hPMSCs protects against aging-induced oxidative damage of CD4+ T cells through activating Akt-mediated Nrf2 antioxidant signaling


 Background: Mesenchymal stem cells (MSCs) was considered as regenerative therapeutic approach in both acute and chronic diseases. However, whether MSCs regulate the antioxidant metabolism of CD4+ T cells and weaken immunosenescence remains unclear. Here, we reported the protective effects of hPMSCs in aging-related CD4+ T cell senescence and identified the underlying mechanisms using a D-gal induced mouse aging model.Methods: In vivo study, 40 male C57BL/6 mice (8 weeks) were randomly divided into four groups: control group, D-gal group, hPMSC group and PBS group. In in vitro experiment, human naive CD4+ T (CD4CD45RA) cells were prepared using a naive CD4+ T cell isolation kit II and pretreated with the Akt inhibitor LY294002 and Nrf2 inhibitor ML385. Then, isolated naive CD4+ T cell were cocultured with hPMSCs for 72 h in the absence or presence of anti-CD3/CD28 Dynabeads and IL-2 as a mitogenic stimulus. Intracellular ROS changes were detected by flow cytometry. The activities of the antioxidant enzymes superoxide dismutase, glutathione peroxidase and catalase were measured by colorimetric analysis. The senescent T cells were detected SA-β-gal stain. The expression of aging related proteins were detected by Western blotting, RT-PCR and confocal microscopy.Results: We found that hPMSC treatment markedly decreased the ROS level, SA-β-gal positive cells number, senescence-associated secretory phenotype (IL-6 and OPN) expression and aging-related protein (P16 and P21) expression in senescent CD4+ T cells. Furthermore, hPMSC treatment effectively upregulated Nrf2 nuclear translocation and the expression of downstream target genes (HO-1, CAT, GCLC and NQO1) in senescent CD4+ T cells. Moreover, in vitro studies revealed that hPMSCs attenuated CD4+ T cell senescence by upregulating the Akt/GSK-3β/Fyn pathway to activate Nrf2 functions. Conversely, the antioxidant effects of hPMSCs were blocked by the Akt inhibitor LY294002 and Nrf2 inhibitor ML385 in senescent CD4+ T cells.Conclusions: Our results indicate that hPMSCs attenuate D-gal induced CD4+ T cell senescence by activating Nrf2-mediated antioxidant defenses and that upregulation of Nrf2 by hPMSCs is regulated via the Akt/GSK-3β/Fyn pathway.


Introduction
With aging process, the immune system function gradually declines, leading to alterations in innate and adaptive immunity in older individuals, which is designated 'immunosenescence' [1]. Immunosenescence affects function and compartment of T cells, leading to age-related immune function decline and increases the susceptibility of elderly individuals to cancers and infectious diseases [2]. It is evident that aging decline alterations in CD4 + T cell immunity was accompanied by aging [3]. CD4 + T cell senescence is characterized by oncogene, reactive oxygen species (ROS) activation or tumor suppressor genes inactivation, leading to irreversible proliferation arrest [4,5]. Therefore, improving the antioxidant capacity of CD4 + T cells and reducing oxidative stress damage may delay the process of immune aging.
In aging, increasing oxidative stress was induced by various metabolites, on the contrary, as the the primary lines of defense the activity of antioxidant enzymes decreased. Nuclear factor erythroid-2-related factor 2 (Nrf2) is a crucial regulator of the cellular antioxidant system, which regulates the expression of a variety of key antioxidant enzymes [6,7]. There is evidence that activied Nrf2 protection against the phenotypic changes and mitochondrial function in memory T cells, relieve aged-related oxidant injury [8]. Increasing evidence suggests that Nrf2 pathway is essential in regulating the innate immune system function [9]. Together these data suggest that interfering with Nrf2 antioxidant signal provides a rational approach to alleviate cellular immune dysfunction during aging.
Mesenchymal stem (or stromal) cells (MSCs) was considered as regenerative therapeutic approach in both acute and chronic diseases [10,11]. Recently, it has been found that MSCs possess potent immunomodulatory and anti-in ammatory properties [12][13][14]. Interestingly, a recent study indicated that treatment with MSCs increased Nrf2 expression and activated the downstream antioxidant HO-1, leading to inhibition of oxidative stress, cell apoptosis and the in ammatory response in lung tissue [15]. However, whether MSCs regulate the antioxidant metabolism in senescent T cells via Nrf2-mediated exogenous antioxidant defenses and its in uence on aging-related T cell dysfunction remain to be elucidated.
Our ndings provided evidence that supports an important role for hPMSCs in attenuating D-galactose (D-gal)-triggered CD4 + T cell senescence by activating Nrf2-mediated exogenous antioxidant defenses.
And revealed the protective effect of hPMSCs in reducing oxidative damage in senescent CD4 + T cells, thereby clinically alleviating immunosenescence.

