Sprague-Dawley rats, purchased from the Laboratory Animal Centre, Huazhong Agricultural University, were artificially caged at a male-female ratio of 1:2 to obtain rats at 19 days of gestation. Caesarean section was performed after the rats were euthanized with an intraperitoneal injection of pentobarbital sodium in 150 mg/kg to collect the umbilical cord tissue, and vascular tissue was subsequently removed. The remaining Wharton's jelly tissue was then cut up and digested with Dulbecco's modified Eagle's medium (DMEM, HyClone; GE Healthcare Life Sciences) containing 0.1% type IV collagenase (Gibco; Thermo Fisher Scientific, Inc.) for 10 min, followed by 0.25% trypsinization for 20 min at 37°C. Following purification, the cells were resuspended in DMEM containing 1% penicillin-streptomycin-amphotericin B solution (Bioswamp, Myhalic Biotechnology Co., Ltd.) and 10% fetal bovine serum (FBS, Gibco; Thermo Fisher Scientific, Inc.) at a concentration of 2×105 cells/ml, seeded into a culture flask pre-coated with polylysine (Sigma-Aldrich; Merck KGaA) and incubated under constant humidity conditions at 37°C in an atmosphere containing 5% CO₂. The culture medium was changed the following day and every 3 days thereafter. Passaging was carried out after the cells reached 85% confluence. UC-MSCs in the third passage were examined for CD29, CD44, CD73, CD105, CD90, CD34, CD45 and HLA-DR expression by flow cytometry (Beckman Coulter, Inc.) and observed under an inverted fluorescence microscope (Leica Microsystems GmbH). The obtained UC-MSCs were then treated with 0.5 μM ATRA (Shanghai Aladdin Biochemical Technology Co., Ltd.) for 12, 24 or 48 h, as previously described (21). Untreated UC-MSCs served as the control (CON).

MTT assay was performed to evaluate the viability of the ATRA-treated UC-MSCs. Briefly, 180 μl of cells in the logarithmic phase were plated into 96-well plates at a concentration of 5×103 cells/well and incubated at constant humidity at 37°C in an atmosphere containing 5% CO₂ overnight. Following treatment with ATRA (0.5 μM) for 12, 24 or 48 h, the cells were treated with 20 μl of MTT (5 mg/ml; Bioswamp, Myhalic Biotechnology Co., Ltd.) for 4 h, followed by the addition of 150 μl of dimethyl sulfoxide. The absorbance was measured using a microplate reader (Thermo Fisher Scientific, Inc.) at 490 nm after 10 min of shaking.

Flow cytometry was carried out to identify the UC-MSCs and evaluate the apoptosis of UC-MSCs with or without ATRA treatment. For cell identification, the cells were collected at a concentration of 1×106 cell/ml into 8 Eppendorf tubes, with 100 μl of cells in each tube, wherein 2 μl of CD29-fluorescein isothiocyanate (FITC) (85-11-0291-80; eBioscience, Inc.), CD44-FITC (MA5-16906; eBioscience, Inc.), CD90-FITC (554894, BD Biosciences), CD34-FITC (MA1-10204; eBioscience, Inc.), CD45-FITC (11-0460-82; eBioscience, Inc.), CD73-FITC (PAB45829; Bioswamp, Myhalic Biotechnology Co., Ltd.), CD105-FITC (MA1-19231; Invitrogen; Thermo Fisher Scientific, Inc.), or 1 μl of HLA-DR antibody (ab175085, Abcam) were added followed by incubation for 45 min in the dark (one tube acted as the control). After washing with phosphate-buffered saline (PBS, Bioswamp, Myhalic Biotechnology Co., Ltd.), the cells were resuspended with 100 μl flow cytometry staining buffer and incubated with 2 μl FITC-conjugated Affinipure goat anti-rabbit IgG(H+L) (PAB160016; Bioswamp, Myhalic Biotechnology Co., Ltd.) for 45 min in the dark, 400 μl of flow cytometry staining buffer was added. The cells were analyzed, and data were acquired using a flow cytometer (Beckman Coulter, Inc.). For the evaluation of apoptosis, cells were harvested by trypsin, resuspended in a binding buffer at a concentration of 1×105 cells/ml, labeled with 10 μl of Annexin V-FITC and 10 μl of propidium iodide (PI) for 30 min at room temperature, and analyzed by flow cytometry. A minimum of 2×104 cells was acquired for each sample.

