Human umbilical vein endothelial cells (HUVECs) were purchased from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in low-glucose Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing 5.5 mmol/l D-glucose, supplemented with L-glutamine, 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin at 37°C in an incubator containing 5% CO₂. Cells were seeded into cell culture dishes and cultured until confluent.

For control and high-glucose treatment, HUVECs were cultured with DMEM containing 5.5 and 30 mmol/l D-glucose, respectively. For drug treatment, the cells were incubated with high-glucose DMEM, and at the same time, Rg1 (>98% pure; DiDa Kexiang Biological Co., Ltd., Guizhou, China) was added to the culture at concentrations ranging from 10⁻⁸ to 10⁻⁵ mol/l for 1 or 3 days.

Total RNA was extracted from HUVECs using TRIzol reagent (Thermo Fisher Scientific, Inc.). RNA was reverse-transcribed into complementary (c)DNA using the RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Inc.). The mRNA levels of heparanase (HPSE) were quantified with an RT-qPCR system (Mastercycler realplex2; Eppendorf, Hamburg, Germany), using SYBR Green SuperMix (Roche Diagnostics, Mannheim, Germany) with appropriate primers pairs. Sequences of primers used in the present study were as follows: HPSE forward, 5′-CCAAAGTTGCTGCTTGCATC-3′ and reverse, 5′-AGTGTCCCAGTGTCTCTCAA-3′; GAPDH forward, 5′-CTGGGCTACACTGAGCACC-3′ and reverse, 5′-AAGTGGTCGTTGAGGGCAATG−3′. The reaction was started by pre-incubation at 95°C for 10 min, followed by 40 cycles of amplification (95°C for 15 sec, 65°C for 15 sec and 72°C for 20 sec). Gene expression levels of HPSE were normalized to those of the reference gene GAPDH measured in the same sample and the results were analyzed by the 2^−∆∆Cq method (28).

Total proteins were prepared using the lysis buffer (20 mmol/l Tris, pH 7.4, 150 mmol/l NaCl, 1 mmol/l EDTA, 1 mmol/lEGTA, 1% Triton X-100, 2.5 mmol/l deoxycholic acid, 1 mmol/l β-glycerophosphate and 1 mmol/l Na₃VO₄), supplemented with protease inhibitors. Protein concentration was determined using a BCA protein assay kit (Beyotime Institute of Biotechnology, Haimen, China) according to the manufacturer's protocol. The protein of each sample (25 µg) was separated using 10% SDS-PAGE and then transferred to polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA). After blocking with 5% fat-free milk for 1 h at room temperature, the membranes were incubated with the following respective primary antibodies: Syndecan-1 monoclonal antibody (1:1,000 dilution; cat. no. ab128936; Abcam, Cambridge, MA, USA), glypican-1 polyclonal antibody (1:1,000 dilution; cat. no. NBP1-33197; Novus Biologicals LLC, Littleton, CO, USA) and GAPDH monoclonal antibody (1:1,000 dilution; cat. no. sc32233; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) at 4°C overnight. The membranes were then gently washed for three times (5 min each) and incubated with horseradish peroxide-conjugated goat anti-rabbit secondary antibody (1:3,000 dilution; cat. no. ZB2301; Zhongshan Goldenbridge Bio, Beijing, China) at room temperature for 1.5 h. After washing, the protein bands were visualized with chemiluminescent substrate (EMD Millipore) for 1 min, and capturing of images and densitometric analysis were performed using an imaging station (Bio-Rad Laboratoris, Inc., Hercules, CA, USA). The relative protein expression levels were normalized to GAPDH and the ratio was compared with that of the control group.

Wheat germ agglutinin (WGA) from Triticum vulgaris binds to sugar moieties of glycocalyx present on the cell surface, the majority of which are likely to be PG constituents of glycocalyx. Therefore, endothelial surface glycocalyx was labeled by fluorescein isothiocyanate (FITC)-conjugated WGA lectin as previously reported (29). Cells were grown to confluence on glass coverslips and fixed in 4% paraformaldehyde for 10 min. After being washed for three times, the cells were incubated with FITC-WGA lectin (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) at 2 µg/ml for 30 min. Coverslips were mounted and visualized using a fluorescence microscope (Leica Microsystems, Wetzlar, Germany).

