Liquiritigenin reverses skin aging by inhibiting UV‐induced mitochondrial uncoupling and excessive energy consumption

The accumulation of reactive oxygen species (ROS) generated by UV radiation can lead to lipid, protein, nucleic acid, and organelle damage, one of the core mechanisms mediating skin aging. In the photoaging process, how ROS drives the imbalance of the body's complex repair system to induce senescence‐like features is not fully understood.

ultraviolet (UVC, 200-280 nm), of which UVA and UVB can penetrate the ozone layer and reach the Earth's surface. UVA accounts for 95% of sunlight, and it can penetrate deep into the basal layer of the epidermis and the dermis. Long-term exposure to UVA can cause various morphological and histological changes in the skin, resulting in relaxation, dryness, increased wrinkles, abnormal pigmentation, telangiectasia, and functional disorders that can even develop into precancerous lesions and skin malignancies. 4,5 Over recent years, with the increasing amount of ultraviolet radiation in the sun and the deterioration of the ecological environment, more and more attention has been paid to preventing or delaying skin aging, especially photoaging. However, photoaging is a complicated and not yet fully understood process. The exploration and elucidation of this process will help formulate corresponding countermeasures and develop safe and efficient antiphotoaging substances.
Reactive oxygen species (ROS) are continuously produced as a by-product of the electron transport chain of mitochondrial aerobic metabolism and are considered to be the leading cause of intrinsic aging in addition to genetic factors. 6 Approximately 1.5%-5% of oxygen consumed by the mitochondria consumption is reduced by the bifurcated electrons to form superoxide, which is subsequently converted to hydrogen peroxide and other ROS in the skin. 7 UV induces ROS production that impairs the structure and function of cells, which activates a myriad of signaling pathways, leading to reduced collagen production and decreased synthesis and activation of matrix metalloproteinases (MMPs), which is responsible for degrading connective tissue, secreting Senescence-Associated Secretory Phenotype (SASP), and ultimately promoting skin aging. 8 The body has evolved a sophisticated system to eliminate the excessive accumulation of ROS, such as glutathione reductase (GSH), nicotinamide adenine dinucleotide phosphate (NADPH), and superoxide dismutase (SOD), which directly eliminated ROS. The Keap-1-Nrf2 [Kelch-like ECH-associated protein 1-nuclear factor(erythroidderived 2)-like 2] regulatory complex present in the cytoplasm regulates the expression of HO-1 or γ-GCLC proteins, which exerts anti-inflammatory, antiapoptotic, and antiproliferative properties to protect against oxidative stress caused by UVA radiation and maintain skin cell homeostasis. 9 The mitochondrial machinery of oxidative phosphorylation consists of four protein complexes (I to IV) in the electron transport chain (ETC) in addition to complex V (ATP synthase), all located in the inner mitochondrial membrane. 10 According to Mitchell's chemiosmotic theory, mitochondrial ATP production relies on the coupling between the proton gradient on either side of the inner mitochondrial membrane and the use of this proton-motive force to feed the complex V-a process known as proton leak-generating heat instead of ATP, that is an important mechanism for energy dissipation that accounts for up to 25% of the basal metabolic rate. 11,12 Mitochondrial uncoupling proteins (UCPs) can be activated by exogenous and endogenous superoxide and lipid peroxidation products, thereby reducing the efficiency of mitochondrial oxidative phosphorylation and the production of ROS. 13,14 Therefore, mitochondrial uncoupling has been identified as a cytoprotective strategy under conditions of oxidative stress, including diabetes, drug resistance in tumor cells, ischemia-reperfusion (IR) injury, or aging. Moreover, evidence indicated that UVA stimulation is a crucial contributor to mitochondrial disorder and affects the growth and death of keratinocytes. 15 However, there is no clear conclusion on whether the large amount of ROS generated by UV can activate mitochondrial uncoupling and how it affects the process of photoaging.
Glycyrrhiza uralensis (Gan-Cao) has been widely used in traditional herbal medicines prescriptions and daily necessities, which is often added to cosmetics because its whitening and sunscreen effect on the crude extract. Liquiritigenin (LQ) is a flavonoid extract from the root of Glycyrrhiza uralensis (Gan-Cao), commonly used in food sweeteners, has been proven by research to have various pharmacological activities such as antioxidant, antitumor, antiinflammatory, and antibacterial. [16][17][18][19] In this study, we established a UVA-irradiated photoaging model and employed RNA-seq and O2K mitochondrial function assay to systematically explore the effects and consequences of UVA on skin keratinocytes regarding mitochondrial uncoupling. Considering that the antiphotoaging active substance, LQ, prevents the massive loss of skin nutrients by inhibiting the excessive uncoupling of mitochondria stimulated by UV and improves the net energy output of cells, thereby promoting skin cells proliferation, reversing the photoaging process, and providing an experimental basis for its use, we spotted to study it as a potential photoaging protective agent. After a 24 h incubation, the culture medium was replaced with phosphate-buffered saline (PBS) and then exposed to UVA irradiation at a distance of 7.5 cm using two UVA lamps (312 nm, 20 W; Phillip). UV strength was measured with a UVA radiometerbiosteritron (Lingshang). Non-UVA cells were kept in the same culture conditions without UVA exposure. Following irradiation, the cells were incubated in DMEM without FBS and penicillin/streptomycin for 24 h.

