SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

Peripheral blood mononuclear cells (PBMCs) from patients with systemic lupus erythematosus (SLE) have increased expression of genes typically induced by type I interferon (IFN). However, it has been difficult to identify and quantify the factors responsible for activation of the IFN pathway in SLE. To characterize these mediators, we developed an assay that measures the functional effects of plasma or serum components on the gene expression of cultured target cells.

Methods

WISH epithelial cell line cells were cultured with medium, with recombinant IFNα, IFNβ, or IFNγ, or with 50% plasma from SLE patients (n = 73), rheumatoid arthritis (RA) patients (n = 19), or healthy donors (n = 30). Real-time quantitative polymerase chain reaction was used to determine WISH cell expression of IFN target genes, including PRKR, IFIT1, IFI44, MX1, and C1orf29 (preferentially induced by IFNα) and CXCL9 (Mig) (preferentially induced by IFNγ).

Results

IFNα-regulated genes were induced by SLE plasma samples, but not by most of the RA or healthy control plasma samples. The activity in SLE plasma was inhibited >90% by anti-IFNα antibody, but not by anti-IFNβ or anti-IFNγ antibodies. The expression of each IFNα target gene induced by SLE plasma correlated with the expression of that gene studied ex vivo in PBMCs from the same patients and with the titer of anti–RNA binding protein (anti-RBP)–specific autoantibodies. Plasma activity paralleled PBMC expression of IFNα-inducible genes over time.

Conclusion

IFNα in SLE plasma is a major stimulus of IFN target gene expression and is related to expression of those genes in PBMCs from SLE patients and to the titer of anti-RBP autoantibodies. These data provide additional support for the view that IFNα mediates immune system activation and dysregulation in SLE.

Systemic lupus erythematosus (SLE) is a multisystem disease characterized by autoimmunity directed toward nuclear particles and profound alterations of the immune system that contribute to inflammation and tissue damage. The production of pathogenic autoantibodies in SLE is T cell and antigen dependent. A current paradigm suggests that apoptotic cells provide a source of self antigens, but it is not clear why some individuals generate an immune response to those self antigens and some do not. Investigators in our laboratory are exploring the hypothesis that production of type I interferon (IFN) contributes to immune dysregulation and autoimmunity in SLE.

A role of IFNα, the prototype type I IFN, in SLE has been suggested for at least 25 years. Importantly, increased levels of IFNα have been detected in the serum of patients with active SLE (1, 2). In addition, recombinant IFNα administered to patients with viral infection or malignancy can induce autoantibodies with specificities similar to those seen in lupus patients and has sometimes triggered clinical lupus, thyroiditis, or arthritis (3, 4). Lupus serum has been shown to increase allogeneic stimulatory activity of monocytes, an effect that was partially inhibited by anti-IFNα antibody, suggesting that IFNα may promote increased antigen-presenting cell function and T cell activation (5). Recent microarray and real-time polymerase chain reaction (PCR) analyses of peripheral blood mononuclear cells (PBMCs) from lupus patients have demonstrated increased expression of a broad spectrum of IFN-inducible genes (6–13).

Although these data collectively suggest a role of IFNα in the activation of genes and immune functions that are dysregulated in SLE, it has been difficult to document a direct role of IFNα in the expression of genes that are overexpressed in PBMCs from lupus patients (6, 7). Measurement of IFNα by enzyme-linked immunosorbent assay (ELISA) has not been sufficiently quantitative, and data from assays that depend on activation of an IFN-responsive promoter construct have not been compared with data quantifying PBMC gene expression (14).

While a direct contribution of IFN to the “IFN signature” seen in microarray studies seems likely, the IFN family includes several type I isoforms in addition to IFNα, including IFNβ, IFNκ, IFNτ, and IFNω, as well as type III IFN (IFNλ) and type II IFN (IFNγ) (15–17). Although IFNα is the dominant type I IFN present in the setting of viral infections, it is possible that other isoforms are active in SLE. In addition, recent efforts to determine the factors that induce IFN in SLE have supported a role of immune complexes containing either DNA or RNA (18–23). Interaction of nucleic acid components of immune complexes with Toll-like receptors (TLRs) or RNA helicases could induce transcription of type I IFNs but might also directly induce the expression of some IFN target genes (24–26).

