Effect of human umbilical cord mesenchymal stem cell exosomes on aerobic metabolism of human retinal pigment epithelial cells

To investigate the effect of exosomes secreted by human umbilical cord mesenchymal stem cells (HUCMSC-Exo) on aerobic metabolism of cobalt chloride (CoCl2)-induced oxidative damage in the human retinal pigment epithelial cell line (ARPE-19), and to explore the protective mechanism of HUCMSC-Exo on oxidative damage in ARPE-19 cells. HUCMSC-Exo were extracted and identified; CCK-8 assay was used to established the oxidative damage mode of ARPE-19 cells induced by CoCl2; JC-1 flow cytometry was used to detect the effects of exosomes with different concentrations (0, 25, 50, or 100 μg/mL) on the mitochondrial membrane potential (MMP) of oxidatively damaged ARPE-19 cells. The effects of exosomes with different concentrations on the activity of oxidative metabolic enzymes (oxidative respiratory chain complexes I, III, IV, and V) and ATP synthesis in oxidatively damaged ARPE-19 cells were detected by spectrophotometry. Under transmission electron microscope, HUCMSC-Exo were round or oval membrane vesicles with diameters of about 40–100 nm. Western blot results showed that HUCMSC-Exo expressed specific marker proteins CD63 and CD81. CCK-8 dates showed that the cell viability of ARPE-19 cells was significantly decreased with increasing CoCl2 concentration, and the concentration of 400 μmol/L CoCl2 was chosen to be the optimal concentration for oxidative damage. MMP was increased in exosomes intervention group (25, 50 or 100 μg/mL), and the dates were statistically different from 0 μg/mL exosome intervention group (P < 0.05). The activities of mitochondrial complexes I, IV, and V in exosomes intervention groups (100 μg/mL) were higher than those in 0 μg/mL exosome intervention group. In 50 μg/mL and 100 μg/mL exosome intervention group, ATP synthesis was significantly different from the 0 μg/mL exosome intervention group (P < 0.05). HUCMSC-Exo had a certain protective effect on ARPE-19 cells induced by CoCl2 in vitro. The protective mechanism of HUCMSC-Exo on oxidative damage ARPE-19 cells might be through saving its aerobic metabolic function, restoring cell ATP synthesis, and improving the ability of cells to repair damage and deal with the hypoxic environment.


Introduction
Retina is one of the tissues with high oxygen demand. The balance between oxygen supply and oxygen consumption can maintain retinal homeostasis. Once this balance is destroyed, it will lead to many retinal diseases [1]. Oxidative stress produced by this large amount of reactive oxygen species (ROS) is associated with the pathogenesis of various eye diseases, such as glaucoma, diabetic retinopathy (DR), agerelated macular degeneration (AMD), and the retinopathies [2]. For a variety of reasons, the retina is particularly vulnerable to oxidative damage [3]. Oxidative stress may be the key factor leading to retinal pigment epithelium (RPE) dysfunction-related retinal diseases [4]. Current cellular strategies for protection against oxidative damage include antioxidants, molecular repair (removal or repair of oxidation-modified biomolecules to counteract functional effects), and cell replacement (using stem cells or progenitor cell populations) [5].
Exosomes have attracted the attention of researchers due to their powerful biological functions, such as tissue repair, inhibition of inflammation, and regulation of immunity. Exosomes are vesicles with the lipid bilayer structure of about 40 ~ 100 nm in diameter [6]. It is secreted by a variety of cell lines and cell types, including tumor cell lines, stem cells, and neurons. Human umbilical cord mesenchymal stem cells (HUCMSCs) are a kind of multifunctional stem cells that exist in neonatal umbilical tissue. Compared with other sources of mesenchymal stem cells, such as bone marrow and adipose tissue, HUCMSCs have obvious advantages because of low cost, are easy obtained, non-invasive procedure to the donors, and representing the noncontroversial source of mesenchymal stem cells [7]. HUCMSCs have a stronger capacity of expansion than bone marrow mesenchymal stem cells (BMSCs) [8].
Because the mesenchymal stem cells (MSCs) and their exosomes (MSCs-Exo) had a similar function [9], this study used HUCMSC-Exo to explore the protection mechanism against oxidative damage in ARPE-19 cells.