| Cell viability assay
Cell proliferation/cell viability was determined using a cell counting kit-8 (CCK-8; Meilunbio). The CCK-8 was used to count living cells by measuring dehydrogenase activities in cells. 22 First, to detect the cytotoxicity of LQ on NHEKs, cells were seeded at a density of 3 × 10 3 cells/well in a 96-well plate overnight and pretreated with various concentrations (5-200 μM) of LQ for 24 h without UV radiation. The cells were added 10 μl/well of CCK-8 solution and incubated at 37 °C for 1-2 h, then analyzed at 450 nm using a microplate reader (PerkinElmer). Second, to determine the optimal dose of UVA radiation, cells were seeded at a density of 3 × 10 3 cells/well in a 96-well plate for 24 h, a culture medium was replaced with phosphatebuffered saline (PBS) and then exposed to 4-12 J/cm 2 UVA irradiation at a distance of 7.5 cm using two UVA lamps. Non-UVA cells were kept in the same culture conditions without UVA exposure. After exposure to 12 J/cm 2 UVA, the cells were incubated in DMEM without  After cells were irradiated and incubated in a serum-free medium for 24 h, then washed with PBS and added 1 ml fixative solution to a well for 15 min at room temperature. The fixative solutions were removed and washed three times with PBS, and the fixed cells were stained in 1 ml staining solution and kept in 37 °C incubators without containing CO 2 overnight. Then the images were captured using an optical microscope (Leica). In addition, the SAβ-gal enzyme contents were quantificationally detected at 400 nm using a microplate reader referring to instructions about the β-galactosidase Assay Kit.

| Measurement of reactive oxygen species (ROS) Production
Intracellular ROS formation was measured by adding 2′,7′-dichlo rofluorescein diacetate (DCF-DA; Sigma-Aldrich). DCF-DA is nonfluorescent until it is hydrolyzed by intracellular esterase and oxidized into the highly fluorescent 2′,7′-dichlorofluorescein (DCF) in ROS presence. 20 Following treatment with LQ (20, 50, 100 μM) for 24 h, NHEKs were given 12 J/cm 2 UVA irradiation. After 24 h, 20 μM DCF-DA was added to the serum-free DMEM for 30 min at 37 °C and dark environment, and the medium containing DCF-DA was removed by washing with a serum-free DMEM. The ROS were, respectively, captured using Laser-scanning confocal microscopy (Leica) and detected by flow cytometry (Beckman Coulter) using a fluorescence-activated cell sorter (FACS-Calibur) at excitation/emission wavelengths of 488/535 nm.

| Antioxidant enzyme activity
To evaluate intracellular antioxidant enzymes activity, after treatment of LQ and UVA irradiation, NHEKs were collected, washed, and resus-

| High-Resolution FluoRespirometry
The Oroboros O2k for High-Resolution FluoRespirometry (Oroboros Instruments, Innsbruck, Austria) was used to measure sample O2 consumption using two 2-ml Duran glass chambers under continuous stirring and temperature at 37 ± 0.001°C. 23 To analyze mitochondrial function, we used a SUIT-003 O2 ce D028, a coupling-control pro- and treated with 20, 50, and 100 μM LQ for 24 h before 12 J/cm 2 UVA irradiation. After 24 h, the cells were collected, washed, and resuspended in DEME. According to the D028 protocol, we add ce1-ce6 reagents in order and process results.

