Hyperoside Suppresses Renal Inflammation by Regulating Macrophage Polarization in Mice With Type 2 Diabetes Mellitus

Accumulating evidence reveals that both inflammation and lymphocyte dysfunction play a vital role in the development of diabetic nephropathy (DN). Hyperoside (HPS) or quercetin-3-O-galactoside is an active flavonoid glycoside mainly found in the Chinese herbal medicine Tu-Si-Zi. Although HPS has a variety of pharmacological effects, including anti-oxidative and anti-apoptotic activities as well as podocyte-protective effects, its underlying anti-inflammatory mechanisms remain unclear. Herein, we investigated the therapeutic effects of HPS on murine DN and the potential mechanisms responsible for its efficacy. We used C57BLKS/6J Lep db/db mice and a high glucose (HG)-induced bone marrow-derived macrophage (BMDM) polarization system to investigate the potentially protective effects of HPS on DN. Our results showed that HPS markedly reduced diabetes-induced albuminuria and glomerular mesangial matrix expansion, accompanied with a significant improvement of fasting blood glucose level, hyperlipidaemia and body weight. Mechanistically, pretreatment with HPS effectively regulated macrophage polarization by shifting proinflammatory M1 macrophages (F4/80+CD11b+CD86+) to anti-inflammatory M2 ones (F4/80+CD11b+CD206+) in vivo and in bone marrow-derived macrophages (BMDMs) in vitro, resulting in the inhibition of renal proinflammatory macrophage infiltration and the reduction in expression of monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor (TNF-α) and inducible nitric oxide synthase (iNOS) while increasing expression of anti-inflammatory cytokine Arg-1 and CD163/CD206 surface molecules. Unexpectedly, pretreatment with HPS suppressed CD4+ T cell proliferation in a coculture model of IL-4-induced M2 macrophages and splenic CD4+ T cells while promoting their differentiation into CD4+IL-4+ Th2 and CD4+Foxp3+ Treg cells. Taken together, we demonstrate that HPS ameliorates murine DN via promoting macrophage polarization from an M1 to M2 phenotype and CD4+ T cell differentiation into Th2 and Treg populations. Our findings may be implicated for the treatment of DN in clinic.


INTRODUCTION
Diabetic nephropathy (DN) is a serious microvascular complication in patients with diabetes mellitus (DM) and is increasingly regarded as an inflammatory process (1)(2)(3). Macrophages are important cells involved in the initiation of inflammatory responses and play a critical role in the pathogenesis of DN by secreting various proinflammatory mediators. They can differentiate into proinflammatory M1 macrophages through the classical activation and antiinflammatory M2 macrophages via the alternative activation (4). In addition, Th1, Th2, Th17 and cytotoxic T cells are also involved in the development and progression of DN (5,6). Maintaining the M1/M2, Th1/Th2 and Th17/Treg immune balances can reduce the production of proinflammatory cytokines and improve DN (7). Therefore, we hypothesized that the imbalance in both innate (M1/M2) and adaptive immunities (Th1/Th2 and Th17/Treg) could play a crucial role in the pathogenesis of DN, while rebalancing these immune responses might represent a novel approach for the treatment of DN.
As a traditional herbal ingredient, Hyperoside (HPS) is one of the main active components in the Chinese herb Tu-Si-Zi (8) (Supplementary Figure 1), which exhibits the antiinflammatory, anti-oxidant and anti-cancerous properties. Because of its high efficacy and low toxicity, HPS is commonly used in treating a variety of ischemic brain and heart diseases. In recent years, emerging evidence has shown the antiinflammatory effects of HPS on various diseases, such as nonalcoholic fatty liver disease (9), osteoarthritis (10), DM-induced cognitive dysfunction (11) and pulmonary fibrosis (12) in vivo, and on lipopolysaccharide (LPS)-stimulated cell injury in a model of in vitro experiments (13)(14)(15) by suppressing activation of the NF-kB signaling pathway. Studies have also confirmed that HPS can inhibit the high glucose (HG) induced inflammatory injury in vitro and in vivo. For example, it can suppress vascular inflammation caused by HG in the human umbilical vein endothelial cells (HUVECs) in vitro and in mice (16,17). And it alleviates HG-induced apoptosis and inflammatory responses of HK-2 cells through the miR-499a-5p/NRIP1 axis (18). However, very few studies have explored the mechanisms underlying the effects of HPS on renal inflammatory injury in diabetic condition. Thus, the effects HPS on DN and its molecular mechanisms of action remain unclear and warrant further investigation.
In the present study, we aimed to investigate the potentially therapeutic effects of HPS on DN and to delineate the mechanisms underlying the therapeutic effects of HPS in a type-2 DN model of db/db mice and in vitro by focusing on its immunologically regulatory mechanisms responsible for the differentiation and activation of macrophages and CD4 + T cells. We found that HPS indeed attenuated DN in a mouse model of type 2 diabetes mellitus by regulating macrophage polarization.

