Balb/c and C57BL/6 mice (6–8 weeks old, 19.9 $\pm$ 2.8 g) were purchased from the Laboratory Animal Center of Xi’an Jiaotong University (Shaanxi, China). All procedures were carried out in accordance with NIH Guidelines, and animal experiments were reviewed and approved by the Biomedical Ethics Committee of Xi’an Jiaotong University.

The induction and culture of imDCs derived from mouse bone marrow HSCs was performed as described previously [15]. Briefly, bone marrow was obtained from the femurs of C57BL/6 mice and the erythrocytes were lysed. Lin-Sca1${}^{+}$cKit${}^{+}$ HSCs were then sorted using a mouse lineage cocktail and fluorochrome-conjugated antibodies (Miltenyi, Cologne, Nordrhein-Westfalen, Germany). Magnetic cell sorting (MACS; Miltenyi, Cologne, Nordrhein-Westfalen, Germany) was also used to isolate CD3${}^{+}$ T cells and CD4${}^{+}$ T cells.

HSCs were treated with different concentrations of Nar (purity $>$98%, Solarbio, Cat#SN8020, Beijing, China) (0, 50, 100, 200, 300, and 400 $\mu$M). To optimal concentration of Nar was determined using the Cell Counting Kit-8 (CCK-8) assay (MedChemExpress, Rocky Hill, NJ, USA) to measure cytotoxicity. The optimal concentration was defined as the maximum Nar concentration at which cell viability was $>$95%. Finally, HSCs were cultured for 3 days in RPMI-1640 with 10% fetal bovine serum (FBS, Cat#12664025, Gibco, Carlsbad, CA, USA) and 20 ng/mL of granulocyte-macrophage colony-stimulating factor (GM-CSF, Cat#31503, Perprotech, Rocky Hill, NJ, USA), 10 ng/mL IL-4 (Perprotech, USA) and 200 $\mu$M Nar (optimal concentration). Flow cytometry was used to detect the hematopoietic cell marker antigens Flk-1 and CD45, and of the monocyte antigen CD11b. This allowed analysis of the molecular phenotypic changes that occured during HSC differentiation into DCs. Following cell culture for a further 3–4 days, some cells were then treated for 48 h with 5 $\mu$g/mL Lipopolysaccharide (LPS, Cat#L5293, Sigma, Germany) for the subsequent experiment.

Flow cytometry was also used to evaluate the phenotype of HSCs-imDCs, Nar-HSCs-imDCs, sorted CD3 and CD4 T cells, the effect of HSCs-imDCs and Nar-HSCs-DCs on Treg cells, apoptosis, and the phagocytic capacity of cells. HSCs-imDCs and Nar-HSCs-imDCs (1 $\times$ 10${}^5$) were stained with FITC-CD80 (Cat# 104705), FITC-CD86 (Cat# 105005), FITC-MHC-II (Cat# 125507), and FITC-CD11c (Cat# 117305), (Biolegend, San Diego, CA, USA). Multicolor staining for Tregs was carried out with FITC-CD4 (Cat# 100405), PE-CD25 (Cat# 113703) and APC-FoxP3 (Cat# 420201) (Biolegend, San Diego, CA, USA). Apoptosis was detected using an Annexin V-FITC Apoptosis Detection Kit (C1062M; Beyotime, Beijing, China), and the data analyzed using FlowJo software (version 10.0, FlowJo LLC, Ashland, OR, USA).

CD4${}^{+}$ T cells were isolated from the spleen of Balb/c mouse using MACS immunomagnetic beads (Miltenyi, Germany). HSCs-imDCs and Nar-HSCs-imDCs were treated with 25 $\mu$g/mL mitomycin. RPMI-1640 medium containing 5 $\mu$g/mL LPS was used as a control. HSCs-imDCs, Nar-HSCs-imDCs, or controls were co-cultured for 72 h at varying proportions with CD4${}^{+}$ T cells (DCs: T cells at 1:80, 1:40, 1:20, or 1:10). The inhibition of CD4${}^{+}$ T cells was analyzed using the CCK-8 assay, with the extent of inhibition was calculated as follows: (1 – OD of the experimental group/OD of the control group) $\times$ 100%.

Enzyme-linked immunoassay (ELISA; Elabscience, Hubei, China) was used to quantify the levels of IL-2 (Cat#E-ELM0042c), IL-4 (Cat#E-EL-M0043c), IL-10 (Cat#E-ELM0046c), IFN-$\gamma$ (Cat#E - EL - M0048c), and TGF-$\beta$ (Cat#E-EL-M1191c) in the culture supernatant of HSCs-imDCs and Nar-HSCs-imDCs before and after 48 h of stimulation with LPS (5 $\mu$g/mL). ELISA was also used to analyze cytokine levels in the serum of graft recipients at 14 days post graft. The detection limits for IL-2, IL-4, IL-10, IFN-$\gamma$ and TGF-$\beta$ varied between 15.63 to 1000 pg/mL.

