A total of 108 patients with SLE and 46 normal healthy donors were recruited from the Department of Rheumatology, RenJi Hospital, which is affiliated with Shanghai Jiao Tong University School of Medicine (Table 1). All SLE patients fulfilled the diagnostic criteria of the American College of Rheumatology. All of study participants signed a patient material and informed consent form, approved by Renji Hospital Institutional Review Board.

Human PBMCs were isolated by Ficolle-Hypaque density gradient separation. IFIT4 mRNA levels were detected by quantitative RT-PCR in total RNA, isolated from the PBMCs of 108 SLE patients and 46 healthy donors with Trizol reagent. IFIT4 protein expression levels were determined by Western blotting in protein samples from the PBMCs of 24 patients with SLE and 24 healthy control individuals.

Blood monocytes from three patients with SLE and three healthy control individuals were isolated using an immuno-magnetic bead method (Miltenyi Biotec, Bergisch Gladbach, Germany) and were used to examine the expression of IFIT4 protein via Western blotting. CD4⁺ lymphocytes were purified from PBMCs via an indirect magnetic labeling system, namely the CD4 T Cell Isolation Kit (Miltenyi Biotec, Bergisch Gladbach, Germany). Purities were generally in excess of 95%. Freshly isolated T lymphocytes were used to measure mixed leucocyte responses (MLRs).

For quantitative analysis of gene expression, total RNA was isolated using a Trizol reagent kit (Invitrogen, Carlsbad, CA, USA). cDNA was synthesized and fluorescence real-time RT-PCR was performed by using SuperScript™ III Platinum^® SYBR^® Green Two-Step qRT-PCR Kits (Invitrogen, Carlsbad, CA, USA) via the ABI PRISM 7900 system (Perkin-Elmer, Foster City, CA, USA). The following primers were used: IFIT4 forward: 5'-AACTACGCCTGGGTCTACTATCACTT-3'; IFIT4 reverse: 5'-GCCCTTTCATTTCTTCCACAC-3'; GAPDH forward: 5'-GAAGGTGAAGGTCGGAGTC-3'; GAPDH reverse: 5'-GAAGATGGTGATGGGATTTC-3'. All data were analyzed using ABI PRISM SDS 2.0 software (Applied Biosystems, Foster, CA, USA). Using the ΔCt method, GAPDH was co-amplified to normalize the amount of RNA 39.

Western blotting was performed as described previously 3439. Protein from PBMCs and monocytes were harvested and transferred to polyvinylidene fluoride membranes. The membranes were incubated with monoclonal anti-human IFIT4 antibody (BD Clontech, Palo Alto, CA, USA) followed by horseradish peroxidase-linked secondary antibodies (Cell Signaling, Beverly, MA, USA). Equal protein loading for Western blots was confirmed by immunoblotting for β-actin.

The cDNA fragment encoding IFIT4 was amplified by PCR from a plasmid template (ORFEXPRESS™ Gateway^® ORF Shuttle Clones; GeneCopeia, Frederick, MD, USA), into which the full-length human IFIT4 cDNA was inserted. PCR products were digested by HindIII restriction endonuclease and cloned into the pEGFP-C1 vector, generating plasmid pEGFP-IFIT4. An endotoxin-free kit (EndoFree Plasmid isolation Kit, Qiagen, Chatsworth, CA, USA) was used to purify the vectors. pEGFP-C1 was used as a negative control (BD Clontech, Palo Alto, CA, USA). For transient transfections, the plasmids mentioned above were transfected into THP-1 cells using Lipofectamine 2000 (Invitrogen, USA) or Cell Line Nucleofector Kits V (Amaxa Biosystems, Cologne, Germany).

To observe the subcellular location of IFIT4, THP-1 cells were transfected with pEGFP-IFIT4 or pEGFP-C1 empty vector. Forty eight hours later, cells were stimulated with 3000 μ/ml IFN-α for 3 days. The cells were examined under a confocal microscope (Leica TCS SP2, Leica Microsystems, Heidelberg, Germany), using DAPI for nuclear staining.

The human monocytic cell line THP-1 (ATCC, Manassas, VA, USA) 40 was cultured in RPMI 1640 (Hyclone, Logan, UT, USA) supplemented with 10% foetal bovine serum and 2 mmol/l L-glutamine at 37°C in 5% carbon dioxide. The following human recombinant cytokines were used as stimulators: recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF; 50 ng/ml), IL-4 (20 ng/ml; R&D Systems, Minneapolis, MN, USA), and lipopolysaccharide (LPS; 100 ng/ml, Escherichia coli 0111:B:4; Sigma, St. Louis, MO, USA).

