Netrin-1 attenuates the progression of renal dysfunction by blocking endothelial-to-mesenchymal transition in the 5/6 nephrectomy rat model

A netrin-1-expressing adenovirus was created using the AdEasy Vector system (Qbiogene, Nottingham, UK) as previously described [17] . The PacI-linearized recombinant plasmid was then transfected into HEK293 cells (ATCC, Manassas, VA, USA). Adenovirus titers were determined by a plaque-forming assay, and expressed as the number of plaque-forming units (PFU). Virus stocks were amplified by culturing HEK293 cells with low-passage virus stocks, and amplification was continued until the titer reached 10¹⁰ PFU/ml. Aliquots of recombinant adenoviruses were then frozen at−80 °C until further use.

All animal procedures were conducted according to the animal care and ethics laws and were approved by the Animal Care Committee of the General Hospital of Shenyang Military Area Command. The 5/6 Nx rat model of chronic renal failure was established according to the method published by Ghosh et al. [18]. Thirty male Sprague–Dawley rats (8–10 weeks old, weighing 250–300 g) were obtained from the General Hospital of Shenyang Military Area Command Animal Centre (Shenyang, China), and maintained at 22 ± 1 °C, with relative air humidity at 60 ± 5 % and with free access to food and water. They were then randomly divided into one of three groups (n = 10 rats/group): sham-operated rats treated with the control adenovirus (empty vector); 5/6 Nx rats treated with the control adenovirus (empty vector); and 5/6 Nx rats treated with the netrin-1-expressing recombinant adenovirus (Ad-netrin-1). The rats in the sham-operated group were anesthetized by intraperitoneal injection of 10 % chloral hydrate (3 ml/kg) followed by surgical exposure and removal of the renal capsule. The 5/6 Nx rats first underwent a partial left nephrectomy (removal of 1/3 of the superior pole and 1/3 of the inferior pole) under anesthesia. The animals were then returned to the vivarium to recover. One week later, they underwent a total right nephrectomy, thus accomplishing a 5/6 nephrectomy. During the second surgery, the 5/6 Nx rats were injected with 1 × 10⁸ PFU of control adenovirus (empty vector) or Ad-netrin-1 via the vena caudalis as reported in previous studies [19]. All animals were sacrificed at the 13th week of the experiment, by an intraperitoneal injection of sodium pentobarbital (50 mg/kg). Blood and kidney tissues were immediately collected and kept at−80 °C for further analysis.

Blood samples were collected from all animals at weeks 12 after surgery, for analysis of blood urea nitrogen (BUN) and serum creatinine (Scr). BUN and Scr concentrations were assayed with an Olympus 400 clinical chemistry analyzer.

Renal tissues were fixed in 4 % buffered paraformaldehyde and then embedded in paraffin. Then, 4-μm thick sections were generated and stained with hematoxylin and eosin (H&E) and Masson’s trichrome. The H&E-stained slices were examined by optical microscopy for pathological changes, and the slides stained with Masson’s trichrome were analyzed for interstitial fibrosis as previously described [20]. To determine the percentage of interstitial space in the kidney, five consecutive fields were randomly selected in the renal cortex and evaluated under a 400× lens on a 10 × 10 grid-imprinted reticule. All points not counted within tubular cells, lumen, glomerulus, or vascular spaces were considered as interstitial points. This fraction represented the relative interstitial volume. Results were expressed as percentages of the measured area, representing the interstitial space and being the relative volume of the interstitium.