FITC-dextran was used to determine the endocytic ability of HSCs-imDCs and Nar-HSCs-imDCs before and after stimulation with 5 $\mu$g/mL LPS for 48 h at 37 ℃. Post-stimulation HSCs-imDCs and HSCs-DCs were used as the control groups.

The Tunel Apoptosis Assay Kit (Cat#1086, Beyotime, Beijing, China) was used to evaluate the apoptosis of Nar-HSCs-imDCs and Nar-HSCs-DCs cells. Briefly, cells were treated with protein kinase K and 3% H${}_2$O${}_2$ and then incubated with Tunel detection solution and Streptavidin-HRP solution. DAB solution was added before observation and photography with the BX41 fluorescence microscope (Olympus Optical Co. Ltd., Tokyo, Japan; amplification: $\times$400). Cell apoptosis was analyzed using Image-Pro Plus analysis software 6.0 (Media Cybernetics, San Diego, CA, USA). The proportion of apoptotic cells was calculated as follows: number of positive cells/total number of cells $\times$ 100%.

C57BL/6 mice served as donors, while Balb/c mice served as the recipients. Balb/c mice were randomly divided into 8 groups, with 6 mice in each group. ① Sham group; allogeneic mouse skin graft model without treatment; ② phosphate-buffered saline (PBS group); 0.3 mL PBS was infused intravenously into recipients 7 days before grafting; ③ 10${}^5$ HSCs-imDCs cell group (10${}^5$ HSCs-imDCs/mouse); ④ 10${}^6$ HSCs-imDCs cell group (10${}^6$ HSCs-imDCs/mouse); ⑤ 10${}^5$ Nar-HSCs-imDCs cell group (10${}^5$ Nar-HSCs-imDCs/mouse); ⑥ 10${}^6$ Nar-HSCs-imDCs cell group (10${}^6$ Nar-HSCs-imDCs/mouse): ③–⑥ group: corresponding cells for each group suspended in 0.3 mL PBS were infused intravenously into recipients 7 days before transplantation; ⑦ Nar group; ⑧ Cyclosporin A (CsA) group; ⑦–⑧ group: 30 mg/kg Nar or CsA were infused into recipients through the tail vein once a day for 3 d before grafting. Allogeneic skin grafting was performed as described previously [16]. Starting from 4 days post-transplant, skin grafts were assessed for rejection once daily for 24 days. Kaplan-Meier analysis was used to evaluate graft survival. The graft together with the recipient’s spleen were dissected for analysis on day 14 after grafting.

Spleen cells from each group were collected after mincing. The effect of Nar-HSCs-imDCs on Tregs in the spleen was evaluated by flow cytometry. Unrelated C3H mice were used to evaluate specific reactivity with lymphocytes.

Grafted skin sections (4-mm thickness) were stained with hematoxylin and eosin (H&E). Images were captured using a BX41 fluorescence microscope (amplification: 200$\times$; Olympus Corporation, Tokyo, Japan).

Quantitative data were shown as the mean $\pm$ standard error (SEM). Experiments were repeated six times. Intergroup deviations were analyzed using one-way analysis of variance (ANOVA) and GraphPad Prism 9.0 software (Dotmatics, Boston, MA, USA), with p 0.05 considered to represent statistical significance.

The chemical structure of naringenin is shown in Fig. 1A. Morphological changes in Nar-HSCs-imDCs were documented in order to assess the role of Nar in promoting HSC differentiation into imDC, while the cell phenotype during differentiation was examined by flow cytometry. The optimal concentration of Nar (maximum concentration at which cell viability was $>$95%) was determined to be 200 $\mu$M (Fig. 1B). Phase contrast microscopy was used to evaluate morphological changes in Nar-HSCs-imDCs after 3 d and 5 d of culture (Fig. 1C). Burr-like protrusions on the cell surface were observed by scanning electron microscopy. Mitochondria, vesicles, and lysosomes were observed by transmission electron microscopy (Fig. 1C). Elevated expression of the hematopoietic cell markers antigen Flk-1 and CD45 and of the monocyte marker antigen CD11b was observed on the Nar-HSCs-imDC surface (Fig. 1D).

Fig. 1.

