Neutralizing interferon‐α blocks inflammation‐mediated vascular injury via PI3K and AMPK in systemic lupus erythematosus

Abstract

Plasmacytoid dendritic cells (pDCs) play a key role in the initiation and amplification of systemic lupus erythematosus (SLE)‐associated vascular injury. In this study, we found that dsDNA induced dose‐ and time‐dependent increase in IFN‐α and Toll‐like receptor 7 (TLR7), TLR9 and IRF7 expression in pDCs. Co‐cultured circulating endothelial cells (ECs) with activated pDCs significantly decreased proliferation, tube formation and migration in ECs. The elevated level of cellular IFN‐α increased cell adhesion, promoted cell apoptosis, induced cell senescence and arrested cells at G0/G1 phase of endothelial progenitor cells (EPCs). Additionally, the co‐culture system activated MAPK and inactivated PI3K. Pristane was used to establish a in vivo SLE‐like mouse model. Importantly, we showed that INF‐α‐neutralizing antibody (IFN‐α‐NA) rescued all the changes induced by IFN‐α in vitro and prevented vascular injury in pristane‐induced SLE model in vivo. In conclusion, we confirmed that activated pDCs promoted vascular damage and the dysfunction of ECs/EPCs via IFN‐α production. IFN‐α‐neutralizing antibody may be a clinical implication for preventing vascular injury. PI3K signalling and AMPK signalling were associated with SLE‐associated vascular functions.

Pristane can cause obvious damage to the blood vessels of ApoE−/− mice, the pathological morphology of the arterial tissue of the SLE mouse model is disordered, and the basement membrane is thickened. The treatment of IFN‐α‐NA reverses these phenotypes. The expression of IFN‐α receptor in the vascular endothelium is increased, and that of SA‐β‐Gal‐positive EPCs in vascular endothelium is enhanced.

Abbreviations

ECs

circulating endothelial cells

EPCs

endothelial progenitor cells

IFN‐α

interferon‐α

MAPK

mitogen‐activated protein kinase

NA

IFN‐α‐neutralizing antibody

PBMCs

peripheral blood mononuclear cells

pDCs

plasmacytoid dendritic cells.

PI3K

phosphatidylinositol‐3 kinase

SLE

systemic lupus erythematosus

TLR7

Toll‐like receptor 7

INTRODUCTION

Cardiovascular disease is a leading cause of morbidity). Plasmacytoid dendritic cells (pDCs) are important immune orchestrator in heart and vessels [1]. Activated pDCs by immune stimuli are potential participants in the inflammatory processes that contribute to vascular and organ injury [2, 3, 4]. pDCs are the main producers of type I interferon (IFN‐I), such as IFN‐α, under the stimulation of Toll‐like receptors (TLRs) 7 and 9 [5, 6]. Activation of pDCs through TLR7/9 triggers the production of type I IFN and differentiation of DCs [5]. Elevated secretion of IFN‐α by pDCs is considered as an inducement of SLE [7]. In SLE patients, up‐regulation of TLR7‐mediated IFN‐α production by pDCs is related to the disease activity of SLE [8]. The mechanism of excessive IFN production has been well established in autoimmune diseases including SLE pathogenesis [2, 9, 10].

Patients with immune disorders or inflammatory diseases always show high level of circulating endothelial cells (ECs) and endothelial progenitor cells (EPCs), especially those with cardiovascular disease and vascular injury [4, 11, 12, 13, 14]. EPCs favour angiogenesis and repair of endothelial injury. Excessive production of IFN‐I by TLR7/9 stimulation accelerates atherosclerosis in mouse with lupus‐like disease, and the proliferation of EPCs is modulated by type I IFN [15]. The mouse model of SLE mimics the clinical presentation of autoimmunity in SLE patients. Two main categories of mouse models of SLE have established: spontaneous and induced. NZB×NZW F1, MRL/MpJ‐Fas^lpr (MRL‐lpr) and BXSB/Yaa mice develop SLE spontaneously, and genetically susceptible animals of SLE show severe organ damage [16]. The bm12 transfer model is a relative easy and efficient way to study the cellular process of SLE [17]. Additionally, intraperitoneal injection of pristane is a standard method to establish an induced SLE mouse model [16]. Pristine induces IFN‐I production through TLR7/MyD88 pathway [18]. A previous studies show that anti‐tumour necrosis factor (TNF)‐α antibody and targeting IFN‐α ameliorate the arthritis symptoms [19, 20].

