Sirtuin 6 mediates the therapeutic effect of endometrial regenerative cell-derived exosomes in alleviation of acute transplant rejection by weakening c-myc-dependent glutaminolysis

All human ERCs used in this study were isolated from menstrual blood provided by healthy female volunteers aged 20–30 years, and the procedure was approved by Tianjin Medical University General Hospital (IRB2024-YX-013-01, Tianjin, China). First, menstrual blood is filtered to remove blood clots, and the bottom red precipitate is obtained by centrifugation. Then, these red precipitates were resuspended by PBS, slowly added to the same volume of human peripheral blood lymphocyte separation solution, and ERCs were obtained by density gradient centrifugation. Finally, the extracted ERCs were cultured in Dulbecco's modified Eagle's medium/Ham's F12 (DMEM/F12) nutrient medium supplemented with 10% fetal bovine serum (FBS) and 1% P/S solution in a 37 °C 5% CO₂ incubator. After being incubated for approximately 1–2 ° weeks, ERCs were harvested and phenotypically identified using flow cytometry. The identified ERCs were then used for further studies.

In order to further identify the stem cell characteristics of ERCs, the three-line differentiation experiments were carried out. 1. Osteogenic differentiation: 0.1% gelatin was added into a 12-well plate, with 500 μL in each hole, so that it covered the whole bottom surface. Then the gelatin coated 12-well plate was placed in a 37 °C incubator for at least 30 min. After 30 min, the gelatin was discarded and used to inoculate cells after the Petri dish was dried. ERCs were inoculated into a 12-well plate coated with gelatin in the amount of 4 × 10⁴ cells per well, and cultured in serum-free medium, and the medium was changed once a day. When the confluence degree of ERCs reached 70%, the medium was discarded by suction, and 1 mL of human mesenchymal stem cells osteogenic differentiation medium was added to each well to start the induced differentiation. After 21–30 days of ERC osteogenesis induction, the precipitation of calcium nodules was observed under the microscope, and the osteogenic differentiation was identified by alizarin red S staining. 2. Chondrogenic differentiation: 3 × 10⁵ ERCs were transferred to a 15 mL centrifuge tube, and the centrifuge was 300 g for 5 min. After centrifugation, the supernatant was poured out, 500 μL of human mesenchymal stem cells were added into chondrogenic differentiation medium, and the cells were resuspended. Centrifuge: 300 g for 5 min. The centrifuge tube cover was unscrewed for gas exchange, and it was kept in a cell incubator for 24 h. After 24–48 °h, when the cells aggregated, the bottom of the centrifuge tube was flicked to make the cartilage ball separate from the bottom of the tube and suspend in the liquid. The cell culture medium was changed every 2–3° days to prevent the cartilage ball from being scattered. After 21–28° days of chondrogenic differentiation of ERCs, the differentiation medium was discarded, and the chondrocytes were washed twice with 3 mL PBS, and then fixed with 3 mL 4% paraformaldehyde. The fixed chondrocytes were embedded in paraffin, and the chondrogenic differentiation was identified by alixin blue staining. 3. Adipogenic differentiation: ERCs were inoculated into 12-well plates with the number of 4 × 10⁴ cells per well, and cultured in serum-free medium. When the confluence of ERCs reached 100%, the medium was discarded, and 1 mL of adipogenic differentiation medium was added to each well to start differentiation induction. Adipogenic differentiation was induced for 9–35° days. After the formation of lipid droplets was observed under the microscope, the adipogenic differentiation was identified by oil red O staining.

To knock down SIRT6 in ERCs and further obtain ERC-Exos with low expression of SIRT6 (SIRT6-KD-ERC-Exos), We bought shRNA-SIRT6 lentivirus containing green fluorescent protein (GFP) and puromycin resistance gene from Gene-Chem company (Shanghai, China). The sequence of shRNA-SIRT6 was as follows: 5′-CGAGGATGTCGGTGAATTATTCAAGAGATAATTCACCGACATCCTCG-3'. Specifically, the SIRT6-Knockdown virus consisted of Lv-hU6-MCS (shRNA-SIRT6 sequence insertion site) -CBh-gcGFP-IRES-puromycin. Similarly, the empty vector virus had the same architecture. Empty vector virus inserted an unintentional sequence (5′-TTCTCCGAACGTGTCACGT-3′) at MSC site, which would not affect the expression of SIRT6. Lentiviral transduction was carried out following the instructions from Gene-Chem, using an optimal multiplicity of infection (MOI = 20). ERCs at passage 2 were plated onto a 6-well plate with a density of 15,000/well, and lentiviral transduction was carried out when the confluence reached 50%–60%. The transduction time was 12–16 °h, and the transduction is terminated according to the cell condition. After transduction, the cells were washed by PBS and replaced with fresh medium without lentivirus. To select for successfully transduced cells, puromycin (2 μg/mL, Solarbio, China) was incorporated into the culture medium. At the same time, puromycin was added to the ERCs of un-transduced lentivirus with the same generation and density, and the screening was finished when it died completely. After the cell screening, the GFP positive cells were detected by fluorescence microscope to be more than 90%, and then Western blot was carried out to detect SIRT6 expression level in ERCs. After successfully constructing SIRT6-KD-ERCs, the supernatant was collected to isolate SIRT6-KD-ERC-Exos. Besides, SIRT6-NC-ERCs could be obtained by successfully transducing ERCs with empty vector virus. Similarly, the further obtained exosome was SIRT6-NC-ERC-Exos.

