G Protein-coupled Receptors and Proopiomelanocortin Expression After Light Emitting Diode Irradiation in Diabetic Wound Healing

Objective. The purpose of this study is to investigate whether light-emitting diode (LED) light at different wavelengths can improve wound healing in both diabetic and normal fibroblasts. Materials and Methods. Both diabetic and normal fibroblast cell lines were cultured and artificial wounds were created on the cultured cells in petri dishes as a streak line with pipette tips. Then, both cells were separately irradiated with 635 nm (red), 520 nm (green), and 465 nm (blue) LED lights at 0.67 J/cm² for 10 minutes. Immediately after LED irradiation, messenger ribonucleic acid samples of each cell line were isolated for microarray analysis. Results. The investigator found that G protein-coupled receptors (GPR) class A, a rhodopsin-like structure gene, were significantly upregulated in all treated groups by transferring the signal to proopiomelanocortin (POMC) genes in diabetic cells. In addition, for normal cell groups, the expression of other genes relevant to viral defense responses markedly increased. However, in diabetic cells, genes relating to acute inflammatory response and mitotic cell cycle were highly expressed. The investigator also found that diabetic cells responded significantly better to wound healing attempts than normal cells because inflammatory response, cell migration, cell proliferation, cell adhesion, and regulation of mitosis pathways could be instantly activated by red and green LED lights. Conclusions. These lights activate the expression of GPR class A gene, which receives photons and transfers this signal to other downstream pathways inside the cell, specifically to the POMC gene, which will recover the wound-healing process to the normal stage. 

Diabetes mellitus is a major cause of chronic wound problems. Patients with diabetes have significantly higher amputation rates than the nondiabetic population.¹ One factor involved in the development of the classic “diabetic” ulcer is the frequency of peripheral vascular disease in patients with diabetes, which interferes with blood supply; another key factor is neuropathy.² The lack of protective sensation can result in deeper wounds. The loss of autonomic nerves leads to decreased sweating and dry, cracking skin, causing an increased risk of skin breakdown.³ 

Previous studies^4,5 have demonstrated a number of genes related to diabetes, especially the precursor protein proopiomelanocortin (POMC) gene. Expression of the POMC gene is crucial in pituitary melanotroph and corticotroph cells in order to control body weight.⁶ In addition, mutation of the POMC gene leads to early-onset type 2 diabetes and obesity. Furthermore, diabetes mellitus leads to erroneous immune function, which means a small wound has a higher propensity for infection.

Wound healing pathways
There are 8 important pathways (eFigure 1) promoting wound healing at molecular levels: (1) collagen synthesis, (2) cell migration, (3) cell cycle and differentiation, (4) angiogenesis and growth hormone, (5) blood clotting, (6) extracellular matrix and focal adhesion, (7) calcium ion for signaling, and (8) immune and inflammatory response. However, for diabetic cells, with the exception of collagen synthesis, cell migration, and cell cycle and differentiation, all of these pathways malfunction, causing an improper wound healing process. 

Light-emitting diode (LED) phototherapy as photobiomodulation
Phototherapy is a useful technique to modulate molecular and cellular responses.^7,8 There are a number of cellular molecules that can receive photon energy from light, such as rhodopsin, cytochrome c, and chlorophyll. Due to their molecular porphyrin structures, these molecules serve as antennas to trap photon energy, which can consequently activate the activities of downstream molecules.

Light-emitting diode is a light source typically used in phototherapy. Previous studies^9,10 reported that LED light at 660 nm and 890 nm significantly accelerated wound healing in leg ulcers.

Although clinical studies have investigated the results of phototherapy on normal, healthy patients,^11,12,13 to the best of the author’s knowledge, no study has completely evaluated both the benefits of different wavelengths of LED lights and its effects on gene expression of diabetic cells together using microarray technique. This study aims to comprehensively analyze the whole gene expression of normal and diabetic cells irradiated with red (635 nm), blue (465 nm), and green (520 nm) LED lights and determine the appropriate condition that can improve cellular function as related to the wound-healing process by using gene expression array technique. 

Light-emitting diode light source
An apparatus equipped with blue (465 nm), green (520 nm), and red (635 nm) LEDs with a 1-A power supply was constructed for this experiment. Calibrating analysis of the light’s emission spectrum and power was provided by a monochrome meter and a power meter from the Department of Physics, Prince of Songkla University (PSU; Hat Yai, Thailand). The LED array (15 cm x 15 cm), assembled by a technician from the Department of Electrical Engineering, PSU, consisted of 3 different wavelength array panels that could be placed over and completely covered above a 10-cm cell culture plate. In this array, the blue LED emitted light between 440 nm and 500 nm with peak emission at 465 nm, the green LED between 495 nm and 575 nm with peak emission at 520 nm, and the red LED between 610 nm and 660 nm with peak emission at 635 nm.

