Male C57BL/6J mice (8 weeks old; 20–25g) were provided by HFK (Bioscience Co., Ltd., Beijing, China). All mice were maintained under standard feeding conditions (temperature 23°C; 12h light and 12h dark cycle) and had ad libitum access to food and water. Our research was performed in accordance with the Guidelines for the Care and Use of Laboratory Animals and the Animal Welfare Act in China. All protocols were approved by the Animal Care and Use Committee of Tongji Medical College, Huazhong University of Science and Technology (Permit Number: 2019S2570).

All mice were randomly assigned to the control group (Ctrl, n=6) or the diabetic group (n=18). The diabetic group was induced by a single intraperitoneal injection of 150mg/kg streptozotocin (STZ, Boster Biological Technology, Wuhan) (dissolved in citrate buffer), as previously described.12,32 The control group received an equal volume of citrate buffer. Diabetes was diagnosed if the tail blood glucose concentration of the mice was ≥16.7mmol/L for 3 consecutive days. 4 to 6 weeks after STZ treatment, mouse urine was collected to measure the volume and protein concentration. DKD was verified by urine protein≥30mg/24h. Subsequently, all DKD mice were randomly divided into the DKD group (n=6), DKD+ low-dose losartan treated group (DKD+Los-L, n=6), and DKD+ high-dose losartan treated group (DKD+Los-H, n=6). The DKD+ Los-L group and DKD+ Los-H group were intragastrically administered 10mg/kg/d and 50mg/kg/d losartan, respectively. After 12 weeks treatment,33 urine samples were collected to measure 24-h urinary protein and urinary creatinine (Ucr), and blood samples were collected to measure serum creatinine (Scr) and blood urea nitrogen. Mice were euthanized for tissue harvest.

hrGECs were purchased from Beijing Beina Chuanglian Biotechnology Research Institute. hrGECs were cultured in Delbuco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum and 100U/mL penicillin–streptomycin at 37°C and 5% CO2 in a humidified atmosphere. Cells were divided into different groups as follows: normal glucose group (NG): 5.5mmol/L d-glucose; high glucose group (HG): 40mmol/L d-glucose; HG+1μmol/L losartan group (HG+Los-L); HG+10μmol/L losartan group(HG+Los-H).

FITC-labeled albumin uptake was analyzed as previously described.11,12 hrGECs were incubated with 50μg/ml or 100μg/ml FITC-BSA for 1h or 3h after treatment with different agents in NG or HG conditions. The cells were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X-100/PBS for 20min, and stained with DAPI to reveal the nuclei. After covering with a coverslip, images were taken with a fluorescence microscope (Olympus BX51, Tokyo, Japan). ImageJ was used to analyze the integrated fluorescence intensities.

hrGECs were planted in a 24-well Transwell plate at a suitable density and cultured in a constant temperature incubator at 37°C. Upon reaching 100% confluence, the transendothelial electrical resistance (TEER) value of each well was measured by EVOM2™ (WPI, New Haven, CT, USA) as previously described.12 The measurement was repeated 9–12 times in each well, and the average value was taken. The medium group without cells inoculated with the same glucose concentration was used as the blank group. The integrity of the cell monolayer was proven when TEER reached a plateau. The TEER value of hrGECs reached a plateau after 96h of culture, when the TEER value reached approximately 40Ωcm2.

The determined hrGECs were resuspended in 5.5mmol/L or 40mmol/L d-glucose, then moved to another Transwell 24-well plate and incubated for 12h. Then, hrGECs were incubated with 50μg/ml or 100μg/ml FITC-BSA in the upper chambers of the Transwell, while the medium was mixed with BSA in the lower chambers of the Transwell. The fluorescence signals in the upper and lower chambers of the Transwell were measured by a fluorescence spectrophotometer (Infinite F200PRO; T ecan, Männedorf, Switzerland).

Kidney tissue was cut lengthwise along the long axis, fixed with paraformaldehyde, embedded in paraffin, and cut into 4μm thick slices several times. To evaluate the degree of pathological damage to kidney tissues, the sections were stained with hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS).