Antibodies and Reagents
The micro-albumin assay kit was obtained from Abcam Biotechnology (Abcam, Cambridge, UK). Hyperoside with a purity higher than 98% was purchased from Sigma-Aldrich (St Louis, USA) and was suspended in 0.5% carboxymethyl cellulose sodium salt high viscosity (CMCNa) (MP Biomedicals, LLC, USA) solution for animal experiments or dissolved in 0.1% DMSO for cell culture experiments. Glucose was purchased from Sigma-Aldrich (St Louis, MO, USA) while Trizol reagents were manufactured by Invitrogen (California, USA). cDNA Kit was purchased from Promega (Promega Corporation, Madison, WI). 24-well transwell chamber was bought from Corning (Corning, Shenzhen, China). All flow cytometric antibodies, including F4/80, CD11b, CD86, CD206, CD4, IL4, and Foxp3 were purchased from eBioscience or Biolegend. And details of the antibodies used in this study are listed in the Table 1.
A Mouse Model of Diabetic Nephropathy in db/db Mice Male db/db mice (C57BL/KsJ) (19) at the age of 8 weeks were used in this study. The db/db mice and heterozygote agematched db/m mice were originally obtained from Model Animal Research Center of Nanjing University, Jiangsu, China and were maintained following guideline "Principles of Laboratory Animal Care and Use in Research" (Ministry of Health, Beijing, China). Animals were placed in a controlled environment of humidity (about 60%) and room temperature (about 24 ± 1°C) with an alternating 12h light and dark cycle. The animals were allowed free access to standard laboratory tap water and food. The mice were then randomly divided into four groups and treated without or with HPS intraperitoneally for 12 consecutive weeks, as described below: 1) Control db/m mice (n=6) received the same volume of distilling water; 2) Control db/db mice (n=6) were given with the same volume of distilling water; 3) db/db +HHPS mice (n=7) received Hyperoside at a high dose of 80mg/kg/day; 4) db/db +LHPS mice (n=7) were treated with Hyperoside at a low dose of 40mg/kg/day.
Fasting blood glucose levels were measured twice weekly using the Bayer glucose monitor (Bayer, Bergkamen, Germany). Mice were sacrificed 12 weeks after HPS treatment to determine serum concentrations of total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C). Serum and urinary creatinine levels were measured using the enzymatic colorimetric method via an automatic biochemistry analyzer (Toshiba-120FR, Tokyo, Japan). Urinary albumin concentration was measured using the ELISA kit with an anti-mouse albumin antibody (Cusabio, Wuhan, China) and was normalized to the urinary creatinine levels and expressed as urinary mAlb/Cr. UACR was calculated according to the following equation: UACR (mg/g) = urinary albumin(mg)/urinary creatine(g). All animal experiments were approved by the Institutional Animal Care and Use Committee of Guangzhou University of Chinese Medicine.

Renal Pathology
The kidney tissues were fixed in 10% formalin buffer and then embedded in paraffin for light microscopic examination. Serial tissue sections (5mm) were stained with hematoxylin & eosin (HE) and periodic acid-Schiff (PAS). Mesangial matrix expansion was determined by assessing PAS-positive materials in the mesangial region. A percentage of the PAS-positive area was analyzed using Image-Pro Plus (Version 5.1.0.20, Media Cybernetics, Silver Spring, MD, USA) and Leica Q500MC image analysis software. Semi-quantitative analysis was performed with 20 glomeruli randomly selected for each subject (at least five mice in each group) and evaluations were made in coded slides.