ERCs were cultured in 10-cm dishes, washed with phosphate buffered saline (PBS), and maintained in FBS-free medium (Transgen, Beijing, China) for 48 h upon reaching nearly 90%–95% confluence. Then, the exosomes were obtained by centrifugation (Beckman centrifuge, United States) under the following conditions: 3,000 g, 4 °C, 10 °min; 10,000 g, 4 °C, 30 °min; 130,000 g, 4 °C, 70 min. The exosomes were resuspended with PBS and stored at −80 °C. The morphology of ERC-Exos was observed using transmission electron microscopy (TEM) (HT7700-SS, HITACHI, Japan). The size distribution of ERC-Exos was measured using nanoparticle tracking analysis (NTA) (Malvern Instruments Ltd., Malvern, United Kingdom). The concentration of ERC-Exos was determined using a bicinchoninic acid (BCA) protein assay kit (Solarbio, Beijing, China). In addition, exosomal markers, including CD9, CD63, TSG101, and calnexin, were detected using Western blot assay with the corresponding antibodies (dilution at 1:2000, 1:2,500, 1:2,000, and 1:3,000 respectively, abcam).

Adult C57BL/6 and BALB/c mice (male, 6–8 ° weeks old, weighing 23–26 g) were obtained from the China Food and Drug Inspection Institute (Beijing, China). All animals used in this study were housed in a conventional experimental environment with sufficient space, water, food, and appropriate temperature at the Tianjin General Surgery Institute (Tianjin, China). All animal procedures followed the animal use protocol approved by the Animal Care and Use Committee of Tianjin Medical University (Tianjin, China), and all the experiments were performed in accordance with the guidelines of the Chinese Council on Animal Care.

C57BL/6 mouse recipients were randomly assigned to three experimental groups (n = 5): untreated, ERC-Exo-treated, and SIRT6-KD-ERC-Exo-treated groups (exosomes were obtained through ERC transduction by shRNA-SIRT6 lentivirus). As described in our previous study (Hu et al., 2021), an intra-abdominal heterotopic cardiac transplantation model was adopted, in which the hearts of BALB/c mice were transplanted into the abdomen of C57BL/6 mouse recipients. Specifically, the donor aorta and the recipient abdominal aorta, and the donor pulmonary artery and the recipient inferior vena cava anastomosed end to side. After the operation, the recipient mice were fed separately under appropriate conditions. In ERC-Exo and SIRT6-KD-ERC-Exo treatment groups, the corresponding exosome treatment was given respectively by tail vein injection on the first, third and fifth day after operation. In detail, each mouse was injected with 200 μg exosomes at a time on the injection day. The heartbeats of the grafts were recorded daily by the same member of our research team who was blinded to the treatment details. According to the force of the graft heartbeat, the pulsation was divided into three degrees: A, beating strongly; B, noticeable decline in the intensity of pulsation; and C, complete arrest of the graft heartbeat. For the graft survival time (n = 5), the days until the graft heartbeat was completely arrested were recorded. Based on our previous experience (Hu et al., 2021), for the assessment of efficacy and immune microenvironment, the recipient mice were euthanized on post-operative day (POD) 8, and the grafts and spleens were extracted for subsequent analyses.

After being formalin-fixed, dehydrated, and paraffin-embedded, the heart graft samples were sliced into sections with 5-μm thickness. The sections were stained with hematoxylin and eosin (H&E). All the finished products were used to evaluate the severity of rejection under a light microscope. The presence of myocyte necrosis, interstitial hemorrhage, lymphocyte infiltration, vasculitis, and intravascular thrombosis was detected and scored in accordance with previously described criteria (Wang et al., 2003). On comparison with normal tissues, the following scores can be obtained: 0, no rejection; 1, mild interstitial or perivascular infiltrate without necrosis; 2, focal interstitial or perivascular infiltrate with necrosis; 3, multifocal interstitial or perivascular infiltrate with necrosis; and 4, widespread infiltrate with hemorrhage and/or vasculitis. The pathological score was calculated according to the degree of heart graft injury in each recipient mouse.