For immunohistochemistry staining, kidney sections were incubated in EDTA at 120°C for 5min and blocked with 3% H2O2 for 15min. Then, nonspecific sites were blocked with 5% normal goat serum for 30min at room temperature. The slices were incubated overnight at 4°C with anti-angiopoietin 2 (ab153934; Abcam, Cambridge, MA, USA) and anti-caveolin-1 (CST3238; CST, Danvers, MA, USA) antibodies. Then, the sections were incubated with a biotinylated goat anti-rabbit lgG secondary antibody (Beyotime, Jiangsu, China) for 20min. After counterstaining with 3,3′-diamnobenzidine (DAB, EnVision Detection Kit), slices were dehydrated via an alcohol gradient, and images were observed by a light microscope.

To knockdown CAV1, small interfering RNA (siRNA)-CAV1 (si-CAV1) and a normal control siRNA (si-NC) were designed and synthesized by RiboBio (Guangzhou, China). When hrGECs reached 30–40% confluence, the medium was replaced with Opti-MEM (Gibco, Thermo Fisher Scientific, MA, USA). Subsequently, cells were transfected with these siRNAs using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) diluted in OPTI-MEM. After 4–6h, OPTI-MEM was replaced with normal medium and incubated for 48h with 5.5mmol/L or 40mmol/L d-glucose. Then, the cells were harvested for further experiments.

To knockdown ANGPT2, lentiviruses expressing shRNA targeting ANGPT2 (shANGPT2) and the corresponding control vector (scra) were designed and synthesized by Genechem (Shanghai, China). Cells were transfected according to the manufacturer's instructions. After 48h of transfection, fluorescence microscopy (Nikon, Tokyo, Japan) was used to observe the transduction efficiency.

Kidney tissue and hrGECs were collected and lysed in ice-cold radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime, Jiangsu, China). The protein concentration was measured using a BCA assay kit. Approximately 50μg protein samples were loaded in 8–12% SDS–PAGE gels and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA, USA). After blocking with 5% nonfat dried skimmed milk (Beyotime) for 1h at room temperature, the membranes were incubated overnight at 4°C with antibodies against GAPDH (1:5000, ab8245, Abcam), angiopoietin 2 (1:1000, ab153934; Abcam) and caveolin-1 (1:1000, CST3238; CST). Finally, the membranes were incubated with corresponding secondary antibodies (1:2500; Eric Biotechnology, Wuhan, China) for 1h at room temperature before detection.

All experiments were performed more than three times. The results are expressed as the mean±standard error of the mean (SEM) by GraphPad Prism 7.0 software. Student's t test was used to analyze statistical significance between the two conditions. One-way analysis of variance (ANOVA) was used for multiple group comparisons followed by Bonferroni's post hoc test. Significance was set at P<0.05.

In vitro, we used the albumin transcytosis model to measure the transport of 50μg/ml and 100μg/ml FITC-BSA across the monolayer of hrGECs cultured under normal glucose and high glucose conditions, respectively. As Fig. 1A shows, albumin transcytosis increased with incubation time and the concentration levels of FITC-BSA. The transcytosis of albumin was significantly increased under high glucose stimulation in hrGECs. Losartan administration reduced the transcytosis of albumin across hrGECs under HG conditions in a concentration-dependent manner (Fig. 1B).

Fig. 1.

High glucose increased albumin transcytosis across hrGECs and losartan inhibited this process. (A) Concentration-dependent transcytosis of FITC-albumin at 1 and 3h in NG (5.5mmol/L d-glucose) condition and HG (40mmol/L d-glucose) condition. **P<0.01 vs. 1h NG 50μg/ml group, #P<0.05 vs. 3h NG 50μg/ml or NG 100μg/ml, &P<0.05 vs. 1h HG 50μg/ml or HG 100μg/ml group. (B) Transcytosis of FITC-albumin in hrGECs incubated with NG, HG, HG+Los-L(1μmol/L), and HG+Los-H(10μmol/L). ****P<0.0001 vs. NG group, #P<0.05 or ###P<0.001 vs. HG group. Data are presented as means±SEM.