To generate DC-like cells (DCLCs), THP-1 cells were treated with GM-CSF and IL-4 (GM-CSF/IL-4) 4041424344 at the indicated concentrations for up to 6 to 12 days. To generate IFIT4-primed DCLCs, THP-1 cells were transfected with pEGFP-IFIT4; 36 hours later the cells were stimulated with GM-CSF/IL-4 for up to 5 to 7 days. The DCLCs derived from THP-1 transfected with pEGFP-C1 were designated controls. Mature and active DCLCs were generated from THP-1 cells that were treated with GM-CSF/IL-4 for 6 to 12 days and further cultured with LPS for an additional 2 days. Cultures were fed every 2 or 3 days by removing half of the medium and adding fresh medium with full doses of cytokines.

To assess the effect of IFIT4 on monocyte differentiation, the morphology of DCLCs primed by transfection with pEGFP-C1 or pEGFP-IFIT4 was examined every day using an Olympus IX-70 inverted microscope

Purified mouse IgG₁, κ isotype control (Catalogue no: 400101); PE mouse IgG_2b, κ isotype control, PE anti-human CD86 (Catalogue no: 305406); CD83 (Catalogue no: 305308); HLA-DR (Catalogue no: 307606); CD1a (Catalogue no: 300106); and FITC anti-human CD14 (Catalogue no: 325604) were obtained from Biolegend (San Diego, CA, USA). FITC anti-human CD40 (Catalogue no: FAB617F) and PE anti-human CD80 (Catalogue no: FAB140P) were from R&D Systems, Minneapolis, MN, USA.

DCLCs were incubated with saturating concentrations of fluorochrome-conjugated monoclonal antibodies at 4°C for 30 minutes and then washed twice in phosphate-buffered saline containing 2% foetal bovine serum and fixed in 1% paraformaldehyde. Cells were analyzed with a FACSort (BD Biosciences, Mountain View, CA, USA). Appropriate fluorochrome or isotype control monoclonal antibodies of each antibody species were used as negative controls.

Mature and active DCLCs, generated as described above, were harvested and γ-irradiated (3,000 rads) followed by incubation with 5 × 10⁴ allogeneic CD4⁺ T cells/well, at ratios of DCLCs to T cells of 1:10, 1:20 and 1:40. Three days later, [³H]thymidine was added (0.5 μCi/well) and the cells were incubated for another 18 hours. The cells were harvested and the incorporated radioactivity measured using a β counter (model LS3801; Beckman Coulter, Brea, CA, USA). Responses were reported as the mean of triplicate samples (mean [counts/minute cpm] ± standard deviation).

Mature and active DCLCs were generated as described above. Total IL-12 (p40 and p70) secreted by DCLCs primed with pEGFP-IFIT4 or pEGFP-C1 transfection was measured with the BIOSOURCE IL-12+p40 ELISA kit (Biosource Europe S.A., Nivelles, Belgium) 45.

To examine the effect of IFIT4-primed DCLCs on T-cell polarization, T cells (2 × 10⁵/well) were plated and cultured with γ-irradiated IFIT4-primed DCLCs at a ratio of DCLCs to T cells of 1:10 for 6 days. 10 ng/ml of phorbol 12-myristate 13-acetate (PMA) was added and the cells cultured for 1 more day. The supernatants were harvested to measure the concentrations of IL-4 and IFN-γ by using the Human IL-4/IFN-γ ELISA Kit (Biosource Europe S.A.).

Two group comparisons of gene expression were assessed using unpaired t-test, or the nonparametric Mann-Whitney test when the data did not have a normal distribution 3184647. Results are presented as the mean ± standard deviation unless specified otherwise. Correlations of IFIT4 mRNA expression levels with titre of auto-antibodies and SLE Disease Activity Index (SLEDAI) scores were determined using Spearman's rank correlation coefficient 47.

