The positive role of vitronectin in radiation induced lung toxicity: the in vitro and in vivo mechanism study

The human fibroblasts WI-38 and IMR-90 cell lines were purchased from American type culture collection (ATCC), and cultured at 37 °C in a humidified atmosphere containing 5% carbon dioxide with F-12K (Hyclone, Beijing, China) supplemented with 20% FBS (Gibco, Invitrogen, Carlsbad, CA, USA).

The human fibroblasts WI-38 and IMR-90 cells were irradiated with ¹³⁷Cs at the doses of 0 Gy (control), 4, 6, 8, 10 and 12 Gy respectively. Dose rate was 77 cGy/min and SSD was 10 cm.

According to standard procedures of western blotting, the protein samples extracted from the fibroblasts WI-38 and IMR-90 cells after receiving irradiation which were separated by 10% SDS-PAGE and then transferred to PVDF membranes (Millipore, Billerica, MA). After blocking with 5% non-fat milk in TBST for 1 h at room temperature, these membranes were incubated with the following primary antibodies (1:1000) at 4 °C overnight: anti-VTN monoclonal antibody (mAb) (Abcam, Cambridge, MA, US), anti-collagen I mAb (Abcam, Cambridge, MA, US), anti-collagen III mAb (Abcam, Cambridge, MA, US), anti-α-SMA mAb (CST, Beverly, MA, USA), anti-TGF-β mAb (Biolegend, San Diego, CA, USA), anti-p-ERK mAb (Santa, Santa Cruz, CA, USA), anti-ERK mAb (Pierce, Cruz, CA, USA), anti-p-AKT mAb (CST, Beverly, MA, USA), anti-AKT mAb (CST, Beverly, MA, USA), anti-p-JNK mAb (CST, Beverly, MA, USA), anti-JNK mAb (CST, Beverly, MA, USA), anti-GAPDH mAb (Sigma-Aldrich, St. Louis, MO, USA) and anti-β-actin mAb (Bioworld Technology, Inc. St. Louis, MO, USA). After washing three times with TBST, the membranes were incubated with secondary antibodies conjugated with HRP (1:10,000) for 1 h at room temperature. At last, these membranes were visualized using enhanced chemiluminescence reagents (Pierce, Cruz, CA, USA) according to the manufacturer’s instructions.

The protein levels of VTN, collagen I, collagen III, Hydroxyproline and α-SMA in WI-38 and IMR-90 cells after receiving irradiation were detected with TMB enzyme-linked immunosorbent assay (ELISA) kit (Invitrogen, GIBCO, Carlsbad, CA, USA) according to the manufacturer’s instructions. The absorbance was measured by a microplate reader (Bio-Rad, Hercules, CA, USA).

Total RNA from cultured WI-38 and IMR-90 cells or tissue samples after received irradiation was extracted by using RNA-Trizol reagent (Invitrogen, Gibco, Carlsbad, CA, USA), and reversely transcribed into cDNA using extraction kit (Amresco, Solon, HO, USA). Primers for quantitative RT-PCR (qRT-PCR) were listed in Additional file 1: Table S1. PCR process and data collection were performed on the 7900HT Fast Real-Time PCR system (Applied Biosystem, Carlsbad, CA, USA) according to the manufacturer’s protocol and GAPDH was used as the reference gene.

The cDNA sequence of VTN was amplified from pGEM-VTN and constructed into the vector pCDH to generate pCDH-VTN. The vector pCDH-VTN went through initial bacterial colony, PCR filtering, doubled digestion and gene sequencing assessment, and then was used to prepare lentivirus by co-transfecting into 293T cells with liposome. Three siRNA sequences targeted at VTN were designed by RNAi designer, and synthesized as follows in Additional file 1: Table S2. The siRNA sequences were inserted into PLVX vector to generate PLVX-si-RNA-VTN. A mixture of pGEM-VTN or PLVX-si-RNA-VTN, psPAX2 and pMDG2 were co-transfected into 293T cells using lipofectamine 2000 reagent to produce lentivirus. WI-38 cells were infected with the recombinant lentivirus-transducing units and 8 μg/mL polybrene (Sigma, St. Louis, MO, USA).

