Enhancing therapeutic efficacy of human adipose-derived stem cells by modulating photoreceptor expression for advanced wound healing

Human adipose-derived stem cells (hADSCs) have been widely used for regenerative medicine because of their therapeutic efficacy and differentiation capacity. However, there are still limitations to use them intactly due to some difficulties such as poor cell engraftment and viability after cell transplantation. Therefore, techniques such as photobiomodulation (PBM) are required to overcome these limitations. This study probed improved preclinical efficacy of irradiated hADSCs and its underlying molecular mechanism. hADSCs were irradiated with green organic light-emitting diodes (OLEDs). Treated cells were analyzed for mechanism identification and tissue regeneration ability verification. Expression levels of genes and proteins associated with photoreceptor, cell proliferation, migration, adhesion, and wound healing were evaluated by performing multiple assays and immunostaining. Excision wound models were employed to test in vivo therapeutic effects. In vitro assessments showed that Opsin3 (OPN3) and OPN4 are both expressed in hADSCs. However, only OPN4 was stimulated by green OLED irradiation. Cell proliferation, migration, adhesion, and growth factor expression in treated hADSCs were enhanced compared to control group. Conditioned medium containing paracrine factors secreted from irradiated hADSCs increased proliferation of human dermal fibroblasts and normal human epidermal keratinocytes. Irradiated hADSCs exerted better wound healing efficacy in vivo than hADSCs without OLED irradiation. Our study introduces an intracellular mechanism of PBM in hADSCs. Our results revealed that photoreceptor OPN4 known to activate Gq-protein and consequently lead to reactive oxygen species production responded to OLED irradiation with a wavelength peak of 532 nm. In conclusion, green OLED irradiation can promote wound healing capability of hADSCs, suggesting that green OLED has potential preclinical applications.

evidence supporting the possibility of using hADSCs for promoting tissue regeneration [1][2][3]. Although several studies have explored the innate therapeutic potency of hADSCs for various diseases and their modes of action [4], there still exist several drawbacks of hADSCs such as low survival rate, heterogeneity, and inconsistent therapeutic outcomes [5][6][7]. Therefore, many recent studies have used various techniques to improve the therapeutic efficacy of hADSCs [8,9]. One strategy to augment the therapeutic potential of hADSCs is to perform photobiomodulation (PBM). PBM has been widely used to treat diseases by introducing a light source such as a laser or a light-emitting diode (LED) to target cells or tissues [10,11]. Growing evidence suggests that PBM may lead to physical and chemical effects in target cells by modulating various biological processes such as cell proliferation and inflammatory responses [12,13]. Furthermore, PBM can be applied to achieve enhanced expression of desired biomolecules that contribute to regeneration of injured tissues.
Previous studies have elucidated that cellular changes caused by PBM are due to production of adenosine triphosphate (ATP), which is related to mitochondrial cytochrome c oxidase (CCO) [14,15]. Electron transfer through CCO can generate a proton gradient that provides energy to ATP synthase for catalysis of ATP production [16]. Importantly, CCO is known to increase reactive oxygen species (ROS) levels in treated cells [17]. Suh et al. have demonstrated that PBM can mediate cell signaling pathways related to expression of opsins (OPNs) known to be G-protein-coupled receptors (GPCRs) [18,19]. Activating transient receptor potential (TRP) channels is one of action mechanisms of OPNs [20]. When TRP channels are activated, they allow calcium ions (Ca 2+ ) to flow into the inner membrane, thereby raising intracellular Ca 2+ concentration [21]. Other signaling pathways of OPNs involve G q -protein [20]. G q activation can induce Ca 2+ release from the endoplasmic reticulum (ER) into the cytoplasm of cells [22]. As reported by Gordeeva et al. [23], intracellular Ca 2+ level elevation can trigger ROS generation. Therefore, it is important to determine the proper mechanism associated with light source used for PBM.
In previous studies of PBM, experiments were conducted by short exposure of target cells to high-intensity light using lasers or LEDs as power sources [10,24]. However, excessive intensities of those lights may generate heat that may serve as additional exogenous stimuli. Thus, a big question when using lasers or LEDs as light sources for PBM is whether alterations of biological processes and changes in phenotypic profiles of exposed cells are caused by irradiation or heat generated from light sources [25,26]. Because there are contrasting results when cells are irradiated with light of similar wavelengths, the application of PBM remains controversial, due to the lack of investigations regarding the correlation between wavelengths and corresponding cellular responses [27,28]. In previous experiments using PBM as exogenous stimuli for hADSCs, researchers have focused on applying PBM particularly for modulating differentiation of cells [29,30]. PBM has also been widely adopted to study differences in apoptosis, cell proliferation, migration, and paracrine factor secretion of hADSCs with different wavelength bands [31]. Based on recent studies reporting that PBM encourages the secretion of paracrine factors from stem cells [32], therapeutic effects of conditioned medium (CM) derived from hAD-SCs exposed to blue, green, and red light are now being actively studied.
In the present study, we used organic LEDs (OLEDs) with a low heat generation and relatively weak intensity for PBM to evaluate cellular activities in response to a specific wavelength. OLED applied in our experiment has been commercialized with officially confirmed uniform properties. It has not been applied to stem cell-related research previously. Based on previous studies showing that responses in cells irradiated with green light can improve wound healing without obvious cytotoxicity [33,34], herein we utilized green light with a wavelength peak of 532 nm. Due to the lack of studies investigating the expression of OPNs in hADSCs in response to PBM, we explored cellular mechanisms involved in the effects of PBM by determining the expression of OPNs. Furthermore, we determined effects of OLED-based PBM on secretion of growth factors, proliferation, migration, and adhesion of hADSCs, all of which are known to play important roles in cutaneous wound healing [35,36]. Subsequently, we optimized irradiation conditions of PBM for hADSCs. Such conditions were used to treat hADSCs which were then applied to mouse cutaneous wound models to evaluate their therapeutic effects in vivo (Fig. 1).