To begin to characterize the mediators most responsible for activation of the IFN pathway in SLE, we developed an assay that measures the functional effects of plasma or serum components on the gene expression of cultured target cells. Our data indicate that IFNα, rather than IFNβ or IFNγ, is present in many SLE plasma samples. Moreover, the IFN activity measured in lupus plasma samples shows a moderate correlation with the expression of messenger RNAs (mRNA) encoded by interferon-inducible genes (IFIGs) quantified in PBMCs from the same patients and a strong correlation with the titer of autoantibodies specific for RNA binding proteins (RBPs) present in those patients. Taken together, these data support a functional link between IFNα activity in lupus plasma, activation of IFIG in lupus PBMCs, and targeting of the lupus immune response to particles containing small RNAs and RBPs.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Patients and controls.

Seventy-three SLE patients, 19 rheumatoid arthritis (RA) patients, and 30 healthy donors were studied. Forty-eight SLE patients, 19 RA patients, and 28 healthy donors provided blood samples for the initial phase of the study. These donors of plasma samples were the same SLE patients, RA patients, and healthy donors who had previously been studied for IFIG expression in PBMCs (10, 11). Samples were obtained from some additional donors not previously studied, including 25 with SLE and 2 healthy subjects.

SLE and RA patients received followup care at the Hospital for Special Surgery (HSS) and met the American College of Rheumatology (formerly, the American Rheumatism Association) classification criteria for SLE or RA (27, 28). Most SLE patients were recruited through the HSS Autoimmune Disease Registry, and many were also studied in a concurrent study of accelerated atherosclerosis in SLE (29). The clinical characteristics and medical therapies of the lupus patients have previously been described (11). Study subjects signed an informed consent form approved by the Institutional Review Board of HSS that described laboratory investigation of patient material as well as review of concurrent and previous clinical and laboratory data. Some SLE patients were tested on multiple occasions.

Plasma samples.

Twenty milliliters of heparinized blood was centrifuged, and the plasma was removed and stored at –70°C. Clinical data were available on all SLE patients. Complete serologic data were available for 59 SLE patients (43 from the initial cohort and 16 new SLE patients), and a composite score for the titer of autoantibodies specific for RBPs (anti-RBP antibodies, including anti-Ro, anti-La, anti-Sm, and anti-RNP) was calculated.

WISH cell culture and stimulation.

Human WISH epithelial cell line cells (product no. CCL-25; American Type Culture Collection, Manassas, VA) were grown in minimum essential medium supplemented with L-glutamine (2 mM), HEPES (20 mM), penicillin (100 units/ml), streptomycin (100 μg/ml), and 10% fetal bovine serum at 37°C in an atmosphere containing 5% CO2. To measure IFIG-inducing activity in patient plasma, WISH cells were cultured at a density of 0.5 × 105/0.1 ml in 96-well flat-bottomed plates containing medium, recombinant human IFNα (rHuIFNα) (IFNαA; BioSource International, Camarillo, CA), rHuIFNβ, rHuIFNγ, or rHuIFNω (R&D Systems, Minneapolis, MN), or 50% donor plasma or serum, in the presence or absence of neutralizing antibodies to IFNα (polyclonal antibody [PBL Biomedical, Piscataway, NJ]; monoclonal antibody [Chemicon, Temecula, CA]), IFNβ, or IFNγ, or isotype control antibodies (R&D Systems). After 6 or 20 hours of incubation, WISH cells were lysed and stored at –70°C.

Real-time quantitative PCR.