As described above, 5-μm heart tissue slices were obtained from the paraffin-embedded grafts. After programmed dewaxing, the slices were first incubated with ethylene diamine tetra-acetic acid (EDTA) antigen repair solution (Solarbio, Beijing, China), and antigen repair was conducted for 15 min at 100 °C in a microwave oven. They were then cultured with 3% hydrogen peroxide to eliminate endogenous peroxides. After being blocked with 10% goat serum, the sections were incubated with rabbit anti-mouse CD4 (dilution at 1:1,000, Abcam), IFN-γ (dilution at 1:100, ABclonal), IL-17A (dilution at 1:100, ABclonal), and Foxp3 (dilution at 1:100, affinity) antibody overnight at 4 °C and then with 100 μL of enhanced enzyme-labeled goat anti-rabbit IgG polymer (DAB kit, ZSGB-BIO, Beijing, China) for 20 min the next day at room temperature. Finally, the slices were stained with hematoxylin for 1 min and then observed under an optical microscope to evaluate the severity of CD4⁺ T cell infiltration. The percentage of the area of the section occupied by positive staining using ImageJ software.

The serum levels of interferon gamma (IFN-γ), IL-17, and IL-10 in the recipient mice were measured using their corresponding ELISA kits (DAKEWE, Shenzhen, China) in accordance with the manufacturer's instructions. In this experiment, 100 μL serum samples were required for each test well, and three representative cytokines were tested in this experiment. CD4⁺ T cells were isolated from the C57BL/6 mice and co-cultured with or without ERC-Exos or SIRT6-KD-ERC-Exos in 12-well plates. Then, the levels of IFN-γ, IL-17, and IL-10 in the cell supernatants were detected using their corresponding ELISA kits (DAKEWE, Shenzhen, China) in accordance with the manufacturer's instructions.

The extracted mouse naïve CD4⁺ T cells were placed in 24-well plates, with about 10⁶ cells per well, which were set into three groups, namely, untreated group, ERC-Exo treatment group, and ERC-SIRT6-KD-Exo treatment group. Each group was added with the same components necessary for CD4⁺ T cell activation, including but not limited to glutamine (2 mM), CD3 (5 μg/mL), CD28 (3 μg/mL), and IL-2 (50 U/mL). After 24 h of different treatments, supernatant and cells were collected separately. Glutamine contents of naïve CD4⁺ T cells were measured by use of a glutamine content measuring kit (Solarbio, Cat. NO.: BC5305, Beijing) in accordance with the manufacturer's protocol. The OD value of samples at A450 was tested by using a multifunctional enzyme label instrument. Then, the glutamine content was calculated according to the manufacturer's protocol. Additionally, the content of glutamine in the supernatant was also detected. Because the total glutamine content added to the culture medium among the three groups is the same, the difference of glutamine uptake by naïve CD4⁺ T cells can be obtained by comparing the glutamine content in supernatant and cells, respectively.

Naïve CD4⁺ T cells were collected after 24 h of different treatments, as in the previous treatment of glutamine content detection. α-KG contents of naïve CD4⁺ T cells were measured by use of an α-KG content measuring kit (Solarbio, Cat. NO.: BC5425, Beijing) in accordance with the manufacturer's protocol. The OD value of samples at A340 was tested by using a multifunctional enzyme label instrument. Then, the α-KG content was calculated according to the manufacturer's protocol.

Flow cytometry was performed to identify the phenotype of the ERCs and detect the immune cell population in the spleens of the recipient mice. The collected cells were divided into 100 μL single cell suspensions and then stained with fluorescent-labeled antibodies, including Zombie Dye, anti-CD14-FITC, anti-CD90-PE, anti-CD-79a-PE, anti-HLA-DR-FITC, anti-CD73-FITC, anti-CD44-APC, anti-CD4-FITC, anti-CD25-PE, anti-IL-17A-Percp-cy5.5, anti-FOXP3-APC, and anti-IFN-γ-PE, which were purchased from eBioscience (Thermofisher, United States) and BioLegend (Biolegend, United States). Finally, the percentage of different cells was analyzed using FlowJ software. The columnar statistical chart was made by Graphpad V8.0 software.

ERCs with a good growth state at passage 5 were adopted. The day before, the suspended ERCs were inoculated in a 12-well plate containing round coverslips, with a density of about 5 × 10⁴ cells/mL/well. When the cell aggregation reached 80%–90%, the cells were taken out and washed twice with PBS, then 1 mL of 4% paraformaldehyde fixed solution was added to each well for cell fixation at room temperature for 15 min, then PBS was washed twice, and 1% BSA blocking solution was added for nonspecific antigen blocking at room temperature for 1 h. Subsequently, the excess BSA blocking solution was sucked off, and CD73 antibody (dilution at 1:100, Abcam) was added for incubation at 4 °C overnight. The next day, the incubated antibody was taken out and returned to room temperature for 30 min. After PBS washing twice, the corresponding cy3-labeled secondary antibody (dilution at 1:50, Proteintech) was incubated for 1 h at room temperature in the dark. After PBS washing twice, the round coverslip was taken out and dripped with neutral resin containing 4,6-diamino-2-phenyl indole (DAPI) staining for sealing. The cells were observed under a fluorescence microscope.