Green OLED irradiation
An OLED panel was purchased from Kaneka (Osaka, Japan). This OLED panel was connected to a regulated direct current power supply with the voltage set at 4 V. hADSCs-seeded tissue culture dishes and plates were placed on the OLED panel inside an incubator. After washing with phosphate-buffered saline (PBS, Gibco BRL) followed by a medium exchange, cells were irradiated with a wavelength peak of 532 nm for an optimized period. Relative gene expression levels of human GAPDH,  OPN3, OPN4, PLC-β, IP3R, C-MYC, CXCL12, CXCR4,  β-CATENIN, INTEGRIN α5, INTEGRIN β1, VEGF, and HGF in green OLED-treated hADSCs, human GAPDH, KI67, PCNA, and C-MYC in conditioned media-treated hDFs and NHEKs, and mouse β-Actin, Acta2, Cd31, and Laminin in in vivo wound regions were quantified by quantitative real-time polymerase chain reaction (qRT-PCR). The sequences of primers used for qRT-PCR are listed in Table 1. hADSCs, hDFs, and NHEKs were seeded into 6-well tissue culture plates at a density of 1 × 10 5 cells/well. Ribonucleic acids (RNAs) were then extracted from cells and in vivo samples using 1 mL of TRIzol (Ambion, Austin, TX, USA) and 200 μL of chloroform (Sigma-Aldrich, St. Louis, MO, USA). RNA pellets were obtained by centrifugation at 12,000 rpm for 10 min at 4 °C. After dissolving the pellets in ribonuclease-free dH 2 O (TaKaRa, Kusatsu, Japan) and PrimeScript RT Master Mix (TaKaRa), PCR amplification was performed using synthesized complementary DNA as template. iQ SYBR green supermix (Bio-Rad, Hercules, CA, USA) and CFX connect RT-PCR detection system (Bio-Rad) were used for qRT-PCR analysis.

