Porcine circovirus type 2 upregulates endothelial-derived IL-8 production in porcine iliac artery endothelial cells via the RIG-I/MDA-5/MAVS/JNK signaling pathway

Background Dysfunction of endothelial cells and vascular system is one of the most important pathological changes of porcine circovirus disease (PCVD) caused by porcine circovirus type 2 (PCV2). PCV2-infected endothelial cells can upregulate the production of endothelial-derived IL-8, which can inhibit the maturation of dendritic cells. Endothelial-derived IL-8 has different structural and biological characteristics compared with monocyte-derived IL-8. However, the mechanism of endothelial-derived IL-8 production is still unclear. Results Key molecules of RIG-I-like signaling pathway RIG-I, MDA-5, MAVS and a key molecule of JNK signaling pathway c-Jun in PCV2-infected porcine iliac artery endothelial cells (PIECs) were upregulated significantly detected with quantitative PCR, Western blot and fluorescence confocal microscopy, while no significant changes were found in NF-κB signaling pathway. Meanwhile, the expression of endothelial-derived IL-8 was downregulated after RIG-I, MDA-5, or MAVS genes in PIECs were knocked down and PIECs were treated by JNK inhibitor. Conclusions PCV2 can activate RIG-I/MDA-5/MAVS/JNK signaling pathway to induce the production of endothelial-derived IL-8 in PIECs, which provides an insight into the further study of endothelial dysfunction and vascular system disorder caused by PCV2.

Here, the study focused on the signal pathway of endothelial-derived IL-8 production in PIECs, providing a basis for the further study of endothelial dysfunction and vascular system disorder caused by PCV2.

PCV2 induces RIG-I, MDA-5 and MAVS activation in PIECs
In order to investigate whether PCV2 could activate the RIG-I-like signaling pathway in PIECs, qPCR was employed to measure the expression levels of RIG-I, MDA-5 and MAVS mRNA. Polyriboinsine-polyribocyaidylic acid (Poly(I:C), which could activate RIG-I and MDA-5 to produce IL-8 in mesothelial cells [18] and some epithelial cells [19], was used as the positive control in this study. The expression levels of RIG-I, MDA-5 and MAVS mRNA in PCV2-PIECs group were higher than those in the PIECs group at 12 and 24 h post-infection (hpi) (P < 0.05) (Fig. 1a, b, c) while substantially increased at 24 hpi (P < 0.01). Western blot showed that the protein expression levels of RIG-I, MDA-5 and MAVS were significantly upregulated in PCV2-PIECs group compared with the PIECs group (P < 0.01) ( Fig. 1d and Fig. 2). All results above indicated that the RIG-I-like signaling pathway was activated after PCV2 infection.

PCV2 does not activate NF-κB signaling pathway in PIECs
In addition to RIG-I-like signaling pathway, NF-κB signaling pathway also plays an important role in the production of inflammatory-related factors. Based on the previous results [9], we chose to test whether NF-κB signaling pathway was activated in PIECs following PCV2 infection. The qPCR results showed that the mRNA expression level of IκBα was significantly upregulated at 12 and 24 hpi in PCV2-PIECs group (P < 0.01) (Fig. 3a). Western blot showed that the protein expression level of IκBα was also upregulated in PCV2-PIECs group ( Fig. 1d and Fig. 3b). But the expressions of NF-κB p65 and NF-κB p-p65 had no significant changes (P > 0.05) ( Fig. 1d and Fig. 3c, d, e). Fluorescence confocal microscopy further confirmed that there was no significant nuclear translocation of NF-κB in PIECs (Fig. 4), suggesting that NF-κB was not activated. All results above indicated that PCV2 did not activate NF-κB signaling pathway in PIECs.

PCV2 activates JNK signaling pathway in PIECs
JNK signaling pathway is an important part of MAPK pathway, which plays an important role in cell cycle, reproduction, apoptosis and cell stress. Based on our microarray results (unpublished data), whether JNK signaling pathway could be activated in PIECs infected with PCV2 were investigated. The mRNA expression levels of c-Jun in the PCV2-PIECs at two time points were both higher than those in the PIECs group (P < 0.05) (Fig. 5a). The protein levels of JNK and c-Jun in the PCV2-PIECs group were both higher than those in the PIECs group at 24 hpi (P < 0.05) ( Fig. 1d and Fig. 5b, c). In the fluorescence confocal experiment, a significant nuclear translocation of c-Jun in the PCV2-PIECs group was observed as compared to the control group (Fig. 6). All indicated that JNK signaling pathway in PIECs was activated after PCV2 infection.