Both the phosphatidylinositol‐3 kinase (PI3K) and mitogen‐activated protein kinase (MAPK) pathways are essential for the proliferation and survival and angiogenesis of ECs/EPCs [21, 22, 23]. It has been well established that pDCs mainly express PI3K p110δ subunit, which is uniquely involved in the production of type I IFN [5]. Both PI3K and MAPK regulate type I IFN production through modulating the nuclear translocation of IRF7 in human pDCs [5, 24, 25]. However, the association of pDC activity with the activation of PI3K and MAPK in ECs/EPCs is not yet well elucidated.

In the present study, we used in vitro and in vivo model to explore the effects of INF production on vascular injury. ECs and EPCs were co‐cultured with activated pDCs, and the results showed that IFN‐α inhibition prevented the vascular injury in SLE‐like mouse model. Inflammation response, as well as PI3K and MAPK activity, was involved in the progress of vascular injury.

MATERIALS AND METHODS

Isolation, culture and activation of pDCs

Animal experimental protocols were approved by the Committee of Animal Experimentation in Central South University. The peripheral blood was collected from eyeballs of a C57BL/6 mouse (Shanghai SLAC Laboratory Animal Co., Ltd.; 2 months old), and peripheral blood mononuclear cells (PBMCs) were isolated. CD11c+pDCs were isolated and purified using immune magnetic bead sorting kits with CD11c‐PE mAb (EasySep™ Mouse CD11c Positive Selection Kit II, Stemcell Technologies) according to the manufacturer's recommendations. pDCs were cultured in the complete Dulbecco's modified Eagle's medium (DMEM) supplementing with 0 μg/L (Control), 20 μg/L (low), 40 μg/L (moderate) and 80 μg/L (high) anti‐ds DNA antibodies or neutralizing antibody (NA) against IFN‐α (5mg/kg) in triplicate for 24 and 48 h. The expression of TLR7, TLR9, IRF7, pIRF7, IFN‐I and IFN‐α in pDCs was examined by Western blotting analysis. The concentration of IFN‐α in the cellular medium was measured using the enzyme‐linked immunosorbent assay (ELISA) to determine the activation of pDCs.

Isolation, culture and identification of EPC cells

Circulating endothelial cells and EPCs were isolated from peripheral blood as previously reported [26]. EPCs were identified using EPC surface markers including CD14 conjugated to fluorescence in isothiocyanate (FITC), CD34 conjugated to phycoerythrin (PE) and KDR‐PerCp conjugated to cyanine dye 5·5 (Cy5·5) using flow cytometry. The uptake of DiI‐acetylated low‐density lipoprotein (DiI‐Ac‐LDL) and UEA was detected using immunofluorescence [27].

Animal model and treatment

Twenty‐two female ApoE^−/− mice (8‐week old, Shanghai SLAC Laboratory Animal Co., Ltd.) were adaptively fed for 1 week and then randomly divided into three groups. Mice in the control group (n = 10) were intraperitoneally (ip.) injected with 0·5 ml of normal saline (NS, single injection), followed by oral treatment of 0·1 ml of NS daily for 6 months. Mice in the model group (n = 6) were injected (ip.) with 0·5 ml of pristane (single injection), followed by tail vein injection of 0·1 ml of NS daily for 6 months. Mice in the treatment group (n = 6) were injected (ip.) with 0·5 ml of pristane (single injection), followed by oral treatment of IFN‐α NA (50 μg/kg body weight) per day for 6 months. High‐fat diet and water were available ad libitum during the experimental period. Blood samples were collected under anaesthetization (3·5 ml/kg of 10% chloral hydrate, intraperitoneally). Aorta were collected and divided into three parts: the first part was used for histological examination, the second part was prepared for freezing microtome section and immunofluorescence analysis, and the third part was used for the Western blotting analysis.

Histopathological examinations

Aorta tissues were cut into pieces and embedded into paraffin. The sections (5 μm) of aortic tissues were cut onto slides, deparaffinized and then used for haematoxylin and eosin (HE) staining examination. HE staining was performed using Gill‐2 haematoxylin and eosin Y (Sigma‐Aldrich) following the manufacturer’s instruction. The morphology of aortic tissues was detected using a light microscopy (BX51; Olympus Optical).

