1. Secretory ability of ADSCs for anti-inflammation-related factors was enhanced following diabetic microenvironment preconditioning
The characteristics of ADSCs were evaluated under diabetic microenvironment preconditioning (LPS, 1 µg/ml; AGE, 1 µg/ml; and glucose, 2.5 mg/ml; for 24 h). FACS analysis showed that diabetic microenvironment pre-ADSCs were negative for CD34, CD11a, and HLA-DR and positive for CD90, CD73, and CD105. In addition, the pre-ADSCs were found to have a spindle-shaped morphology when observed under an optical microscope (Fig. 1A). Moreover, pre-ADSCs showed the ability to differentiate into adipocytes and osteoblasts (Fig. 1B), and the apoptotic rate of the ADSCs did not increase significantly after preconditioning with the diabetic microenvironment as observed by Annexin V/7-AAD staining (Fig. 1C). These results showed that the diabetic microenvironment pre-ADSCs had similar characteristics to the original ADSCs. Previous studies confirmed that IL-6, monocyte chemoattractant protein-1 (MCP-1), IL-10, TGF-β, TNF-α, and other cytokines secreted by MSCs are closely related to their immunoregulatory effects. Moreover, our previous study showed that IL-6 and MCP-1 are responsible for MSC-induced M2 macrophage polarisation [21]. ADSCs were pretreated with different concentrations of the diabetic microenvironment at different times in our study, and the secretion of IL-6, MCP-1, IL-10, TGF-β, and TNF-α was detected at the transcription and protein levels. According to the results of qRT-PCR, the transcription levels of IL-6 and MCP-1 were highest when ADSCs were pretreated in a low-concentration diabetic microenvironment for 24 h (Fig. 1D). The results of ELISA, which was used to detect the cytokines secreted by cells, were consistent with those of qRT-PCR (Fig. 1E). Therefore, pretreatment with low-concentration diabetic microenvironment for 24 h was selected as the optimal pretreatment condition.
2. Compared to ADSCs, co-culture with pre-ADSCs further promoted M2 macrophage polarisation in vitro
Next, we sought to determine whether pre-ADSCs would further promote the transformation of macrophages in vitro. Cell-climbing immunofluorescence and PCR were used to detect M1 and M2-related markers (M1: iNOS; M12: Arg1). Macrophages were extracted from peritoneal lavage, and immunofluorescence staining showed that more than 90% of the cells were F4/80 positive (Fig. S1A). LPS stimulation markedly increased the ratio of iNOS+ (M1 macrophage marker) cells to 65.5%, which was reduced to 27.3% by culturing with ADSCs in a transwell system and further reduced to 11.8% following pre-ADSC co-culture (Fig. 2A-C). Additionally, qRT-PCR analysis revealed that compared to the ADSC group, pre-ADSC treatment resulted in higher expression of genes encoding M2 macrophages and anti-inflammatory molecules (Arg1, CD206, CD163, and IL-10) and lower expression of genes encoding M1 macrophages (iNOS, TNF-α, IL-1β) and pro-inflammatory molecules (Fig. 2D). In summary, these results suggest that the ability of ADSCs to promote the polarisation of macrophages from M1 to M2 was significantly enhanced following diabetic microenvironment preconditioning in vitro.
3. Compared to ADSCs, multiple pre-ADSC infusions further improved glucose homeostasis by ameliorating insulin resistance and enhancing recovery of pancreatic islets
To examine the effect of pre-ADSCs on glucose homeostasis and long-term T2D complications, we used an HFD diet combined with a one-time injection of low-dose STZ to induce long-term T2D complications in a rat model, as we previously reported (Fig. S1B). In brief, eight-week-old male SD rats were fed an HFD for 8 weeks. A single dose of STZ (25 mg/kg) was intraperitoneally injected. Rats with more than three random glucose level measurements ≥ 16.7 mmol/L were considered diabetic. Then, the T2D rats were fed a HFD for 24 weeks. The rats were divided into four groups: normal group (Nor group), diabetic untreated group (T2D group), normal ADSC treatment group (ADSCs group), and diabetic microenvironment-preconditioned ADSC treatment group (pre-ADSCs group). Late-stage T2D rats (ADSCs and pre-ADSCs groups) were infused with ADSCs or pre-ADSCs once a week through the tail vein for a total of 6 months.
Approximately two months after ADSC infusion, the random blood glucose levels were found to gradually decrease and reached 15.6 ± 1.3 mmol/L at the end of treatment, whereas six-month infusion of pre-ADSCs resulted in nearly normal random blood glucose level (9.8 ± 1.7 mmol/L) (Fig. 3A). Compared to ADSC treatment, IPGTT showed much more improved glucose clearance after pre-ADSC treatment (Fig. 3B). Insulin sensitivity was enhanced following ADSC infusion, as indicated by improvements in IPITT, homeostatic model assessment of insulin resistance (HOMA-IR), and GIR, which were further improved in the pre-ADSC group (Fig. 3C-E). Moreover, compared to the ADSC group, multiple infusions of pre-ADSCs significantly promoted the regeneration of pancreatic β-cells (based on islet numbers) and the proportion of β-cells/islet (Fig. 3F-H). Overall, these results show that compared to ADSCs, multiple pre-ADSC infusions further improved glucose homeostasis by ameliorating insulin resistance and enhancing the recovery of pancreatic islets.