RNA was extracted from each cell lysate using the RNeasy Mini kit (Qiagen, Chatsworth, CA), and 0.4 μg of this RNA was reverse-transcribed to complementary DNA (cDNA) in a 20-μl reaction using SuperScript III RNase H reverse transcriptase (Invitrogen, Carlsbad, CA). The cDNA obtained from each sample was diluted 1:40, and 10 μl was amplified in a 25-μl real-time quantitative PCR reaction using 0.4 μM sense and antisense primers and the 2× iQ SYBR Green Supermix (Bio-Rad, Hercules, CA). Hypoxanthine guanine phosphoribosyltransferase 1 (HPRT-1) was used as a housekeeping gene control.

Primer sequences for the 5 IFIGs and for HPRT1 were as follows: for IFIT1, 5′-CTCCTTGGGTTCGTCTATAAATTG-3′ (forward) and 5′-AGTCAGCAGCCAGTCTCAG-3′ (reverse); for IFI44, 5′-CTCGGTGGTTAGCAATTATTCCTC-3′ (forward) and 5′-AGCCCATAGCATTCGTCTCAG-3′ (reverse); for PRKR, 5′-CTTCCATCTGACTCAGGTTT-3′ (forward) and 5′-TGCTTCTGACGGTATGTATTA-3′ (reverse); for C1orf29, 5′-AATCAGACAGAACAGTTAATCCTC-3′ (forward) and 5′-TCAACCATATCTTCAATGCTACC-3′ (reverse); for MX1, 5′-TACCAGGACTACGAGATTG-3′ (forward) and 5′-TGCCAGGAAGGTCTATTAG-3′ (reverse); for Mig, 5′-CATCATCTTGCTGGTTCTG-3′ (forward) and 5′-AGGATTGTAGGTGGATAGTC-3′ (reverse); and for HPRT1, 5′-TTGGTCAGGCAGTATAATCC-3′ (forward) and 5′-GGGCATATCCTACAACAAAC-3′ (reverse).

WISH cells cultured with medium were included in each assay to provide a basis for normalization across experiments. Results for each culture condition are reported as the relative expression compared with WISH cells cultured with medium. Details of the real-time quantitative PCR method and data analysis have been described in detail previously (10, 30).

Statistical analysis.

Two-group comparisons of continuous data that had a normal distribution were assessed using t-tests. The Kruskal-Wallis nonparametric analysis was used to compare the 3 study groups for the values of the 5 IFIGs because the data were not normally distributed. Correlation and linear regression analyses were performed to detect relationships between WISH cell data and IFN scores, anti-RBP autoantibody titers, or anti–double-stranded DNA (anti-dsDNA) autoantibody titers.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Time course and dose response for the induction of IFIG in WISH cells by rHuIFN and SLE plasma.

To develop an in vitro assay for quantification of IFIG-inducing activity in patient plasma or serum, WISH cell line cells, which were previously demonstrated to be IFN-responsive (31), were cultured with rHuIFNα, rHuIFNγ, or SLE plasma. The expression of 5 genes that had been selected for preferential induction in PBMCs by IFNα in our previous studies was measured in cultured WISH cells. The mRNA for IFIT1, IFI44, PRKR, C1orf29, and MX1 were preferentially induced by IFNα, with MX1 being the most sensitive to stimulation with rHuIFNα (Table 1). Both IFIT1 and MX1 showed rapid expression after 6 hours of WISH cell culture with rHuIFNα (Figure 1A), and rHuIFNω and rHuIFNβ (other type I IFNs) induced these IFIGs with similar kinetics (data not shown). IFIT1 and MX1 were induced in WISH cells by rHuIFNα in a dose-related manner (Figure 1B). One IFIG, CXCL9 (Mig), was strongly induced in WISH cells by rHuIFNγ, but was not induced by rHuIFNα, after a 20-hour incubation period (Table 1).

Table 1. Genes preferentially induced in WISH cells by IFNα or IFNγ*
 IFIT1IFI44PRKRC1orf29MX1Mig
  • *

    Recombinant human interferon-α (IFNα) or IFNγ (500 units/ml) was incubated with WISH cells for 20 hours, and the expression of interferon-inducing genes was determined by real-time quantitative polymerase chain reaction. Values are the relative expression compared with WISH cells cultured with medium alone. Results are from 1 of 3 similar experiments.