Immunohistochemistry
Immunohistochemical staining was performed on paraffin sections using a microwave-based antigen retrieval technique, which involved heat-induced antigen retrieval (HIAR) with sections incubated in ethylenediaminetetraacetic acid (Tris-EDTA) buffer (pH 9.0). The sections were further incubated with the following primary antibodies at 4°C overnight: anti-CD68 (catalog no. ab201340, Abcam) and anti-MCP-1 (catalog no. ab7202, Abcam), followed by the appropriate secondary antibody. The sections were then developed with 3, 3diaminobenzidine (DAB) to produce a brown product and were counterstained with hematoxylin.

Real-Time Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)
Total RNA was isolated from kidney tissues or HG-induced macrophages in vitro using Trizol reagent (Qiagen, San Diego, CA, USA) following the manufacturer's instructions. 1ug RNA was reversely transcribed to cDNA using Oligo(dT) 15 primer and superscript reverse transcriptase (Promega Corporation, Madison, WI). Target gene expression was quantified by realtime PCR using SYBR Green Supermix and the ABI Real-Time PCR Reaction System (Bio-Rad Laboratories). PCR conditions were described as the following: denaturation at 95°C for 3 min, followed by 40 cycles at 95°C for 15 s, 57°C for 30 s, and 72°C for 30 s, with final elongation at 72°C for 10 min. A mouse housekeeping gene b-actin was selected as the internal control. Three independent experiments were performed by calculating the relative mRNA levels using 2-△△Ct methods with values normalized to the reference gene b-actin. The primer sequences (Sangon Biotech, Shanghai, China) that were used are listed in Table 2.

Purification of Splenic T Cells
The purification of CD3 + T cells was carried out by nylon wool columns as described previously (21). Briefly, spleens from 6week male C57/BL6 mice were gently and mechanically disassociated through a 70um cell strainer and lysed with RBC lysis buffer (BOSTER, Wuhan, China) to obtain splenic cells suspension. Autoclaved and sterile nylon fibers (0.5g) were put into a 10mL column, and then the column was equilibrated by 20mL warm RPMI-1640 medium, sealed and incubated at 37°C and 5% CO2 for 1 hour. Spleen cells (1x10^8) were added to the column, sealed and incubated at 37°C and 5% CO2 for 1 hour. Cells subjected to nylon wool purification were resuspended with 2mL warm RPMI-1640 medium, and the column was washed with 2mL warm RPMI-1640 twice. Finally, cells were collected, centrifuged and resuspended for further experiment.

Co-Culture of M2 Macrophages and T Cells
Transwell coculture of mouse M2 macrophages and T cells was performed using a 24-well multi-well chamber and polycarbonate membranes (0.4-um porous). Briefly, T cells were collected and seeded in the upper 24-well transwell plate (5x10^4 cells/well) with 10ng/mL IL-2 (Peprotech, USA) and 1x PMA/ionomycin (Multi-Science Biotechnology, China) in complete RPMI-1640 medium, while M2 macrophages were cultured in the lower 24-well transwell plate (5x10^5 cells/well) with complete RPMI-1640 medium in the absence or presence of different concentrations of HPS for 72 hrs. After co-culture, T cells were collected and centrifuged (400g, 4°C, 5min) for flow cytometric analysis.

Cell Proliferation Assays
T cell proliferation in the coculture was determined using Carboxyfluorescein Diacetate Succinimidyl Ester (CFSE) Cell Proliferation Assay and Tracking Kit (Beyotime, China). For CFSE staining, T cells were isolated, purified and resuspended to 1x10^6 cells/ml in 1x CFSE cell labeling solution at 37°C for 10 min, followed by centrifuging and washing with RPMI-1640 complete medium. After the coculture, CFSE-labeled T cells were collected to perform co-further culture with M2 macrophages for 72 hrs, and finally analyzed using a flow cytometer (Novocyte Quanteon). Data were interpreted as the percentage of proliferated cells.

Flow Cytometric Analysis
Single-cell suspensions from spleens were prepared as the following. Briefly, spleens from the db/m and db/db mice were minced and filtered through 40mm nylon meshes, and then the splenic cells were suspended in PBS and lysed with red blood cell-lysis buffer, followed by centrifuging (400g, 4°C) for 5min. As for the bone marrow-derived macrophages (BMDMs), they were first isolated and cultured as described above.