HUCMSCs-Exo purification and identification
Briefly, HUCMSCs (Salial, Guangzhou, China) were cultured in a 5% CO 2 incubator (Thermo Fisher, Shanghai, China) at 37 ℃ in DMEM/F12 (Gibco, Guangzhou, China) medium containing 10% fetal bovine serum (FBS)(Gibco, Australia) and 1% penicillin-streptomycin solution (Gibco, Guangzhou, China). 48 h before extraction of the 4 ~ 10 generation hUCMSCs with good growth, the original medium was discarded and replaced with DMEM/F12 medium containing 10% exosome-free FBS (SBI, Guangzhou, China) and 1% penicillin-streptomycin solution. After 48 h of cell culture, the culture medium was collected and put into the 15-ml centrifuge tube. The cells and cell fragments were removed by 1000 rpm centrifugation for 10 min at 4 ℃. The supernatant was prepared and filtered to the ultrafilter tube through a 0.22μm aseptic membrane. The exosome concentration was collected by centrifugation at 4000 rpm for 8 ~ 10 min at 4 ℃. After adding exosome extraction reagent ExoQuick-TC (SBI, Guangzhou, China), the mixture was taken out after standing for at least 12 h and centrifuged at 10000 rpm for 30 ~ 40 min at 4 ℃. The bottom precipitate was collected, resuspended with 100-500 μl phosphate buffer saline (PBS) (Gibco, Guangzhou, China), and stored in the -80℃ ultra-low temperature refrigerator for later use.

Characterization of HUCMSCs-Exo
The exosome stored in -80℃ was quickly moved into a 37 °C thermostat water bath for about 1 min, and the freezing tube was gently shaken until it melted completely. The exosome suspension was prepared by diluting the exosome samples with PBS buffer at 1:10 ~ 1:20. The copper mesh was dipped in a small amount of diluted exosome sample and was placed in a 3% phosphotungstic acid solution for negative staining for about 5 min. The excess dye solution was removed with a filter paper. The morphological characteristics of exosomes were detected with transmission electron microscopy (TEM, JEOL 2100F, Beijing, China).

Quantification and detection of exosomes protein
The exosomes were lysed with RIPA lysis buffer (Solarbio, Beijing, China) containing protease inhibitors, and A562 was determined by Microplate Reader (Bio-Tek, Guangzhou, China). The protein concentration of the exosomes was calculated according to the standard curve. Western blot was used to detect the expression of CD63 (proteintech, Wuhan, China) and CD81 (proteintech, Wuhan, China) in the exosome after quantification.
Establishment of oxidative damage model of ARPE-19 cells ARPE-19 cells (ATCC, Beijing, China) in the logarithmic growth phase were digested and centrifuged to form a single-cell suspension and inoculated in 96-well plates. After the cells were adherent overnight, the original medium was absorbed and replaced with the medium containing different concentrations (0, 50, 100, 200, 400, 800) μmol/L of CoCl 2 (Sigma, Guangzhou, China). Each group was replicated 5 times and incubated at 37 ℃ and in 5%CO 2 incubator for 24 h. Each well was incubated with 10 μl CCK-8 (MCE, Guangzhou, China) for 30 min at 37 °C in 5% CO2 incubator. A450 was determined by Microplate Reader. The survival rate of ARPE-19 cells was calculated, and the concentration of 50% cell survival rate was used as the optimum concentration of (CoCl 2 )-induced oxidative damage of ARPE-19 cells.

Activity detection of oxidative respiratory chains complex
The activity of I, III, IV, and V (I: NADH dehydrogenase, III: Cytochrome C reductase, IV: Cytochrome C oxidase, V: F1F0-ATP synthasein) the mitochondrial respiratory chains complex was determined by spectrophotometry with reference to the method of Vyatlina et al. [11] 10 ~ 20 μg of mitochondrial protein was added into the buffer solution with a final volume of 2 mL, and distilled water was used as a blank tube to correct the absorbance to 0 point. The changes of absorbance values at wavelength of 340 nm and 550 nm for 3 min were measured, respectively. The unit of enzyme activity was nmol/min/ 10^4 cell.

Cellular extraction of ATP
First, the cells were collected into the centrifuge tube, the supernatant was excluded, and 1 ml of the acid extract was added into the 5 million cells at a ratio of 500 ~ 1000:1. The cells were crushed by ultrasonic for 1 min (ice bath, intensity 20% or 200 W, ultrasonic 2S stopped for 1 s). The cells were centrifuged at 8000 g at 4 ℃ for 10 min. The supernatant was integrated into another centrifuge tube, and an equal volume of alkaline extract (Solarbio, Beijing, China) was added to neutralize and mix. The supernatant was centrifuged at 8000 g at 4 ℃ for 10 min; then, the supernatant was taken and placed on ice to be measured.