RIG-I/MDA-5/MAVS is involved in the upregulation of endothelial-derived IL-8 expression induced by PCV2 in PIECs
To study the role of RIG-I/MDA-5/MAVS in the production of endothelial-derived IL-8 induced by PCV2, siRNA was designed to knockdown the expression of RIG-I, MDA-5 or MAVS. The expression levels of endothelial-derived IL-8 mRNA and protein in the PCV2-siRIG-I group were significantly downregulated compared to those in the PCV2-PIECs group (P < 0.01) (Fig. 7). Similar results were also obtained in the PCV2-siMDA-5 and the PCV2-siMAVS groups (Fig. 7). These results strongly suggested that the RIG-I/MDA-5/MAVS pathway was involved in the upregulation of endothelialderived IL-8 expression induced by PCV2 in PIECs.

JNK signaling pathway contributes to the upregulation of endothelial-derived IL-8 expression induced by PCV2 in PIECs
To determine whether the expression of endothelialderived IL-8 induced by PCV2 was regulated by JNK signaling pathway, a JNK inhibitor was used to treat PIECs. The expression level of endothelial-derived IL-8 in the inhibitor treatment group was significantly downregulated compared to the control group (P < 0.01) (Fig. 8). It suggested that JNK pathway contributed to the upregulation of endothelial-derived IL-8 expression induced by PCV2 in PIECs. Discussion RIG-I and MDA-5, the sensory molecules of viral infection [20,21], can recognize RNA viruses and some DNA viruses by RNA polymerase III [22] and induce the production of different cytokines by identifying exogenous viral molecules [21,23]. RIG-I mainly recognizes Paramyxoviridae (Newcastle Disease Virus, Sendai Virus, and Respiratory Syncytial Virus) [20], Flavivirus [24] and Rhabdoviridae (Rabies Virus) [25]. MDA-5 mainly recognizes Coronaviridae (Mouse Hepatitis Virus) [26]. RIG-I and MDA-5 can simultaneously recognize Measles Virus [27] and West Nile Virus [28]. In both VR1BL and PK-15 cells, RIG-I and MDA-5 also recognize PCV2 together [15,29], which was similar to that of our study in PIECs. All these indicated that PCV2 could infect different kinds of cells in piglets via RIG-I/MDA-5 pathway. MAVS, which is located on the outer membrane of mitochondria, is a key common adaptor of retinoic acidinduced gene I-like receptors (RLRs) [30]. Caspase recruitment domains (CARDs) are released from the signal suppression state and recruit MAVS by CARD-CARD interaction after RIG-I and MDA-5 identify the viral components in the cytoplasm [31][32][33]. The activation of MAVS can effectively promote the downstream signaling pathway and induce the productions of interferon and proinflammatory factor [34]. Once MAVS is degraded, (See figure on previous page.) Fig. 3 The relative expression levels of key proteins in NF-κB signaling pathway. PIECs were infected with PCV2 (MOI = 0.5) for 1 h and then cultured for 12 and 24 h. The IκBα mRNA expression was assessed by qPCR at two time points (a). Quantitative analyses of the immunoblot data of IκBα (b), NF-κB p65 (c) and NF-κB p-p65 (d) was performed at 24 h. The ratio of NF-κB p-p65/ NF-κB p65 (e). The uninfected PIEC was used as a negative control and Poly(I:C) treatment was used as a positive control. Data are results of three independent experiments and presented as the mean and standard deviation (error bars) for each group. **, P < 0.01 Fig. 4 Nuclear translocation of NF-κB after PCV2 infection in PIECs. PIECs were infected with PCV2 (MOI = 0.5) for 1 h and then cultured for 24 h. Fluorescence confocal microscopy was used to measure the cellular localization of NF-κB p65 in PCV2-infected PIECs. The localization of NF-κB p65 (red) was observed with a fluorescence microscope using immunofluorescence staining with anti-NF-κB p65 and Alexa Fluor®647 conjugate anti-mouse IgG. Nuclei were stained with DAPI. The uninfected PIEC was used as a negative control and Poly(I:C) treatment was used as a positive control. Bar = 20 μm its downstream signal will also be blocked [35,36]. In this study, we confirmed that PCV2 could activate the adapter protein MAVS in PIECs. siRNA (siRIG-I, siMDA-5, siMAVS) transfection experiments further confirmed the roles of these signaling pathway above in the production of endothelial-derived IL-8 in PIEC after PCV2 infection. Taken together, PCV2 could activate the signaling pathway of RIG-I/MDA-5/MAVS in PIECs and therefore induced the production of endothelialderived IL-8.
Based on the previous results [9] and microarray results (unpublished data), we chose to investigate whether NF-κB and JNK signaling pathway were the key downstream pathway in this process. The results demonstrated that PCV2 infection could activate JNK but not NF-κB signaling pathway in PIECs. It indicated that PCV2 could activate JNK signaling pathway to upregulate the production of endothelial-derived IL-8 in PIECs, which was different from the mechanism of monocytederived IL-8 production in downstream pathway.