A total of 108 SLE patients and 46 healthy donor individuals were matched for both age and sex (Table 1). The results of real-time quantitative RT-PCR revealed that the means of IFIT4 relative mRNA levels were 37.84 ± 3.52 in PBMCs from SLE patients and 10.58 ± 2.64 in those from healthy control individuals, and the difference was statistically significant (P < 0.001; Figure 1a). IFIT4 protein levels, examined using Western blotting, were significantly increased in PBMCs from 24 SLE patients than in those from the 24 healthy control individuals (P < 0.05; Figure 1b,c). In addition, IFIT4 protein levels in the monocytes of SLE patients were significantly increased compared with those in control individuals (P < 0.05; Figure 1d).

Spearman's correlation analysis was carried out to determine the relationship between IFIT4 expression and the clinical characteristics of SLE. We found that IFIT4 mRNA relative expression correlated with ANA titre in 108 SLE patients (r = 0.4783, P < 0.001; Figure 2a), and with anti-dsDNA autoantibody titre in 36 SLE patients (r = 0.3932, P < 0.05; Figure 2b), and with anti-Sm antibody titre in seven SLE patients (r = 0.9088, P < 0.01; Figure 2c). Moreover, SLE patients who were positive for anti-dsDNA autoantibodies had higher IFIT4 expression than did those who were negative (P = 0.0277; Figure 2d). The difference in IFIT4 expression between patients who were positive or negative for anti-Ro (anti-SSA; Figure 2e) or anti-aCL/β2-GP1 (Figure 2f) was not statistically significant (P > 0.05). Our findings suggest that higher IFIT4 expression is associated with a greater likelihood of having autoantibodies against ribonucleoproteins.

In the SLE population as a whole, we found that SLE patients with hypocomplementaemia or leucopenia or/and thrombocytopenia usually had higher IFIT4 mRNA relative expression than did SLE patients without these disorders (P = 0.0121, Figure 2g; P = 0.0301, Figure 2h). However, IFIT4 expression in PBMCs from SLE patients with or without nephritis was not statistically different (P > 0.05; Figure 2i). We found no correlation between IFIT4 gene expression and scores of the SLEDAI-2K in 108 SLE patients (r = 0.0574, P > 0.05; Figure 2j).

IFIT4 was examined using real-time quantitative RT-PCR and was found to be predominantly expressed in seven out of 14 normal human tissues, namely spleen, lung, leucocytes, lymph nodes, placenta, bone marrow and foetal liver (in order of the highest to the lowest), most of which belong to the immune system (P < 0.05; Figure 3a). Rather than being restricted to one immune cell, IFIT4 was expressed in many immune cells, including CD19⁺, CD8⁺, CD14⁺ and CD4⁺ cells (P < 0.05; Figure 3b).

To determine the the subcellular localization of IFIT4 protein, THP-1 cells were transfected with pEGFP-C1 or pEGFP-IFIT4 plasmid and examined by confocal microscopy (Figure 3c). In contrast to the pEGFP-C1 group, in which EGFP (enhanced green fluorescent protein) was dispersed throughout the cell (Figure 3c [subpanel a]), EGFP-IFIT4 fusion protein was specifically localized to the cytoplasm of THP-1 cells transfected with pEGFP-IFIT4 (Figure 3 [subpanel b]), and no translocation of IFIT4 into the nucleus was observed after 3 days of stimulation with IFN-α2a (Figure 3c [subpanel c]).

IFIT4 mRNA and protein in normal PBMCs was preferentially induced by IFN-α2a (3,000 units/ml), as observed using real-time quantitative RT-PCR, fluorescence-activated cell sorting (FACS), and Western blotting. The time course study revealed strong induction of IFIT4 mRNA by IFN-α2a as early as 24 hours after treatment (Figure 3d). FACS findings showed that IFIT4 protein was expressed in about 48.7% of normal PBMCs, with mean fluorescence intensity of 121.18. However, after exposure to IFN-α2a for 3 days, up to 99.47% PBMCs expressed IFIT4 protein with a mean fluorescence intensity of 327.97 (P < 0.05; Figure 3e). Upregulation of IFIT4 expression by IFN-α2a was confirmed by Western blotting (Figure 3f).