Calcium assay
Intracellular calcium concentration in green OLEDtreated hADSCs was measured using a calcium detection assay kit (ab102505; Abcam, Cambridge, UK). In this study, 1 × 10 6 of hADSCs were treated with green OLED and lysed with calcium assay buffer. After centrifugation at 27,000×g for 5 min at 4 °C, supernatants were collected. Samples were mixed with chromogenic reagent and incubated at room temperature for 10 min. Optical density of each sample was then measured at 575 nm using a microplate reader (Infinite F50; Tecan, Männedorf, Switzerland).

Cellular ROS and ATP assay
Relative levels of ROS in green OLED-treated hADSCs were evaluated by cellular ROS assay using 2′,7′-dichlorodihydrofluorescein diacetate (H2DCFDA, Invitrogen, Carlsbad, CA, USA). hADSCs for cellular ROS assay were seeded into 96-well tissue culture plates at a density of 3 × 10 3 cells/well. After cells were incubated in 10 μM H2DCFDA at 37 °C for 20 min, the fluorescence intensity (Ex/Em = 494 nm/524 nm) was measured using a microplate reader (Varioskan LUX multimode microplate reader; Thermo Fisher Scientific, Waltham, MA, USA). hADSCs for DCFDA staining were seeded into 6-well tissue culture plates at a density of 7 × 10 4 cells/well. After these cells were incubated in 10 μM H2DCFDA at 37 °C for 2 min, they were examined using a fluorescence microscope (DFC 3000 G; Leica Microsystems GmbH, Wetzlar, Germany). ATP assay kit (ab83355; Abcam) was used for evaluating the amount of ATP in green OLEDtreated hADSCs, following the manufacturer's protocol. Each cell lysate sample was prepared from 1 × 10 6 of hADSCs.

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining
Cytotoxicity of green OLED irradiation was analyzed by TUNEL staining using an ApopTag fluorescein in situ apoptosis detection kit (Merck KGaA, Darmstadt, Germany). Briefly, hADSCs were seeded into 6-well tissue culture plates at a density of 7 × 10 4 cells/well. After green OLED irradiation, cells were fixed with 4% paraformaldehyde (Biosesang, Seongnam, Korea). Equilibration buffer, working strength TdT enzyme, stop/wash buffer, and working strength anti-digoxigenin conjugate were applied to fixed cells followed by PBS washing after each step. For counterstaining, 4' ,6-diamidino-2-phenylindole (DAPI, Vector Laboratories, Burlingame, CA, USA) was used. Hydrogen peroxide solution (Sigma-Aldrich) was used for a positive control of TUNEL analysis [37].  Stained cells were examined using a fluorescence microscope (DFC 3000 G).

Fluorescein diacetate (FDA) and ethidium bromide (EB) staining
FDA/EB staining was performed to assess cytotoxicity of green OLED irradiation. hADSCs were seeded into 6-well tissue culture plates at a density of 7 × 10 4 cells/ well. The staining solution was prepared using 10 mL of FDA (Sigma-Aldrich) stock solution (1.5 mg/mL of FDA in dimethyl sulfoxide), 5 mL of EB (Sigma-Aldrich) stock solution (1 mg/mL of EB in PBS), and 3 mL of PBS. Green OLED-treated cells were washed with PBS and incubated in the staining solution at 37 °C for 2 min. After washing with PBS, stained cells were examined using a fluorescence microscope (DFC 3000 G).

Immunocytochemistry
hADSCs were seeded into 6-well tissue culture plates at a density of 5 × 10 4 cells/well. After green OLED irradiation, cells were fixed with 4% paraformaldehyde. These fixed cells were blocked with 10% (v/v) normal goat serum (Jackson Immuno Research Laboratories, West Grove, PA, USA) and 0.6% (v/v) Triton X-100 (Sigma-Aldrich) in PBS at room temperature for 1 h. Cells were then immunofluorescence stained with anti-KI67 antibody (ab16667; Abcam) and fluorescein (FITC)-conjugated AffiniPure goat anti-rabbit IgG (H + L) (Jackson Immuno Research Laboratories). DAPI was used as a counterstain. Stained cells were examined using a fluorescence microscope (DFC 3000 G).