Differentiation of HSCs into imDCs under the action of Nar. (A) Chemical structure of naringenin; molecular formula: C${}_{15}$H${}_{12}$O${}_5$. (B) The optimal concentration of Nar (maximum concentration at which cell viability was $>$95%) was 200 $\mu$M. Compared with 0 $\mu$M, ${}^{*}$p 0.05. (C) Phase contrast microscopy showing morphological changes in Nar-HSCs-imDCs after 3 d and 5 d of culture. Scanning electron microscopy and transmission electron microscopy of Nar-HSCs-DCs after 6 d of culture (red arrow = mitochondria; green arrow = lysosome). (D) Flow cytometry detection of Flk, CD45 and CD11b expression on the surface of Nar-HSCs-imDCs. Each experiment contained six biological replicates.

Cell surface phenotypes were analyzed by flow cytometry in order to determine the effect of Nar on the differentiation of HSCs into imDCs. The results showed significantly greater downregulation of CD80, CD86, and MHC-II expression on the surface of Nar-HSCs-imDCs than HSCs-imDCs (p 0.05). In contrast, the expression of CD11c was significantly upregulated in Nar-HSCs-imDCs (p 0.01) (Fig. 2). These results suggest that Nar inhibits the maturation of DCs derived from HSCs.

Fig. 2.

DC surface phenotype in differentiation culture. The Nar-HSCs-imDCs group is compared with the HSCs-imDCs group. ${}^{*}$p 0.05, ${}^{**}$p 0.01. Each experiment contained six biological replicates.

We next investigated the effects of HSCs-imDCs and Nar-HSCs-imDCs on T cells by isolating CD3${}^{+}$ and CD4${}^{+}$ T cells from the spleen of recipient mice using magnetic beads. The purity of CD3${}^{+}$ T cells was 94.31 $\pm$ 2.57% (Fig. 3A). HSCs-imDCs and Nar-HSCs-imDCs were first stimulated with LPS for 48 h. MLR with sorted T cells was then performed for 72 h. DCs were co-cultured with T cells in varying proportions (1:10, 1:20, 1:40, and 1:80). The Nar-HSCs-DCs group inhibited T cell proliferation more than the HSCs-DCs group. The difference was most significant at a proportion of 1:40 (Fig. 3B) (p 0.01). The purity of CD4${}^{+}$ T cells was 97.95 $\pm$ 1.28% (Fig. 3C). HSCs-DCs and Nar-HSCs-DCs were co-cultured with CD4${}^{+}$ T cells at a ratio of 1:40 for 72 h. Nar-HSCs-DCs again inhibited the proliferation of CD4${}^{+}$ T cells significantly more than HSCs-DCs (Fig. 3D) (p 0.01). We next sorted CD4${}^{+}$CD25${}^{+}$ Treg cells and co-cultured them with HSCs-DCs or Nar-HSCs-imDCs at a ratio of 40:1. Flow cytometry showed that CD4${}^{+}$CD25${}^{+}$FoxP3${}^{+}$ Treg cell proliferation was promoted significantly more by Nar-HSCs-DCs than by HSCs-DCs (Fig. 3E) (p 0.01). Furthermore, IL-2 and IFN-$\gamma$ levels in culture medium from the Nar-HSCs-DCs group were significantly lower than in the Nar-HSCs-DCs group (p 0.05), whereas the levels of IL-4, IL-10 and TGF-$\beta$ were significantly higher (Fig. 3F) (p 0.01). In summary, these results showed that Nar-HSCs-DCs can inhibit the proliferation of CD4${}^{+}$ T cells, enhance the proliferation of CD4${}^{+}$CD25${}^{+}$FoxP3${}^{+}$ Treg cells, suppress the secretion of IL-2 and IFN-$\gamma$, while increasing the secretion of IL-4, IL-10 and TGF-$\beta$.

Fig. 3.

Nar-HSCs-imDCs suppress T cells and enhance the proliferation of Treg cells. (A) Purity of CD3${}^{+}$ T cells after sorting. (B) MLR with different ratios of imDCs to CD3${}^{+}$ T cells. (C) Purity of CD4${}^{+}$ T cells after sorting. (D) Mixed cultures of HSCs-DCs or Nar-HSCs-DCs with CD4+ T cells at a ratio of 1:40. (E) Effect of HSCs-DCs or Nar-HSCs-DCs on Treg cell proliferation. (F) Cytokine levels in supernatants from co-cultured DCs and Treg cells. Nar-HSCs-DCs are compared with the HSCs-DCs group. ${}^{*}$p 0.05, ${}^{**}$p 0.01. Each experiment contained six biological replicates.

imDCs have been reported to have more vital phagocytosis ability than DCs [6]. The present study found that Nar-HSCs-imDCs and Nar-HSCs-DCs had significantly higher endocytic capacity for FITC-dextran at 37 °C than HSCs-imDCs and HSCs-DCs, respectively (Fig. 4) (p 0.05).

Fig. 4.

