DAPI(#4083) and antibodies against Histon H3 (#4499), Nrf2 (#12721), PI3K (#4249), Akt (#9272), β-actin (#4970), HO-1 (#43966), p53 (#2527), p-PI3K (#17366), p-Akt (#4060) and NQO1 (#62262) were obtained from CST corporation (Massachusetts, USA). bpV(HOpic) (SML0884), anti-γ-H2AX (05-636-AF555), and PKH26 (PKH26PCL) were obtained from Sigma-Aldrich (St. Louis., MO, USA). 8-Hydroxy-2’-deoxyguanosine(8-OH-dG) kit (SKT-120-96S) was obtained from StressMarq Biosciences (Victoria, BC, Canada). SOD kit (BC0170) and Total antioxidant capacity (T-AOC) kit (BC1310) were obtained from Solarbio (Qingdao, China). RNA was extracted using a RNeasy Plus Mini kit obtained from Qiagen (#74134, Qiagen Inc, CA, USA).

Male C57BL/6 mice (8 weeks) were obtained from Lvye Pharmaceutical Co., Ltd. Shandong, China (SYXK 2018–0028). Mice were free access to water and chow diet and housed in a standard environment with a regular 12/12 h light/dark cycle. This study was performed following protocols approved by the Binzhou Medical University Institutional Animal Care and Use Committee (permit number: 2020-06).

The mice were randomly divided into four groups (n = 10): (1) control group, treated with saline (20 mL kg^-1 day^-1) as a vehicle for 7 weeks; (2) D-gal group, treated with D-gal (200 mg kg^-1 day^-1) for 7 weeks; and (3) hPMSC-Exo group, D-gal modeling followed by 100 µg hPMSC-Exo week^-1 intervention (from 5 to 7 week); (4) PBS group, D-gal modeling followed by150 μL PBS kg^-1 week^-1 intervention (from 5 to 7 week). The detailed experimental scheme is exhibited in Figure 2A . CD4⁺ T cells were isolated from single-cell suspensions of splenocytes using a magnetic cell separator as described previously (16).

As described previously (17), hPMSCs were isolated from human term placentas of donors. Isolated hPMSCs were identified by the immunophenotype for human leukocyte antigens HLA-DR, CD90, CD14, CD105, CD19, CD34, and CD73 using flow cytometry (FCM) and observed the cells morphology using microscopy. The hPMSCs were cultured in osteogenic and adipogenic induced medium to differentiate into osteoblasts and adipocytes and followed Alizarin Red staining to verify osteogenesis differentiation and Oil Red O staining to verify adipogenic differentiation.

hPMSC-derived exosomes were harvested from passage 3 MSCs grown to 80% confluency in 10-cm plates (Corning) using the VEX Exosome Isolation Reagent (R601, Vazyme Biotech) as described previously (18). In short, the collected hPMSCs culture supernatant was centrifuged at 200 ×g for 30 min at 4°C. After being mixed with exosome isolation reagent overnight at 4°C, the collected supernatant was centrifuged at 10,000 ×g for 1 h and collect the pellet to resuspended in 200 μl of PBS and stored at -80°C or used immediately.

Isolated exosomes were identified by transmission electron microscopy (TEM) (JEOL JEM-1230), western blotting of exosomal surface marker proteins (CD63, CD9, and HSP70, and use Cytochrome-C (Cyt-c) as a negative control), and NanoSight analysis (Malvern Instruments, Malvern, UK).

Human naive CD4⁺ T (CD4CD45RA) cells were isolated by CD4⁺ T Cell Isolation Kit II (Miltenyi, Bergisch Gladbach, Germany). Cells were pretreated with the PTEN inhibitor bpV (HOpic)(100 μM) for 1 h. To over-express or inhibit miR-21 in hPMSCs, miR-21 inhibitor, inhibitor NC, mimics NC, and miR-21 mimics (GenePharma Co. Ltd., Shanghai, China) were used to treat the hPMSCs. By using the IL-2 (2.5 ng/ml) and anti-CD3/CD28 Dynabeads (1 μg/ml) as mitogenic stimulus, 2 µg/ml hPMSC-Exo were added to CD4⁺ T cells (4×10⁶) and cultured at 37°C in 5% CO₂ for 72 h.

Purified hPMSC-Exo were incubated with PKH26 (Sigma-Aldrich) for 5 min at room temperature. The PKH26 labeled hPMSC-Exo were resuspended in PBS and incubated with CD4⁺ T cells at 37°C for 6 h. After counter-stained nuclei with DAPI, the fluorescence image was captured using a laser scanning confocal microscope (FV3000, Olympus Corporation, Japan).