IFNα 500 units/ml12.793.014.4928.92138.20.59
IFNγ 500 units/ml2.792.871.3711.1423.03995.73
thumbnail image

Figure 1. Time course and dose response of interferon-α (IFNα) target gene expression induced by recombinant human IFNα (rHuIFNα) or systemic lupus erythematosus (SLE) plasma. A, WISH cells were incubated for 6 hours or 24 hours with 100 units/ml of rHuIFNα or SLE plasma at 50% volume/volume concentration, and the expression of IFN-inducible genes (IFIGs) was determined by real-time quantitative polymerase chain reaction. Plasma data represent the same sample tested on 3 different occasions. Values are the mean and SEM relative expression (RE) of 2 IFNα-inducible genes (IFIT1 and MX1) from 3 similar experiments. M = medium. B, Three different concentrations of rHuIFNα (100 units/ml, 200 units/ml, and 500 units/ml) or 3 concentrations of plasma or serum from a lupus patient (10%, 25%, and 50%) were incubated with WISH cells for 6 hours, and IFIG expression was assayed. Values are the mean relative expression of 2 IFNα-inducible genes (IFIT1 and MX1) from 1 of 2 similar experiments.

Download figure to PowerPoint

We then cultured WISH cells with SLE patient plasma and again measured IFIG expression. SLE plasma or serum used at 50% volume/volume concentration induced IFIT1 and MX1 with kinetics similar to that seen with rHuIFNα and with activity at a level similar to that of 100 units/ml of rHuIFNα (Figures 1A and B). Lupus plasma and serum induced IFIG expression to a similar degree (Figure 1B). The IFIG-inducing activity of recombinant IFNα and that present in SLE sera were heat-labile, with activity reduced by 62–99% after incubation at 56°C for 30 minutes (data not shown). The results demonstrate that WISH epithelial cell line cells and real-time quantitative PCR can be used to assay components of patient plasma or serum that mediate activation of the IFN pathway.

Correlation of lupus plasma activity with expression of IFIG in SLE PBMCs.

To assess the level of IFIG-inducing activity in plasma samples from a diverse population of SLE patients, as well as from RA patients (rheumatic disease controls) and healthy subjects (healthy controls), WISH cells were cultured with medium or with plasma from 73 SLE patients, 19 RA patients, and 30 healthy donors, and IFIG expression was measured by real-time quantitative PCR. Five IFNα-inducible genes and 1 IFNγ-inducible gene were quantified. As shown in Figure 2, a substantial proportion of SLE patients (about one-third) demonstrated high plasma IFIG-inducing activity for all 5 IFNα-inducible genes tested. (Data for the IFNγ-inducible gene are not shown.) When expression of 1 IFIG in WISH cells cultured with SLE plasma was plotted against the expression of each of the other 4 IFIGs, a significant correlation was demonstrated in each case (r values ranging from 0.36 to 0.89; data not shown). When the 3 study groups were compared, the induction of the IFI44, C1orf29, and PRKR genes was significantly higher in SLE patients than in healthy donors (Figure 2). The mean relative expression of Mig induced by SLE plasma samples was <1, indicating minimal or no functional IFNγ activity in those samples (data not shown).

thumbnail image

Figure 2. Induction of interferon-α (IFNα) target gene expression by plasma from systemic lupus erythematosus (SLE) patients, but not from rheumatoid arthritis (RA) patients or healthy donors (HD). Plasma samples from SLE patients (n = 73), RA patients (n = 19), and healthy donors (n = 30) were incubated with WISH cells at a 50% volume/volume concentration for 20 hours. WISH cells were then lysed and used for RNA isolation, reverse transcription, and amplification by real-time quantitative polymerase chain reaction. Shown is the relative expression (RE) of 5 IFNα-induced genes (IFI44, C1orf29, PRKR, IFIT1, and MX1). Horizontal lines indicate the mean. P values were determined by Kruskal-Wallis nonparametric analysis.