Statistical Analysis
All data were expressed as mean ± standard deviation (SD) and analyzed using Graphpad Prism 8.0 software (San Diego, CA, USA). One-way ANOVA with a one-tailed Student's t-test was used to identify significant differences in multiple comparisons. The post hoc comparisons using the Student-Newman-Keuls tests were performed for inter-group comparisons of multiple variables. A probability of P < 0.05 was considered to be statistically significant.

Hyperoside (HPS) Reduces Albuminuria and Improves Glycolipid Metabolism Dysfunction in db/db Mice
We first determined the effects of HPS on proteinuria as well as glycolipid metabolism in db/db mice. Compared with control db/ m mice, the levels of urine albumin-creatine ratio (UACR), body weight (BW), fasting blood glucose (FBG), total cholesterol (TC), and low-density lipoprotein-cholesterol (LDL-C) were significantly higher in db/db mice ( Figure 1). Administration of HPS (either 40 or 80mg/kg/day) for 12 weeks resulted in a significant reduction of these indicators in db/db mice ( Figure 1 and Table 3).

HPS Improves Renal Morphological Abnormalities in db/db Mice
Mesangial matrix expansion and glomerular basement membrane thickening were observed in the kidneys of db/db mice and were ameliorated in HPS-treated db/db mice (Figure 2A). The mesangial expansion index and glomerulosclerosis index, indicating the progression of the mesangial changes in DN, were significantly increased in db/db mice compared to db/m control mice. However, HPS treatment for 12 weeks significantly decreased both the mesangial expansion index and glomerulosclerosis index, resulting in a reduction of renal mesangial expansion and extracellular matrix accumulation in db/db mice (Figures 2B, C).

HPS Treatment Reduces Total
Macrophages and Chemokine MCP-1 in db/db Mice To determine an impact of HPS on total macrophage numbers in the kidney of db/db mice, immunohistochemistry (IHC) staining was performed. We found that compared to control db/db mice, administration of HPS remarkedly decreased the number of CD68-positive macrophages in the renal tissue, although they were significantly increased in control db/db mice compared to non-diabetic db/m control mice ( Figures 3A-E). On the other hand, IHC staining also showed that HPS significantly reduced MCP-1 expression in the kidney tissue of db/db mice ( Figures 3F-J), indicating that overall, HPS is antiinflammatory in murine DN.
HPS Alters the Balance of Pro-Inflammatory and Anti-Inflammatory Cytokines in db/db Mice Quantitative real-time PCR analysis was performed to measure the gene expression of the inflammatory cytokines and chemokines in the kidney of db/db mice. We found that HPS treatment decreased the gene expression of the proinflammatory cytokines/chemokines, including iNOS, MCP-1, IFN-g, IL-17 and TNF-a, in the renal tissue, whereas it increased the expression of anti-inflammatory cytokines Arg-1 and IL-10 ( Figure 4). Our results suggest that HPS indeed exerts antiinflammatory effects in a murine model of DN.

HPS Promotes M2 Macrophage Polarization in the Kidney of db/db Mice
Immunofluorescence staining was performed to determine the effects of HPS on M2 macrophage polarization in the kidney of db/db mice since M2 macrophages play an important role in controlling renal inflammation. The results showed that expression of F4/80, a marker of total macrophage population, was increased in the kidney of db/db mice compared to db/m mice while HPS treatment significantly reduced its expression ( Figures 5A, C). On the other hand, the downregulated expression of M2 macrophage marker CD206 in the kidney of db/db mice was markedly increased by HPS (Figures 5B, D). The results indicate that HPS reduces renal inflammatory injury in db/db mice via promoting M2 macrophage polarization.
The decreased percentage of F4/80 + CD11b + CD206 + cells in spleen of db/db mice was significantly increased after HHPS treatment ( Figures 6C, F). We also found a downward trend for the percentage of F4/80 + CD11b + CD86 + cells in db/db mice treated with HHPS. However, there were no statistically