Animal models
Male C57BL/6 mice (8 weeks) were obtained from Binzhou Medical University (Yantai, China). Mice were housed in a standard environment with a regular light/dark cycle and free access to water and chow diet. The project was approved by the Ethics Committee of Binzhou Medical University, Yantai, China.

Isolation of hPMSCs
hPMSCs were isolated from human term placentas of donors as described previously [17]. Isolated hPMSCs were identi ed by detection of cell morphology using microscopy and cell surface antigens CD34, CD105, CD90, CD19, CD73, CD14 and HLA-DR using ow cytometry (FCM).
Then, hPMSCs were added at a 1:10 ratio to 4×10 6 CD4 + T cells in direct contact and cultured at 37°C in 5% CO 2 for 72 h in the absence or presence of anti-CD3/CD28 Dynabeads (1 μg/ml) and IL-2 (2.5 ng/ml) as a mitogenic stimulus. Detailed cell experimental designs are presented in Fig. 4A.

Intracellular ROS detection
The intracellular production of ROS was assessed by 2',7'-dichloro uorescein diacetate (H2DCF-DA) (Sigma, St Louis, MO, USA). Brie y, collected CD4 + T cells were washed with PBS. Then, added H2DCF-DA (10 μM) and incubated at 37℃ for 30 min. The levels of intracellular ROS in CD4 + T cells were analyzed by FCM after washed with PBS.

Antioxidant Biomarkers detection
The activities of the antioxidant enzymes superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT) in CD4 + T cells were measured by colorimetric analysis. GSH-Px activity was detected using the DTNB method [18], CAT activity was measured using the ammonium molybdate method [19] and SOD activity was measured using the xanthine-oxidase method [20].

SA-β-Gal staining
SA-β-Gal activity in senescent T cells were tested as described previously [21]. Brie y, T cells were xed with 3% formaldehyde ather washed in PBS, and followed to incubate overnight at 37°C with SA-β-Gal staining solution. After washing with PBS, senescent cells were observed under microscopy (Leica, Germany).

Western Blot Analysis
Western blot analyses were performed as previously described [22]. In short, total proteins were electrotransferred to PVDF membranes after separated by SDS-PAGE. Then, membranes were incubated with indicated primary antibody overnight at 4°C after blocked in 5% BSA dissolved in TBST for 2h at room temperature, followed by incubation with appropriate secondary antibody 2 h at room temperature. ECL plus detection reagents (Beyotime, Shanghai, China) was used for visualized protein bands. The Image J gel analysis software was used for densitometric analysis.

RNA extraction and quantitative real-time PCR
The levels of mRNA was measured using quantitative real-time PCR assay. Relative abundance of genes was calculated using 2 -∆∆CT formula, and β-actin as internal control. Primers attached in the additional le 1.

Immuno uorescence assay
CD4 + T cells were xed in 4% paraformaldehyde for 10 min after washed with cold PBS. Then, cells were permeabilized for 15 min using 1% Triton X-100. After washed with PBS, cells were incubation with primary antibody overnight. After washed with PBS, cells were incubation in uorescence-tagged secondary antibody for 1 h. Nuclei were counter-stained with DAPI. Laser scanning confocal microscope (FV3000, Olympus Corporation, Japan) was used for uorescence images capture.