This study was approved by the Ethic Committee of Shanghai Chest Hospital, Shanghai Jiao Tong University (050432-4-1008A) and all the experiment procedures involved in animals were conducted according to the Guideline of Animal Care and Use Committee of Shanghai Chest Hospital, Shanghai Jiao Tong University. Sixty male C57 black 6 (C57 BL/6) mice with 3 months old which were purchased from the Shanghai Model Organisms Center, Inc., (Shanghai, China) were randomly divided into 5 groups, including normal group (for the mice without irradiation), model group (for the mice with irradiation only), sc-RNA group (for the mice with empty virus with irradiation), OE-VTN group (for the mice infected with VTN-overexpressing lentivirus with irradiation) and the siRNA group (for the mice infected with VTN-si-RNA lentivirus with irradiation). Saline (for model group), empty lentivirus (the vacant group as negative control). VTN-overexpressing lentivirus (for the OE-VTN group) and VTN-si-RNA lentivirus (for the siRNA group) were given to C57 BL/6 mice (12-week-old C57 BL/6, male, 12 mice/group, 1 × 10⁷ TU/mL) via tail vein injection, respectively. After transfected for 48 h, the mice in the 4 irradiation groups received a single dose of 12 Gy irradiation at the whole semi-thorax by using 6 MeV linear accelerator (ClinaciX, Varian), once only a mouse could receive irradiation. SSD was 100 cm, dose rate was 600 cGy/min, and irradiation length was 1 cm. After 8 or 12 weeks of irradiation, mice were sacrificed. Their lungs were obtained surgically and then subjected to hematoxylin and eosin (H&E) staining. The levels of hydroxyproline were evaluated by alkaline hydrolysed sample assay according to protocol.

Immunohistochemical staining was operated according to standard procedures. In brief, the lung tissues from all sacrificed rats were firstly fixed with 4% polyformaldehyde (PFA). Then they were produced into 3.5 μm sections, respectively. Tissue slides were incubated at 4 °C overnight with anti-VTN mAb (1:50) (Abcam, Cambridge, MA, US), anti-collagen I mAb (1:50) (Abcam, Cambridge, MA, US). Lastly, these sections were observed under a microscope and five random fields were chosen to evaluate the fibrosis score. Scores were measured by the cell cytoplasm staining patterns of the lung tissues as described: score 0, absent staining; score 1, light yellow staining; score 2, light brown staining; score 3, dark brown staining. Fibrosis scores were analyzed by two senior and experienced pathologists in single-blind review and evaluated the pathological changes according to the method proposed by Phillips et al. [21].

All experiments were repeated at least three times in this study. The statistical analysis was performed by using GraphPad Prism 5 software (GraphPad Software, Inc., La Jolla, CA, USA). All data were shown as mean ± standard deviation (SD) unless otherwise noted. The main statistical methods were t-test and one-way ANOVA. p < 0.05 was considered as statistically significant.

In order to investigate the change in the expression of VTN under irradiation, ELISA assay was undertaken at 6, 12, 24, 36, 48, 60 h after exposure to 0, 4, 6, 8, 10, or 12 Gy of irradiation in WI-38 and IMR-90 cells to detect the secretion level of VTN (Fig. 1a, b). The results exhibited that, it needed 12 h of irradiation with at least dose of 6 Gy for WI-38 cells (p < 0.05) to present increased VTN expression, and IMR-90 cells (p < 0.05) needed at least 8 Gy dose in the same duration for increased VTN expression. In fact, the expression levels of VTN protein reached the highest value after irradiation with 8 Gy for 48 h in both cell lines.

Furthermore, the mRNA and protein expression levels of VTN, collagen I, and collagen III in post-irradiation WI-38 and IMR-90 cells after being treated with 8 Gy dose of irradiation were detected by using quantitative RT-PCR and western blotting, respectively (Fig. 1c–f). The results showed that mRNA expression levels of collagen I and collagen III, which were consistent with the VTN mRNA level, increased with irradiation exposure time and reached the peak at 48 h post-irradiation both in WI-38 and IMR-90 cells (Fig. 1c, d). A similar tendency was revealed in the results of western blot analysis: protein expression levels of VTN, collagen I and collagen III significantly increased with prolonged irradiation time and all the maximal protein expression levels appeared at 48 h post-irradiation (Fig. 1e, f). Pearson analysis (data not shown) showed a positive correlation between protein expression level of VTN and collagen I and III in both WI-38 (r = 0.79, 0.70, respectively; p < 0.01) and IMR-90 cells (r = 0.45, 0.40, respectively; p < 0.05). Considering the changes in protein expression of VTN, collagen I and collagen III induced by irradiation were more obvious in WI-38 cells than that in IMR-90 cells, WI-38 cells were chosen to be treated with irradiation dose of 8 Gy for 48 h as the optimum cell-irradiation condition for our subsequent experiments.