Endocytic capacity of Nar-HSCs-imDCs and Nar-HSCs-DCs. The results are compared with the HSCs-imDCs and HSCs-DCs groups, respectively. ${}^{*}$p 0.05. Each experiment contained six biological replicates.

We hypothesized that LPS stimulation of Nar-HSCs-imDC resulted in apoptosis. To test this, flow cytometry was used to quantify the level of apoptosis in Nar-HSCs-DCs. The apoptosis rate of Nar-HSCs-DCs was significantly higher than that of HSCs-DCs (Fig. 5A). TUNEL staining showed similar results (Fig. 5B).

Fig. 5.

Apoptosis in HSCs-DCs and Nar-HSCs-DCs. (A) Apoptosis in HSCs-DCs and Nar-HSCs-DCs was measured by flow cytometry. (B) Apoptosis calculated by TUNEL staining. Nar-HSCs-DCs was compared with the HSCs-DCs group. ${}^{**}$p 0.01. Each experiment contained six biological replicates.

H&E staining showed that lymphocyte infiltration was reduced in the HSCs-imDCs and Nar-HSCs-imDCs groups compared to the control group, especially in the 10${}^6$ Nar-HSCs-imDCs group (Fig. 6A). The graft rejection score was lower in the Nar-HSCs-imDCs group compared to the HSCs-imDCs group, particularly in the 10${}^6$ Nar-HSCs-imDCs group (Fig. 6B). Moreover, the allograft survival time was significantly prolonged in the 10${}^6$ Nar-HSCs-imDCs group (Fig. 6C).

Fig. 6.

Nar-HSCs-imDCs induce immune hypo-responsiveness in the mouse skin transplant model. (A) H&E staining. (B) Graft scores. (C) Graft survival curves. Each experiment contained six biological replicates.

The Nar-HSCs-imDCs group showed a higher percentage of CD4${}^{+}$CD25${}^{+}$ Treg/CD4${}^{+}$ T cells than the HSCs-imDCs and Nar groups, while the percentage of CD4${}^{+}$CD25${}^{+}$FoxP3${}^{+}$ Treg/CD4${}^{+}$CD25${}^{+}$ Treg showed the same trend. The percentage was also significantly higher in the 10${}^6$ Nar-HSCs-imDCs group compared to the 10${}^5$ Nar-HSCs-imDCs group (Fig. 7A,B) (p 0.05). Cytokine levels in the serum of all groups were measured on day 7 post-transplant. IL-2 and IFN-$\gamma$ levels were lower, whereas IL-4, IL-10, and TGF-$\beta$ levels were higher in the serum from 10${}^6$Nar-HSCs-imDCs immunization recipients compared with 10${}^5$ Nar-HSCs-imDCs and 10${}^6$ HSCs-imDCs immunization recipients (Fig. 7C) (p 0.05). On day 7 post-transplant, spleen lymphocytes were taken from donor mice and unrelated third-party C3H mice, treated with 20 $\mu$g/mL mitomycin (MMC), and then used as stimulatory cells. Lymphocytes from the spleen of recipient mice were used as reaction cells. The in vitro MLR model was used to test whether the effect was specific. The responsiveness of spleen lymphocytes was found to be significantly lower in donor mice than in C3H mice in both the HSC-imDCs and Nar-HSCs-imDCs groups, and was lowest in the 10${}^6$ Nar-HSCs-imDCs group (Fig. 7D) (p 0.05). This result demonstrates that the infusion of donor-derived Nar-HSCs-imDCs can induce donor-specific immune tolerance, especially in the 10${}^6$ Nar-HSCs-imDCs group.

Fig. 7.

Nar-HSCs-imDCs inhibited T cells and promoted Treg cells proliferation in allografts. (A) The ratios of CD4${}^{+}$CD25${}^{+}$Treg cells/CD4${}^{+}$ T cells in the spleen in all groups. (B) CD4${}^{+}$CD25${}^{+}$FoxP3${}^{+}$ Treg cells/CD4${}^{+}$CD25${}^{+}$ Treg cells in the spleen in allgroups. (C) Cytokine levels in the serum in all groups were measured on day 7 post-transplant. (D) Reactivity of recipient lymphocytesto the donor or unrelated third-party lymphocytes. Compared with the PBS group, ${}^{*}$p 0.05. Compared with the 105 HSCs-imDCsgroup, ${}^{\mathrm{\#}}$p 0.05. Compared with the 106 HSCs-imDCs group, ${}^{\mathrm{^}}$p 0.05. Compared with the 106 Nar-HSCs-imDCs group, ${}^{\mathrm{†}}$p 0.05. Compared to third-party lymphocytes, ${}^{\mathrm{▲}}$p 0.05. Each experiment contains six biological replicates.