Cell counting kit-8 (CCK-8) assay
Cell viability was evaluated using CCK-8 assay kit (Dojindo Molecular Technologies, Inc., Kumamoto, Japan). Briefly, hADSCs were seeded into 24-well tissue culture plates at a density of 1.5 × 10 4 cells/well. Green OLED-treated cells were washed with PBS and incubated with 10% (v/v) CCK-8 solution in culture medium at 37 °C for 2 h. Optical density of each sample was then measured at 450 nm using a microplate reader (Infinite F50).

Cell scratch assay
Cell migration was analyzed by cell scratch assay [38]. Briefly, hADSCs were seeded into 24-well tissue culture plates at a density of 1.5 × 10 4 cells/well. SPLScar scratcher (SPL Life Sciences, Pocheon, Korea) was used to scratch cells prior to green OLED irradiation. Cellcovered area was estimated as follows: [1 − (pixel size of uncovered area at each time point)/(pixel size of initially scratched area)] × 100% [39].

Enzyme-linked immunosorbent assay (ELISA)
Secretion levels of SDF-1, VEGF, and HGF proteins into medium of green OLED-treated hADSCs were quantified by ELISA. Briefly, hADSCs were seeded into 6-well tissue culture plates at a density of 1 × 10 5 cells/well. Supernatants were collected and analyzed using ELISA kits (R&D Systems). Diluted capture antibody was added to plates and incubated at room temperature overnight. After rinsing with a washing buffer, plates were blocked with reagent diluent. Samples, diluted detection antibody, streptavidin-HRP, and substrate solution were added to plates and incubated at room temperature. Plates were washed with a washing buffer before each step. After adding stop solution to the plate, optical density of each sample was then measured at 450 nm using a microplate reader (correction 540 nm, Infinite F50).

Phalloidin staining
Phalloidin staining was performed to assess cell adhesion. hADSCs were seeded into 6-well tissue culture plates at a density of 5 × 10 4 cells/well. These cells were then fixed with 4% paraformaldehyde after green OLED irradiation. Antifade mounting medium with phalloidin (Vector Laboratories) was used for actin filament (F-actin) staining, and DAPI was used as a counterstain. Stained cells were examined using a fluorescence microscope (DFC 3000 G).

CM treatment
Effects of paracrine factors secreted from green OLEDtreated hADSCs were evaluated by CM treatment. Briefly, hADSCs, hDFs, and NHEKs were seeded in 6-well tissue culture plates at a density of 1 × 10 5 cells/ well. Conditioned media were collected from supernatants of hADSCs at 48 h after green OLED irradiation for 24 h and diluted with serum-free media to reduce influences of fetal bovine serum. Media for culturing hDFs and NHEKs were replaced with 25% and 50% (v/v) CM. Cells were sampled at 48 h after media exchange.

Full-thickness skin wound model
Five-week-old female athymic nude mice (BALB/c-nude; OrientBio, Seongnam, Korea) were anesthetized by intraperitoneal injection of 100 mg/kg ketamine (Yuhan, Seoul, Korea) and 10 mg/kg xylazine (Bayer, Leverkusen, Germany). Two 10-mm-diameter full-thickness skin wounds were made on the dorsal side of each mouse. Wounds were then fastened to the underlying tissue with 6-0 silk sutures (AILEE, Busan, Korea). After excision, mice were divided into three groups: no treatment (NT, 100 μL of PBS was injected into each wound) served as control, hADSCs (SC, 2.5 × 10 5 of hADSCs without green OLED irradiation in 100 μL of PBS were injected into each wound), and green OLED-treated hADSCs (GTC, 2.5 × 10 5 of hADSCs were harvested at 24 h after green OLED irradiation for 24 h in 100 μL of PBS and injected into each wound) groups. Every wound was covered with a dressing (Tegaderm; 3M Health Care, St. Paul, MN, USA) after injection [40]. All procedures were approved by the Institutional Animal Care and Use Committee of Sungkyunkwan University (approval No. SKKUIACUC2020-06-11-1).