ATP synthesis assay
Preheat spectrophotometer for more than 30 min; adjust the wavelength to 700 nm and distilled water zero. One blank tube and one standard tube are made, respectively, and one pair of care is set for each measuring tube. Add relevant reagents in each tubes according to the instructions (Table1). Preparation of chromogenic agent: according to the volume of proposed chromogenic agent (sample number *0.87 ml) (Solarbio, Beijing, China), reagent 4(ml) (Solarbio, Beijing, China): reagent 5(ml) (Solarbio, Beijing, China) = 1:5 should be prepared before use. Sample determination: The mixture was thoroughly mixed and was incubated in water bath at 37℃ for 20 min. After the water bath at 37℃ for 20 min, the absorbance value of each tube was measured at 700 nm. Cal- Standard liquid concentration, 1 μmol/mL; V1: solution in the reaction system, 0.03 mL; V2: extraction liquid volume, 2 mL; V3: serum (slurry) volume, 0.1 ml.

Statistical analysis
The statistical software SPSS 19.0 (IBM, Armonk, NY, USA) and GraphPad Prism 8.0 (GraphPad Software, California, USA) was used to analyze data. The results are represented as mean ± SD. For comparison of different groups and evaluation multiple comparisons, statistical comparisons were performed by oneway ANOVA and Tukey. In these analyses, P < 0.05 was considered statistically significant.

Characteristics of exosomes
Exosomes were extracted by ExoQuick-TC, and TEM demonstrated that HUCMSCs-Exo were round or oval membranous vesicles with diameters between 40 ~ 100 nm (Fig. 1A-D), which was consistent with the basic morphology of exosomes. Western blot was used to detect the expression of CD63 and CD81 in HUCM-SCs-Exo (Fig. 1E As shown in Fig. 2, the cell viability of ARPE-19 cells was significantly decreased with increasing CoCl2 concentration compared with the normal group. The survival rate of ARPE-19 cells treated with 400 μg/ mL and 800 μg/mL CoCl 2 concentrations was statistically significant reduced compared with the normal group (P < 0.01). When the CoCl 2 concentration is 400 μmol/L, the survival rate of ARPE-19 cells is 50%, which was chosen to be the optimal concentration for oxidative damage.

Effect of HUCMSCs-Exo on the morphology of ARPE -19 cells injured by CoCl 2
On observation under an inverted microscope, normal ARPE-19 cells were fusiform or polygonal monolayer adherent cells with a clear outline. The cytoplasm might contain pigment, namely lipofuscin, which was brown and mostly located in the inner side of the cell (Fig. 3A). When 400 μg/mL CoCl 2 was added, the number of ARPE-19 cells decreased, the space widened, the arrangement was disordered, and the cells shrank and the nuclei aggregated ( Fig. 3B-C). After treated with exosomes (25 μg/mL, 50 μg/mL and 100 μg/mL), respectively, ARPE-19 cells morphology was similar to the normal group, with clear cell boundaries and increased cell numbers (Fig. 3D-F). The cell morphology of the 50 μg/mL and 100 μg/mL exosome intervention groups was similar to that of the normal group, with clear cell boundaries and increased cell number.  (Fig. 4). These results indicate that CoCl 2 may initiate the mitochondrial apoptotic pathway in ARPE-19 cells. However, MMP was increased in exosomes intervention group (25, 50 or

Effect of HUCMSCs-Exo on the activity of oxidative respiratory chains complex in ARPE-19 cells injured by CoCl 2
Respiratory chain complex I, namely NADH-Co Q reductase or NADH dehydrogenase, is the main part of O 2.32− generated in the respiratory electron transport chain. Its activity reflects the state of respiratory electron transport chain and reactive oxygen species (ROS) production. Based on statistical analysis (Fig. 5A), the enzyme activities of respiratory chain complex I in normal group, control group, and exosomes intervention group (0, 25, 50 or 100 μg/mL) were (0.0183 ± 0.0037, 0.0548 ± 0.0037, 0.0150 ± 0.0030, 0.0183 ± 0.0037, 0.0256 ± 0.0037, 0.0657 ± 0.0001) nmol/min/10^4 cell, respectively. The activities of respiratory chain complex I in 100 μg/mL exosome intervention group were (0.0657 ± 0.0001) nmol/ min/10^4 cell, which had statistical difference with the normal groups and 0 μg/mL exosomes intervention group (P < 0.001).