Cell culture assay
Both the type 2 diabetic fibroblast cell line (DMCL) from a donor diagnosed with type 2 diabetes (untransformed, Caucasian, male, maturity onset diabetes of the young, aged 23 years, Cat.#AG06083) and the healthy normal fibroblast cell line (NCL; untransformed, Caucasian, male, leg skin, Cat.#GM03440) were obtained from Coriell Institute for Medical Research (Camden, NJ) and cultured at 37°C 5% carbon dioxide in Dulbecco’s Modified Eagle’s medium (supplemented with 10% fetal bovine serum and 100 µg/mL penicillin-streptomycin; GIBCO Laboratories, Gaithersburg, MD) in 10-cm plates with an initial seeding density of 2 000 000 cells. During the entire experiment, these cells were maintained and used no more than 5 passages. The DMCL and the NCL were cultured in the normal blood glucose level, assuming that they were from well-glycemic control patients. The DMCL and the NCL served as the treatment group and control group, respectively.

Light-emitting diode irradiation assay and messenger ribonucleic acid (mRNA) isolation
Before LED light exposure, the cells were cultured until 90% confluence, and a number of artificial wounds were created randomly (to avoid bias of wound location) with a 1-mL plastic pipette tip. Then, in the 6 treatment groups, the 3 groups of the DMCLs and the 3 of the NCLs were exposed to the light at 0.67 J/cm² for 10 minutes at different wavelengths. Each group was pentaplicated (5 petri dishes) and 5 artificial wounds were created on each petri dish. One diabetic fibroblast and 1 normal fibroblast received no light exposure to serve as controls. The labeled names and conditions for each treated group are shown in Table 1. The energy power of the light that the cells received was carefully calibrated by an optical power meter. The irradiation distance from this LED array to media surface was 10 cm. This apparatus had been placed in a lamina flow (Labculture Class II [Low Noise] Biosafety Cabinet; Esco Technologies, Inc, Horsham, PA), and a cooling airflow fan maintained a consistent temperature during light exposure.  Afterwards, all cells were immediately trypsinized and their mRNA was extracted.  All extractions were performed with a GeneJET RNA Purification Kit (Thermo Fisher Scientific, Waltham, MA). Each condition was pentaplicated for reliable and valid results. 

Messenger ribonucleic acid microarray assay
All mRNA samples were shipped to an Agilent-certified microarray service in India, and the purity of the mRNA samples was tested by Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA) with RNA integrity number > 8 before hybridizing with microarray chips. SurePrint G3 Gene Expression Microarrays v2 (8 x 60k; Agilent) were used for this experiment.

Statistical and microarray data analysis
All raw data were analyzed by GeneSpring 13 (Agilent). All gene expression profiles with 2-fold differences were selected and followed by differential gene expression analysis, gene ontology analysis, and pathway analysis. The cut-off level for all analysis was P < .05. In addition, raw data were submitted to the National Center for Biotechnology Information (NCBI) database for other researchers to use.

Differential gene expression results
After raw data of all fibroblast cells were analyzed, all results were normalized (intensity values shown in eFigure 2). Total gene expression profiles and raw data can be accessed through the Gene Expression Omnibus database of NCBI at GSE78017 and GSE78018. There were a number of differential gene expressions of each treatment group that varied significantly up and down more than twofold; some genes were expressed higher in the diabetic groups (eFigure 2). There was a significant difference between the DMCLs and NCLs that were grouped based on their conditions and genes (eFigure 3).

The complete number of differential gene expression between each condition and the NCL control is shown in Table 2. Although the percent changes of gene expression of 3% to 10% were about the same between the groups, the set of affected genes were different when the data were further analyzed into their functions, gene ontology, and pathways. 

Gene ontology analysis against normal healthy cells
In order to explore how light irradiation affects cellular molecules and the biological function of fibroblast cells, all differential expression results were further categorized into gene ontologies. After analysis, the gene ontology of the treated NCL groups demonstrated that the red light significantly affected cellular components up to 35%, especially the cell membrane area, and also induced transmembrane signaling activities at molecular levels up to 14% (eFigure 4). In addition, the green light significantly activated the cell periphery function and the activities of spindle fibers at the kinetochore area to promote chromosomal condensation for cell proliferation. Blue light was able to only trigger the activities of G protein-coupled receptors (GPR) significantly by 9%, which was statistically significat at P < .05. 