Statistical analysis
Data represent as mean ± SEM. Statistical signi cance is determined by unpaired two-tailed Student's ttest (or nonparametric test), one-way analysis of variance (ANOVA) followed by Tukey's post hoc test. Signi cance was de ned as P < 0.05.

hPMSCs treatment attenuates D-gal-induced CD4 + T cells senescence in mice.
The schematic of hPMSC treatment is shown in Fig. 1A. We puri ed CD4 + T cells using immunomagnetic beads and detected ROS changes in CD4 + T cells and the effect of hPMSCs on senescence. The ROS level markedly increased in CD4 + T cells collected from D-gal group in compared with control group, while hPMSC treatment signi cantly decreased the generation of ROS in comparison with that of the PBS treatment group (Fig. 1B-D). In addition, the activities of SOD, CAT and GSH-Px decreased signi cantly in the D-gal group in compared with control group, while hPMSC treatment greatly improved the activities of SOD, CAT and GSH-Px in comparison with that of the PBS treatment group (Fig. 1E-G).
SA-β-gal is considered as a key indicator of aging cells [23,24]. To explore CD4 + T cell senescence in an aging mouse model and the effect of hPMSCs on senescence, we detected the positive rate of CD4 + T cells using an SA-β-gal kit. The number of SA-β-gal-positive CD4 + T cells markedly increased in the D-gal group in comparison with that of control group. Conversely, the number of SA-β-gal-positive CD4 + T cells was markedly declined in hPMSCs group compared with that of PBS treatment group (Fig. 1H-I).
Moreover, the expression of aging-related protein expression of P21 and P16 signi cantly improved in the D-gal group in compared with control group, while hPMSCs treatment greatly declined the levels of P16 and P21 in comparison with that of the PBS treatment group (Fig. 1J-K).
Secretion of proin ammatory factors is thought to be a key feature of SASP phenotype, including IL-8, IL-6 and proin ammatory chemokines [25], and increased expression osteopontin (OPN) was found in senescent CD4 + T cells [26]. The mRNA levels of IL-6 and OPN was markedly improved in the D-gal group, while hPMSCs treatment greatly declined the levels of IL-6 and OPN in comparison with that of the PBS treatment group (Fig. 1L). These data indicated that hPMSCs treatment attenuates CD4 + T cell senescence in a D-gal-induced aging mouse model.

hPMSCs promote the expression of Nrf2-mediated antioxidant genes.
Nrf2 plays an important role in regulating in ammation, senescence and intracellular redox balance [27].
Here, we found no signi cant change in total Nrf2 protein and/or Nrf2 mRNA expression in CD4 + T cells under any treatment condition ( Fig. 2A-B, D). However, when treated with D-gal, the ratio of nuclear/cytoplasm Nrf2 was markedly declined in CD4 + T cells ( Fig. 2A, C). Conversely, hPMSC treatment markedly improved the ratio of nuclear/cytoplasm Nrf2 in CD4 + T cells compared with those of the PBS treatment group ( Fig. 2A, C). These results suggested that hPMSCs upregulates of Nrf2 expression in the nucleus rather than increase total Nrf2 expression. In addition, we found that the protein and/or mRNA expression of the Nrf2 target antioxidant genes NQO1, CAT, HO-1 and GCLC was markedly lower in the Dgal group compared with control group (Fig. 2A, E-H). Conversely, hPMSC treatment greatly increased the expression of these proteins and/or mRNAs in comparison with that of the PBS treatment group ( Fig. 2A, E-H). These data indicated that hPMSCs treatment activates Nrf2 transcriptional functions.
Previous studies indicated that the activation of Nrf2 is regulated by adjusting Fyn-mediated degradation and nuclear export of Nrf2 [22]. To investigate the mechanisms by which hPMSC treatment activates Nrf2 transcriptional functions in CD4 + T cells, the total and phosphorylated GSK-3β, Akt and nuclear Fyn levels was measured. As shown in Fig. 3B-E, we found markedly decreased phosphorylation of GSK-3β and Akt and increased nucleus Fyn level in the D-gal group. hPMSC treatment markedly raised the phosphorylation of GSK-3β and Akt and decreased the level of Fyn in the nucleus in comparison with those of the PBS group. These results were also supported by examining Nrf2 and Fyn nuclear localization in the different groups (Fig. 3A). These results suggest that hPMSCs attenuate CD4 + T cell senescence by upregulating Nrf2 functions via Akt/GSK-3β/Fyn pathway.
We further con rmed whether the Akt/GSK-3β/Fyn pathway was critical in the protective effects observed in senescent CD4 + T cells that were treated with hPMSCs. A total of 4×10 6 human naive CD4 + T cells (CD4CD45RA cells) were isolated and pretreated for 1 h with the Akt inhibitor LY294002 (30 μM). Then, hPMSCs were added at a 1:10 ratio to CD4 + T cells and cocultured for 72 h in the presence or absence of anti-CD3/CD28 Dynabeads and IL-2 as a mitogenic stimulus. The schematic showing the experimental design is shown in Fig. 4A. As shown in Fig. 4B-C, with the increase of culture time, the expression of P16 and P21 raised in activited naive CD4 + T cells. This indicated that the senescence level of activated CD4 + T cells increased with the increase of culture time in vitro.
As shown in Fig. 4D-F, the ratio of nuclear/cytoplasm Nrf2 markedly improved in activated CD4 + T cells after cultured for 72 h, while hPMSCs treatment further improved the ratio of nuclear/cytoplasm Nrf2 in activated CD4 + T cells. Furthermore, we found that the mRNA and/or protein expression of the Nrf2 target antioxidant genes NQO1, HO-1, CAT and GCLC were also markedly improved in activated CD4 + T cells ( Fig. 4H-O). hPMSC treatment further increased the expression of these proteins and/or mRNAs in activated CD4 + T cells (Fig. 4H-O). Moreover, we found markedly increased phosphorylation of GSK-3β and Akt and decreased nucleus Fyn level in activated CD4 + T cells (Fig. 5B-E). hPMSC treatment signi cantly increased the phosphorylation of GSK-3β and Akt and decreased the nuclear Fyn level (Fig.  5B-E). These results were also supported by examining Nrf2 and Fyn nuclear localization in the different groups (Fig. 5A).
Subsequently, we explored the impact of LY294002 on the expression of Nrf2 and its downstream target genes in the presence of hPMSCs. Signi cantly reduced nuclear Nrf2, NQO1, HO-1, CAT and GCLC expression was observed in hPMSCs treated CD4 + T cells after LY294002 supplementation ( Fig. 4D-O). These data revealed that the effects of hPMSCs on Nrf2-mediated antioxidant signal are downstream of the Akt/GSK-3β/Fyn pathway in senescent CD4 + T cells.