The uptake of exosomes by naïve CD4⁺ T cells was determined as follows. ERC-Exos and SIRT6-KD-ERC-Exos were incubated with 1 μM PKH26 (Solarbio, Beijing, China) for 15 min at 37 °C and then centrifuged at 100,000 g for 70 min to remove unbound dye. Naïve CD4⁺ T cells were acquired from C57BL/6 mouse spleens and co-cultured with PKH-26 labeled exosomes for 6 h. The treated cells were washed twice with PBS and fixed with 4% paraformaldehyde. Following DAPI staining, the cells were observed under a confocal microscope.

Naïve CD4⁺ T cells were extracted from C57BL/6 mice and co-cultured with or without ERC-Exos or SIRT6-KD-ERC-Exos in 96-well plates to investigate whether SIRT6-expressing ERC-Exos can influence CD4⁺ T cell differentiation in vitro. In brief, a suspension of naïve CD4⁺ T cells was prepared. The sacrificed spleens were ground and individually filtered using sterilized meshes (100 meshes). After the red blood cells were lysed, the remaining cells were resuspended and counted. Based on the number of cells, CD4⁺ (L3T4) T cell-sorting magnetic beads (Miltenyi biotec, Germany) with corresponding volumes were added and allowed to stand for 10 min for complete mixing. The mixed liquid was then filtered twice by gravity under a magnetic field. Subsequently, the filter column was washed with culture medium to obtain naïve CD4⁺ T cells. Each well was covered with 2 × 10⁵ naïve CD4⁺ T cells, and the cells were cultured in 200 μL of Roswell Park Memorial Institute 1,640 (RPMI-1640) medium. Finally, ERC-Exos or SIRT6-KD-ERC-Exos were added to different groups for co-culture, and flow cytometry was conducted after 72 h.

To explore whether SIRT6-expressing ERC-Exos affect naïve CD4⁺ T cell activation in vitro, we extracted naïve CD4⁺ T cells from C57BL/6 mice and co-cultured them with or without ERC-Exos or SIRT6-KD-ERC-Exos in 96-well plates for 12–24 h. After being co-cultured, T cells were collected for flow cytometry. Specifically, the collected cells were divided into 100 μL single cell suspensions and then stained with fluorescent-labeled antibodies, including anti-CD4-FITC and anti-CD25-PE, which were purchased from eBioscience (Thermofisher, United States) and BioLegend (Biolegend, United States). Finally, the percentage of different cells was analyzed using FlowJ software.

CD4⁺ T cells in vitro from the untreated, ERC-Exo-treated, and SIRT6-KD-ERC-Exo-treated groups were washed and then lysed in radio immunoprecipitation assay (RIPA) lysis buffer containing a protease inhibitor. After the lysis, the lysate was centrifuged, then the supernatant was sucked, and the concentration of the sample was measured by the BCA protein assay kit. Subsequently, loading buffer (Solarbio, Beijing, China) was added and the protein sample was denatured by boiling. The equal mass protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to a polyvinylidene fluoride (PVDF) membrane. The protein-containing membrane after transfer was blocked with bovine serum albumin solution and then incubated with primary antibodies at appropriate dilution (SIRT6, 1:2000, abcam; c-Myc, 1:3,000, Affinity; ASCT2, 1:750, ABclonal; GLS1, 1:1,000, Affinity; p-S6K1, 1:500, ABclonal; S6K1, 1:1,000, ABclonal) at 4 °C overnight. The next day, the membrane was washed in tris-buffered saline with tween-20 (TBST) and then incubated in the corresponding secondary antibody dilution (goat anti-rabbit, 1:2000, Affinity) at room temperature. Finally, the membrane was washed with TBST and incubated with an enhanced chemiluminescence solution (Sparkjade, Jinan, China) for several seconds. The membrane was photographed and analyzed on an exposure platform. The quantitative analysis was performed by comparing the ratio of target proteins/internal proteins in different groups using ImageJ.

Experimental data are presented as mean ± SD. Differences between multiple groups were calculated using one-way analysis of variance (ANOVA). The survival curve of the heart graft was constructed using the Kaplan–Meier cumulative survival method, and differences among groups were analyzed using the log-rank (Mantel–Cox) test. In the data charts, one asterisk represents p ≤ 0.05, two asterisks represent p ≤ 0.01, and three asterisks represent p ≤ 0.001. Differences with p ≤ 0.05 were considered statistically significant. GraphPad Prism (version 8.0) was used for the statistical analyses.