| Mitochondrial index detection
The mitochondria membrane potential was detected using a mitochondrial membrane potential assay kit with JC-1 (C2006, Beyotime Institute of Biotechnology). After LQ and UVA irradiation treatment, NHEKs were incubated with the JC-1 probe in the culture media for 30 min in the dark. After washing with PBS, cells were observed under a fluorescence microscope. The fluorescence images were captured by Laser-scanning confocal microscopy. Based on the luciferase-catalyzed oxidation of D-luciferin, the ATP level was evaluated according to an ATP assay kit (S0026, Beyotime Institute of Biotechnology). The NAD + /NADH Assay Kit (S0175, Beyotime Institute of Biotechnology, Shanghai, China) was employed to determine the NAD+/NADH ratio at 450 nm wavelength.

| Western blot assay
NHEKs were seeded in a 6 cm culture dish at a density of 5 × 10 6 cells per well and treated with 20, 50, and 100 μM LQ for 24 h before 12 J/ cm 2 UVA irradiation. After 24 h, the cells were rinsed with PBS and lysed in ice-cold RIPA buffer containing phosphatase inhibitors and protease inhibitors for 45 min. Thirty micrograms of protein were separated on 12% and 8% SDS-PAGE and transferred to a PVDF membrane for Western blotting. PVDF membrane was blocked with 5% nonfat milk-supplemented PBST (1% Tween-20) for 1 h at room temperature. Subsequently, the primary antibodies (Nrf2, Keap-1, SOD2) and secondary antibodies (β-tubulin) were used for 4°C overnight incubation and 2 h treatment. The blots were captured using enhanced chemiluminescence (Bio-Rad). Additionally, densitometry analysis was conducted by Image J software (National Institutes of Health). The primary antibodies used were obtained from ABclonal and Boster.

| Oil Red O Stain
The frozen section was taken out from the refrigerator, staying at 37°C for 30 min. Then, the sample was incubated at 60% isopropanol for 10 s followed by oil red for 12 min without light. Next, the sample was put in 60% isopropanol for several seconds, washed for 2 min, and restained with hematoxylin for 30 s, washing for 1-3 min.
After that, the sample was put into 1% hydrochloric acid for 3 s, washed until it returned to blue, and examined under a microscope.
At last, the section was sealed with glycerogelatin, and the images were captured using an optical microscope.

| LC-MS analysis
To evaluate intracellular linoleic acid contents, after treatment of LQ and UVA irradiation, cells were collected and treated with an ultrasonic crusher. Add 0.1% 2,6-di-tert-butyl-4-methyl phenol (BHT) in ice methanol for centrifugation, and the supernatant was obtained. Add the linoleic acid isotope internal standard to 15-20 μl of the supernatant, put it in for 25-35 min, and then centrifuge. The supernatant obtained by centrifugation is dried with nitrogen and then derivatized with 50-60 μl of n-butanol at 60-65°C. Treat for 20-30 min, then blow dry with nitrogen, and redisperse and dissolve with 100-150 μl of mobile phase to obtain the sample loaded and tested. The columns were conditioned using 6 ml of methanol followed by 6 ml of water. Samples were mixed with at least 12 times their volume of ice-cold water and loaded onto the column. The columns were then rinsed with 10% methanol, dried with a vacuum (approximately 200 mmHg) for 2 min, and eluted with methanol containing 0.0004% (v/v) BHT into a glass culture tube containing 10 μl 30% glycerol in methanol.
The resulting eluate was dried under nitrogen gas and reconstituted in HPLC grade methanol, and the content of linoleic acid was measured by LC-MS.

| Immunofluorescence (IF)
NHEKs were seeded in a confocal dish at a density of 1 ×

| Differential expression analysis
Total cDNA derived from the sample was sequenced on an Illumina NovaSeq 6000 system at Novogene Technology. Differential expression analysis of two conditions/groups (two biological replicates per condition) was performed using the DESeq2 R package (1.20.0). DESeq2 provides statistical routines for determining differential expression in digital gene expression data using a model based on the negative binomial distribution. The resulting p values were adjusted using the Benjamini and Hochberg approach for non-UVAling the false discovery rate p adj ≤ 0.05 and |log2(foldchange)| ≥ 1 were set as the threshold for significantly differential expression.