Assessment of wound healing efficacy
Skin covered areas of wounds were quantified by assessing photographs taken every 3 days. Pixel sizes of areas were adjusted according to the ruler taken with and calculated using Adobe Photoshop CC (Adobe systems, CA, USA). Skin covered areas were estimated as follows: [1 − (a pixel size of the wound at each time point)/(a pixel size of the wound on day 0)] × 100% [41].

Histological analysis
Tissue samples of wound regions were retrieved on day 9 and fixed in 4% paraformaldehyde. Fixed samples were soaked in 30% (w/v) sucrose in PBS and then embedded in optimum cutting temperature compound (SciGen Scientific, Chicago, IL, USA). Frozen samples were sliced vertically into 10-μm-thick sections. Thereafter, vertical slices were blocked and incubated with anti-laminin (ab11575; Abcam, 1:200) and anti-involucrin (ab53112; Abcam, 1:100) antibodies at 4 °C overnight. After washing with PBS, sections were incubated with fluorescein (FITC)-conjugated AffiniPure goat anti-rabbit IgG (H + L) (1:50) at room temperature for 1 h. Stained tissues were examined using a fluorescence microscope (DFC 3000 G). In vivo specimens were stained with hematoxylin and eosin (H&E) to identify tissue regeneration and structures. Masson's trichrome staining was performed to examine tissue fibrosis.

Statistical analysis
Quantitative data are presented as mean ± standard deviation using GraphPad Prism (GraphPad Software, San Diego, CA, USA). All data were analyzed using Student's t test. Statistical significance was considered at p < 0.05.

Cellular changes in hADSCs induced by green OLED irradiation
Cellular changes induced by green OLED irradiation in terms of photoreceptor expression, ROS generation, and ATP amount in green OLED-treated hADSCs compared to those in cells without light irradiation were assessed. The opsin family gene expression was evaluated to identify receptors stimulated by green OLED irradiation. As shown in Fig. 2a, b, qRT-PCR results showed significant increases in OPN3 and OPN4 gene expression levels after irradiating hADSCs with green OLED for 24 h compared to the NT group. Expression levels of OPN3 and OPN4 at 0 and 24 h post-irradiation were significantly elevated in hADSCs irradiated with green OLED for 24 h. In contrast with the group of hADSCs with 24 h of green OLED irradiation, only OPN4 gene expression was slightly increased in hADSCs irradiated with 4 h green OLED than in hADSCs without light irradiation. To find out the exact photoreceptor stimulated by green OLED irradiation, Western Blot analysis was performed. Only OPN4 protein was found to be upregulated (Fig. 2c). Furthermore, PLC-β, and IP3R gene expression levels in green OLED-treated hADSCs were analyzed to investigate the mechanism. They were found to be significantly enhanced at 0 and 24 h after irradiation compared to those in the NT group (Fig. 2d). Amounts of intracellular Ca 2+ , ROS and ATP were evaluated to determine whether molecular behaviors between the green OLED-treated group and the NT group were different. All assays were performed right after irradiation. There was no difference in Ca 2+ concentration (Fig. 2e), but cellular ROS assay showed a statistically significant difference in intracellular ROS level between the two groups (Fig. 2f ). Green OLED-treated hADSCs exhibited higher levels of ROS than untreated cells. Additionally, H2DCFDA staining showed the same tendency (Fig. 2g). This increment might have beneficial effects on treated hADSCs according to other experimental results of this study. However, the amount of ATP in the treated group did not statistically differ from that in the NT group (Fig. 2h). These results indicate that green OLED irradiation stimulates OPN4 expression and consequently enhances intracellular ROS level without ATP production (Fig. 2i).