Because it is usually difficult to transfect primary monocytes, we chose an appropriate monocytic cell line, namely THP-1, which has a suspended, rounded appearance and is widely used as a model for monocyte-macrophage differentiation 4243444849. We confirmed that THP-1 cells have the potential to differentiate into mature DCLCs with various characteristics of DC morphology (data not shown) after 12 days of treatment with GM-CSF/IL-4 followed by stimulation with LPS for another 2 days. An effect of IFIT4 on monocyte differentiation into DCs in terms of morphological changes was observed. Upon treatment with GM-CSF/IL-4 for 86 hours, no DC appearance was observed in THP-1 cells transfected with pEGFP-C1 vector (Figure 4a). Specifically, THP-1 cells remained dispersed and rounded (left lane: ×20) and were similar in size to intact THP-1 cells (right lane: ×40). In contrast, THP-1 cells primed with pEGFP-IFIT4 transfection partially acquired DC morphology after as little as 48 hours of GM-CSF/IL-4 treatment (Figure 4b). These IFIT4-primed DCLCs not only aggregated into clusters and became half adherent (left lane: ×20) but they also developed cytoplasmic protrusions or dendrites around the cell surface and increased in size (right lane: ×40). Moreover, THP-1 cells primed with pEGFP-IFIT4 transfection and stimulated with GM-CSF/IL-4 for 6 to 8 days plus 2 days of LPS exhibited the distinct mature and active morphology of DCs, manifesting as numerous processes and long veils, and dendrites (Figure 4c, left lane). However, 12 days of stimulation with GM-CSF/IL-4 was necessary for THP-1 cells primed with pEGFP-C1 transfection to differentiate into mature DCLCs (Figure 4c, right lane). These findings indicate that IFIT4-primed DCLCs adapted DC morphology sooner and had a greater resemblance to DCs than did THP-1 transfected with control plasmid upon the same GM-CSF/IL-4 treatment.

We first assessed the expression background of related surface markers of THP-1 cells by FACS. We found that CD14 was expressed at high levels, whereas CD80, CD86, CD1a and CD1b were expressed at lower levels on the surface of intact THP-1 cells (Figure 5a), confirming that THP-1 cells were more consistent with a monocytic than a dendritic phenotype. To determine the effect of IFIT4 on DC differentiation in terms of cell phenotypic changes, THP-1 cells primed with pEGFP-C1 or pEGFP-IFIT4 transfection were treated with GM-CSF/IL-4 for 90 hours; the cell surface markers were analyzed by flow cytometry. We found that DCLCs primed with pEGFP-IFIT4 transfection expressed higher levels of CD40, CD80, CD86 and HLA-DR, but lower levels of CD14 than did those DCLCs primed with pEGFP-C1 transfection upon the same GM-CSF/IL-4 stimulation (P < 0.05; Figure 5b). There was no change in CD1a expression between the two groups (Figure 5b).

To examine further the effect of IFIT4 on modulating the ability of DCLCs to activate T-cell proliferation, an allogeneic MLR assay was conducted. Mature DCLCs primed with pEGFP-IFIT4 or pEGFP-C1 were generated with 6 days of GM-CSF/IL-4 stimulation and further treatment with 2 days of LPS, and then these DCLCs were cultured with allogeneic CD4⁺ cells with different ratios of DCLC to T cells. As shown in Figure 6a, the extent of T-cell proliferation induced by DCLCs primed with pEGFP-IFIT4 transfection was significantly higher than those induced by DCLCs primed with pEGFP-C1, indicating that IFIT4-primed DCLCs had a greater ability to induce T-cell proliferation than controls (Figure 6a; P < 0.05).

To determine the effect of IFIT4 on IL-12 secretion by activated DCLCs, THP-1 cells transfected with pEGFP-IFIT4 or pEGFP-C1 were stimulated with GM-CSF/IL-4 for 6 days and further treated with LPS for 2 days (1 × 10⁵ cells/ml) to generate mature, activated DCLCs. We found that the mature and activated IFIT4-primed DCLCs produced more IL-12 than did DCLCs primed with pEGFP-C1 transfection (Figure 6b; P < 0.05).

To address whether IFIT4 modulates the capacity of DCLCs to direct T-helper cell differentiation, mature and activated DCLCs were generated as described above. Then these mature DCLCs were γ-irradiated followed by incubation with T cells for 6 days. The ELISA findings revealed that T cells that were stimulated with IFIT4-primed mature DCLCs secreted more IFN-γ (about 2.4-fold) than did those stimulated with DCLCs primed with pEGFP-C1 transfection (P < 0.05; Figure 6c). We did not find that any of the DCLCs affected IL-4 secretion to a significant degree (Figure 6c; P > 0.05).