Following treatment with ATRA, the cell status was detected by immunofluorescence. Incubated cells were washed twice with PBS (pH 7.4) and fixed in 4% paraformaldehyde at room temperature for 30 min, followed by 3 washes in PBS. The cells were then permeabilized using 0.5% Triton X-100 (Bioswamp, Myhalic Biotechnology Co., Ltd.) in PBS for 20 min and blocked in 5% bovine serum albumin (BSA) for 1 h. The cells were subsequently incubated with primary antibodies against Nestin (1:200, ab92391, Abcam), neuron-specific enolase (NSE; 1:200, ab53025, Abcam) and glial fibrous acidic protein (GFAP; 1:200, ab16997, Abcam) for 1 h at room temperature, and thereafter with AlexaFluor 488-conjugated Affinipure goat anti-rabbit secondary antibody (1:10,000, SAB43742, Bioswamp, Myhalic Biotechnology Co., Ltd.) for 1 h at room ttemperature. The nuclei were stained with DAPI (Bioswamp, Myhalic Biotechnology Co., Ltd.) and stained cells were examined under a fluorescence microscope (Leica, Microsystems GmbH).

The mRNA expression levels of Nestin, NSE and GFAP in UC-MSCs following treatment with ATRA for 24 h were evaluated by RT-qPCR. Total RNA was extracted using TRIzol reagent (Ambion) and reversed transcribed into cDNA according to the following system: Total RNA (500 ng) combination with 4 μl 5X PrimeScript II buffer (Takara), 2 μl, 10 mM of dNTP Mix (Solarbio), 1 μl Oligo DT18 primer (Takara), 1 μl, 40 U/μl of Recombinant RNase Inhibitor (Takara), 1 μl, 200 U/μl of PrimeScript II RTase (Takara) and 11 μl ddH₂O, followed by amplification using SYBR FAST qPCR Master Mix (KM4101, Kapa Biosystems). The thermocycling conditions are as follows: 95°C for 3 min; 40 cycles of denaturation at 95°C for 50 sec, annealing at 56°C for 10 sec, and extension at 72°C for 20 sec; and final extension from 65°C to 95°C at the rate of 5 sec/0.5°C. The primer sequences were as follows: NSE forward, 5′-GCCTCTACGGGCATCT-3′ and reverse, 5′-ATC AGG TTG TCC AGT TTC-3′; GFAP forward, 5′-GAC GCA TCA CCT CCG CT-3′ and reverse, 5′-CGC CTT GTT TTG CTG TTC-3′; Nestin forward, 5′-AGG GGC AGA CAT CAT TGG-3′ and reverse, 5′-CAG AGA CTA GCG GCA TTC C-3′; GAPDH forward, 5′-CCA CTC CTC CAC CTT TG-3′ and reverse, 5′-CAC CAC CCT GTT GCT GT-3′. The 2^−ΔΔCq method was utilized to calculate the relative mRNA expression levels (24). GAPDH served as an internal reference gene.

A total of 40 Sprague-Dawley rats (weighing 120±7.0 g, 7 weeks old) were obtained from the Laboratory Animal Centre, Huazhong Agricultural University, and raised in a specific-pathogen-free room with a humidity of 50-60% and a temperature of 23±2°C with standard water and diet. Diabetes was induced with a single intraperitoneal injection of freshly prepared STZ (Sigma-Aldrich; Merck KGaA) solution in citrate buffer (pH 4.5, Sigma-Aldrich; Merck KGaA) at 60 mg/kg as previously described (22). The rats were confirmed to be diabetic when the fasting blood glucose levels were measured to be ≥16.7 mmol/l at 1 week after the injection, as previously described (25). Diabetic rats were fed a standard diet and unlimited water for 2 months. The fasting blood glucose levels were continuously monitored each week to ensure the models were successful. Insulin was not used during the study for the treatment of any extreme glycemic levels. All animal protocols were approved by the Institutional Review Board of Wuhan Myhalic Biotechnology Co., Ltd. based on the ethical Guidelines for Animal Care and Use of the Model Animal Research Institute (HLK-20180802-01) and were performed in accordance with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research.

After 2 months, 30 diabetic rats with lens opacities (milky-white colored lens) were selected to establish the DR model and were randomly divided into 3 groups (n=10 per group) as follows: The model (MOD), ATRA and UC-MSC group. The rats in the MOD group were intravitreally injected with 2 μl of saline. Those in the ATRA and UC-MSC groups were administered an intravitreal injection of ATRA-treated or normal UC-MSCs (2 μl, 1×105-6 cells), respectively (12). The control rats (CON group, n=10) were not subjected to STZ induction. Prior to the injection, UC-MSCs with or without ATRA treatment were labeled with CM-Dil (Invitrogen; Thermo Fisher Scientific, Inc.) solution. After 8 weeks of treatment, the lenses of the rats were observed, and the rats were then euthanized with an intraperitoneal injection of a lethal dose of pentobarbital sodium at 150 mg/kg; death was confirmed with no heartbeat and respiration. The eyeballs were collected and fixed with 4% paraformaldehyde for 24 h. The harvested retinal tissues were immersed in 10% formalin buffer.

