EG was resuspended in PBS at a concentration of 1 × 10 6 cells mL −1 . Next, the EG suspension underwent sequential extrusion through 5-and 2-µm pore-sized PC membranes three times each using a miniextruder. The sizes of the EMV EG were measured using dynamic light scattering (ELS0Z, Ostuka electronics, Japan). The morphology of the EMV EG was observed by high-resolution TEM (JEM-3010, Jeol, Japan) and confocal laser scanning microscopy (TCS SP8 HyVolution, Leica, Germany). For the CLSM visualization, the β-glucans in EG were stained with aniline blue for 10 min before the extrusion and the outer membrane of the EMV EG was stained with Texas red-DHPE for 10 min.
Quantitative Analysis of β-Glucan in EMV EG : The amount of β-glucan was determined by the aniline blue assay. To prepare the working solution, 6.7 × 10 −3 m aniline blue diammonium salt, 1 m glycine, 1.25 m NaOH, and 1.5 mL of 2 n HCl were mixed with 18.5 mL of deionized water. This solution was stored in a dark environment overnight until the dark blue color was decolorized to yellow. Thereafter, 300 µL of the EMV EG was incubated with 30 µL of 0.9 m NaOH at 70 °C for 30 min for the denaturation of β-glucan. In this process, NaOH disrupted the EMV EG membrane and facilitated conjugation of the glucan with the aniline blue dye. Next, 100 µL of the EMV EG was added to 900 µL of the working solution and incubated at 50 °C for 30 min. Finally, the solution was transferred to a 96-well plate and the fluorescence intensity was measured at 430/535 (excitation/emission) nm using a microplate reader (Spark, Tecan, Switzerland).
Cell Culture and Viability Assay: The HaCaT cells were incubated in DMEM with 10% FBS and 1% penicillin-streptomycin, and 1% www.advmatinterfaces.de gentamicin at 37 °C in a humidified atmosphere containing 5% CO 2 . CELLOMAX Viability Assay (PRECARE GENE, Korea) was used to evaluate the cell viability. The HaCaT cells were seeded in the 96-well plate at a density of 1 × 10 4 cells per well and incubated for 24 h. Next, the cells were treated with different concentrations of the β-glucan and EMV EG for another 24 h. The cells were washed with PBS, and the viability assay solution diluted with DMEM was added to each well. The plate was incubated for 2 h at 37 °C, and then the absorbance was measured at 450 nm using the microplate reader.
Analysis of Cellular Uptake: The HaCaT cells were seeded in the 96-well plate at a density of 1 × 10 4 cells per well and incubated for 24 h. Thereafter, Texas red-DHPE-tagged EMV EG in DMEM was treated for 24 h. After washing with PBS twice, the fluorescence intensity was measured at 485/625 (excitation/emission) nm using the microplate reader. To visualize the EMV EG in the cells, the HaCaT cells were seeded in a confocal 6-well plate at a density of 1 × 10 4 cells per well and incubated for 24 h. Next, the Texas red-DHPE-tagged EMV EG was treated for 0 and 24 h. The cells were fixed with 4% paraformaldehyde for 10 min at 37 °C. After washing with PBS, the cell nuclei were stained with DAPI for 5 min at 25 °C. To visualize cellular uptake, fluorescence images were obtained using CLSM.
ICC Staining for BrdU: The HaCaT cells were seeded in the 96-well plate at a density of 5000 cells per well and incubated for 24 h. The β-glucan and EMV EG in DMEM (v/v, 5/5) were treated with 0.5% FBS for 24 h, and 50 µM BrdU (ab142567, Abcam) was added before the last 2 h. Next, the cells were fixed with 4% paraformaldehyde for 10 min and permeabilized www.advmatinterfaces.de with 0.25% Triton X-100 for 15 min. For DNA hydrolysis, 2 n HCl was treated for 30 min and then washed with PBS. The cells were blocked with a blocking solution (1% bovine serum albumin in PBS with 0.1% of Tween 20) for 1 h and then stored with an anti-BrdU antibody (ab6326, Abcam) overnight at 4 °C. Subsequently, the cells were treated with an Alexa Fluor-488-conjugated secondary antibody (ab150157, Abcam), and DAPI was counterstained for the cell nuclei. Immunostaining images were obtained using a fluorescence microscope, and the percentage of the BrdU-positive cells was calculated.
In Vitro Wound Scratch Assay: The HaCaT cells were seeded in a 12-well plate at a density of 1 × 10 5 cells per well and incubated until the cells reached 100% confluence. Next, cell monolayers were scratched using sterile 200-µL pipette tips. After gentle washing with PBS, the β-glucan and EMV EG in DMEM were treated with 0.5% FBS and incubated for 0, 24, and 48 h. The gap areas were recorded using an optical microscope (Axio Vert.A1 Bio, ZEISS, Germany), and the relative wound area was calculated.
Quantification of Collagen Synthesis in Fibroblasts: CCD-986sk cells were seeded in the 12-well plate at a density of 5 × 10 4 cells per well and incubated for 24 h. Next, the β-glucan and EMV EG in DMEM were treated with 0.5% FBS for 72 h. The cells were washed with PBS and fixed with ice-cold 70% ethanol overnight at 4 °C. Next, Sirius-red solution (VitroVivo Biotech, USA) was added to each well and stored overnight at 4 °C to stain the synthesized collagen. The unbound dye was removed by washing with PBS. Subsequently, the absorbance of the dye-conjugated collagen was measured at 540 nm using the microplate reader. Furthermore, the synthesized collagen was visualized using the optical microscope.