On the other hand, once the investigator further analyzed for the affected biological functions (eFigure 5), it was discovered that 4 major factors promoting wound healing were significantly activated: cell signaling, cell response, cell proliferation, and inflammatory response. For example, the red light boosted cell signaling (up to 13%) and cell response to cytokine stimulus (5%), while green light notably improved cell response to type I interferon and viral infection, cell proliferation (up to 11%), and inflammatory response. In short, for NCLs, the green light seemed to promote wound healing related to viral infection better than the red light. 

The gene ontology of the treated DMCLs are shown in eFigure 6.  After analysis, the investigator found all 3 colors of light could significantly affect cell components up to 35%, especially most proteins on the membrane. 

However, only red light could significantly improve the molecular functions of calcium-ion binding, cytokine activities, and receptor-binding proteins, of which about 11% were affected. eFigure 7 shows the affected biological functions from the treated diabetic groups. Only the green light and the red light could improve 5 major factors related to wound healing: cell signaling, cell response, cell migration, cell proliferation, and inflammatory response. The effects would improve the ability of nearby cells to proliferate, migrate, and induce immune system activities at the wound area. This could help promote wound healing in diabetic cells. In contrast, blue light had little effect on cell response, cell migration, and cell proliferation. Therefore, in diabetic cells, the red light and the green light might be effective for improving the impaired wound-healing process in diabetic cells. 

Pathway analysis
After gene ontology analysis, highly expressed genes from each condition were analyzed in terms of their pathways and natural language processing interactions. As previously observed from the gene ontology analysis, GPR might serve as the first receiver of the light irradiation. eFigures 8 and 9 list the significantly affected GPR genes; the following downstream pathways from each light condition were further analyzed.

Red light irradiation-treated group
In comparison with the normal control group, only genes relating to response to virus and defense response to virus from the red light-treated normal cells (NCL RED) group were significantly upregulated (eFigure 5). This antiviral response pathway started from the myxovirus (influenza virus) resistance 1 (MX1) gene (eFigure 10); then, it regulated toll-like receptor (TLR) 3 genes, which belong to the TLR families and induce other interferon regulatory factors. However, calcium-ion binding, regulation of acute inflammatory response, and blood clot formation genes were greatly increased in the red light-treated diabetic cells (DMCL RED) group (Table 3). This pathway started from chemokine ligand (CXCL) 8 and GPRC6A genes (eFigure 11). Then, these genes regulated the endothelin receptor type A and somatostain genes. Next, they regulated POMC and corticotropin-releasing hormone (CRH) genes, respectively, and these also triggered other molecules to activate the wound-healing effect. Homeostasis, cell survival, and immunoglobulin production effects also were observed. Moreover, parathyroid hormone (PTH) genes were also affected. 

Nevertheless, when DMCL RED was compared with NCL control, the activities of angiogenesis, cell adhesion, growth factor activity, extracellular matrix, and hair cycle obviously decreased. In contrast, in NCL RED, most of the other important pathways related to the wound-healing process were downregulated, such as the response of wounding, collagen fibril assembly, extracellular matrix, elastic fiber formation, and formation of fibrin clot.

Green light irradiation
The comparison of the green light-treated groups with the NCL control are shown in Table 3. For NCL GREEN, the expressions of cell proliferation, cell response, and inflammatory response increased (eFigure 5). And, it was surprising that immune system process, cellular response to type I interferon, and defense response to virus also increased. This antiviral response pathway of NCL GREEN started in a similar process to NCL RED (eFigures 10, 12). On the other hand, extracellular matrix signaling and response to wounding were apparently down (data not shown). For DMCL GREEN, like the expression of DMCL RED as represented in Table 3, the expressions of calcium-ion binding, regulation of acute inflammatory response, and inflammatory response were increased as well. This pathway started from the CXCL2 gene (eFigure 13). These genes then regulated the calcitonin related polypeptide alpha, neurotension, atypical chemokine receptor 3, cholecystokinin, POMC, and CRH genes, which trigger other molecules to activate the wound-healing effect. Moreover, the follicle-stimulating hormone receptor gene was also affected. However, when DMCL GREEN responses were compared with the DMCL control, cell division, mitotic cell division, endochondral ossification, metabolism of angiotensinogen to angiotensin, and cell proliferation were upregulated (eFigure 6). 