Download figure to PowerPoint

The expression of each of these 5 IFNα-inducible genes in WISH cells cultured with SLE plasma was correlated with the level of mRNA encoded by IFNα-inducible genes in PBMCs collected on the same day from the same patients (expressed as an IFNα score) (Figure 3). The IFNα score was calculated based on real-time quantitative PCR analysis of the expression of 3 IFNα-inducible genes (IFIT1, IFI44, and PRKR) in SLE patient PBMCs, as described in detail in a previous publication (11). For each of the 5 IFNα-inducible genes, the P value was less than 0.0001, suggesting that much of the IFIG expression detected in SLE PBMCs may be attributed to stimuli present in patient plasma.

thumbnail image

Figure 3. Correlation between systemic lupus erythematosus (SLE) plasma activity and the interferon-α (IFNα) score in peripheral blood mononuclear cells (PBMCs). Relative expression (RE) of 5 IFNα-induced genes (IFI44, C1orf29, PRKR, IFIT1, and MX1) in WISH cells cultured with 50% SLE plasma was plotted against the IFNα score, previously assayed based on real-time quantitative polymerase chain reaction analysis of patient PBMCs. The r and P values were determined by Spearman's correlation analysis. The method for calculating the IFNα score, based on the expression of IFIT1, IFI44, and PRKR in PBMCs, has been described previously (10, 11).

Download figure to PowerPoint

Inhibition of SLE plasma activity with anti-IFNα antibodies.

To identify the components in SLE plasma responsible for IFIG expression in WISH cells, we first tested a rabbit polyclonal anti-IFNα antibody, as well as monoclonal anti-IFNα, anti-IFNβ, and anti-IFNγ antibodies, for inhibitory activity against the relevant recombinant IFNs. Polyclonal anti-IFNα antibody effectively inhibited the induction of IFIT1 and MX1 by rHuIFNα, as did monoclonal anti-IFNα antibody. Monoclonal anti-IFNβ and anti-IFNγ antibodies were also specific for IFNβ and IFNγ, respectively, and were only minimally cross-reactive with the other IFNs (data not shown).

We then assessed the capacity of those antibodies to inhibit the activity in SLE plasma samples. Both polyclonal and monoclonal anti-IFNα antibodies significantly inhibited IFIT1 expression induced by SLE plasma (Figure 4A) and nearly ablated MX1 expression (Figure 4B). Dose-response studies showed that 1 μg/ml of monoclonal anti-IFNα antibody resulted in >50% inhibition of the SLE plasma–induced IFIG expression (Figure 4C). In contrast to the results with anti-IFNα antibodies, anti-IFNβ and anti-IFNγ antibodies had minimal effect on WISH cell IFIG expression induced by SLE plasma (Figures 4A–C). In addition, preliminary experiments using a monoclonal antibody confirmed to be specific for IFNλ1 (IL-29), a member of the type III IFN family, showed no inhibition of IFIG-inducing activity in SLE plasma (data not shown).

thumbnail image

Figure 4. Inhibition of systemic lupus erythematosus (SLE) plasma activity by polyclonal and monoclonal anti–interferon-α (anti-IFNα) antibodies. A and B, WISH cells were incubated for 6 hours with 100 units/ml of recombinant human IFNα (rHuIFNα) or with 50% SLE plasma together with medium (M), 2.5 μg/ml of rabbit polyclonal anti-IFNα antibody (PαAb), 29.6 μg/ml of monoclonal anti-IFNα antibody (MαAb), 10 μg/ml of monoclonal anti-IFNβ antibody (MβAb), or 3 μg/ml of monoclonal anti-IFNγ antibody (MγAb), and the expression of IFN-inducible genes (IFIGs) (IFIT1 in A and MX1 in B) was determined. The doses used for all antibodies maximally inhibited the induction of IFIGs by 500 units/ml of their corresponding rHuIFN and at the same time showed only minimal cross-reactivity with the other IFNs. Values are the mean and SEM relative expression (RE) in 3 different SLE patients. = P < 0.05; ∗∗ = P < 0.01, by t-test. C, WISH cells were cultured with 50% SLE plasma in the presence of various concentrations of polyclonal anti-IFNα antibody, control rabbit IgG (Piso), monoclonal anti-IFNα antibody, monoclonal anti-IFNβ antibody, or mouse isotype control IgG1 (Miso), and the relative expression of IFIT1 was determined. Results are expressed as the percentage inhibition compared with the expression induced by 50% SLE plasma. Values are the mean and SEM of 3 lupus patients. D, WISH cells were preincubated with 10 μg/ml of cycloheximide (CHX) prior to addition of SLE plasma. After 24 hours of culture, IFIT1 expression was determined. Values are the mean and SEM of 5 experiments.