Cells were seeded onto microporous polyester membranes (0.4-µm pore size) of Transwell filter inserts (Corning Inc., Corning, NY, USA). The cells were grown onto the upper chamber of the Transwell until confluent. The HUVECs were stimulated with vehicle, 30 mmol/l high glucose or Rg1 for 24 h as indicated. The TEER of the monolayer of HUVECS was measured using a Millicell ERS-2 Volt-Ohm meter (EMD Millipore) according to the protocol of a previous study (30). After subtraction of the value determined using a blank, cell-free filter, the mean value of the TEER was expressed in common units (Ωcm²). TEER values of a vehicle-treated monolayer of endothelial cells were designated as baseline values. The percentage of TEER relative to baseline value was calculated via the following formula: TEER%=(TEER of experimental wells/baseline TEER of experimental wells) ×100%.

The transendothelial passage of albumin was analyzed by measuring the passage of FITC-labeled bovine serum albumin (BSA; Sigma-Aldrich; Merck KGaA) across the monolayer as described previously (31). In brief, cells were seeded onto the upper chambers of Transwells and allowed to grow to confluence as described above. The medium in the insert was replaced with serum-free medium (SFM) containing 0.5 mg/ml FITC-labelled BSA and the medium in the well was replaced by SFM only. After 1, 2 and 3 h of incubation, the medium was collected from each well, and the fluorescence of the aliquots was measured on a fluorometer with excitation at 495 nm and emission at 520 nm. The amount of albumin passing the endothelial cell monolayer was calculated with a standard curve generated from a set of FITC-labeled BSA dilutions.

All data were analyzed using SPSS software (version 17.0; SPSS, Inc., Chicago, IL, USA). Values are expressed as the mean ± standard deviation. The independent Student's t-test and one-way analysis of variance (ANOVA) with a least significant difference (LSD) post hoc test, were used for comparison between groups. P<0.05 was considered to indicate a statistically significant difference.

PG core proteins are important constituents of glycocalyx on the cell surface. Expression of PG core proteins syndecan-1 and glypican-1 was analyzed by western blot analysis after exposure to high glucose for different durations (Fig. 2). It was demonstrated that syndecan-1 was gradually decreased in HUVECs incubated with high glucose from 1–4 days. Over the same duration, the expression of glypican-1 was also decreased by high-glucose stimulation, with the changes being significant at 3 and 4 days (Fig. 2A). HUVECs under high-glucose stimulation were then treated with different concentrations of Rg1 for 3 days. It was observed that treatment with Rg1 increased PG core proteins in HUVECs at concentrations ranging from 10⁻⁸ to 10⁻⁵ mol/l and a significant difference was identified at 10⁻⁵ mol/l (Fig. 2B).

To assess the effects of high glucose and Rg1 treatment on HPSE expression, HUVECs were incubated with 30 mmol/l glucose and different concentrations of Rg1 as indicated. The expression of HPSE mRNA was detected by RT-qPCR. It was observed that HPSE mRNA expression in high glucose-treated cells was rapidly increased and reached a peak at 1 day, and then it returned to baseline levels at 3 and 4 days (Fig. 3A). To observe the effect of Rg1, HUVECs were first incubated with high glucose, followed by different concentrations of Rg1 for 1 day. It was indicated that HPSE expression in HUVECs was reduced by treatment with Rg1 at concentrations ranging from 10⁻⁸ to 10⁻⁵ mol/l, and a significant difference was observed at 10⁻⁷ mol (Fig. 3B).

WGA-FITC, which binds to sugar residues on cell surface, was used to quantify the expression of glycocalyx in HUVECs. A marked reduction in WGA-FITC binding was observed in high glucose-induced cells compared with untreated controls, suggesting the disruption of endothelial glycocalyx. As expected, the expression of glycocalyx was increased by treatment with Rg1, indicating that Rg1 prevented the loss of glycocalyx and attenuated the disruption of the glycocalyx in HUVECs incubated with 30 mmol/l glucose (Fig. 4).

TEER is associated with the integrity of cell monolayers. Its decline represents an increase in the passage of water and small molecules across the cell monolayer and an impaired cell barrier function. To study the effects of Rg1 treatment on TEER, HUVECs were stimulated with high glucose in the presence or absence of Rg1. High glucose caused a modest reduction in the mean TEER by up to 20% relative to that in the controls, which was inhibited in the presence of Rg1 (Fig. 5A).

Since 30 mmol/l high glucose leads to decreased TEER, it was examined whether Rg1 was able to preserve the permeability of cultured HUVEC monolayers. The permeability was assessed by measuring the passage of FITC-labeled BSA across the monolayer, i.e., the transendothelial albumin passage. As presented in Fig. 5B, stimulation with 30 mmol/l high glucose significantly increased transendothelial albumin passage across HUVECs, while this effect was inhibited in the presence of Rg1.