Cell co‐culture system

ECs and EPCs were incubated with activated or inactivated pDCs (control) in triplicate in the transwells (0·4 μm). In brief, ECs and EPCs were seeded into the lower chambers and pDCs were seeded into the upper chambers and incubated at 37°C and 5% CO₂.

Cell viability of ECs

Cell Counting Kit‐8 (CCK‐8) assay kit (Beyotime Institute of Biotechnology) was used to determine the cell viability of ECs under different treatments. ECs and pDCs at the final density of 5 × 10³ cells/well were seeded into lower and upper chambers of the transwells in triplicate, respectively, and incubated for 12, 24, 48 and 72 h at 37°C and 5% CO₂. The upper chambers and pDCs were removed before cell viability assay, and the ECs were washed with PBS. For cell viability assay, 10 μl of CCK‐8 solution was added into each well and then maintained for 60 min. Cell viability at 450 nm was detected using a Bio‐Rad microplate absorbance reader (Bio‐Rad).

ECs in vitro cell migration assay

The invasion ability of ECs after incubation with pDCs for 48 h was detected using 24‐well transwell chambers (5 μm; Coring, NY, USA). In brief, serum‐free DMEM and 1 × 10⁵ ECs were placed in triplicate into the upper chambers of transwell, while activated or inactivated pDCs (control) were seeded into the lower chamber supplemented with complete DMEM. Transwells were maintained at 37°C with 5% CO₂ for 48 h. ECs adhered to the upper surface of upper chambers were removed, and the migrated cells adhered to undersurface were fixed and stained with crystal violet. The digital photographs of migrated cells were captured using an Olympus microscope (BX51). Five randomly and arbitrarily selected (non‐overlapped) fields were selected for cell counting.

Transwell assay of ECs

ECs (1·0 × 10⁴ cells/well) were placed into the lower chambers of transwell (0·4 μm) coated with 50 μl of Matrigel (BD Biosciences) [28]. Cells were incubated at 37°C and 5% CO₂ for 48 h and then washed with PBS. Photomicroscopy was performed using an Olympus microscope (BX51, Olympus). Five non‐overlapped and arbitrarily selected fields were selected for the counting of vascular tubules.

ELISA measurement

Serum IFN‐α level in ApoE−/− mice was detected at 6 months post‐treatment, and the cellular IFN‐α level in the lower chamber of co‐culture systems was detected at 48 h post‐incubation. IFN‐α content was determined using the ELISA kit (Elabscience Biotechnology) and a microplate absorbance reader (Bio‐Rad).

Immunofluorescence analysis

The uptake of DiI‐acetylated low‐density lipoprotein (DiI‐AcLDL) and UEA in EPCs, the expression of EPC surface markers (including CD14‐FITC, CD34‐PE and KDR‐PerCp‐Cy5.5) in EPCs, and the expression of IFN‐α receptor (R) in EPCs/ECs and vascular endothelial cells were detected using immunofluorescence. Cells at the final density of 2 × 10⁴ cells/well were placed into lower chambers for 48 h at 37°C and 5% CO₂. The slides of vascular tissues were deparaffinized. Cells or tissue slides were treated with 4% paraformaldehyde (Beyotime) for 30 min and Triton‐X‐100 for 20 min, followed by incubation with primary antibodies at 4°C overnight and secondary antibody at 37°C for 2 h. DAPI was used for nuclear staining at 4°C for 10 min. Fluorescent images were captured using a laser scanning confocal microscope (LSM710; Zeiss).

Flow cytometric analysis for EPCs

The apoptotic percentage and cell cycle distribution of EPCs after incubation with pDCs for 48 h were detected using flow cytometric assay. EPCs in the lower chambers of transwells were harvested and fixed. Cells were treated with propidium iodide (PI) for 20 min. The cell cycle distribution of EPCs was analysed using a BD FACSCalibur flow cytometry (BD Biosciences). The apoptosis of EPCs was determined using an Annexin V‐FITC (BD Biosciences) according to the methods of manufacturers’ instruction. PI was used for the staining of nuclear. Apoptotic EPCs were determined using FACSCalibur flow cytometry.