(0.1MB).

Immunohistochemical staining results showed that the expression of ANGPT2 in the glomerulus was increased significantly in DKD mice (Fig. 2A, B). In vitro, the expression of ANGPT2 was also increased in hrGECs under HG conditions (Fig. 2C, D). Losartan treatment significantly reduced the expression of ANGPT2 in hrGECs under HG stimulation (Fig. 2E, F). These results indicate that ANGPT2 is involved in the pathological process of GECs under high glucose stimulation, and losartan could regulate this process.

Fig. 2.

HG up-regulated ANGPT2 expression and losartan inhibited this process. (A) Representative immunohistochemical (IHC) staining of ANGPT2 in kidney. (B) Statistical data of IHC staining of ANGPT2 (n≥5). **P<0.01 vs. Control. (C) Representative western blot of ANGPT2; (D) Densitometric analysis of ANGPT2 western blot signals (n≥3). *P<0.05 vs. NG. (E) Western blot of ANGPT2 of hrGECs incubated with NG, HG, HG+Los-L, HG+ Los-H. (F) Densitometric analysis of western blot signals of ANGPT2 (n≥3). **P<0.01 vs. NG group, ##P<0.01 vs. HG group. Data are presented as means±SEM.

(0.31MB).

Previous studies have found that CAV1 is closely related to the production of albuminuria in DKD.12,34 Compared with control mice, the expression of CAV1 was increased significantly in DKD mice (Fig. 3A, B). Western blot results showed that the expression of CAV1 increased significantly under HG stimulation (Fig. 3C, D). After high-dose losartan intervebtion, the expression of CAV1 was decreased in hrGECs under HG stimulation (Fig. 3E, F). These findings suggest that losartan may inhibit HG-related kidney injury by regulating CAV1 expression.

Fig. 3.

HG up-regulated CAV1 expression and losartan inhibited this process. (A) Representative immunohistochemical (IHC) staining of CAV1 in kidney. (B) Statistical data of IHC staining of CAV1 (n≥5). ***P<0.001 vs. Control. (C) Representative western blot of CAV1; (D) Densitometric analysis of CAV1 western blot signals (n≥3). *P<0.05 vs. NG, **P<0.01 vs. NG. (E) Western blot of ANGPT2 of hrGECs incubated with NG, HG, HG+Los-L, HG+ Los-H. (F) Densitometric analysis of western blot signals of ANGPT2 (n≥3). *P<0.05 vs. NG group, #P<0.05 vs. HG group. Data are presented as means±SEM.

(0.28MB).

To identify the role of ANGPT2 in albumin transcytosis, hrGECs were transfected with lentiviruses expressing shRNA targeting ANGPT2 to specifically knock down ANGPT2 expression (Fig. 4A, B), which caused a decrease in albumin transcytosis both in NG and HG groups (Fig. 4C). To explore whether CAV1 is involved in HG-related kidney injury by regulating albumin transcytosis, hrGECs were transfected with siCAV1 to specifically knock down CAV1 expression in NG and HG conditions (Fig. 4D, E). Under HG condition, the decrease in CAV1 expression reduced albumin transcytosis in hrGECs. However, there was no significant effect of albumin transcytosis in the NG group (Fig. 4F). Thus, we concluded that ANGPT2 and CAV1 could promote albumin transcytosis of hrGECs during HG exposure.

Fig. 4.

ANGPT2 and CAV1 downregulation inhibited albumin transcytosis in hrGECs. (A) Representative western blot of ANGPT2 with shANGPT2 for 72h. (B) Densitometric analysis of ANGPT2 western blot signals (n≥3). *P<0.05 vs. NG+scra (scramble shRNA), ###P<0.001 vs. NG+shANGPT2, &&P<0.01 vs. HG+ scra. (C) FITC-albumin uptake analysis. **P<0.01 vs NG+ scra, ###P<0.001 vs. NG+shANGPT2, &&P<0.01 vs. HG+ scra. (D) Representative western blot of CAV1 with si-CAV1 for 48h. (E) Densitometric analysis of CAV1 western blot signals (n≥3). *P<0.05 vs. NG+si-NC, **P<0.01 vs. NG+si-NC, #P<0.05 vs. HG+si-NC. (F) FITC-albumin uptake analysis. ***P<0.001 vs NG+ scra, ###P<0.001 vs. HG+si-NC. Data are presented as means±SEM.