| KEGG analysis
KEGG is a database resource for understanding high-level functions and utilities of the biological system, such as the cell, the organism, and the ecosystem, from molecular-level information, especially large-scale molecular data sets generated by genome sequencing and other high-throughput experimental technologies (http://www. genome.jp/kegg/). We used the cluster profiler R package (3.8.1) to test the statistical enrichment of differential expression genes in KEGG pathways.

| Statistical analysis
All numerical data were presented as mean ± standard deviation  Figure 1A, compared with the non-UVA group, with the increase in UVA radiation dose, the cell viability decreased by 20%, 34%, and 48%, respectively, with significant differences (p < 0.05). It was shown that UVA radiation would produce cytotoxicity to NHEKs, thereby inhibiting their proliferation, and in a dose-dependent manner, so 12 J/cm 2 was chosen as the radiation dose for subsequent experiments.

| UVA induces intracellular ROS accumulation and cellular senescence in NHEKs
According to the theory of free radical aging, reactive oxygen species (ROS) are mainly derived from cellular oxidative metabolism Next, we verified the senescence rate of NHEKs in this state. β-Galactosidase (SAβ-Gal) is the earliest and most widespread aging biomarker and is considered the "gold indicator" for demonstrating senescent cellular and tissue aging. When H 2 O 2 induced cellular senescence, it appeared in NHEKs but not in HaCaT keratinocytes. 26 The cells in each group were stained with β-galactosidase, the senescent cells were stained blue, and the proportion of blue-stained cells in the whole cell was the senescence rate. As shown in Figure 1D, the bluestained area of cells in the UVA group was significantly increased compared with the non-UVA group. The color was considerably darker, and the morphology was markedly changed. Therefore, we successfully exploited the UVA-induced senescence injury model of NHEKs.

| UVA reduces ATP production and stimulates the mitochondrial uncoupling of oxidative phosphorylation processes in NHEKs
To investigate mitochondrial function and energy consumption, we used the O2K mitochondrial respiration function system to determine mitochondrial respiration rate, oxygen consumption, and ATP production in NHEKs ( Figure 3A). We found that after UVA irradiation, the mitochondrial respiration rate was reduced. The normal oxidative phosphorylation pathway was inhibited, decreasing both ATP production ( Figure 3B) and oxygen consumption ( Figure 3C).
These observations indicated that NHEKs accelerated the metabolic rate and nutrient consumption after UVA irradiation but did not ultimately convert to ATP.
To further understand these phenomena, we deeply analyzed the transcriptomic data and found that mitochondrial uncouplingrelated gene expression was elevated in NHEKs after UVA irradiation, as shown in the heatmap ( Figure 3D), which suggests that oxidative and phosphorylation processes are uncoupled in response to damage from UVA-generated superoxide following irradiation of NHEKs. This explains why the energy consumed manifested by the increase in metabolic rate does not generate a large amount of ATP, as this part of the energy is very likely to be released in the form of heat. Therefore, we speculated that the skin is exposed to ultraviolet rays from the sun for many years, which accelerates the loss of nutrients (such as carbohydrate macromolecules and various lipid components) from keratinocytes making the skin lackluster and wrinkling.

| LQ can improve the antioxidant capacity of NHEKs and inhibit UVA-induced ROS generation in NHEKs
We next verified the antioxidative stress and skin aging capacity of LQ. By detecting the contents of MDA, GSH, and SOD in cells, it was determined that LQ could protect against UVA-induced oxidative stress in NHEKs. Figure 4A Next, we analyzed the accumulation of ROS in cells after LQ intervention. As shown in Figure 4D, NHEKs were pretreated with 20, 50, and 100 μM LQ for 24 h before irradiation. The results showed that the fluorescence intensity was significantly weakened, and ROS generation was reduced. The changes in ROS content were detected by flow cytometry. As shown in Figure 4E This trend is consistent with the western blot results( Figure 5A).
In addition, compared with the non-UVA group, the expression of