Download figure to PowerPoint

It was possible that inducers of IFNα, rather than IFNα itself, were indirectly responsible for WISH cell IFIG expression. To determine whether production of new protein was required for the expression of IFIGs in WISH cells cultured with SLE plasma, we precultured WISH cells in cycloheximide to inhibit protein synthesis, which was followed by culture with SLE plasma for 24 hours. Cycloheximide did not reduce the capacity of SLE plasma to stimulate IFIG expression in WISH cells, and in fact, it increased IFIG expression, suggesting that IFNα protein itself, rather than a stimulus for IFNα production, directly accounted for the activation of IFIGs in WISH cells by lupus plasma (Figure 4D).

Longitudinal study of IFNα activity in SLE plasma.

To determine whether IFN activity in SLE plasma varies over time and parallels the IFNα score, reflecting IFNα-inducible gene expression in PBMCs, we collected sequential plasma samples from 2 SLE patients over 5 or 7 months. The relative expression of IFNα-inducible genes induced in WISH cells by patient plasma varied over time and was associated with a similar pattern of IFNα-inducible gene expression in the patients' PBMCs, as reflected in the IFNα score (Figures 5A and B). Of the 5 IFIGs measured, MX1 most closely tracked the IFNα score. Documentation of the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) score suggested a relationship between IFIG-inducing plasma activity and the IFNα score and disease activity in the patient data shown in Figure 5A. Administration of intravenous pulse glucocorticoids ablated IFN pathway activity in that patient, as determined by both measures. Fluctuations in the dosage of prednisone or addition of hydroxychloroquine therapy may have complicated the interpretation of a relationship between plasma IFN activity and the SLEDAI score for the patient data shown in Figure 5B.

thumbnail image

Figure 5. Relationship between the interferon-α (IFNα) activity in systemic lupus erythematosus (SLE) patient plasma and the IFNα score in peripheral blood mononuclear cells (PBMCs) and the anti–RNA binding protein (anti-RBP) autoantibody titer. Blood samples from A, SLE patient 749 and B, SLE patient 775 were collected on the indicated dates in 2003. IFN activity in plasma was quantified in the WISH cell assay, the relative expression (RE) of 5 IFNα-inducible genes (MX1, C1orf29, IFI44, IFIT1, and PRKR) in plasma was determined, and the IFNα score was determined by analysis of PBMCs. On the same dates, the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) was determined, and the daily prednisone (Pred) dose was recorded, along with other medication changes. For patient 749, intravenous pulse glucocorticoid (ivGC) therapy was given as 1 gm of intravenous methylprednisolone 4 days prior to the last time point indicated (12/9/03). HCQ = hydroxychloroquine. C, and D, Relative expression of 2 IFNα-inducible genes (MX1 in C and C1orf29 in D) in WISH cells induced by 50% plasma from 59 SLE patients is plotted against the anti-RBP autoantibody titer in the same patients. The r and P values were determined by Spearman's correlation analysis.

Download figure to PowerPoint

Correlation of IFIG expression with anti-RBP–specific autoantibody titer.