Cell viability of hADSCs after green OLED irradiation
Apoptotic activities of hADSCs induced by green OLED irradiation were evaluated by TUNEL assay, FDA/EB staining, and Western Blot analysis. As shown in Fig. 3a, b, TUNEL and FDA/EB images did not reveal any apoptotic or dead cells in any group. Western Blot analysis showed no statistical difference in BAX or Bcl-2 expression between green OLED-treated and NT groups of hADSCs (Fig. 3c). Overall, these results  (n = 4). f Quantification of ROS in hADSCs after irradiation for 24 h as evaluated by cellular ROS assay (n = 6). g Representative H2DCFDA staining images of hADSCs after green OLED irradiation for 24 h (scale bar = 250 µm). h Quantification of ATP in hADSCs after green OLED irradiation for 24 h as evaluated by ATP assay (n = 4). i Schematic of OPN4 molecular mechanism. *p < 0.05; **p < 0.001 compared to no treatment group with unpaired t test Lee et al. Stem Cell Research & Therapy (2022) 13:215 indicated that the irradiation did not elevate apoptotic activities in 24 h green OLED-treated group.

Upregulation of hADSC proliferation induced by green OLED irradiation
As a cellular marker, KI67 was investigated to determine cell proliferation of hADSCs. Intracellular KI67 protein expression was analyzed by two kinds of assays. First, KI67 proteins were directly detected through immunocytochemistry (Fig. 4a). KI67 protein expression levels in both groups were increased at 24 h but decreased at 48 h. Meanwhile, there were significantly more KI67 proteins in treated hADSCs at 0 and 24 h post-irradiation than in untreated cells. Protein expression was analyzed qualitatively using Western Blot. Results showed the same tendency to KI67 staining (Fig. 4b). Moreover, C-MYC gene expression levels in green OLED-treated hADSCs were higher than in untreated controls with statistically significant differences at 0 and 24 h post-irradiation as assessed by qRT-PCR (Fig. 4c). Furthermore, the CCK-8 assay showed increased cell proliferation in green OLEDtreated groups compared to each control (Fig. 4d). These data indicate that green OLED irradiation for 24 h can enhance the proliferation of hADSCs for at least 24 h after irradiation.

Enhanced hADSC migration induced by green OLED irradiation
Cell motility was examined by cell scratch assay. As shown in Fig. 5a, cell migration morphology was observed at 0, 24, and 48 h after green OLED irradiation. Covered area by treated cells was significantly larger than that of untreated control at 24 and 48 h after irradiation. Additionally, CXCL12 and CXCR4 gene expression levels in hADSCs were evaluated by qRT-PCR (Fig. 5b). Their gene expression levels in treated hADSCs were significantly higher than those in the NT group at every time point except for CXCL12 at 48 h post-irradiation. Green OLED-treated hADSCs also secreted a significantly higher amount of SDF-1 than untreated cells at every time point (Fig. 5c). These data indicate that green OLED irradiation for 24 h could boost migration of hADSCs for approximately 48 h after irradiation.

Enhancement of cell adhesion by green OLED irradiation
Phalloidin staining was performed to detect F-actin related to cell adhesion. Fluorescent stained hADSCs and F-actin were observed at 0, 24, and 48 h after green OLED irradiation (Fig. 5d). F-actin in the treated group occupied a significantly larger area than that in the NT group at 24 and 48 h after irradiation. Furthermore,  (Fig. 5e). These data indicate that green OLED irradiation can improve adhesion of hADSCs for approximately 24 h after the irradiation.

Effects of paracrine factors secreted from green OLED-treated hADSCs on cell proliferation
VEGF and HGF are known to play important roles in wound healing. In this study, VEGF and HGF gene expression levels in hADSCs were evaluated by qRT-PCR (Fig. 6a). Expression levels of VEGF at 0 and 24 h and HGF at 48 h post-green OLED irradiation were significantly increased compared to those in the NT group. However, they showed no significant difference at other time points. VEGF and HGF proteins secreted from hADSCs were detected using ELISA (Fig. 6b). This result was expected based on gene expression data. Compared to the NT group, green OLED-treated group showed significantly higher VEGF concentration at every time point. However, HGF concentration in the treated group was only higher at 48 h post-irradiation. Therefore, conditioned media retrieved at 48 h after green OLED irradiation were used to demonstrate effects of paracrine factors on cell proliferation (Fig. 6c,  d). Expression levels of KI67, PCNA, and C-MYC genes in hDFs treated with 25% and 50% CM for 48 h, but not C-MYC in 25%-CM-treated hDFs, were significantly upregulated than those in the NT group. In case of NHEKs, 50%-CM-treated cells had significantly enhanced gene expression levels of PCNA and C-MYC compared to control cells.