EPC adhesion assays

Endothelial progenitor cell adhesion ability was detected using a cell adhesion detection kit (BestBio Company) after 48‐h incubation with pDCs. Before experiments, the lower chambers of transwells were coated with coating buffer at 4°C overnight. The lower chambers were then washed and filled with EPCs in triplicate. Cells were incubated at 37°C and 5% CO₂ for 30 min, followed by washing with fresh DMEM. The lower chambers were then filled with complete DMEM and staining solutions and incubated for 2 h. The OD₄₅₀ values of EPCs in chambers were read using a Bio‐Rad microplate reader. Adhesion ability was calculated as: Adhesion rate = (OD_test–OD_empty)/(OD_Control–OD_empty).

Ageing assay of EPCs

Senescence‐associated β‐galactosidase (SA‐β‐Gal) in aged EPCs was detected using a β‐galactosidase staining kit (Jiancheng Biological Company) [29]. After treatment for 48 h, cells in the lower chambers were harvested and then subjected to ageing assay according to the recommendation from manufactures. For the ageing of cells in aorta, the frozen section of aorta tissues was stained using SA‐β‐Gal. Cellular and tissue images were captured using a microscope (Olympus microscope BX51).

Western blotting analysis

Cellular and tissue proteins were extracted using lysis buffer (Beyotime) and then quantified using a Bradford protein assay kit (Thermo Fisher Scientific Inc.). Protein samples were separated using 10% SDS‐PAGE and then transferred onto PVDF membranes (Millipore) following the standard methods. The expression of intracellular adhesion molecule 1 (ICAM1), C‐X‐C motif chemokine 10 (CXCL10) and C‐X‐C chemokine receptor 3 (CXCR3) proteins in EPCs and PI3K, pPI3K, MAPK and pMAPK in ECs under different treatments; TLR7, TLR9, IFN regulatory factor 7 (IRF7), IFN‐I and IFN‐α protein in the PBMCs, and ICAM1, CXCL10, CXCR3, PI3K, pPI3K, MAPK and pMAPK protein in aorta of ApoE^−/− mice was detected using Western blotting analysis with specific primary antibodies and HRP goat anti‐rabbit/rat IgG secondary antibodies. Antibodies were purchased from Cell Signaling Technology, Inc., Abcam or Boster Biotechnology. Enhanced chemiluminescence detection system (Millipore) and Image‐Pro Plus 6.0 software (Media Cybernetics Inc.) were used for the detection of protein expression.

Quantitative real‐time PCR

Total RNA was isolated from PBMCs of ApoE^−/−mice using TRIzol reagent (TaKaRa). The first strand cDNA was synthetized using Bestar^® qPCR RT Kit (DBI Bioscience). The relative expression level of TLR7, TLR9, IRF7, IFN‐I and IFN‐α in the PBMCs derived from blood was detected in triplicate using specific primer pairs (Table 1) and a Bestar^® SYBR Green qPCR Master Mix Kit (DBI Bioscience). An Agilent Stratagene Mx3000 qRT‐PCR machine (Agilent) was employed for amplifying the PCR product. The reaction conditions are as follows: denaturation at 94°C for 3 min; 40 cycles of 94°C for 20 s, 58°C for 20 s and 72°C for 20 s. The $2^{‐\mathrm{\Delta }\mathrm{\Delta }{C}_{\mathrm{t}}}$ method was used to calculate the relative expression level of mRNAs. The internal reference gene used in this study was GAPDH.

TABLE 1.

Primer sequences

ID	Sequence (5′‐3′)	Product length (bp)
M‐β‐actin F	CATTGCTGACAGGATGCAGA	139
M‐β‐actin R	CTGCTGGAAGGTGGACAGTGA
TLR7.F	GGAGCTGGTGGCAAAATTGG	122
TLR7.R	TGCTGAGCTGTATGCTCTGG
TLR9.F	TGTGAGCTGAAGCCTCATGG	159
TLR9.R	GGACAGGTGGACGAAGTCAG
IRF‐7.F	TGCTTTCTAGTGATGCCGGG	134
IRF‐7.R	GGGTTCCTCGTAAACACGGT
IFN‐α.F	ATTTCCCCTGACCCAGGAAGATG	167
IFN‐α.R	CTCTCAGTCTTCCCAGCACATT

Statistical analysis

All data were expressed as the mean ± standard deviation (SD). Statistical analysis was analysed using the GraphPad Prism 6.04. Differences between two groups were analysed using the unpaired t‐test, and those among more than three groups were analysed using the one‐way analysis of variance (ANOVA) followed by the Tukey test. P < 0·05 was considered as statistically significant difference.