The ERCs' morphology at passages 3–5 was used to assess their purity after being extracted from the volunteers' menstrual blood. Results showed that the ERCs were spindle shaped, fibroblast-like, and colony-forming (Figure 1A). Furthermore, a significant rate of proliferation was indicated by the doubling time, which was roughly 24 h. The stem cell characteristics of the ERCs were further evaluated using common MSC surface markers. At passage 5, the ERCs were detached and stained with the MSC surface markers CD14, CD79a, HLA-DR, CD44, CD73, and CD90. Consistent with our previous reports, the ERCs showed high expression levels of CD44, CD73, and CD90 but no expression of CD14, CD79a, and HLA-DR (Figure 1B). Among these MSC surface markers, CD73, a representative ERC positive marker, was also selected for immunofluorescence staining (Figure 1C) to further verify the stem cell characteristics of ERCs for supporting these flow cytometry analyses. Additionally, ERCs' strong osteogenic, chondrogenic, and adipogenic differentiation abilities were confirmed by the three-lineage differentiation experiment (Figure 1D). These findings demonstrated that the ERCs used in our study were MSCs with stable stem cell properties, the ability to cling to the culture dish, and the capacity to multiply quickly.

In order to determine whether ERC-Exos can express SIRT6, we first identified the subcellular localization of SIRT6 in ERCs. The results showed that, consistent with previous reports, SIRT6 could be expressed in the nucleus and cytoplasm of ERCs (Supplementary Figure S1). The expression of SIRT6 in the ERC-Exos was measured using Western blot analysis. Results showed that SIRT6 did express in ERC-Exos (Figure 2A). To construct ERC-Exos with low SIRT6 expression, ERCs were transduced using a shRNA-SIRT6 lentivirus. At 72 h after transduction, GFP expression was detected in more than 90% SIRT6-KD-ERCs, and knocking down SIRT6 in ERCs would not cause changes in cell morphology (Figure 2B). Further examination showed that the expression of SIRT6 in ERCs was significantly decreased after transduction (Figure 2C). Subsequently, the supernatants were centrifuged to obtain SIRT6-KD-ERC-Exos. The result showed that SIRT6 expression was significantly knocked down and almost undetectable in SIRT6-KD-ERC-Exos (Figure 2D). Subsequently, exosomes purified from the supernatant of ERCs were identified using TEM and NTA. As previously described, TEM found that ERC-Exos exhibited a bilayer membrane structure with a typical cup-shaped morphology (Figure 2E). NTA showed that the majority of ERC-Exos were in the size range of 30–200 nm (Figure 2F). Moreover, the characteristics of SIRT6-KD-ERC-Exos were identified, and the findings were consistent with those of exosomes obtained from un-transfected ERCs. The results suggested that knocking down SIRT6 in ERC-Exos did not affect the morphological characteristics or size distribution of these exosomes (Figures 2E,F). Meanwhile, Western blot analysis identified exosome-specific markers, including CD9, CD63, and TSG101, but not the non-exosome marker calnexin in the ERC-Exo and SIRT6-KD-ERC-Exo groups (Figure 2G). These data indicated that ERC-Exos could express SIRT6 and that knocking down SIRT6 did not alter the basic characteristics of ERC-Exos.

Then, at the beginning of subsequent experiments, we set up three groups (ERC-Exos, SIRT6-NC-ERC-Exos, and SIRT6-KD-ERC-Exos) to identify exosomes and compare their functions. Results, as shown in the Supplementary Figure S2A, there was no difference in exosome morphology, particle size distribution, and surface protein markers among the three groups. Subsequently, we also compared the functions of the three groups. We executed the naïve CD4⁺ T cell differentiation experiment in vitro. The results showed that there was no difference in differentiation of CD4⁺ T cells between ERC-Exos and SIRT6-NC-ERC-Exos groups after treatment (Supplementary Figure S2B). Taken together, these data indicated that ERC-Exos and SIRT6-NC-ERC-Exos were identical in shape, size, and function.