The CD4⁺ T cells of each group were harvested and lysed by ultrasonication in the presence of a protease inhibitor. The supernatant was collected for analysis after centrifugation at 3,000 × g for 10 min. The levels of T-AOC and SOD in the supernatant were measured using appropriate kits following the manufacturer’s instructions (Solarbio, Qingdao, China).

The intracellular production of ROS was assessed by 2’,7’-dichlorofluorescein diacetate (H2DCF-DA) (Sigma, St Louis, MO, USA) (19). The detection procedure was carried out following the instructions provided by the manufacturer.

The CD4⁺ T cells were re-suspended and the DNA was isolated using the method recommended by ESCODD. The 8-OH-dG in the DNA was detected using an ESA Coulochem II electrochemical detector in line with a UV detector as previously described (20).

The percentage of SA-β-Gal positive CD4⁺ T cells was measured as described previously (21). In short, CD4⁺ T cells were incubated with SA-β-Gal staining solution overnight at 37°C after being fixed with 3% formaldehyde. After washing with PBS, SA-β-Gal positive CD4⁺ T cells were determined under microscopy (Leica, Germany).

Western blot analyses were performed as described previously (22). Briefly, after SDS-PAGE, proteins were electro-transferred to PVDF membranes and blocked in 5% BSA dissolved in TBST for 2h at room temperature. After incubated with primary antibody overnight at 4°C, PVDF membranes were incubated with secondary antibody 2 h at room temperature. The protein bands were visualized using developer (NCI4106, Pierce, Rockford, IL, USA) and performed densitometric analysis by the software.

The potential binding site between PTEN and miR-21 was identified using TargetScan (http://www.targetscan.org). The miR-21 mimics/inhibitor and corresponding negative control (NC) were synthesized by Guangzhou RiboBio Co., Ltd. Amplicons were inserted between SacI and XbaI cleavage sites of pmirGLO vector (Promega, U.S.A.). The human embryonic kidney cell line (HEK) 293T cells (purchased from Procell Life Science & Technology Co., Ltd., Wuhan, China) were selected based on the low endogenous miRNA expression. 800 ng wild-type or mutant reporter and 20 μM miR-21 mimic, inhibitor (GenePharma Co., Ltd, China) were co-transfected into HEK293T cells using Lipofectamine 2000 (Invitrogen, U.S.A.). 24 h after transfection, Renilla and Firefly luciferase activity of lysed HEK293T cells was determined by Dual-Luciferase Reporter Assay System (Promega).

miRNA and mRNA was measured using a quantitative real-time PCR assay. The relative abundance of genes was calculated using the 2^-ΔΔCT formula. U6 served as the internal reference for miR-21, and β-actin was the internal reference of mRNA. Primers are attached in Additional File 1 .

Data were expressed as mean ± SEM. Statistical significance is determined by unpaired two-tailed Student’s t-test (or nonparametric test), one-way ANOVA followed by the Tukey’s test for multiple comparisons (23). A value of P< 0.05 was statistically significant.

As shown in Figure 1A , the hPMSCs showed typical fibroblastic morphology. The hPMSCs were cultured in adipogenic and osteogenic induction medium to differentiate into osteoblasts( Figure 1B ) and adipocytes ( Figure 1C ), respectively. In addition, the FCM results suggested that more than 95% of isolated hPMSCs expressed CD105, CD90, CD73 and but not HLA-DR, CD19, CD14, or CD34 ( Figure 1D ).

Electron microscopy exhibited revealed typical circular particles ranging from 30 to 200 nm in diameter, and NTA revealed a similar size distribution of hPMSC-Exo ( Figures 1E, F ). Western blot analysis showed that hPMSC-Exo were positive for CD63, CD9, and HSP70 (positive markers for exosomes), but negative for Cyt-C (negative markers for exosomes) ( Figure 1G ).

The animal experiment schematic of hPMSC-Exo intervention is shown in Figure 2A . We purified CD4⁺ T cells using immunomagnetic beads and tested the SA-β-gal positive rate of CD4⁺ T cells. As shown in Figures 2B, C , the rate of SA-β-gal-positive CD4⁺ T cells significantly improved in the D-gal group compared with the control group. Conversely, the rate of SA-β-gal-positive CD4⁺ T cells was significantly reduced in hPMSC-Exo group compared with the PBS group ( Figures 2B, C ).

To evaluate the effects of hPMSC-Exo treatment on the redox metabolism of CD4⁺ T cells during aging, we detected the change of oxidative stress biomarkers and antioxidant capacity in CD4⁺ T cells. The ROS and 8-OH-dG levels of CD4⁺ T cells markedly improved in the D-gal group compared with the control group, while hPMSC-Exo intervention markedly reduced the levels of ROS and 8-OH-dG compared with the PBS group ( Figures 2D, E ). In contrast, the activities of T-AOC and SOD declined markedly in the D-gal group compared with the control group, while hPMSC-Exo intervention significantly increased the activities of T-AOC and SOD compared with the PBS group ( Figures 2F, G ).