Inhibition of the Akt/GSK-3β/Fyn pathway impairs the protective effects of hPMSCs on senescent
CD4 + T cells.
To further elucidate the protective effects of hPMSCs on senescent CD4 + T cells by Akt-mediated Nrf2 antioxidant signaling, we conducted SA-β-gal staining and detected changes in ROS in CD4 + T cells after coculture with hPMSCs. As shown in Fig. 6A-E, hPMSC treatment signi cantly decreased the percentage of SA-β-gal-positive CD4 + T cells and the level of ROS in activated CD4 + T cells. Although the antioxidant enzyme activity of SOD, CAT and GSH-Px were improved signi cantly in CD4 + T cells after activated, hPMSC treatment could further improve the antioxidant enzyme activity in activated CD4 + T cells (Fig. 6F-H). Furthermore, the signi cantly decreased percentage of SA-β-gal-positive CD4 + T cells and ROS levels and signi cantly increased antioxidant enzyme activity were abolished by LY294002 or ML385 (Nrf2 inhibitor) supplementation. Moreover, the hPMSC-mediated reduction in the expression of aging-related mRNA and/or protein for IL-6, OPN, P16 and P21 were also abrogated by LY294002 or ML385 supplementation (Fig. 6I-K). These data revealed that the protective effects of hPMSCs on senescent CD4 + T cells were dependent on Akt-mediated Nrf2 antioxidant signaling.