HPS Also Reduces Protein Expression of Proinflammatory Cytokines Associated With CD4+ T-Cells but Increases FoxP3 Expression in the Kidney of db/db Mice
To explore whether HPS restores the balance of Th cell subsets or cytokines, we investigated the biological effects of HPS on the protein expression of IFN-g, IL-17 and FoxP3 in the renal tissue using immunofluorescence (IF) staining. We found that HPS treatment significantly downregulated IFN-g and IL-17 expressions ( Figures 7A, B) but upregulated the expression of FoxP3 in the kidney of db/db mice compared to that of control db/db mice ( Figure 7C). The same results were confirmed when MFI was analyzed statistically (Figures 7D-F), indicating that HPS can alter the balance of Th1/Th17/Treg cells.
Compared with the HG group, however, HG plus HPS groups reduced mRNA levels of M1 macrophage-associated  pro-inflammatory mediators iNOS, TNF-a, IL-17, MCP-1 and IFN-g ( Figures 8A-E), while the mRNA levels of antiinflammatory cytokines Arg-1 and IL-10 ( Figures 8F, G) were increased in the group with high concentrations of HPS. These data suggest that HPS can diminish pro-inflammatory responses stimulated by HG in BMDMs.

Hyperoside Promotes M2 Macrophage Formation in High Glucose (HG)-Induced BMDMs
We then examined the effects of HPS on macrophage polarization in vitro. BMDMs were pretreated with HG to mimic the in vivo diabetic condition. M1/M2 macrophage polarization was determined via flow cytometry and represented as mean fl uorescence intensity (MFI). The percentage of F4/80 + /CD206 + BMDMs was significantly decreased in the presence of HG, but increased after HPS treatment ( Figures 9B, D), indicating that HPS promotes M2 macrophage polarization, which otherwise is inhibited by HG. Meanwhile, we found that the proportion of F4/80 + /CD86 + cells had an upward trend in HG-induced BMDMs and a downward one with a high concentration of HPS treatment (HHPS) ( Figures 9A, C), yet without a statistical significance. These results indicate that the promotion of M2 macrophage formation may be critical for HPS mediated anti-inflammatory effects in BMDMs in the face of HG.

HPS Inhibits T Cell Proliferation and Drives T Cell Differentiation Towards Th2 and Treg Populations in a Co-Culture Model of T Cells/M2 Macrophages
Given that M2 macrophages affect Treg/Th cell differentiation, we examined the effect of HPS on T cell proliferation and Th2/ Treg generation in the coculture of T cells and M2 BMDMs. . The data are presented as the mean ± SD (n = 5 per group, ## P < 0.01 vs db/m; *P < 0.05 and **P < 0.01 vs db/db, with each symbol color representing a particular group).
As shown in the Figure 10, T-cell proliferation was inhibited by addition of M2 macrophages to the coculture (MT) as compared with T cell culture alone (T) without M2 macrophages ( Figures 10A, D), while compared with the coculture group of M2 plus T cells (MT), HPS showed more effective inhibition of T cell proliferation (Figures 10A, D). Hence, these results indicate that M2 macrophages suppress T cell proliferation and that HPS plus M2 macrophages can further enhance their suppression of T-cell proliferation. Furthermore, we asked whether HPS also affected the differentiation of T cells co-cultured with M2 macrophages. As shown in Figure 10, more T cells co-cultured with M2 macrophages expressed Th2 cytokine IL-4 (CD4 + IL4 + ) ( Figures 10B, E) and FoxP3 (CD4 + Foxp3 + ) ( Figures 10C, F) than T cells without M2 macrophages. Interestingly, compared with the M2 and T cell coculture group (MT) alone, the groups treated with HPS (MT+LHPS or +HHPS) showed further upregulation of the frequency of Th2 (CD4 + IL4 + ) ( Figures 10B, E) and CD4 + Foxp3 + Treg cells (Figures 10C, F). , and anti-inflammatory cytokines Arg-1 (F) and IL-10 (G) in the kidney tissue in four groups following HPS treatment: db/m, db/db, db/db +LHPS, and db/db +HHPS mice. Relative mRNA expression levels were normalized to b-actin. Data are representatives of three independent experiments and presented as means ± SD (n = 6 per group, # P < 0.05 and ## P < 0.01 vs db/m; *P < 0.05 and **P < 0.01 vs db/db; ns, not significant).
These results indicate that HPS may enhance the ability of M2 macrophages to suppress inflammation by promoting T cell differentiation into Th2/Treg subsets.