In addition, the aging-related protein expression of p53 and γ-H2AX markedly improved in the D-gal group compared with the control group, while hPMSC-Exo intervention significantly reduced the levels of p53 and γ-H2AX compared with the PBS group ( Figures 2H, I ).

Secretion of pro-inflammatory factors is considered to be a key feature of the SASP phenotype, including OPN, IL-6, and proinflammatory chemokines (24, 25). As shown in Figure 2J , D-gal treatment markedly increased the mRNA levels of IL-6 and OPN compared with the control group, while hPMSC-Exo intervention significantly reduced the levels of IL-6 and OPN compared with the PBS group. These results suggested that hPMSC-Exo treatment weakens CD4⁺ T cell senescence in aging mice.

Nrf2 plays a crucial role in regulating intracellular redox balance, senescence, and inflammation (26, 27). Previous research suggested that the activation of Nrf2 is regulated by triggering PTEN/PI3K-mediated degradation and nuclear export of Nrf2 (27). Here, we found markedly increased expression of PTEN and declined phosphorylation of PI3K and Akt in CD4⁺ T cells from the D-gal group compared with the control group ( Figures 3A–D ). While hPMSC-Exo intervention markedly decreased the expression of PTEN and markedly raised the phosphorylation of PI3K and Akt in CD4⁺ T cells compared with the PBS intervention group ( Figures 3A–D ).

Furthermore, we found that the expression of nuclear Nrf2 and downstream target genes HO-1 and NQO1 were significantly declining in CD4⁺ T cells from the D-gal group compared with the control group ( Figures 3E–G ). Conversely, hPMSC-Exo intervention markedly improved the expression of these antioxidant proteins in comparison with that of the PBS intervention group ( Figures 3E–G ). These data indicated that hPMSCs treatment promoted the nuclear transfer of Nrf2 and the expression of Nrf2 target antioxidant genes. Therefore, the obtained data indicated that hPMSC-Exo attenuates CD4⁺ T cell senescence by targeting Nrf2 functions via PTEN/PI3K pathway.

We found that miR-21 is expressed at a significantly higher level in hPMSC-Exo than in hPMSCs themselves, suggesting that MSCs package the majority of the miR-21 produced in exosomes ( Figure 4A ). In addition, we found markedly increased expression of miR-21 in CD4⁺ T cells from the D-gal group compared with the control group. Conversely, hPMSC-Exo intervention further increased the expression of miR-21 in CD4⁺ T cells compared with the PBS intervention group ( Figure 4B ).

To validate whether PTEN is a target gene of miR-21, we performed a dual-luciferase reporter assay and identified the miR-21 binding site in PTEN, which was predicted by TargetScan to be in the 3’-UTR ( Figure 4C ). miR-21 mimics treatment downregulated the luciferase activity compared to the control group, whereas miR-21 inhibitor treatment markedly improved the relative luciferase activity. In addition, the effect of miR-21 on luciferase activity was not detected when the seed sequence of the miR-21 binding sites was mutated ( Figure 4D ). In addition, we modified the expression of miR-21 in 293T cells and using qRT-PCR and Western blot analysis demonstrated decreased PTEN mRNA and protein levels in cells treated with miR-21 mimics, while opposite trends were observed in cells treated with miR-21 inhibitor ( Figures 4E–G ). These data indicated that PTEN was targeted and negatively regulated by miR-21.

In the CD4⁺ T cells of D-gal treated mice, the level of miR-21was significantly decreased in the hPMSC-Exo miR-21 inhibitor treatment group compared with hPMSC-Exo group ( Figure 4H ). Furthermore, the miRNA level of PTEN was markedly increased, while the mRNA levels of HO-1 and NQO1 were significantly decreased after hPMSC-Exo miR-21 inhibitor treatment ( Figure 4H ). In addition, the aging-related protein expression of p53 and γ-H2AX were also improved significantly in the hPMSC-Exo miR-21 inhibitor treatment group compared with hPMSC-Exo group ( Figures 4I, J ). These results suggested hPMSCs ameliorate CD4⁺ T cells senescence by exosomes transmitted miR-21.