To determine whether SIRT6 mediates the therapeutic effect of ERC-Exos on AR, C57BL/6 recipient mice were transplanted with heart grafts from BALB/c donor mice and divided into three groups, namely, untreated, ERC-Exo-treated, and SIRT6-KD-ERC-Exo-treated group. As shown in Figure 3A, C57BL/6 recipient mice were injected with ERC-Exos or SIRT6-KD-ERC-Exos on the first, third and fifth day after operation, and at the same time, the untreated group mice were injected with the same volume of PBS. The results showed that ERC-Exos significantly prolonged allograft survival, and this effect was related to SIRT6 expression (Figure 3B). Specifically, the allograft survival time was significantly longer (p < 0.05) in the ERC-Exo-treated group (20.2 ± 2.4° days) than in the untreated group (8.2 ± 1.6° days). However, the allograft survival time was significantly shorter (p < 0.05) in the SIRT6-KD-ERC-Exo-treated group (14 ± 2° days) than in the ERC-Exo-treated group (20.2 ± 2.4° days). Graft pathology (Figure 3C) revealed that the grafts in the untreated group showed severe rejection with vasculitis and massive cell infiltration. After ERC-Exo treatment, the lesions in the grafts were significantly attenuated, with almost normal pathology on POD 8. However, SIRT6 knockdown reduced the therapeutic effect of ERC-Exos in mediating allograft protection, and the grafts showed typical features of AR. Graft pathology was scored according to accepted diagnostic criteria for heart rejection (Stewart et al., 2005). According to the findings, the untreated group had the highest pathological lesion score, followed by the SIRT6-KD-ERC-Exo-treated group. According to Figure 3D, the group that received ERC-Exo had the lowest pathological lesion score. In addition, ERC-Exo treatment significantly reduced intra-graft CD4⁺ T cell infiltration, but this effect was markedly inhibited when SIRT6 was knocked down in ERC-Exos (Figures 3E,F).

In order to further verify whether the therapeutic effect of ERC-Exos is related to its regulation of CD4⁺ T cell differentiation in the graft, we performed immunohistochemical staining in different (untreated, ERC-Exo treated, and SIRT6-KD-ERC-Exo treated) groups of mouse heart grafts. Different from the detection of CD4⁺ T cell infiltration level, we paid more attention to the different subtypes and cytokine levels of CD4⁺ T cells in the grafts. Results showed that in the untreated group, more inflammatory cytokines such as IFN-γ and IL-17A were found in the transplanted heart of mice, and the levels of these two cytokines decreased significantly after ERC-Exo treatment (Figures 4A,B,D,E). However, when SIRT6 was knocked down in ERC-Exos, this trend was obviously reversed (Figures 4A,B,D,E). In addition, we also detected the infiltration of anti-inflammatory cell Tregs in the graft. The infiltration distribution of Tregs showed an opposite trend. Treg infiltration was less in untreated grafts, but the proportion of Tregs in grafts increased significantly after ERC-Exo treatment. Remarkably, compared to the ERC-Exo treatment group, Treg infiltration in these grafts dramatically diminished after SIRT6-KD-ERC-Exos were administered (Figures 4C,F). These findings indicated that via modulating the proportion between different CD4⁺ T cell subtypes in the graft, SIRT6 mediated the therapeutic benefits of ERC-Exos in alleviating AR of cardiac allografts.

The above results showed that the SIRT6-expressing ERC-Exos affected the CD4⁺ T cell infiltration and the degree of pathological damage in these grafts. To explore the reasons behind this therapeutic effect, we determined the proportions of CD4⁺ T cell subtypes, including Th1, Th17, and Tregs, in the peripheral lymphoid organs of recipient mice. As shown in Figures 5A–D, the proportions of Th1 and Th17 cells were significantly lower in the ERC-Exo-treated and SIRT6-KD-ERC-Exo-treated groups than in the untreated group. Compared with the ERC-Exo-treated group, the SIRT6-KD-ERC-Exo-treated group had significantly higher proportions of Th1 and Th17 cells in the spleens of the recipient mice. Meanwhile, the Treg proportion was significantly higher (p < 0.001) in the ERC-Exo-treated group than in the untreated group (Figures 5E,F). Similarly, the Treg population was significantly lower (p < 0.01) in the SIRT6-KD-ERC-Exo-treated group than in the ERC-Exo-treated group (Figures 5E,F). Subsequently, the serum levels of cytokines, including IFN-γ, IL-17, and IL-10, were examined in the different groups. Compared with the other groups, the ERC-Exo-treated group had the lowest levels of the pro-inflammatory cytokines IFN-γ and IL-17 (IFN-γ: untreated group vs. ERC-Exo-treated group, p < 0.001; ERC-Exo-treated group vs. SIRT6-KD-ERC-Exo-treated group, p < 0.001; IL-17: untreated group vs. ERC-Exo-treated group, p < 0.001; ERC-Exo-treated group vs. SIRT6-KD-ERC-Exo-treated group, p < 0.01) and the highest anti-inflammatory cytokine IL-10 (untreated group vs. ERC-Exo-treated group, p < 0.001; ERC-Exo-treated group vs. SIRT6-KD-ERC-Exo-treated group, p < 0.01, Figure 5G). Although the SIRT6-KD-ERC-Exo-treated group had lower IFN-γ and IL-17 secretion and higher IL-10 secretion than that in the untreated group, the therapeutic effect was still stronger in the ERC-Exo-treated group than in the SIRT6-KD-ERC-Exo-treated group (IFN-γ: ERC-Exo-treated group vs. SIRT6-KD-ERC-Exo-treated group, p < 0.001; IL-17: ERC-Exo-treated group vs. SIRT6-KD-ERC-Exo-treated group, p < 0.01; IL-10: ERC-Exo-treated group vs. SIRT6-KD-ERC-Exo-treated group, p < 0.01, Figure 5G). These data indicated that SIRT6 mediated ERC-Exos to remodel the proportion of different CD4⁺ T cell subtypes and serum cytokine levels in transplant recipients.