Discussion
CD4 + T cells play a central role in the persistence and development of immune responses. Evidence indicates that increased oxidative damage is related to the aging-related decline in immune functions [4,28]. Based on the antioxidant and anti-aging effects stem cells have been thought to be the source of seed cells for tissue engineering and biological therapeutics [15,29]. However, there is no evidence showing the protective effect of MSCs against aging-induced T cell dysfunction. Therefore, the present study shows that (1)  The balance between ROS generation and antioxidant capacity is necessary to ensures physiological levels of intracellular ROS and T cell-mediated immune response. Accordingly, excessive ROS generation in T cells will destroy the intracellular redox balance and leads to metabolism disorder and immune response dysfunction [30,31]. Here, we also observed markedly raised ROS level and decreased antioxidant enzyme activity of SOD, CAT and GSH-Px in the D-gal-treated group. Senescent cells cause serious damage to the tissue microenvironment by secreting senescence associated secretory phenotype, which is characterized by a markedly upregulating in the secretion of proin ammatory cytokines, matrix remodeling factors, chemokines and proangiogenic factors [32,33]. In this study, we also found markedly increased SASP expression (IL-6 and OPN) in senescent CD4 + T cells. Furthermore, the levels of P16 and P21 and SA-β-gal-positive cells increased markedly in senescent CD4 + T cells. Our data are consistent with previous ndings of the gradual accumulation of P16 and P21 expression and increased activity of SA-β-gal during several aging-associated diseases and physiological aging [34,35]. Based on the antioxidant capacity, MSCs and the MSC secretome derived from distinct tissue origins have been tested for the treatment of many diseases. Yan et al found that hfPMSCs protected against H 2 O 2 -induced cell oxidative damage and apoptosis by upregulating Nrf2/Keap1/ARE antioxidant signaling [36]. Shalaby et al reported that MSC injection was effective in modulating oxidative stress in E. coli-induced acute lung injury [37]. Here, we reported that hPMSC treatment markedly decreased the levels of ROS, SASP (IL-6 and OPN), aging-related protein (P16 and P21) and the number of SA-β-gal-positive cells in senescent CD4 + T cells. These results indicated that hPMSCs attenuate age-associated CD4 + T cell senescence. As a critical redox sensor, Nrf2 plays a vital role in antioxidant response in most tissue cells [38]. Numerous studies have suggested that disrupted activation of Nrf2 antioxidant signaling leading to decreased endogenous antioxidant response during aging [34,39,40]. In this study, we found that aging accompanies markedly The mechanism of Nrf2 regulation has been widely studied [42,43]. Several recent studies have suggested that the activition of Nrf2 was regulated by Akt/GSK-3β/Fyn mediated degradation and nuclear export of Nrf2 [22,44]. In this study, we detected signi cantly decreased activation of the Akt/GSK-3β/Fyn pathway in senescent CD4 + T cells. Similarly, a D-gal-induced decline in Akt phosphorylation was also found in human umbilical vein endothelial cells [45]. Recently, Li et al reported that human amniotic MSCs e ciently ameliorate heat stress-induced skin injury by inhibiting apoptosis in skin cells through activating the Akt signaling pathway [46]. Based on these studies, we hypothesized that Akt/GSK-3β/Fyn pathway is involved in hPMSCs induced Nrf2 activation in senescent CD4 + T cells.
Consistent with this hypothesis, we found that hPMSC treatment not only increased Akt phosphorylation but also inhibited GSK-3β activity, which decreased Fyn nuclear accumulation. Inactivation of Fyn kinase reinforces cell antioxidant defense by abolishes ubiquitination-mediated Nrf2 suppression. Under in vitro, we found that inhibition of the PI3K/Akt pathway by the inhibitor LY294002 downregulated Nrf2regulated antioxidant genes in senescent CD4 + T cells. Moreover, the hPMSC-mediated reduction in the expression of aging-related mRNA and/or protein for IL-6, OPN, P16 and P21 were abolished by Akt inhibitor LY294002 and Nrf2 inhibitor ML385. In addition, the decreased percentage of SA-β-gal-positive CD4 + T cells and ROS levels by hPMSC treatment were also abolished by LY294002 and ML385 supplementation. Our ndings reveal that the protective effects of hPMSCs on senescent CD4 + T cells depend on Akt-mediated Nrf2 antioxidant signaling. However, the direct link between hPMSC treatment and phosphorylation of the Akt pathway in senescent CD4 + T cells remains a key unanswered question.

Conclusions
In summary, as illustrated in Fig. 7, our data demonstrate a novel role for hPMSCs in attenuating D-gal induced CD4 + T cell senescence by activating Nrf2-mediated antioxidant defenses and that upregulation of Nrf2 by hPMSCs is regulated partially via the Akt/GSK-3β/Fyn pathway. Our ndings reveal that the administration of hPMSCs may be a novel therapeutic strategy for immunosenescence treatment.