We further investigated the regulatory effect of hPMSC-Exo miR-21 on PTEN expression in CD4⁺ T cells ( Figure 5A ). The exosomes treated with different plasmids were co-cultured with CD4⁺ T cells, and the expression of miR-21 and PTEN was detected. The expression of miR-21 was markedly improved in CD4⁺ T cells after co-cultured with hPMSC-Exo compared PBS treatment group ( Figure 5B ). Conversely, hPMSC-Exo treatment greatly decreased the expression of PTEN in CD4⁺ T cells compared with the PBS group ( Figure 5C ). Moreover, hPMSC-Exo treated with miR-21 mimics further increasing the expression of miR-21 and decreasing the expression of PTEN in CD4⁺ T cells compared hPMSC-Exo treatment group, while the opposite trend appeared in the hPMSC-Exo treatment miR-21inhibitor group ( Figures 5B, C ). After incubation with CD4⁺ T cells, PKH26-labeled hPMSC-Exo exhibited red fluorescence in the cytoplasm of CD4⁺ T cells, implying the hPMSC-Exo uptake by CD4⁺ T cells ( Figure 5D ).

In addition, we further explored the effect of hPMSC-Exo on Nrf2 and downstream target antioxidant genes NQO1 and HO-1 expression in CD4⁺ T cells. Although hPMSC-Exo treatment did not cause significant changes in Nrf2 expression, the expression of NQO1 and HO-1 was significantly increased in CD4⁺ T cells compared PBS treatment group ( Figures 5E–G ). Moreover, hPMSC-Exo treated with miR-21 mimics further upregulating the expression of NQO1 and HO-1 in CD4⁺ T cells compared hPMSC-Exo treatment group, while the opposite trend appeared in the hPMSC-Exo treatment miR-21 inhibitor group ( Figures 5F, G ). The aforementioned results suggested that hPMSC-Exo was critical in transferring exogenous miR-21 from hPMSCs to CD4⁺ T cells and regulate the expression of PTEN/PI3K-Nrf2 signaling pathway.

We further confirmed whether the PTEN/PI3K-Nrf2 axis plays a vital role in the protective effects observed in senescent CD4⁺ T cells treated with hPMSC-Exo. After immuno-magnetic separation, a total of 4×10⁶ human naive CD4⁺ T cells (CD4CD45RA cells) were pretreated for 1 h with the PTEN inhibitor bpV(HOpic)(100 μM). Then, 2 µg/ml hPMSC-Exo were added to CD4⁺ T cells and co-cultured for 72 h. The schematic of experimental design is shown in Figure 6A . The expression of PTEN significantly decreased in activated CD4⁺ T cells after cultured for 72 h, while hPMSC-Exo intervention further decreased the expression of PTEN in activated CD4⁺ T cells ( Figures 6B, C ). Furthermore, the phosphorylation of PI3K and Akt were significantly increased in activated CD4⁺ T cells, and the hPMSC-Exo intervention further improved the phosphorylation of PI3K and Akt ( Figures 6D, E ). In addition, we found that the expression of nuclear Nrf2 and downstream target antioxidant genes NQO1 and HO-1 was markedly improved in activated CD4⁺ T cells, and the hPMSC-Exo treatment further improved the expression of these proteins ( Figures 6F–H ). Moreover, significantly improve nuclear Nrf2, NQO1, and HO-1 expression were also found in hPMSC-Exo treated CD4⁺ T cells after bpV(HOpic) supplementation ( Figures 6F–H ). These results suggested that hPMSC-Exo promotes the PTEN/PI3K-Nrf2 pathway activation by inhibiting PTEN in senescent CD4⁺ T cells.

We further evaluated the protective effects of hPMSC-Exo on senescent CD4⁺ T cells via PTEN/PI3K -Nrf2 pathway, we conducted SA-β-gal staining and detected changes in redox metabolism biomarkers in CD4⁺ T cells after coculture with hPMSC-Exo. hPMSC-Exo treatment significantly decreased the levels of ROS and 8-OH-dG and the rate of SA-β-gal-positive CD4⁺ T cells in activated CD4⁺ T cells ( Figures 7A, B, E, F ). Although the activities of T-AOC and SOD were increased greatly in CD4⁺ T cells after activation, hPMSC-Exo intervention could further increase the activities of T-AOC and SOD in activated CD4⁺ T cells ( Figures 7C, D ). Furthermore, hPMSC-Exo treatment also significantly decreased the levels of aging-related protein expression of p53 and γ-H2AX and the mRNA levels of OPN and IL-6 in activated CD4⁺ T cells. Furthermore, the hPMSC-Exo mediated downregulation in the expression of aging-related mRNA and/or protein for IL-6, OPN, p53 andγ-H2AX were also enhanced by bpV(HOpic) intervention ( Figures 7G–I ). The aforementioned results suggested that the protective effects of hPMSC-Exo on senescent CD4⁺ T cells relied on PTEN/PI3K pathway mediated Nrf2 antioxidant signaling.