RESULTS

Identification of EPCs and activation of pDCs

The isolated EPCs were identified using the surface biomarkers (CD14, CD34 and KDR‐PerCp), as shown in Figure 1A and Figure S1A; EPCs expressed CD34 (50·36 ± 10·06%) and KDR (92·5 ± 6·55), but did not express CD14. In addition, as shown in Figure 1B, EPSs bind to UEA and the uptake of Dil‐ac‐LDL. Isolated pDCs were treated with dsDNA, and the content of IFN‐α was increased by dsDNA in a dose‐ and time‐dependent manner (Figure 1C). The production of cellular IFN‐α and the expression of IFN‐α and IFN‐I were increased by all concentrations of dsDNA at 48 h post‐treatment (Figure 1C,D,H,I). In addition, interferon regulatory factor 7 (IRF7) was activated by dsDNAs as shown by Western blot (Figure 1D‐E). As type I IFN production by pDCs in response to viral infection requires the activation of TLR7 and TLR9 (15549123), we found that the expression of TLR7 and TLR9 was induced by dsDNA in a dose‐ and time‐dependent manner (Figure 1D,F,G). These results indicated the activation of pDCs by dsDNA.

FIGURE 1.

Identification of EPCs and activation of pDCs. A and B, The identification of isolated EPCs using flow cytometry (CD14, CD34 and KDR‐PerCp; the red lane indicates not activated and the blue lane indicates activated pDCs) and immunofluorescence (Dil‐ac‐LDL and UEA). C, The production of IFN‐α in pDCs in response to dsDNA treatments. D‐I, The expression and fold change of protein expression in pDCs in response to dsDNA treatments. *, ** and *** note p < 0·05, p < 0·01 and p < 0·001 vs. 0 μg/L at the same time interval, respectively. #, ## and ### indicate p < 0·05, p < 0·01 and p < 0·001 vs. 40 μg/L at the same time interval, respectively

pDCs‐derived IFN‐α inhibits EC proliferation, angiopoiesis and migration

To investigate the impact of IFN‐α on EC cellular behaviour, Ecs cells were co‐cultured with activated pDCs (a‐pDCs). We noted an increase in IFN‐α production in EC culture (p < 0·01, Figure 2A and Figure S1B) and the up‐regulated expression of IFN‐α receptor in EC cells compared with inactivated pDCs (ina‐pDCs; Figure 2B and Figure S1C). The increased IFN‐α suppressed cell proliferation (Figure 2C), vascular tube formation (Figure 2D) and cell migration (Figure 2E,F) in a dose‐dependent manner in Ecs. In addition, pDCs activated by 80 μg/L of dsDNA significantly decreased the phosphorylation of PI3K and increased the phosphorylation of MAPK in a dose‐dependent manner, compared with inactivated pDCs (p < 0·01; Figure 2G,H). These data demonstrated pDC‐derived IFN‐α impaired cell proliferation, angiopoiesis and cell migration in Ecs.

FIGURE 2.

Response of ECs to pDC activation. A, The production of IFN‐α in pDC culture. B, The expression of IFN‐α receptor (R) in ECs. C, The cell proliferation of ECs. D, Representative images of the angiogenesis in ECs. E and F, Representative images and statistics of the migrated ECs. G, The expression and fold change of proteins in ECs. *** notes p < 0·001 vs. ina‐pDCs. ## and ### indicate p < 0·01 and p < 0·001 vs. a‐pDCs (20 μg/L) respectively. ina and a represent the inactivated and activated pDCs

pDCs‐derived IFN‐α influences the cell adhesion, apoptosis, cell cycle distribution and ageing in EPCs