| LQ inhibits UVA-induced mitochondrial uncoupling and increases ATP productivity
To investigate whether LQ helps block UVA-stimulated oxidative phosphorylation uncoupling, thereby inhibiting excessive loss of skin nutrients after restoring redox homeostasis in NHEKs, the whole-transcriptomic data were compared between the UVA and the UVA + LQ group. We identified that the expression of genes encoding mitochondrial uncoupling proteins was significantly decreased, and cellular thermogenesis, oxidative phosphorylation, and cellular senescence pathways were inhibited ( Figure 6A,B).
Mitochondrial uncoupling causes an H+ influx, which leads to a decrease in mitochondrial membrane potential. When the membrane potential is high, a polymer is formed in the cell, and red fluorescence is generated; otherwise, it becomes a monomer, resulting in green fluorescence; the red/green fluorescence ratio represents mitochondrial membrane potential. As shown in of that in the non-UVA group, and 20, 50, and 100 μM LQ groups led to a 4x, 17x, and 21x increase in the ratios compared with the UVA group, respectively (p < 0.05). The results showed that LQ prevented the decrease of mitochondrial membrane potential, thereby repairing mitochondrial function. And the O2K data also showed that after the addition of LQ, the ATP productivity and the NAD + /NADH ratio was significantly increased ( Figure 6F,G) without a significant change in the oxygen consumption rate ( Figure 6E). These results suggest that LQ can decouple mitochondrial oxidative phosphorylation by eliminating UVA-induced accumulation of superoxide and lipid peroxidation products, thereby preventing rapid nutrient depletion and thermogenesis and improving ATP generation efficiency and net energy abundance in NHEKs.

| LQ inhibits UVA-induced excessive loss of nutrients and senescence phenotype in NHEKs and promotes cell proliferation
Next, we performed oil red staining and linoleic acid content detection assay on NHEKs. The results showed that the oil red staining of keratinocytes was reduced after UVA stimulation, and the lipid components were largely lost ( Figure 7A,B). The oil red staining became darker after adding LQ, indicating that LQ could increase the lipid content of NHEKs. At the same time, linoleic acid testing proved this.
Moreover, we found that the collagen content of the cells was significantly increased after LQ treatment ( Figure 7C). Previous studies have shown that net energy stress causes cellular senescence and reduces cell proliferation potential. The CCK-8 cell viability experiment confirmed that LQ inhibited the UVA-induced decrease in cell viability and was a more effective proliferative capacity than the non-UVA group ( Figure 7D,E). Finally, β-galactosidase staining was performed on NHEKs in each group ( Figure 7F). Compared with the UVA group, 20, 50, and 100 μM LQ inhibited this change to a certain extent, and the morphological changes were smaller. One hundred micromolar LQ group was the most obvious. Therefore, LQ was able to reduce the UVA-induced senescence rate of NHEKs. Based on the above results, LQ inhibits mitochondrial uncoupling, reduces excessive consumption of nutrients, improves cell net energy, restores cell proliferation vitality, and promotes collagen synthesis to inhibit photoaging. It was shown that LQ is a potential active substance for reversing skin aging.

| DISCUSS ION
The core mechanism of classical skin photoaging suggests that UVA radiation can damage the membrane integrity and cytotoxicity of keratinocytes, leading to excessive accumulation of intracellular ROS overexpression, resulting in cellular oxidative damage, collagen degradation, and ultimately accelerated aging. 27  On this basis, it is shown to reduce the oxidative stress of skin cells, relieve the excessive uncoupling of mitochondria to a certain extent, improve the efficiency of oxidative phosphorylation and ATP production, and block the massive loss of skin nutrients and net energy caused by UV pressure. Under sufficient nutrients and energy, LQ can further promote collagen synthesis, and promote the proliferation of keratinocytes through the PI3K-AKT pathway, [30][31][32] thereby reversing skin photoaging.

| CON CLUS ION
In this study, the vitro skin photoaging model was successfully established by using 12 J/cm 2 UVA. By detecting cell viability, it was found that 20, 50, and 100 μM LQ had strong photoprotective activ-

ACK N OWLED G M ENTS
We are grateful to Prof. Dongming Xing, Cancer Institute of the Affiliated Hospital of Qingdao University and Department of Life Sciences Tsinghua University.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no potential conflict of interest relevant to this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

E TH I C S S TATEM ENT
This study is designed for animal and human experiments and therefore does not require Ethics Committee approval.