Blue light irradiation
For the blue light-irradiated groups compared with NCL, only GPR activity was strongly escalated in the blue light-treated normal cells (NCL BLUE) as shown in eFigure 5. This pathway started from the beta-2 adrenergic receptor, component 5A receptor 1, P2Y purinoceptor 12 (P2RY12), and olfactory receptor 2G3 genes (eFigure 14). Then, these genes regulated some other genes in their pathways, shown in eFigure 14, that they interacted with along their downstream pathway. However, the expression of other downstream pathways was not significantly changed. For blue light-treated diabetic cells (DMCL BLUE), shown in eFigure 15, the key proteins, which were affected after the GPR proteins became activated from light irradiation, were POMC and CRH genes. Homeostasis, cell survival, and immunoglobulin production effects were observed as well; the PTH gene was also affected. Moreover, positive regulation of cell proliferation, epithelial cell differentiation, and cell migration were notably boosted. In contrast, collagen assembly, extracellular matrix, potassium channels, alpha cells, immune regulation, and interleukin-7 signaling were remarkably downregulated. 

Gene expression significantly affects the wound-healing pathway
The investigator found the red-light irradiation significantly activated the function of cell signaling in both normal and diabetic cells, while the green light could trigger cell proliferation and inflammatory response in normal cells. In contrast, the blue light had little effect on GPRs.

This study demonstrated that the red light could trigger the activities of cell proliferation and inflammatory response via GPR class A genes, which are responsible for cell signal transduction from the outside environment of the cell to activate other intracellular molecules. Due to the structure of GPR class A, which is a rhodopsin-like structure located outside the cell membrane, it can directly bind with the photons from the light and transfer these photon signals to other downstream pathways inside the cell. These downstream pathways are directly linked to cell migration, inflammatory response, and cell proliferation. A number of genes in the GPR class A group were found to be significantly responsive to the LED light, including P2RY14 and luteinizing hormone/human chorionic gonadotrophin receptor. These genes activate the activity of POMC and CRH genes, which are highly expressed in pituitary melanotroph and corticotroph cells but normally suppressed by mutation in diabetes.¹⁴ This activity improved cellular homeostasis, cellular metabolism, and other immune responses of wound healing via increased activities of cell migration, cell proliferation, blood clotting, extracellular matrix and focal adhesion, calcium ion signaling, and inflammatory response pathways.

Zhang et al¹⁵ evaluated the effects of the red light and found that it could potentially accelerate the wound-healing process in normal healthy cells. However, no previous study has investigated the effect of LED irradiation on diabetic cells at the gene level. To the best of the author’s knowledge, the present study is the first to evaluate the effect of various wavelengths of LED light on diabetic cells.

In addition, the investigator found that the blue light could not improve any wound-healing process as previously reported.¹⁶ This might be explained by the structure of GPR, which can only be activated by photons from red and green lights.

This study demonstrated that red light might be used as an adjuvant treatment for diabetic ulcers. However, further animal and clinical studies are required. In addition, the investigator revealed that red and green lights could activate antiviral response pathways, which may be for another potential clinical implication such as treatment for herpes infection.

A limitation of this study was the lack of proteomic and metabolomic analysis in this experiment due to limited funding. In order to produce better results, such analyses, animal models, and clinical testing are recommended for future research.

Red and green LED light could improve diabetic wound healing by upregulating gene expression in 6 of 8 wound-healing pathways, including cell migration, cell cycle and differentiation, blood clotting, extracellular matrix and focal adhesion, calcium ion for signaling, and immune and inflammatory response. These lights activate the expression of GPR class A gene, which receives photons and transfers this signal to other downstream pathways inside the cell and specifically to the POMC gene, which will recover the wound-healing process to the normal stage.

The author thanks Professor Chittanon Buranachai, Physics Department, PSU (Hat Yai, Thailand) for measuring the LED light spectrum and power; Mr. Somjade Rong-rueng, Electrical Engineering Department, PSU, for building a prototype of a LED machine; and the Department of Medicine at PSU for funding this research.

Affiliation: Prince of Songkla University, Hat Yai, Songkhla, Thailand

Correspondence:
Pongsathorn Chotikasemsri, PhD 
Biomedical Engineering Department
Faculty of Medicine
Prince of Songkla University
15 Kanjanavanich Road, 
Hat Yai, Songkhla
Thailand
golfgoofy19@gmail.com

Disclosure: The author discloses no financial or other conflicts of interest. This research was funded by the Department of Medicine, Prince of Songkla University.