The status of transplanted cells in the ATRA and UC-MSCs groups was evaluated by immunofluorescence staining. Following fixation in 10% formalin buffer for 48 h, the retinal tissues were embedded in paraffin and sliced at a thickness of 4 μm, followed by dewaxing with water. The tissues were then stained with 4′,6-diamidino-2-phenylindole (DAPI), and examined under a fluorescence microscope (Leica, Microsystems GmbH).

Following fixation in 10% formalin buffer for 48 h, the retinal tissues were embedded in paraffin and sliced at a thickness of 4 μm, followed by dewaxing with water. The sections were stained in hematoxylin solution (Bioswamp, Myhalic Biotechnology Co., Ltd.) for 3 min and subsequently in eosin solution (Bioswamp, Myhalic Biotechnology Co., Ltd.) for 3 min at room temperature. Following dehydration, the tissues were cleared with xylene and mounted with neutral balsam. After staining, the sections were observed under an upright microscope (Leica, Microsystems GmbH), and Leica Application Suite was used to collect and analyze the relevant parts of the samples.

Qualitative analysis of apoptosis in retinal tissues was performed using a TUNEL kit (Roche Diagnosticis) according to the manufacturer's instructions. Following fixation in 10% formalin buffer for 48 h, the retinal tissues were embedded in paraffin and sliced at a thickness of 2-3 μm, followed by dewaxing with water. The tissues were treated with protease K at room temperature for 30 min, 50 μl of TUNEL reaction mixture in a dark humidified chamber at 37°C for 1 h, and 50 μl of converter-POD for 1 h. Finally, diaminobenzidine was added, and the core was counter-stained with hematoxylin. The sections were observed under an upright microscope (Leica, Microsystems GmbH). The obtained images were quantified using Image Pro-Plus 6.0 (Media Cybernetics, Inc.).

The protein expression of CTGF, tissue plasminogen activator (t-PA), plasminogen activator inhibitor (PAI), transforming growth factor-β2 (TGFβ2), and VEGF in retinal tissues was measured by western blot analysis. The collected retinas were homogenized in radioimmunoprecipitation assay lysis buffer (Bioswamp, Myhalic Biotechnology Co., Ltd.) and centrifuged at 12,000 × g for 15 min. The concentration of proteins was detected using a bicinchoninic acid protein assay kit (Bioswamp, Myhalic Biotechnology Co., Ltd.). Total proteins (20 μg) were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes (MilliporeSigma). The membranes were blocked with 5% skim milk for 2 h at room temperature in Tris-buffered saline and incubated with primary antibodies against CTGF (1:1,000, ab6992, Abcam), t-PA (1:2,000, ab227069, Abcam), PAI (1:1,000, ab66705, Abcam), TGFβ2 (1:1,000, ab113670, Abcam), VEGF (1:2,000, ab32152, Abcam), or GAPDH (1:1,000, 2118, Cell Signaling Technology, Inc.) for 1 h at room temperature, followed by incubation with goat anti-rabbit IgG secondary antibody (1:10,000, SAB43714, Bioswamp, Myhalic Biotechnology Co., Ltd.) for 2 h at room temperature. Immunoreactivity was visualized by colorimetric reaction using an enhanced chemiluminescence substrate buffer (MilliporeSigma). Membranes were scanned with a Tanon-5200 imager (Shanghai Tianneng Technology Co., Ltd.) and analyzed using IQTL 8.1 software.

The levels of interleukin (IL)-1 (RA20108), IL-6 (RA20607) and IL-10 (RA20090) in the serum of retinas were detected using ELISA kits (Bioswamp, Myhalic Biotechnology Co., Ltd.) according to the manufacturer's instructions. The absorbance was measured at 450 nm using a microplate reader (Labsystems Multiskan MS, Finland).

The data of the present study are expressed as the means ± standard deviation (SD). Statistical analysis was performed using IBM SPSS statistics 19.0. The statistical significance of differences between >2 groups was evaluated by one-way analysis of variance followed by Bonferroni's test and significant differences between 2 groups were analyzed using a t-test. P<0.05 was considered to indicate a statistically significant difference.