Frozen tissues were cut into 4-μm thick sections and fixed in 100 % acetone at−20 °C for 10 min. After washing with phosphate-buffered saline and blocking with 5 % bovine serum albumin (BSA) for 1 h, the tissues were incubated with primary antibodies specific for CD31 (Santa Cruz Biotechnology, Europe) and α-SMA (Abcam, England) at 4 °C overnight. The tissues were then incubated with a mixture of two secondary antibodies that had been raised in different species and conjugated to different fluorochromes (fluorescein isothiocyanate-conjugated goat anti-mouse and Cy3-conjugated goat anti-rabbit), and were incubated in 1 % BSA for 1 h at room temperature in the dark. For the negative control experiments, the primary antibody was replaced with non-immune IgG, which provided minimal labeling. Images were captured using an LSM5 Image Browser (Zeiss) and analyzed using a laser scanning confocal microscope (LSM 510 META, Zeiss). The areas showing CD31 and α-SMA immunoreaction were measured, and fluorescence intensity was analyzed as previously described [21].

Frozen tissue samples were lysed in TRIzol reagent (Invitrogen, Carlsbad, CA), and total RNA was extracted according to the manufacturer’s instructions. The levels of transcripts were determined by real-time PCR (RT-PCR) using SYBR ®Premix Ex Taq ™ II (TaKaRa, Shiga, Japan) on an ABI 7901HT series PCR cycler (Applied Biosystems, Foster City, CA, USA). The data were normalized to β-actin expression and further normalized to the negative control. Primers were obtained from Sangon Biological Engineering Technology and Services (Shanghai, China), and the following specific primers were designed: β-actin, 5′-CCCACTCTTCCACCT TTG-3′ (sense) and 5′-TAGCCATATTCATTGTCATACC-3′ (antisense); CD31, 5′-CTCCATCCTGTCGGGTAA-3′ (sense) and 5′-TCATTCACGGTTTCTTCG-3′ (antisense); α-SMA, 5′- GGCAACCTCAAGAAGTCCC−3′ (sense) and 5′- GTGCAGCCATCCACAAGC−3′ (antisense); Snail,5′-GGCTGATGGAAGGCAGAG−3′ (sense) and 5′- CCAGTGGGTTGGCTTTAGTT−3′ (antisense).

Total protein was extracted from tissue samples with a commercial kit (GE healthcare Biosciences, Pittsburg, PA, USA). The protein concentrations in the lysates were assayed by the BCA assay using commercial kits (Invitrogen). According to the assay results, the lysates were adjusted to a protein concentration of 1 g/l and kept in Eppendorf tubes (35 μl/tube) at−70 °C before subsequent use. The protein (40 μg) obtained from each lysate was electrophoresed in a gel containing 8 % sodium dodecyl sulfate. The separated fractions were transferred to a nitrocellulose membrane, treated with skimmed milk for 2 h to block non-specific antibody-binding sites, rinsed with Tris-buffered saline (TBS), and incubated overnight at 4 °C with primary antibodies specific to Snail (1:1000; Cell Signaling Technology, Beverly, Massachusetts), CD31 (1:1000; Santa Cruz Biotechnology, Europe) and α-SMA (1:500; Abcam, England). After rinsing with TBS (10 min, three times each), the membrane was incubated with a horseradish peroxidase-labeled secondary antibody (1:5000; Sigma-Aldrich Co.) at room temperature for 1 h, rinsed with TBST (Tris-buffered saline with 10 % Tween-20) and developed using the enhanced chemiluminescence kit (ECL, GE Healthcare Biosciences). Equal protein loading was confirmed by immunostaining against β-actin (1:4000; Sigma-Aldrich Co). The membrane was imaged and analyzed with a Multimage Light Cabinet (Alpha Innotech Corp., San Leandro, California).

Transforming growth factor (TGF)-β serum levels were determined by ELISA (R&D Systems) according to the manufacturer’s instructions. ELISA results were read on a Fusion from PerkinElmer (Boston, MA).

Our previous studies showed that the 5/6 Nx group had substantially reduced netrin-1 expression compared with the sham-operated group, and that netrin-1 was not detectable in the interstitial space. The Ad-netrin-1-treated 5/6 Nx group showed substantially increased netrin-1 expression compared with the 5/6 Nx group [17].