To further explore the role of SIRT6 in mediating ERC-Exo-induced naïve CD4⁺ T cell differentiation in vitro, we co-cultured ERC-Exos with naïve CD4⁺ T cells from the peripheral lymphoid organs of the C57BL/6 recipient mice. Firstly, the purity of isolated naïve CD4⁺ T cells was detected by flow cytometry after isolation. The results (Supplementary Figure S3) showed that the purity of CD4⁺ T cells isolated from C57BL/c mice spleens was more than 98%. Besides, the recognized markers of naïve CD4⁺ T cells (CD44^loCD62L^hi) in mice were also used for staining, and the results showed that the purity of naïve CD4⁺ T cells was over 96% (Supplementary Figure S3). Because CD44CD62L can be used not only to identify naïve CD4⁺ T cells but also to identify effector T cells (CD44^hiCD62L^lo), the data show that effector T cells (0.02%) could be almost ignored in the newly extracted naïve CD4⁺ T cells of mice. Then, on the basis of the experimental grouping in vivo, the T-cell, T-cell + ERC-Exo, and T-cell + SIRT6-KD-ERC-Exo groups were established in vitro. Subsequently, the proportions of different CD4⁺ T cell subtypes were determined in vitro in the three groups. The T-cell group had the highest proportions of Th1 and Th17 cells and the lowest Treg proportion, whereas the T-cell + ERC-Exo group showed the opposite trend. Compared with the T-cell + ERC-Exo group, the T-cell + SIRT6-KD-ERC-Exo group had significantly higher proportions of Th1 and Th17 cells and lower Treg proportion (T-cell + ERC-Exo group vs. T-cell + SIRT6-KD-ERC-Exo group, Th1 cells: p < 0.05; Th17 cells: p < 0.05; Tregs: p < 0.01; Figures 6A–F). Finally, the secretion levels of cytokines IFN-γ, IL-17, and IL-10 were detected in the different groups. As shown in Figure 6G, the results were consistent with the experimental results in vivo; that is, ERC-Exo treatment induced cytokines to tilt in the direction of anti-inflammation, while the absence of SIRT6 in exosomes weakened this effect. These results indicated that SIRT6 mediated ERC-Exos to modulate naïve CD4⁺ T cell differentiation and cytokine secretion in vitro.

Subsequently, we further detected the level of cell death and proliferation during CD4⁺ T cell differentiation in vitro. We explored it by Annexin V/7-AAD dye and the results showed that there would be some cell death in the process of T cell differentiation. When ERC-Exo treatment was given, the proportion of apoptotic cells in T cells decreased obviously, and when SIRT6 was knocked down, the decreasing trend increased again (Supplementary Figure S4A,B). These data indicated that SIRT6 also played a certain role in mediating ERC-Exos to inhibit T cell death. As to proliferation during in vitro T cell differentiation experiments, we used Ki67 dye to detect the proliferation of different treatment groups. The results, as shown in the Supplementary Figure S4C, showed that ERC-Exo treatment obviously inhibited the excessive proliferation of T cells under inflammatory conditions, while the knockdown of SIRT6 in exosomes reversed this trend, which indicated that SIRT6 mediated its regulation of T cell proliferation in ERC-Exos.

To further explore the mechanisms by which SIRT6 mediates ERC-Exos to modulate CD4⁺ T cell differentiation, we conducted an exosome phagocytosis experiment. Both ERC-Exos and SIRT6-KD-ERC-Exos were phagocytosed by naïve CD4⁺ T cells, with no significant difference in the level of phagocytosis (Figure 7A). Subsequently, the SIRT6 protein level in naïve CD4⁺ T cells were measured after co-culturing with ERC-Exos or SIRT6-KD-ERC-Exos for 12 h. Western blot analysis showed that ERC-Exo treatment significantly increased SIRT6 protein expression in naïve CD4⁺ T cells (Figures 7B,C). However, treatment with SIRT6-KD-ERC-Exos did not significantly change SIRT6 protein expression level in naïve CD4⁺ T cells (Figures 7B,C).