DISCUSSION
Proteinuria or albuminuria is a clinical risk of the onset and development of diabetic nephropathy (DN) (22,23). The results of this present study showed that Hyperoside (HPS) dramatically attenuated albuminuria, which is usually accompanied by dyslipidemia and obesity in db/db mice. We also investigated the mechanisms underlying effects of HPS on DN and revealed that HPS regulated inflammation by promoting M2 macrophage polarization in addition to shifting Th cell balance towards Treg and Th2 cells, thus attenuating the pathogenesis of DN. This study provides a rationale for developing HPS as a drug for the treatment of DN.
In this study, a significantly higher urine albumin-tocreatinine ratio (UACR), hyperglycemia, and an increase in body weight, LDL-C and TC in db/db mice suggested that the animals developed DN. When compared with db/db mice, significant decreases in UACR were observed in HPS-treated db/db mice, suggesting that HPS ameliorated proteinuria in diabetic mice. In addition, the levels of fast blood glucose, LDL-C, TC, and body weights were significantly reduced after treatment with HPS, indicating that it exerted an effect on hyperglycemia and hyperlipidemia. However, the parameters of the renal function, such as Scr and BUN, were not significantly improved by HPS. The effects of HPS on hyperglycemia, hyperlipidemia and renal function index is not totally consistent with the report by Zhang J et al (24), which showed that HPS decreased the Scr in DN mice, although it did not affect the glucose and lipid metabolism. This discrepancy may be due to the differences in animal models used in these two studies. The db/db mice used in our study have a background of C57BLKS/J strain, which is a widely used mouse model for type 2 diabetes and an approved model of albuminuria associated with DN, and the renal function in these diabetic mice declines at 15-18 weeks without a significant change of serum biochemical indicators, whereas Zhang et al. used low-dose of STZ to induce type 1 diabetes in rats, resulting in obvious deterioration in serum renal function indicators. In the present study, we confirmed the potential protective effects of HPS on glomerulosclerosis in DN. When db/db mice were treated with HPS, their histopathology was remarkably improved, which was still polarization in DN progression has been paid more attention (27,28). M1 macrophages produce large amounts of proinflammatory cytokines iNOS, TNF-a, MCP-1, and other proinflammatory mediators that amplify inflammation, resulting in further damages during DN pathogenesis (29,30). On the other hand, M2 macrophages inhibit renal inflammation and ameliorate injury by secreting anti-inflammatory cytokines, such as IL-10 and Arg-1 (31,32). Accumulating evidence has suggested that the levels of proinflammatory factors (proinflammatory cytokines and chemokines) increase with the development of DN and are independently associated with urinary albumin excretion in DN (33,34). Proinflammatory cytokines that are synthesized and secreted by macrophages in the local microenvironment can directly damage the renal architecture and then trigger the renal fibrosis (35). Therefore, regulation of M1/M2 macrophage phenotypes exhibits anti-proteinuric and renoprotective effects on DN progression (36)(37)(38). However, there have been few studies exploring drugs that can regulate macrophage polarization in DN. In this study, we demonstrated that HPS effectively regulated macrophage polarization by shifting proinflammatory M1 macrophages to M2 ones, resulting in the inhibition of proinflammatory macrophage infiltration in the kidney, and thus altered the balance of pro-inflammatory and anti-inflammatory cytokines in db/db mice and in BMDMs in vitro. Taken together, our results indicated that the antiinflammatory effects of HPS via the regulation of M1/M2 macrophage polarization may be critical for the direct attenuation of proteinuria and improvement of renal tissue damage.
As reported previously, the inflammation and its progression result from not only innate immune responses dominated by macrophage-mediated effects, but also the adaptive immune responsiveness mediated by lymphocytes. Consistently, in addition to the regulation of macrophage polarization, our study showed that HPS also altered the balance of Th1/Th17/ Treg cells. Th1 and Th17 cells have been positively associated with renal damages in DN (39). The hallmark of Th1 and Th17 cells is the production of two cytokines interferon g (IFNg) and interleukin (IL)-17, which are abundant in diabetic kidneys and play important roles in the development and progression of inflammatory injury in DN (40,41). Targeting Th17 cells by mycophenolate mofetil or IL-17A neutralizing antibody could attenuate albuminuria and tubulointerstitial