To determine the role of netrin-1 in CKD, renal function was determined by measuring BUN and Scr in all three groups of rats. As expected, the 5/6 Nx rats (n = 10) had significant elevations of Scr and BUN. The BUN and Scr levels were significantly increased in the 5/6 Nx groups compared with the sham-operated group (n = 10) at 7 weeks and 12 weeks; however, Ad-netrin-1 treatment (n = 10) had a significantly beneficial effect on reducing Scr and BUN levels (Fig. 1, P < 0.05). In H&E-stained images of kidney sections, the kidneys of the sham-operated rats showed no significant pathologic changes (Fig. 2, n = 10). However, focal interstitial fibrosis with tubular dilatation and shrinkage was observed in the vehicle-treated 5/6 Nx group. The pathological changes were alleviated in the Ad-netrin-1-treated 5/6 Nx group. Masson staining showed that the area of interstitial fibrosis in the vehicle-treated 5/6 Nx group was significantly greater than that in the sham-operated group. A smaller interstitial space was observed in the Ad-netrin-1-treated 5/6 Nx group. These data suggest that netrin-1 protected renal function in the 5/6 Nx rats.

Previous studies have shown that EndoMT contributes to the accumulation of fibroblasts in tumors [22]. Here, we tested the hypothesis that EndoMT occurs in 5/6 Nx rat.

Renal tissues from the sacrificed rats were double-labeled using antibodies against the endothelial marker CD31 and the fibroblast marker α-SMA, which are indicators of EndoMT. Confocal microscopy revealed colocalization of CD31 and α-SMA (Fig. 3). Further, we observed that CD31 immunoreactivity decreased and some CD31-positive vessels acquired α-SMA staining in the vehicle-treated 5/6 Nx rats compared with sham-operated rats. α-SMA positive fibroblast populations were substantially increased in the 5/6 Nx rats compared with the sham-operated controls, which suggests that EndoMT had occurred in 5/6 Nx rats. The administration of Ad-netrin-1 markedly reduced the number of CD31+/α-SMA+ double-labeled cells in the Ad-netrin-1-treated 5/6 Nx rats. EndoMT involves a progressive loss of the endothelial cell marker CD31 associated with a reciprocal gain of the mesenchymal marker α-SMA. These findings suggest that EndoMT may account for a considerable portion of the fibroblasts in 5/6 Nx rats, and that EndoMT can be attenuated by netin-1.

Next we investigated the expression levels of CD31 and α-SMA in renal tissues by Western blotting and real-time PCR. Mesenchymal mRNA and protein expression profiles showed increased expression of α-SMA and decreased expression of the endothelial marker CD31 in the vehicle-treated 5/6 Nx group compared with the sham-operated group (Fig. 4, P < 0.05). However, α-SMA expression was markedly decreased and CD31 expression was increased in the Ad-netrin-1-treated 5/6 Nx group compared with the vehicle-treated 5/6 Nx group. There are a number of key factors required for EndoMT. These include Snail and E-cadherin. Snail is a master regulator of EndoMT and can repress E-cadherin expression [23]. Snail levels were up-regulated in the 5/6 Nx rats. This was down-regulated following administration of netrin-1. These findings indicate disruption of EndoMT by Netrin-1 in 5/6 Nx rats.

TGF-β is a potent inducer of endothelial to mesenchymal transitions. In addition, it is important in EndoMT. We investigated the expression levels of TGF-β by ELISA. The serum of the vehicle-treated 5/6 Nx group showed stronger TGF-β expression compared to the sham-operated group; however, TGF-β expression in the Ad-netrin-1-treated 5/6 Nx group was significantly decreased compared with the vehicle-treated 5/6 Nx group (Fig. 5, P < 0.05). These findings indicate that Netrin-1 repressed TGF-β expression in the 5/6 Nx rats.

All animal procedures were conducted according to the animal care and ethics laws and were approved by the Animal Care Committee of the General Hospital of Shenyang Military Area Command.

Not applicable, there is no patients participate within the study.

Not applicable.