We had previously observed that elevated IFIG expression in SLE PBMCs was highly associated with the presence of autoantibodies specific for RBPs (11). In the present study, we found a strong correlation between the level of expression of MX1 and C1orf29 mRNA in WISH cells cultured with SLE plasma and the aggregate anti-RBP autoantibody titer (titer of anti-Ro plus anti-La plus anti-Sm plus anti-RNP) in the serum of SLE patients (Figures 5C and D).

To strengthen support for a relationship between IFN activity and production of anti-RBP autoantibodies, WISH assay data were separately analyzed in the 43 patients who were previously studied and reported to show a significant association between a high IFNα score and the presence of anti-RBP autoantibodies (11). In that initial cohort, a strong correlation was observed between MX1 expression in the WISH assay and the anti-RBP titer (r = 0.6468, P < 0.001). To provide confirmation of that relationship in a second patient group, we analyzed data from 16 patients who were not previously studied. The correlation between MX1 mRNA expression induced by plasma in the WISH assay with the anti-RBP titer in this second patient cohort was equally significant (r = 0.6595, P < 0.001).

Our previous study of IFIG expression in PBMCs showed a negative correlation between the IFNα score and the serum C3 level. Similarly, expression of MX1 in WISH cells cultured with SLE plasma showed a negative correlation with serum C3 levels (r = –0.3564, P = 0.003).

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Data presented by many investigators over more than 25 years have built a strong case for an important role of IFNα in the pathogenesis of SLE (32, 33). Most recently, microarray and real-time quantitative PCR analyses of PBMCs from SLE patients have documented the activation of gene products responsive to type I IFN as a dominant molecular pathway in this disease (6–13). While one or more members of the type I IFN family are presumed to be responsible for the gene expression pattern observed, other classes of stimuli, including immune complexes comprising nucleic acids and lupus autoantibodies and other ligands of TLRs, are also candidate activators of IFN target genes (18–26).

Definitive characterization of the biologic factors that stimulate this gene expression signature in vivo has been challenging. Type I IFN mRNA is usually present at only trace levels in PBMCs from healthy individuals, and the cells that produce IFNα at relatively high levels, the plasmacytoid dendritic cells, are present at low frequency in healthy subjects and are apparently recruited from the peripheral blood to sites of immune system activity in lupus patients (34–36). Although blood and other body fluids might be expected to reflect the recent production of cytokines when studied ex vivo, current ELISAs have proved either insensitive or unreliable in the experience of several groups of investigators (refs.6 and7, and Hua, et al: unpublished observations). A modified immunoassay, termed the dissociation-enhanced lanthanide fluoroimmunoassay, has been used to demonstrate increased levels of IFNα in SLE serum but has not been shown to reflect the full complement of IFIG-activating IFNs or to provide significant correlative data with clinical or serologic features of disease (34).

We describe herein a new functional assay that allows characterization of the plasma components responsible for IFN pathway activation in SLE. The assay is based on the capacity of plasma or serum to induce IFIG expression in responsive target cells. Numerous reports in the literature have used WISH or other cell line cells to measure the effect of IFN on cell viability in the setting of virus infection (37). While sensitive, these viral assays do not allow direct comparison of IFN activity in plasma with activation of IFIG in PBMCs, nor do they permit analysis of other components of plasma, in addition to IFN, that might affect IFIG expression. The assay described herein, using real-time quantitative PCR analysis of human WISH epithelial cell line cells cultured with patient plasma, was shown to be sensitive and reliable for the detection of IFN activity in SLE plasma and serum. The WISH cell line expresses high levels of IFIG upon in vitro stimulation with IFNα, IFNβ, and IFNγ, and the level of this expression is dose related. Previous data from other investigators indicate that various cell lines are characterized by different sensitivities to distinct IFN isoforms and show variable patterns of IFN target gene expression (31). The WISH cell line responds similarly to rHuIFNα and to SLE plasma, supporting its relevance for the quantification of IFN in SLE patients. We have also shown that PBMCs from healthy controls can serve as targets for assay of plasma factors that induce IFIG (10).