Conclusions
Our study demonstrated that PCV2 could upregulate the production of endothelial-derived IL-8 via RIG-I/MDA-5/MAVS/JNK signaling pathway in PIECs (Fig. 9). It elucidated the mechanism of endothelial-derived IL-8 production in PCV2 infected endothelial cells, which was different from the mechanism of monocyte-derived IL-8 production. Our study provides an insight into the further study of endothelial dysfunction and vascular system disorder caused by PCV2.  6 μl). The conditions were as follows: 95°C for 5 min, followed by 40 cycles of 95°C for 40 s, 56°C for 30 s, and 72°C for 30 s. Water was used to replace template DNA in the negative control. All results were analyzed using MxPro Software. Primer sequences for qPCR are listed in Table 1.   Western blot analysis 70% confluent PIECs in a 6-well plate was infected with PCV2 (MOI = 0.5) with Poly(I:C) (5 μg/ml) for 1 h in an atmosphere of 5% CO 2 at 37°C. The cells were further incubated for 24 h in RPMI1640 supplemented with 10% FBS. Total proteins were extracted from the collected cells according to the manufacturer instructions of total protein extraction kit (Keygen, Nanjing, China) and quantified by BCA protein assay kit (Cwbio, Beijing, China). Quantitative protein was subjected to SDS-PAGE and then transferred onto a polyvinylidene difluoride (PVDF) membrane. The membrane was further blocked with Tris Buffer Solution Tween-20 (TBST) containing 5% BSA for 2 h and then incubated overnight in a 1:2000 dilution of primary antibody at 4°C with gentle shaking. After washed four times in TBST, the membrane was blotted with the second antibody (1: 5000) for 1 h. Actin was used as the internal reference.
The blot was visualized with Omega Aplegen scanner (Aplegen, San Francisco, CA, USA). The gray matter volume of protein was quantified by importing the images into an ImageJ analysis software. The original blots can be referred in the Additional file 1.

Fluorescence confocal microscopy
The cells were treated in the same manner as Western blot analysis. At 24 hpi, the cells were digested with 0.25% trypsin. All cells were centrifuged and the supernatant was discarded. The cells were suspended in 200 μl of RPMI1640 supplemented with 10% FBS, and then dripped into a confocal chamber treated with polylysine. After adhering for 4 h, the supernatant was discarded, PIECs were fixed with 2% paraformaldehyde for 15 min at room temperature. After washed three times with phosphate-buffered saline (PBS), they were blocked with blocking solution for 1 h. Then, the diluted primary antibody (p65 1:800, c-Jun 1:400) was added and incubated overnight at 4°C. The diluted secondary antibody (1:500) was added in the dark and incubated at room temperature for 1 h. The nucleus was stained with DAPI solution at 37°C for 20 min. Anti-fluorescence attenuating seals were added dropwise protected from light at 4°C. The cells were observed with a laser scanning confocal microscopy (Olympus FV100, Hamburg, Germany).

Inhibitor treatment
PIECs were inoculated into 6-well plates. When the cells covered 50-70% of the surface, they were treated with JNK inhibitor SP600125. The same dose of dimethyl sulfoxide (DMSO) was used as the control (The final concentration of DMSO was less than 0.1%). After 1 h of treatment, the cells were washed with PBS and then infected with PCV2 (MOI = 0.5) for 1 h. The cells and supernatants were sampled after 24 h. The expression of endothelial-derived IL-8 was detected by qPCR and ELISA.

ELISA
The expression of endothelial-derived IL-8 in the supernatant of each group was determined according to the manufacturer instructions of porcine IL-8 ELISA kit (CUSABIO, Wuhan, China). The OD value was read at 490 nm using a microplate reader (BioTek, Winooski, VT, USA). Standard curves were obtained according to Curve Expert software using the dilutions of a sample

Statistical analysis
We performed all statistical analyses using the SPSS Statistics 17 (IBM Corporation). The results are expressed as the mean ± standard deviation (SD). Differences between the experimental groups and the control groups were analyzed using ANOVA followed by the Duncan's multiple range test for multiple comparisons. P < 0.05 was considered to be statistically different and was denoted by *. P < 0.01 was considered to be significantly different and was denoted by **. P < 0.001 was considered to be extremely significantly different and was denoted by ***. Unless indicated otherwise, the experiments were performed in triplicate (n = 3).
Additional file 1. The original blots for the figures. For our gel data, the target proteins analyzed by western blot were transferred from SDS-PAGE and the PVDF membrane would be sliced into different strips according to their protein size identified with the loading marker. Then the panel of antibodies targeting these proteins were used for western blot analysis. Our SOP for processing the gel data only collected the pictures with target protein regions instead of the whole gel, so here are the original gels for the data we presented in the manuscript.