The co‐culture of EPCs with activated pDCs significantly increased the adhesion ability (p < 0·001, Figure 3A). The cell apoptosis was determined by Annexin V/PI double staining to measure the early and late apoptosis. As shown in Figure 3B, activated pDCs induced cell apoptosis in EPCs in a dsDNA dose‐dependent manner. Increased IFN‐α level increased the percentage of cell at G0/G1 cell cycle phage and decreased that at S stage (Figure 3C). The SA‐β‐Gal‐positive EPCs were increased by dsDNA treatment (Figure 3D). Additionally, CXCL10, an INF‐γ‐inducible CXC chemokine, plays an important role in inflammation, migration and invasion. We found that the expression of CXCL10 was significantly induced by activated pDCs (Figure 3E). As an adhesion molecule, the expression of ICAM1 was increased (Figure 3E) under co‐culture. CXCR3 has shown to inhibit angiogenesis, and our results revealed that activated pDCs significantly enhanced the level of CXCR3 (Figure 3E). These results revealed that co‐cultured EPCs with activated pDCs promoted cell adhesion, apoptosis and cell cycle distribution through regulating the expression of CXCL10, CXCR3 and ICAM1.

FIGURE 3.

Response of EPCs to pDC activation. A, The adhesion rate of EPCs. B and C, The cell apoptosis (calculated the summary of early and late apoptosis) and cell cycle distribution of EPCs in response to treatments, respectively. D, Representative images of senescence‐associated β‐galactosidase (SA‐β‐Gal) staining in EPCs. Magnification, ×200. (E) The expression and fold change of proteins. *, ** and *** note p < 0·05, p < 0·01 and p < 0·001 vs. ina‐pDCs, respectively. #, ## and ### indicate p < 0·05, p < 0·01 and p < 0·001 vs. a‐pDCs (20 μg/L), respectively. ina and a represent the inactivated and activated pDCs

Activated pDCs inhibit EC proliferation, angiopoiesis and migration in an IFN‐α‐dependent way

To further determine the effects of activated pDCs on the proliferation, migration and angiopoiesis of ECs through IFN α, we treated active pDCs with IFN‐α‐NA and found that IFN‐α‐NA significantly reversed the expression of IFN‐α and IFN‐α receptor (Figure 4A,B and Figure S1D‐E). In addition, the cell proliferation, migration and angiopoiesis of ECs reduced by activated pDCs were recovered by IFN‐α‐NA (Figure 4C‐F). Moreover, we found the IFN‐α‐NA administration rescued the phosphorylation of PI3K and MAPK in ECs (Figure 4G,H). These data indicated that the cell proliferation, angiopoiesis and migration in ECs were mediated by activated pDCs in an IFN‐α‐dependent way.

FIGURE 4.

Effects of IFN‐α‐neutralizing antibody (NA) on ECs. (A) The production of IFN‐α in pDC cell culture. (B) The expression of IFN‐α receptor (R) in ECs. (C) The cell proliferation of ECs under different treatments. (D) Representative images of the angiogenesis in ECs. (E and F) Representative images and statistics of the migrated ECs. (G and H) The expression and fold change of proteins in ECs. *and *** note p < 0·05 and p < 0·001 vs. ina‐pDCs, respectively. ## and ### indicate p < 0·01 and p < 0·001 vs. a‐pDCs (80 μg/L) respectively. ina and a represent the inactivated and activated pDCs

Activated pDCs inhibit EPC adhesion, apoptosis and ageing in an IFN‐α‐dependent way

Next, we showed that the addition of IFN‐α‐NA eliminated EPC adhesion, decreased pDC activation‐induced apoptosis and disturbance of cell cycle in EPCs (Figure 5A‐C). The SA‐β‐Gal‐positive EPCs were decreased by IFN‐α‐NA, suggesting IFN‐α‐NA‐suppressed active pDCs induced cellular senescence (Figure 5D). The expression of cell adhesion‐, migration‐ and angiogenesis‐related protein, CXCL10, CXCR3 and ICAM1, was induced by activated pDCs, but reversed by IFN‐α‐NA (Figure 5E). These data revealed that pDC activation‐reduced cell viability in EPCs was mediated by IFN‐α.

FIGURE 5.