To confirm that UC-MSCs were successfully extracted, cells were visually evaluated using a microscope, whereby the typical spindle-shaped fibroblast-like morphology of UC-MSCs was observed (Fig. 1A). Flow cytometry was then carried out to detect the expression of surface antigens of UC-MSCs. As shown in Fig. 1B, the extracted cells strongly expressed the mesenchymal cell markers, CD90 (97.37%), CD44 (96.58%), CD29 (96.82%), CD73 (97.58%) and CD105 (97.98%), while the hematopoietic markers, HLA-DR (1.80%), CD45 (1.24%), and CD34, (1.72%) were absent (20,21). These findings indicate that UC-MSCs were successfully isolated without the inclusion of other cell populations.

Following treatment of the UC-MSCs with ATRA for 24, 48 and 72 h, cell proliferation and differentiation were determined. Compared to the untreated cells, the viability of the ATRA-treated cells significantly increased (P<0.01, Fig. 2A). In addition, ATRA exerted minimal effects on the apoptosis of UC-MSCs (Fig. 2B). Immunofluorescence staining was performed to examine the differentiation of UC-MSCs treated with ATRA. Strong positive staining was observed for the glial cell marker, GFAP, the nerve cell marker, Nestin, and the neuron-specific marker, NSE (26), in the ATRA-treated UC-MSCs compared to that of signals in the untreated UC-MSCs (Fig. 2C). In addition, the mRNA expression levels of Nestin, NSE and GFAP in the untreated UC-MSCs were lower than those in the ATRA-treated UC-MSCs (Fig. 2D), suggesting that ATRA promoted UC-MSC differentiation into different types of nerve cells. Moreover, reticular neuron-like cells were observed (red boxes) after the UC-MSCs were treated with ATRA for 24 h (Fig. 2C). Thus, 24 h of ATRA treatment was selected as the final treatment duration for UC-MSCs in the transplantation experiment.

The blood glucose levels of all selected rats with STZ-induced DR were >16.7 mmol/l (data not shown). At 8 weeks following transplantation with CM-Dil-labeled UC-MSCs, the UC-MSCs in retinal tissue were visualized by immunofluorescence. A shown in Fig. 3A, retinal tissues contained a large number of well-distributed UC-MSCs (white arrow pointing towards red area). The transplantation of untreated or ATRA-treated UC-MSCs improved the lens opacity of rats with DR macroscopically (Fig. 3B).

Furthermore, the pathological morphology of retinal tissues was investigated by H&E staining (Fig. 4). The surface of the retina was smooth, and cells at the outer nuclear layer (ONL, upward arrow), inner nuclear layer (INL, downward arrow) and retinal ganglion cells (RGCs) were all regularly arranged in the CON group. However, cells at the ONL and INL appeared in irregular arrangements in the MOD group. Additionally, RGCs protruded through the inner limiting membrane with a rough retinal surface (red arrows). With UC-MSC treatment, the cells at the ONL and INL exhibited a regular arrangement. This was accompanied by greater RGC protrusion, which was even more prominent in the ATRA group and closely resembled that of the CON group. These findings suggest that UC-MSC transplantation attenuated DR in rats and that ATRA further accentuated the effects of UC-MSCs.

TUNEL assay was performed to detect apoptosis in retinal tissues (Fig. 5). The number of TUNEL-positive cells (brown) was significantly higher in the MOD group compared to the CON group (P<0.01). However, the number of TUNEL-positive cells in the UC-MSC group decreased compared to the MOD group (P<0.01) and further decreased in the ATRA group (P<0.01).

The expression levels of the angiogenesis-related proteins, VEGF, t-PA and PAI, and that of the fibrosis-associated proteins, CTGF and TGFβ2, were assessed by western blot analysis. As shown in Fig. 6, the levels of all the above-mentioned proteins were upregulated in the MOD group compared to those in the CON group (P<0.01). Comparatively, UC-MSC transplantation decreased the expression of these proteins (P<0.01) and this decrease was further enhanced in the ATRA group compared to the UC-MSC group (P<0.01, apart from t-PA). Therefore, UC-MSC transplantation inhibited the neovascularization of retinal tissues in rats with DR, and treatment of UC-MSCs with ATRA further promoted this inhibition.

As shown in Fig. 7, compared to the CON group, the secretion levels of IL-6 and IL-1 in the serum of retinas were notably increased in the MOD group (P<0.01). In addition, compared to the MOD group, the levels of IL-6 and IL-1 were markedly decreased in the UC-MSC group (P<0.01) and were even further decreased in the ATRA group (P<0.01). The expression IL-10 exhibited an opposite trend to that of IL-6 and IL-1, indicating that UC-MSC transplantation attenuated inflammation in the retinal tissue of rats with DR, which was further suppressed by treatment of the UC-MSCs with ATRA.