(0.22MB).

To further explore the relationship between ANGPT2 and CAV1 in hrGECs, we knocked down ANGPT2 to detect the expression of CAV1 under high glucose conditions. Upon transfection with shANGPT2 to specifically knock down ANGPT2, the expression of CAV1 was reduced (Fig. 5A). Statistical analysis confirmed this result (Fig. 5B). When we knocked down CAV1 in hrGECS under HG conditions, the expression level of ANGPT2 was not significantly changed (Fig. 5C, D). Thus, by summarizing the experimental results, we concluded that ANGPT2 could positively regulate the expression of CAV1. In summary, under HG stimulation, the expression of ANGPT2 in GECs increases, which leads to an increase in albumin transcytosis by promoting the expression of CAV1 and ultimately increasing the excretion of urinary albumin (Fig. 5E).

Fig. 5.

Regulation between ANGPT2 and CAV1 in hrGECs. (A) Western blot of ANGPT2 and CAV1 after transfection with shANGPT2 in HG condition. (B) Densitometric analysis of western blot signals of ANGPT2 and CAV1 (n≥3). **P<0.01 vs. scra. (C) Western blot of ANGPT2 and CAV1 after transfection with si-CAV1 in HG condition. (D) Densitometric analysis of western blot signals of ANGPT2 and CAV1 (n≥3). *P<0.05 vs. si-NC. (E) Summary chart of relationship between ANGPT2 and CAV1 in GECs during HG exposure. Data are presented as means±SEM.

(0.27MB).

Compared with control mice, the levels of blood glucose (Fig. 6A), ratio of kidney weight/body weight (Fig. 6B), and urine albumin/creatinine ratio (Fig. 6D) were obviously higher in DKD mice treated with losartan. There were no detectable changes in the blood glucose of DKD mice treated with losartan compared with DKD mice (Fig. 6A). However, compared with DKD mice, mice administered losartan showed an improved ratio of kidney weight/body weight, serum creatinine and urine albumin/creatinine ratio (Fig. 6B–D). PAS staining revealed that glomerular and tubular basement membrane thickening, mesangial matrix accumulation, and mesangial expansion were reduced by losartan treatment in a dose-dependent manner (Fig. 6E).

Fig. 6.

Losartan inhibited albumin transcytosis by down-regulating ANGPT2 and CAV1 expression in vivo. (A) Blood glucose level (BG). (B) Ratio of the kidney weight/body weight (KBWR). (C) Serum creatinine (Scr). (D) Urine albumin/creatinine ratio (UACR). (E) Representative PAS (periodic acid-Schiff) staining images of kidney. (F) Western blot of ANGPT2 and CAV1 of kidney in Ctrl group, DKD group, DKD+ Los-L(10mg/kg/d) group and DKD+ Los-H(50mg/kg/d) group. (G) Densitometric analysis of western blot signals of ANGPT2 and CAV1 (n≥3). *P<0.05 vs. Ctrl group, #P<0.05 vs. DKD group. Data are presented as means±SEM.

(0.38MB).

Then, we examined the effects of losartan on the expression of ANGPT2 and CAV1. Western blot results showed that the expression of ANGPT2 and CAV1 was increased significantly in DKD mice, which was consistent with the previous results of immunohistochemistry staining. Compared with DKD mice, the expression of ANGPT2 and CAV1 was decreased by losartan treatment in a dose-dependent manner (Fig. 6F, G). Based on the above results, we concluded that losartan reduced albumin transcytosis in GECs during HG exposure by regulating ANGPT2 and CAV1 expression.