Effects of IFN‐α‐neutralizing antibody (NA) on EPCs. A, The adhesion rate of EPCs. (B and C) The cell apoptosis (calculated the summary of early and late apoptosis) and cell cycle distribution of EPCs in response to treatments respectively. D, Representative images of senescence‐associated β‐galactosidase (SA‐β‐Gal) staining in EPCs. Magnification, ×200. E, The expression and fold change of proteins. *, ** and *** note p < 0·05, p < 0·01 and p < 0·001 vs. ina‐pDCs respectively. #, ## and ### indicate p < 0·05, p < 0·01 and p < 0·001 vs. a‐pDCs (80 μg/L) respectively. ina and a represent the inactivated and activated pDCs

IFN‐α‐NA suppresses vascular damage in pristane‐induced SLE mouse model

Pristane administration induced obvious vascular damage in ApoE ^−/− mice, the pathological morphology of arterial tissues in SLE mouse model was disordered, and the basilar membrane was thickened, while IFN‐α‐NA reversed these phenotypes (Figure 6A). The expression of IFN‐α receptor in vascular endothelium was increased (Figure 6B), and that of SA‐β‐Gal‐positive EPCs in vascular endothelium was enhanced (Figure 6C). The production of IFN‐α in serum was induced in SLE mouse model and reversed by IFN‐α‐NA (Figure 7A). Additionally, SLE induced mRNA and protein expression of TLR7, TLR9, IRF7, IFN‐I and IFN‐α in PBMCs was repressed by IFN‐α‐NA (Figure 7B,C). Moreover, IFN‐α‐NA also recovered the expression of ICAM1, CXCL10 and CXCR3 in aorta of SLE model (Figure 7D). In SLE mouse model, PI3K pathway was activated and MAPK pathway was inactivated, while IFN‐α‐NA repressed these effects (Figure 7D). These in vivo results showed that IFN‐α‐NA recovered the damage effects in SLE mouse, suggesting that IFN‐α‐NA might be a potential therapeutic method for SLE.

FIGURE 6.

Histology and immunochemistry in aorta of ApoE ^−/− mice. The representative images of haematoxylin and eosin (HE) staining examination (A), expression of IFN‐α receptor (B) and senescence‐associated β‐galactosidase (SA‐β‐Gal) staining in aorta of ApoE^−/− mice (C) respectively. Magnification, ×200. NA, neutralizing antibody

FIGURE 7.

Expression of factors in response to SLE and IFN‐α‐neutralizing antibody (NA). A, The serum content of IFN‐α in ApoE^−/− mice. B and C, The expression level of genes and proteins in PBMCs respectively. D, The expression of proteins in aorta of ApoE ^−/− mice. ** and *** note p < 0·01 and p < 0·001 vs. control respectively. #, ## and ### indicate p < 0·05, p < 0·01 and p < 0·001 vs. model, respectively

DISCUSSION

The pathogenic mechanisms leading to SLE‐associated vascular disease are complex and related to the immune dysregulation [30, 31]. Immune stimulus activated pDCs are potential participants in the SLE‐associated vascular injury through inflammatory processes [4, 32]. Our present study here showed that ECs activated by pDCs‐mediated immune stimuli contributed to vascular injury via TLR/IFN‐α‐mediated inflammatory processes. These processes were related to the PI3K and MAPK signalling pathways.

pDCs are the main producers of type I IFN in humans and mice [5]. The elevated secretion of IFN‐α in human body is considered as an inducement of SLE and a pathogenesis of vasculopathy [7, 9, 10]. Type I IFN impairs the proliferation of EPCs and accelerates atherosclerosis in mouse with lupus‐like disease [15]. Here, we confirmed the significant elevation of IFN‐α and the activation of TLR7/9/IRF7/IFN‐α signalling in activated pDCs. The proliferation, migration and formation of vascular tubules in ECs were inhibited by activated pDCs. The elevated production of IFN‐α also promoted the apoptosis and ageing and inhibited the expression of chemotactic cytokines. IFN‐α increased ICAM1 expression and therefore increased the adhesion ability of EPCs. The fact that the inhibition of IFN‐α induced decrement in inflammation and injury in ECs/EPCs and cardiovascular diseases suggests that IFN‐α blockage may be a potential therapeutic management for SLE‐induced vascular injury.