Improved cutaneous wound healing by injecting green OLED-treated hADSCs
According to previous experimental results, hADSCs harvested at 24 h after green OLED irradiation for 24 h were injected into mouse skin wound sites to promote tissue repair. As shown in Fig. 7a, the wound healing rate of the GTC group began to take the lead on day 6. The GTC group showed significantly higher wound healing rate than the NT group on day 6 and other groups on day 9. Moreover, gene expression levels of Cd31 and Laminin in the GTC group were significantly increased compared to those in other groups (Fig. 7b).
Furthermore, the GTC group showed higher levels of laminin and involucrin protein signals than other groups based on immunohistochemistry (Fig. 7c). Additionally, histological images (Fig. 7d) of the GTC group revealed not only improved tissue regeneration, but also reduced inflammation and fibrosis through H&E and Masson's trichrome staining.

Discussion
By employing various techniques for photo-physical stimulation, PBM has been widely used to study the differentiation of stem cells or to enhance cellular functions such as proliferation and migration [42,43]. Although various research studies on PBM have led to advancements overcoming limitations of stem cells therapy, more studies investigating detailed underlying molecular mechanisms of PBM are required. Technical challenges in application of PBM remained to be addressed. Previous studies have focused on the evaluation of cellular responses based on wavelength variance or tried to determine cellular responses caused by diverse conditions with an identical type of light source [44][45][46]. Notably, contrasting results have been observed when different types of cells are irradiated with light of the same wavelength [47,48]. These studies show that the intracellular mechanisms triggered by PBM rely on the light source of PBM and the type of irradiated cells. In the present study, we used OLED light with a wavelength peak of 532 nm to irradiate hADSCs. We designed experiments to reveal biological changes in hADSCs promoted by green OLED irradiation and their underlying mechanisms. It has been suggested that CCO plays a major role in cellular responses of stem cell to PBM [49]. OPNs besides CCO are also related to responses to PBM [18]. Based on these reports, we examined changes in the expression of OPN family (OPN3 and OPN4) in hADSCs after green light-based PBM. Both OPN3 and OPN4 gene expression levels were upregulated after 24 h of irradiation. However, only OPN4 was increased at the protein level. Although further examination using various cell types is needed, these results may indicate that OPN4 is more sensitive to OLED light with 532 nm wavelength than OPN3 in hADSCs. Further supporting evidence of stimulation of OPN4 was gathered by determining ROS levels in hADSCs after OLED irradiation. Conigrave and Ward have reported that OPN4 can activate G q protein, thereby increasing cytoplasmic Ca 2+ concentration and ROS levels [22,23]. Although we did not observe significant difference in intracellular Ca 2+ concentration, both cellular ROS assay and H2DCFDA staining showed increases in ROS level and gene expression levels of PLC-β and IP3R known to mediate ER Ca 2+ release [50] were elevated in green OLED-treated group of hADSCs. These results indicate that green OLED irradiation does not follow the  5). b VEGF and HGF proteins secreted from hADSCs at 0, 24, and 48 h after green OLED irradiation for 24 h as evaluated by ELISA (n = 5). Gene expression levels of KI67, PCNA, and C-MYC in hDFs and NHEKs cultured in 25% (c) and 50% (d) conditioned media as evaluated by qRT-PCR (n = 5). *p < 0.05; **p < 0.001 compared to no treatment group with unpaired t test molecular mechanism of TRP channels or the CCO pathway, the proposed mechanisms in PBM when using red or near-infrared light in the range of 600-700 nm and 780-1100 nm [49]. As shown in our results, green OLED treatment had no obvious cytotoxic effect on hADSCs despite generation of ROS. These results demonstrate that green OLED irradiation does not induce overexpression of ROS that can increase