Discussion
In recent years, more and more evidence proved MSCs or MSCs-Exo could rescue the aerobic metabolism of damaged cells [12]. Arslan et al. [13] injected MSCs-Exo into the mouse model of myocardial infarction through the tail vein, and the results showed that the levels of ATP and NADH could be restored within 1 h. This study also showed that MSCs-Exo enhanced myocardial viability by reducing oxidative stress and increasing phosphorylated Akt and phosphorylated GSK-3β [13]. Panfoli et al. [14] suggested that the UCMSCs-Exo of full-term neonates could express functional respiratory chain complexes I, IV, and V, which consumed oxygen and produced ATP. Besides, some studies had shown that there were functional expressions of respiratory chains complexes and the tricarboxylic acid cycle (TAC cycle) enzymes in urinary exosomes; its proteomics showed that proteins were concentrated in certain specific functions, one of which was aerobic metabolism [15,16].
The ability of exosomes to restore cellular aerobic metabolism may be due to their oxidative phosphorylation independent of mitochondria. Panfoli et al. [14] showed that ND4L, expressed by exosomes, was a complex I subunit encoded by DNA, indicating that the mechanism of oxidativephosphorylation (OXPHOS) in exosome membrane was the same as that in mitochondria. There was a supramolecular arrangement of the respiratory complex in the mitochondrial membrane, which could be transferred to endoplasmic reticulum (ER), through heterologous fusion between mitochondrial and ER, and eventually transferred to the exosomes body when sprouting in Polyneices [14]. This may be one of the explanations for the origin of the exosome OXPHOS mechanism. Additionally, Islam et al. [17] reported that there was a gap junction between mitochondrial DNA and alveolar cells in human MSCs after acute lung injury, so mitochondrial transferred and exosome transferred might also play a role in restoring the aerobic metabolism of injured cells. If the MSCs-Exo could save the aerobic metabolism and restored the ATP synthesis of cells, it was hoped to improve the ability of cells to repair injuries and deal with the hypoxic environment, thereby promoting cell repair and regeneration and restoring their functions. This would provide a theoretical basis for the therapeutic efficacy of MSCs and MSCs-Exo in a variety of diseases.
Our results showed that the vesicles extracted by ExoQuick-TC showed a 40 ~ 100-nm round or oval bilayer membrane structure, highly expressed CD63 and CD81, and demonstrated that HUCMSC-Exo were successfully isolated. We tested ARPE-19 cells proliferation ability by CCK-8, and results showed that CoCl2 inhibited ARPE-19 cells proliferation. However, compared with the non-intervention group, ARPE-19 cell after exosomes intervention group (50 or 100 μg/mL) had significantly more surviving cells, and the cell morphology was also closer to that of normal cells.
Besides, in the results of MMP detection, the proportion of MMP roughly increased with the increasing of exosomes concentration, and in the 50 μg/mL exosomes intervention group was closest to that of the normal group. The elevated MMP may indirectly reflect mitochondrial functional recovery, which had been demonstrated in the detection of respiratory chain complex and ATP synthesis. The enzyme activities of the respiratory chain complex I and complex V in 100 μg/mL exosomes intervention group were significantly higher than those in other groups. However, there was no significant change in the enzyme activity of respiratory chain complex III and complex IV in exosomes intervention group, and it was lower than that in the normal group. Also, the amount of ATP synthesis in 50 μg/mL and 100 μg/mL exosome intervention group had significant difference with 0 μg/mL exosome intervention group, and the ATP synthesis in 100 μg/mL exosomes intervention group was the highest.
The above results suggested that the addition of exosomes at a certain concentration could restore the aerobic metabolism and ATP synthesis of ARPE-19 cells damaged by oxidation and have a protective effect on the cells. Moreover, the MMP and the activity of the respiratory chain complex I in the control/ non-intervention group were higher than those in the low concentration exosomes intervention group, which might be due to the production of a certain amount of exosomes in ARPE-19 itself after oxidative damage. Due to the replacement of the original culture medium in the 0 μg/mL and 25 μg/mL exosomes concentration group, the exosomes produced by ARPE-19 itself were removed, while the exosome concentration was lower; it might not be enough to have a significant effect on cell function. In this study, hUCMSC was "starved" before exosomes were extracted; exosome extraction can be improved by removing exosome serum for hUCMSC culture. Weiss ML et al. released cells from umbilical cord matrix by enzymatic degradation of extracellular matrix, and this improved isolation method can help to produce a sufficient number of exosomes [7]. Further experiments are needed to determine whether there are other more effective pretreatment methods.
It is remarkable that the therapeutic effect of MSCs is mainly mediated by paracrine signaling of exosome [18]. And compared with MSCs, exosome is a kind of cell-free therapy, which has more advantages [19]. Our results showed that HUCMSC-Exo had a certain protective effect on ARPE-19 cells injured by CoCl 2 in vitro. The protective mechanism may be to restore cellular ATP synthesis and improve the ability of cells to repair damage and deal with hypoxic environment by saving their aerobic metabolic function.