The production of type I IFN in pDCs depends upon PI3K activation, which is activated by TLR/IRF7 signalling in primary human pDCs [5, 33, 34, 35]. PI3K pathway is involved in a variety of biological processes including cell proliferation and survival, B‐ and T‐cell receptor signalling and the activation of chemokine receptors [5, 36, 37, 38]. MAPK signalling is crucial for the IRF7 production, maturation and activation of pDCs [24, 25]. The inhibition of PI3K has been reported to be associated with decreased inflammatory status after artery injury [39] and the CXCR3‐mediated chemotactic response in human airway epithelial cells [40]. Guiducci et al. [5] showed that PI3Kδ was essential for TLR‐mediated IRF7 nuclear translocation and type I IFN production in activated human pDCs. However, the block of PI3K did not affect TLR7/9‐stimulated inflammatory cytokines (including IL‐6 and TNF‐α) or maturation of pDCs. Wang et al showed that p38 MAPK enhanced IRF7 nuclear translocation and IFN‐I expression in pDCs [24]. We demonstrated that PI3K inactivation and MAPK activation were associated with pDC activation. The complex mechanism of type I IFN production in pDCs and ECs/EPCs injury induced by activated pDCs may be associated with PI3K/MAPK switch.

Elevated level of circulating ECs is a proxy for diverse inflammatory diseases with vascular injury [4, 11, 13]. The vascular inflammatory response involves interaction between vascular ECs and inflammatory cells (including monocytes, neutrophils and macrophages). When undergoing inflammatory activation, ECs expressed adhesion molecules including ICAM1 and therefore increased the vascular damage [6, 41, 42]. In addition, elevated blood and urinary ICAM1 is a biomarker for SLE [12]. ICAM1 is an efferocytosis receptor in macrophages [43]. Our present study showed that the adhesion of EPCs was increased by co‐culturing with activated pDCs, accompanied by elevated expression of ICAM1, which was confirmed in aorta of SLE model mouse. These results indicated the increased vascular injury may be mainly resulted from inflammatory status in patients or animal model of SLE.

IFN‐α plays a critical role in the aetiopathogenesis of SLE, making it serve as a promising therapeutic target for the treatment of SLE [44]. Enormous antibodies against IFN‐α or IFN‐α receptor are shown to be associated with decreased disease activity [20]. A previous report found that oral treatment of anifrolumab, an anti‐IFN‐α receptor monoclonal antibody, substantially reduced disease activity compared with placebo across multiple clinical end‐points in the patients with moderate‐to‐severe SLE [45]. Another anti‐IFN‐α antibody, sifalimumab, shows dose‐dependent inhibition of IFN‐I in SLE patients, and the level of IFN‐related proteins in affected skin was also decreased [46]. Here, we showed that oral treatment of IFN‐α‐NA significantly suppressed vascular damage in pristane‐induced SLE mouse model, suggesting that IFN‐α‐NA might be a potential clinical implication for preventing vascular injury.

CONCLUSIONS

In conclusion, we demonstrated that IFN‐α elevation was a crucial mechanism for vascular or ECs/EPC damage induced by activated pDCs. Activated pDCs induced IFN‐α production and inhibited cell proliferation, angiogenesis, migration and PI3K signalling, but activated MAPK pathway in ECs. In addition, IFN‐α elevation increased chemokine cytokines (CXCL10 and CXCR3), EPC adhesion ability, apoptosis, ageing and G₀/G₁ cell cycle arresting. The inhibition of IFN‐α, however, suppressed inflammation in ECs/EPCs, apoptosis, and ageing of EPCs and rescued the proliferation and migration of ECs. Our study demonstrated that inhibition of IFN‐α rescued vascular endothelium from pristane‐induced IFN‐α production and endothelial cell ageing, suggesting IFN‐α NA might be a potential clinical implication for the treatment of SLE.

CONFLICT OF INTEREST

The authors have no commercial or other associations that might pose a conflict of interest.

AUTHORS’ CONTRIBUTIONS

XW‐D, XJ‐H and W‐X conceived and designed the experiments. XW‐D, Y‐R and XY‐H performed the experiments. XW‐D, QY‐L and W‐X analysed the data. XW‐D wrote the manuscript.

ETHICAL APPROVAL AND CONSENT TO PARTICIPATE

All animal protocols in this study were approved by the ethics committee of the Institutional Animal Care and Use Committee (IACUC), the Second Xiangya Hospital, Central South University, China (No. 2020526), and were performed in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals.

CONSENT FOR PUBLICATION

Not applicable.

AVAILABILITY OF SUPPORTING DATA

All data generated or analysed during this study are included in this published article.

Supporting information

Fig S1

Ding X, Xiang W, Yi R, Huang X, Lin Q, He X. Neutralizing interferon‐α blocks inflammation‐mediated vascular injury via PI3K and AMPK in systemic lupus erythematosus. Immunology. 2021;164:372–385. 10.1111/imm.13379