VTN-overexpressed lentivirus and VTN-si-RNA lentivirus were established and transfected into WI-38 cells to construct VTN-overexpressed (OE-VTN) cells and VTN knockdown cells (si-RNA group). WI-38 cells transfected with non-targeting siRNA (si-NC group) were used as control for si-RNA group, while the mock-vehicle transfected WI-38 cells were applied as the negative control groups for OE-VTN group. VTN overexpressed and si-RNA lentivirus vectors were successfully constructed and verified through qRT-PCR and western blot (Fig. 2a–d). Among the 3 lentivirus vectors designed to interfere with VTN expression, the si-RNA3 had the most effective inhibitory effect on VTN expression in mRNA level compared with si-NC, and was chosen for subsequent construction of VTN silencing cell models. ELISA and western blot analysis were adopted to explore the protein expression levels of VTN, collagen I and III in WI-38 cells in each group at 48 h post-irradiation with 8 Gy. As shown in Fig. 2e–h, in OE-VTN cells, the protein expression levels of VTN were significantly higher than that in the normal and negative control group (p < 0.001). On the contrary, in si-RNA cells, the expression levels of VTN were significantly declined compared with the negative control group. Similarly, the protein expression levels of collagen I and III were significantly up-regulated in OE-VTN cells (p < 0.001 vs negative control), and protein expression level of collagen I was significantly down-regulated (p < 0.01 vs negative control) in si-RNA cells, whereas collagen III did not.

Many articles have reported that PI3K/AKT, ERK and JNK signaling pathway were involved in the TGF-β1-induced lung fibrosis, and the phosphorylation form of them could act as the key fibrosis regulators [22‐26]. Therefore, for further researches, we tested phosphorylation levels of fibrosis regulators, including AKT1, ERK1 and JNK, and utilized western blot analysis to detect the levels of phosphorylated and non-phosphorylated proteins, respectively. Results revealed that the levels of p-AKT and p-ERK were increased (p < 0.05) while the expression level of p-JNK remained almost unchanged and there is no significant change in levels of non-phosphorylated ERK, AKT and JNK compared with control. Conversely, knockdown of VTN decreased the levels of p-AKT, p-ERK, and p-JNK (Fig. 3b). These results indicated that VTN could activate the fibrosis process through activating ERK, AKT and JNK signal pathway in vitro.

To investigate effects of different expression levels of VTN on RILT process in vivo, VTN overexpression (OE-VTN group, n = 12) and VTN knockdown (si-VTN group, n = 12) mice models were constructed via tail vein injection of VTN overexpressed and interfering lentivirus, the mice injected with empty lentivirus and the same dosage of saline were used as negative control (sc-RNA group, n = 12) and normal group (n = 12), respectively. 48 h after transfection, all the mice in 4 groups were exposed to a single irradiation in a dose of 8 Gy to induce RILT. Meanwhile, mice without irradiation treatment were used as normal group (n = 12). The expression levels of VTN, collagen I, and collagen III in the lung tissues of all mice were demonstrated in Fig. 4. 8 and 12 weeks after irradiation, 6 mice in each group were sacrificed, and their lung tissues were collected for further detection. The mRNA and protein levels of VTN, collagen I and III were increased in the lung tissue from mice in irradiation groups compared with that in normal group (p < 0.05). Compared with the sc-RNA group, the mRNA and protein levels of VTN, collagen I and III were significantly up-regulated in the OE-VTN group (VTN: p < 0.001; collagen I: p < 0.01; collagen III: p < 0.05), which were significantly decreased in si-RNA group (VTN: p < 0.001; collagen I: p < 0.05; collagen III: p < 0.05), except for the mRNA level of collagen I (p > 0.05). Besides, there was no significant difference in the expression patterns of target factors between 8 and 12 weeks post-irradiation. Consistently, immunohistochemical staining results also confirmed the changes in the level of Collagen I in VTN-overexpressed mice (Fig. 4i). Western blot analysis results further verified that VTN induced up-regulation of collagen I, collagen III, and α-SMA (Fig. 4j).

Hydroxyproline is one of the specific amino acids in the collagen and α-SMA is considered as the biomarker in the activation of fibroblasts. Herein, we found that VTN could also affect the expression levels of hydroxyproline and α-SMA (Fig. 4d, h).

With H&E staining, it could be observed that after irradiation for 8 and 12 weeks, some symptoms of RILT that alveolar walls were thickened, alveolar spaces became narrow and parts of the alveolar walls were damaged occurred. In addition, the development and severity of these pathological changes in morphology were aggravated by VTN overexpression in vivo. By contrast, the process was slower and less severe in the si-VTN group compared with model group (Fig. 5c). The immunohistochemistry of VTN was demonstrated in Fig. 5d. Through microscopical observation, the degrees of pulmonary fibrosis in 5 groups were demonstrated in Fig. 5b (the normal group scored 0). Significantly higher fibrosis extent in OE-VTN group (p < 0.05) and obviously lower fibrosis level in si-RNA group (p < 0.05) were found, when compared with sc-RNA. These results suggested that VTN could exacerbate fibrotic process in mice lung after irradiation.