SIRT6 has been shown in earlier research to be able to block c-Myc transcription and translation in a variety of cell types. The transcription factor c-Myc is essential for T cell proliferation and differentiation. According to prior proteomic researches, c-Myc regulates the expression of important proteinases and amino acid transporters in immunocompetent T cells, with ASCT2 and GLS1 displaying the most notable alterations (Wang et al., 2011; Marchingo et al., 2020). To find out this change, we first used CD3/CD28 to stimulate the activation of naïve CD4⁺ T cells, and then treated them with ERC-Exos/SIRT6-KD-ERC-Exos, respectively, and then detected the transcription level of c-Myc in T cells. The results showed that after ERC-Exo treatment, the transcription level of c-Myc mRNA in naïve CD4⁺ T cells decreased significantly, and this decrease would be reversed with the knock-down of SIRT6 in exosomes (Supplementary Figure S5). Besides, the transcription level of ASCT2 and GLS1 mRNA in naïve CD4⁺ T cells also showed the same trend (Supplementary Figure S5). Then, we further use Western blot to analyze their expression at the translation level. The results showed that ERC-Exo treatment significantly inhibited c-Myc expression, whereas SIRT6-KD-ERC-Exo treatment restored this inhibitory function (Figures 7D,E). The downstream target proteins of c-Myc, including ASCT2 and GLS1, were also detected, and the result was consistent with the change in c-Myc protein expression (Figures 7D,E). In order to further clarify that this change is indeed related to SIRT6, we further explored it by using SIRT6 inhibitor OSS_128,167. In the experimental group where SIRT6 inhibitor interfered with ERC-Exo treatment, the protein levels of c-Myc, ASCT2, and GLS1 increased compared with ERC-Exo treatment (Supplementary Figure S6A,B). Subsequently, we detected the glutamine uptake of CD4⁺ T cells in different treatment groups. These results are consistent with the above changes in protein levels. Compared with the other two groups, the untreated group showed the highest glutamine intake level (Figure 7F). However, after ERC-Exo treatment, the glutamine uptake level of T cells decreased obviously, showing the lowest level (Figure 7F). Compared with ERC-Exo treatment, the uptake level of glutamine in T cells treated with SIRT6-KD-ERC-Exo increased, but it was still lower than that in the untreated group (Figure 7F). In addition, the content of α-KG in CD4⁺ T cells was further detected in the three groups, respectively. Consistent with the level of glutamine intake, the highest level of α-KG content was displayed in CD4⁺ T cells of the untreated group (Figure 7G). After ERC-Exo treatment, the content of α-KG in CD4⁺ T cells decreased significantly, but when SIRT6 in ERC-Exo was knocked down, the content of α-KG in T cells increased (Figure 7G). Similarly, after OSS_128,167 treatment, glutamine uptake and intracellular α-KG production of naïve CD4⁺ T cells were significantly inhibited compared with ERC-Exo treatment group (Supplementary Figure S6C,D). Previous studies have revealed that the activation state of mammalian target of rapamycin complex 1 (mTORC1) is regulated by glutaminolysis in various cells. Considering the decrease of glutamine uptake and its metabolite α-KG, we further explored the activation state of the mTORC1 signaling pathway through Western blot analysis. Similarly, the data showed that the ERC-Exo treatment inhibited the activation of the mTORC1 pathway in naïve CD4⁺ T cells, and this effect was related to the expression of SIRT6 in exosomes (Figures 7H,I). Administration of SIRT6-KD-ERC-Exos significantly reversed the inhibition of the mTORC1 pathway (Figures 7H,I). The Supplementary Figure S6E,F further proved this effect mediated by SIRT6 in ERC-Exos.

In order to further clarify that this effect is mediated by SIRT6, we used SIRT6 activator and SIRT6 inhibitor in Jurkat cell to explore the activated state of c-Myc/glutaminolysis/mTORC1 status respectively. The results showed that the level of c-Myc-dependent glutaminolysis increased after the activation of CD3/CD28 in Jurkat cell, which led to the activation of mTORC1 (). When SIRT6 activator was given, these effects were obviously inhibited (), indicating that SIRT6 did play a key role in inhibiting c-Myc-dependent glutaminolysis of T cells. Similarly, the opposite trend after administration of SIRT6 inhibitor further proves this effect (). These data demonstrated that SIRT6 did regulate the c-Myc/glutaminolysis/mTORC1 status in activated T cells. Supplementary Figure S7 Supplementary Figure S7 Supplementary Figure S7

We measured the activation level of naïve CD4⁺ T cells in light of earlier research showing that the production of important proteins linked to glutaminolysis is suppressed and that T cell activation is linked to glutaminolysis. The results showed that administration of ERC-Exos significantly restrained the activation of naïve CD4⁺ T cells compared with untreated T cells (T-cell + ERC-Exo group vs. T-cell group, p < 0.01, Figures 7J,K; Supplementary Figure S8). However, SIRT6 knockdown in ERC-Exos led to an increased activation of naïve CD4⁺ T cells (T-cell + ERC-Exo group vs. T-cell + SIRT6-KD-ERC-Exo group, p < 0.05, Figures 7J,K; Supplementary Figure S8). Taken together, these data showed that SIRT6 mediated ERC-Exos to inhibit naïve CD4⁺ T cell activation and mTORC1 activity by weakening c-Myc-dependent glutaminolysis.