fibrosis in mice with DN (6,42). On the other hand, Treg cells expressing a specific transcription factor forkhead box P3 (FoxP3) have been implicated in the inhibition of DN progression by suppressing the activation of effector T-cells and exerting anti-inflammatory activity (43). Depletion of Tregs exacerbated diabetic-associated renal injury in db/db mice, whereas the adoptive transfer of Tregs or induction of Tregs had the opposite effect (44)(45)(46)(47)(48). Studies have reported that Th17/Treg imbalance contributed to the development and progression of DN (41,45,49), and reversing the imbalance by Dapagliflozin attenuated albuminuria and , and anti-inflammatory cytokines Arg-1 (F) and IL-10 (G) were shown. Values are presented as means ± SD with n = 5 per group. Differences between experimental groups were evaluated using ANOVA ( # P < 0.05 and ## P < 0.01 vs Control; *P < 0.05 and **P < 0.01 vs HG; ns, not significant). HG, high glucose treated group; HG+LHPS, high glucose and low dose of HPS treatment group; HG+MHPS, high glucose and middle dose of HPS treatment group; HG+HHPS, high glucose and high dose of HPS treatment group.
tubulointerstitial fibrosis independently of glycemic control in db/db mice (50). Therefore, the renoprotective effects of HPS in this study may be also associated with its reversal of Th1/Th17/ Treg imbalance. Interaction of renal tissue macrophages with T cells produces various reactive oxygen species, proinflammatory cytokines, metalloproteinases and growth factors, which in turn enhance the local immune responses and increase inflammation within the kidney in DN (7,51,52). Given that HPS remarkedly promoted M2 macrophage polarization in HG-induced BMDMs, we performed co-culture of M2 macrophages and T cells and found that HPS could enhance the ability of M2 macrophages to promote T cell differentiation into Treg and Th2 subsets. Regulation of T-cell proliferation and differentiation by macrophages are well documented in various disease settings, including DN (29,53,54). Although the mechanisms of crosstalk between these cells are not clear, it has been reported that M2polarized macrophages can produce Th2-type and antiinflammatory cytokines that in turn inhibit T-cell proliferation (55). Therefore, the renoprotective effects of HPS may be attributed to its contribution to the shift of macrophage polarization towards an anti-inflammatory M2 phenotype, which then modulate Th1/Th2 or Th17/Treg balance and thus suppress renal inflammation. BMDMs were treated with high glucose (35mM) for 48 hrs, followed by pretreatment with high-dose HPS (HHPS) (50uM) or low-dose HPS (LHPS) (12.5uM), and then collected for flow cytometric analysis. The results showed that there was no significant difference in the proportion of F4/80 + /CD86 + macrophages between groups with or without Hyperoside treatment (A, C). However, the proportion of F4/80 + /CD206 + macrophages was decreased in BMDMs with high glucose but increased after Hyperoside treatment (B, D). Differences between experimental groups were evaluated using ANOVA ( # P < 0.05 vs Control; *P < 0.01 vs HG; ns, not significant). HG, high glucose treated group; HG+LHPS, high glucose and low dose of HPS treatment group; HG+HHPS, high glucose and high dose of HPS treatment group.
In conclusion, we demonstrated that HPS ameliorates renal inflammatory injury in DN via promoting macrophage polarization from an M1 to M2 phenotype and CD4 + T cell differentiation into Th2 and Treg populations. Moreover, HPS may enhance the ability of M2 macrophages to suppress inflammation by indirectly promoting T cell differentiation into Th2/Treg subsets. Our findings may be implicated for the treatment of DN in clinic. Further studies on how HPS modulates macrophage polarization are warranted.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. For proliferation assays, T cells were first labeled with CFSE before the co-culture, and the CFSE dilution of CFSE-labeled CD4 + T cells was detected by flow cytometry. Flow cytometric analysis showed T-cell proliferation was inhibited by addition of M2 macrophages to the coculture system, and the inhibition was further enhanced by HPS. (B, E) The percentage of IL4-expressing CD4 + T cells was increased by HPS treatment compared with a control group (MT). (C, F) The percentage of CD4 + Foxp3 + T cell subsets was also increased by HPS treatment compared with a control group (MT). Differences between experimental groups were evaluated using ANOVA ( # P < 0.05 and ## P < 0.01 vs T group; *P < 0.05 and **P < 0.01 vs MT group). T, T cells alone; MT, T cells cocultured with M2 macrophages; MT+LHPS, T cells cocultured with M2 macrophages treated with low-dose HPS (12.5 mM); MT+HHPS, T cells cocultured with M2 macrophages treated with high-dose HPS (50 mM).