4. Compared to ADSCs, multiple pre-ADSC infusions in long-term T2D complication rats further ameliorated T2D-induced kidney diseases
DN is one of the most serious and chronic vascular complications of T2D. Therefore, we evaluated the effects of ADSCs and pre-ADSCs on DN progression. Kidney dysfunction was evident in the T2D rats. Serum creatinine levels were almost doubled (Fig. 4A), the ACR reached 830.2 µg/mmol, which was more than ten-fold higher than that in the normal group (Fig. 4B), and blood urea nitrogen (BUN) reached 25.2 mg/dL (Fig. 4C). All indexes in the pre-ADSC group were further reduced compared to those in the ADSC group. Tubulointerstitial fibrosis, glomerulosclerotic changes, hypertrophy of glomeruli, swelling of tubule cells, and infiltration in T2D rats were obvious with H&E, PAS, and Masson’s trichrome staining. As expected, all kidney damage was significantly attenuated following pre-ADSC treatment compared to that in the ADSC group (Fig. 4D-H). Positive expression of tissue fibrosis markers including collagen type 1 (col1) and alpha smooth muscle actin (α-SMA) was readily observed in the T2D group. The infusion of ADSCs slightly decreased the ratio of α-SMA + area to 64.6% in the kidney, whereas pre-ADSCs substantially reversed this trend with the ratio reducing to 30% (Fig. 4I-J). Treatment with ADSCs also decreased the ratio of collagen I + area, and pre-ADSCs showed a tendency to further reduce this index, although the differences were not statistically significant. Preconditioning clearly improves the protective effects of ADSCs on the kidney.
5. Compared to ADSCs, multiple pre-ADSC infusions in long-term T2D complication rats further ameliorated T2D-induced liver diseases
It is well known that liver fibrosis and fat accumulation contribute to the development of advanced T2D. Therefore, we examined the serum indicators of lipid metabolism and changes in hepatic histopathology. H&E, Sirius Red, and Masson’s trichrome staining revealed that the incidence of liver steatosis, fibrosis features, and inflammation in the pre-ADSC group displayed greater improvement than that in the ADSC group (Fig. 5A, D). The non-alcoholic steatohepatitis (NASH) score, which is used to describe NASH-like features (such as inflammation and steatosis), was decreased by up to 30.52% in the pre-ADSC group compared to the ADSC group (Fig. 5C). The liver partial lipid metabolism disorder levels (including ALP and ALT) in the pre-ADSC group were significantly improved compared to those in the ADSC-treated group (Fig. 5B). Moreover, to further determine whether pre-ADSC treatment could significantly ameliorate liver fibrosis, we detected the expression of profibrogenic genes such as tissue inhibitor of metalloproteinases-1 (TIMP-1), matrix metalloproteinase (MMP-8 and 9), col1, and α-SMA. All these parameters were significantly decreased in the pre-ADSC group compared to those in the other groups. Although the statistical difference was not significant, MMP-2 and col3 levels were slightly decreased in the pre-ADSC group compared to the ADSC group (Fig. 5E, F). These results confirmed the advantages of pre-ADSC infusions in mitigating liver damage caused by long-term T2D compared to ADSCs.
6. Compared to ADSCs, multiple pre-ADSC infusions in long-term T2D complication rats further ameliorated T2D-induced cataract and lung diseases
A recent study demonstrated obvious structural and physiological abnormalities of the lung in patients with T2D. Therefore, we next evaluated the lungs in the various groups to detect the effect of pre-ADSCs [19]. Compared to the ADSC group, histopathological damage, including infiltration of inflammatory cells, disordered structure of the lung tissue, and alveolar thickness was mitigated effectively in the pre-ADSC group (Fig. 6A), The comprehensive score of lung injury was significantly improved following pre-ADSC treatment (Fig. 6B). In addition, the amelioration of glycogen granules and pulmonary fibrosis were more significant in the pre-ADSC group than in the ADSC group as detected by the percentage of positive PAS-stained areas and Masson-stained areas of collagen (Fig. 6C, D). These findings were consistent with the results of immunohistochemical staining of α-SMA and detection of the expression of related genes including MMP-2, 8 and 9, col3, TIMP-1, and α-SMA (Fig. 6E-H). All these results confirmed that compared to the ADSC group, the total amount of collagen in the alveolar space was reduced following treatment with pre-ADSCs.
Another common complication of diabetes is cataract. After treatment, both ADSC and pre-ADSC groups showed an improvement in the degree of turbidity of the lenses, but the pre-ADSC group showed significantly improved transparent and clearer lenses (Fig. 6I). Furthermore, the anti-cataract effect of pre-ADSC treatment was confirmed by H&E staining (Fig. 6J).
7. Compared to ADSCs, multiple pre-ADSC infusions further increased M2 phenotype macrophage polarisation and attenuated tissue inflammation
Next, we analysed whether multiple pre-ADSC infusions further promoted M2 macrophage polarisation and attenuated inflammation in vivo. We analysed the phenotype changes in macrophage in the pancreas, kidney, liver, lung, and adipose tissues of T2D rats. The brown dark-stained cells represent Arg1-positive cells, which represent M2 phenotype macrophages. The proportion of M2 infiltration in each tissue from the pre-ADSC treatment group was significantly higher than that from the T2D and ADSC groups (Fig. 7A-B). Likewise, in the pre-ADSC group, the genes encoding CD163, CD206, and Arg1 (M2 macrophage markers) were highly expressed, and the expression of iNOS (M1 macrophage marker) was lower (Fig. 7C-E). Additionally, qRT-PCR analysis of tissues showed that the expression of genes related to anti-inflammation (IL-10) was enhanced whereas those related to fibrosis and inflammation (TGF-β, TNF-α, IL-1β) was decreased in the pre-ADSC group.