the cytotoxicity [51]. It has been reported that moderate level of ROS generation is associated with upregulation of growth factors such as VEGF which affects cell proliferation, migration, and adhesion of mesenchymal stem cells [52][53][54]. Accordingly, proliferation markers KI67 and C-MYC [55,56] were increased at 24 h after OLED irradiation. CCK-8 assay result also showed that cell proliferation was enhanced with green OLED treatment. The reduction of KI67 and C-MYC expression at 48 h after irradiation demonstrates that our PBM-based approach has a 24 h validity. This may be lengthened with a longer irradiation time of highintensity light. However, the possible cytotoxicity should be carefully analyzed. We observed that the increase in Fig. 7 Improved in vivo wound healing in green OLED-treated hADSCs. a Representative photographs of skin wounds at 0, 3, 6, and 9 days after skin wound modeling and various treatments with wound coverage ratio quantification (n = 5). Mice were divided into three groups: no treatment (NT) served as control, hADSCs (SC), and green OLED-treated hADSCs (GTC) groups. b Gene expression levels of Acta2, Cd31, and Laminin in skin wound regions on day 9 as evaluated by qRT-PCR (n = 5). c Laminin (blue: nuclei, green: laminin, scale bar = 100 µm) and involucrin staining (blue: nuclei, green: involucrin, scale bar = 250 µm) of vertical sections of skin wound regions of each group on day 9. d H&E (scale bar = 200 µm) and Masson's trichrome staining (scale bar = 200 µm) of vertical sections of skin wound regions of each group on day 9. *p < 0.05; **p < 0.001 compared to other groups with unpaired t test the motility of hADSCs was due to increased expression levels of CXCL12, CXCR4, and SDF-1 [57][58][59]. Increased level of β-CATENIN, INTEGRIN α5, and INTEGRIN β1 after light irradiation is known to be associated with cell adhesion [60,61], including adhesion of hADSCs. VEGF and HGF are widely known to play important roles in wound healing [62]. They were also upregulated in hAD-SCs irradiated with green OLED. Notably, CM obtained from irradiated hADSCs enhanced the expression of KI67, PCNA [63], and C-MYC genes in hDFs and NHEKs possibly due to increased amounts of VEGF and HGF in the CM of irradiated hADSCs. Here, hDFs and NHEKs are widely used cell types for in vitro assessment of wound healing [64]. For animal experiment, mouse cutaneous wound model was used. We observed the most prominent wound healing in the GTC group. Expression levels of Cd31 gene (angiogenesis) [65] and Laminin gene (skin regeneration) [66] were increased at wound sites of mice injected with green OLED-treated hADSCs compared to those of mice in other groups. Immunohistochemical staining of laminin and involucrin [67] known to be associated with skin tissue regeneration showed significant increases in GTC-injected mice. Overall, our results verified that proliferation, migration, and adhesion of hADSCs as well as paracrine factor expression, all of which known to contribute to wound healing, were significantly enhanced after green OLED irradiation. Our approach suggests that green OLED-based PBM might serve as a novel preconditioning strategy to augment the wound healing efficacy of hADSCs.

Conclusions
The present study demonstrates that a long-term mild PBM has better therapeutic efficacy than conventional methods. Parameters to maximize its efficacy are also suggested. We newly established photoreceptor molecular mechanism to hADSCs corresponding to green OLED used in this study. Importantly, cell proliferation, migration, adhesion, and paracrine factor expression in hADSCs are comprehensively enhanced by green OLED irradiation. These effects lasted for 24-48 h. These results indicate that green OLED can be utilized for cell culture without worrying about differentiation or mutation of cells. Furthermore, preclinical values of green OLEDtreated hADSCs were verified with a mouse wound model compared to conventional cell therapy. Collectively, this report provides a concept of wound healing improved by hADSCs, including OPN4 stimulated by OLED irradiation with a wavelength peak of 532 nm.