Western Blot Analysis: The HaCaT cells were seeded at a density of 1 × 10 6 cells per well in the 6-well plate, and then the cells were treated with β-glucan and the EMV EG . Next, the cells were washed with PBS and lysed with 1× RIPA (CELL BIOLABS, USA) and 1× protease inhibitor (Cell Signaling Technology, USA). The protein concentration was determined using the BCA assay (Takara, Japan). Thereafter, the cell lysates were loaded to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (TGX Precast Midi Protein Gel, Bio-Rad, USA) and transferred onto polyvinylidene fluoride membranes (Millipore, USA). Subsequently, the membranes were blocked with 5% nonfat milk in Tris-buffered saline containing 0.2% Tween 20 (TBS-T) for 1 h at room temperature. Next, the membranes were incubated overnight at 4 °C with primary antibodies: cytokeratin 10 (ab111447), Src (ab47405), Slug (ab106077) from Abcam (UK), and cytokeratin 14 (PA5-16722) from Invitrogen (USA). Thereafter, the proteins were detected using a chemiluminescent reagent, EzWestLumi Plus (ATTO, Japan). The reactive bands were detected with an Odyssey infrared imaging system (ImageSaver6, ATTO, Japan), and the amounts of proteins were measured using the CS Analyzer 4 (ATTO, Japan).
Ex Vivo Skin Irritation Test: The protocol for the Keraskin skin irritation test was based on the procedure of a validated reference method adopted by OECD TG 439 (OECD, 2015). After 22 ± 2 h of preincubation, the tissue was removed from the incubator, and 40 µL of β-glucan and EMV EG was applied immediately to the tissue surface. Next, sterile forceps were used to gently spread liquid and tilt the insert. Thereafter, the plates were incubated for 30 min at 37 °C containing 5% CO 2 . The tissues were then rinsed with PBS. After the tissues were postincubated for 42 ± 2 h, the entire medium was removed. The tissues were blotted and transferred to a 24-well plate that contained MTT (0.4 mg mL −1 ) and incubated for 3 h at 37 °C containing 5% CO 2 . Next, the tissues were transferred to a new 6-well plate, prefilled with 2 mL of isopropanol. Formazan extraction was performed at room temperature for 3 h, and 250 µL of formazan extract per tissue was transferred to the 96-well plate. The optical density (OD) was measured at 570 nm using isopropanol as a blank.
Ex Vivo Live Full-Thickness Analysis of Porcine Skin: The porcine skins were obtained from Apures (Pyeongtaek-si, Gyeonggi-do, Korea) and Optipharm (Cheongju-si, Chungcheongbuk-do, South Korea). The porcine skins were obtained from the ears of a micropig, which was euthanized for research on drug delivery. The ear tissue was rinsed with 70% ethanol and PBS. A punch biopsy was collected from a relatively flat area of the ear tissue, and the cartilage in the middle was removed to obtain only the skin tissue. Then, the core of tissue was excised using a 10-mm biopsy punch (Acu-punch, Acuderm, Fort Launderdale, FL, USA) and inserted into a cell culture insert, so that the epidermal surface of the skin tissue was exposed to air. The gap between the skin tissue and the cell culture insert was sealed with a silicone ring. The tissues were placed in medium-filled 6-well plates and preincubated for 20-24 h at 37 °C in a humidified atmosphere containing 5% CO 2 . Medium was changed every 2 days. After preincubation, a wound was induced in the porcine skin tissue inf necessary and then the tissue was treated with test substances every other day. To evaluate the viability, tissues were transferred to a 48-well plate that contained MTT (0.4 mg mL −1 ) and incubated for 3 H, and 250 µL of formazan extract per tissue was transferred to a 96-well plate. The OD was measured at 570 nm using isopropanol as a blank. The tissue viability of the skin tissue was expressed as the ratio of the skin disk to its weight in milligrams.
Histological Analysis: For the histological examination, after MTT extraction, Keraskin and porcine biopsy skin samples were fixed in 10% neutral-buffered formalin. According to a previous study (Park et al., 2019), [23] preserved tissues from each group were paraffin-embedded, sectioned, stained with H&E, and then examined microscopically under an Olympus DP71 microscope (Center Valley, PA, USA). The tissue images were obtained using a virtual slide system (Aperio Scanscope XT, Vista, CA, USA).
Immunohistochemistry Staining: For immunostaining, paraffinembedded pig skins were cut into 4-µm sections. The sections were deparaffinized in xylene and sequentially rehydrated through a descending graded series (100%, 95%, and 70%) of ethanol. Antigen retrieval (DAKO, S1699, Santa Clara, CA, USA) was conducted using a high-pressure cooker for 15 min. After cooling on ice until the solution became transparent, the sections were incubated in 3% H 2 O 2 for 30 min and washed twice with PBS. To reduce nonspecific signals, the samples were incubated with a serum-free protein-blocking solution (DAKO, X0909, Santa Clara, CA, USA) for 1-2 h at room temperature. Anti-PCNAs (clone PC10; Santa Cruz Biotechnology, Dallas, TX, USA) were incubated overnight at 4 °C in a humidity chamber. After three washes in PBS, the sections were incubated in a horseradish-peroxidaseconjugated antirabbit secondary antibody (DAKO, K4003, Santa Clara, CA, USA) for 15 min at room temperature. For immunohistochemistry, 3,3′-diaminobenzidine (DAB; DAKO, K3468, Santa Clara, CA, USA) was used to develop the antibodies, and Mayer's hematoxylin (DAKO, S3309, Santa Clara, CA, USA) was used for counterstaining. Each experiment was performed using an identical time for DAB development.
Statistical Analysis: All the statistical data were calculated using a one-way analysis of variance followed by Tukey's post hoc test using Origin Software (OriginLab 8.5). The statistical significance was set at p < 0.05.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.

Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.