Although the role of IFNα in lupus has been reported for many years and has been supported by abundant data, most of those studies have only investigated IFNα, considered the prototype type I IFN. In fact, the gene targets of the various IFNα isoforms, as well as those of IFNβ and IFNω, are largely similar. IFNλ has been less well studied than IFNα, but early reports suggest that it activates genes similar to IFNα and IFNβ (16, 17). Type II IFN, or IFNγ, may be expressed locally at sites of tissue inflammation, as in lupus nephritis, but high levels of circulating IFNγ have not been reproducibly observed. In fact, our earlier studies, in which IFIG expression was detected in SLE PBMCs, indicated that few lupus patients show evidence of increased expression of IFIG that are preferentially induced by IFNγ (10, 11).

The findings of the present study strongly support a major role of IFNα, since both polyclonal and monoclonal anti-IFNα antibodies inhibited much of the IFIG-inducing activity in SLE plasma. Monoclonal antibodies that we confirmed to be specific for IFNβ did not inhibit the lupus plasma activity, and preliminary data suggest that IFNλ does not contribute to the activation of WISH cells by SLE plasma. The role of IFNω could not be adequately tested because the commercially available antibody proved nonspecific. The expression of Mig, a gene highly sensitive to induction by IFNγ, was not increased in WISH cells cultured with lupus plasma samples, suggesting that if present at all, IFNγ levels were not sufficient to activate that very responsive gene.

Further supporting a major role for IFNα is the observation that plasma activity across the entire SLE patient population correlated with expression of IFNα-inducible genes in PBMCs from the same patients, as measured by an IFNα score. In addition, the plasma IFIG-inducing activity and PBMC IFNα score showed parallel patterns when measured in plasma and PBMC samples collected longitudinally and showed potential for demonstrating an association with quantitative measures of disease activity. IFNα appeared to be directly responsible for IFIG expression since the protein synthesis inhibitor, cycloheximide, did not abrogate the activity in lupus plasma. Together, these data demonstrate that the WISH assay system can be used to quantify and characterize in patient plasma the IFN isoforms most relevant to activation of the IFN pathway that occurs in vivo in many lupus patients and indicate that in this assay system, IFNα is the major stimulus responsible for IFN target gene expression. These results do not rule out additional contributions of other plasma factors, variability in the expression or function of cell surface receptors, or differences in the efficiency of intracellular signaling pathways to induce IFIG in SLE PBMCs.

In our earlier study of SLE PBMCs, logistic regression analysis of clinical and serologic data showed that the presence of autoantibodies specific for RBPs (Ro, La, Sm, or RNP) was independently associated with high expression of IFNα-inducible genes in lupus PBMCs. Consistent with the proposed functional link between plasma activity and IFIG expression in SLE PBMCs, IFIG expression in WISH cells cultured with SLE plasma was highly correlated with the titer of anti-RBP autoantibodies (the sum of the titers of the 4 component specificities). Among them, anti-Ro antibody showed the strongest correlation. This observation was further supported when WISH assay data were analyzed separately in 2 patient cohorts, those previously characterized in our studies of IFIG expression in PBMCs and a second cohort of newly recruited SLE patients. In both groups, MX1 expression induced by patient plasma was significantly correlated with anti-RBP autoantibody titer. In contrast to anti-RBP, the anti-dsDNA autoantibody titer showed no significant correlation with IFIG expression assayed in WISH cells induced by SLE plasma (data not shown).

The pathophysiologic mechanisms that underlie the correlation between plasma IFN activity and anti-RBP autoantibodies have not been clarified. We have proposed a potential role of innate immune system activation through TLRs that encounter single-stranded or double-stranded RNA, followed by activation of an adaptive immune response that becomes focused on the protein components of RNA-associated complexes (33). In recent studies showing the capacity of immune complexes in SLE sera to trigger IFNα production by dendritic cells, a contribution of DNA- or RNA-associated material derived from apoptotic or necrotic cells was required for induction of IFNα expression (22). The assay described here, using WISH cells as sensitive targets of type I IFNs in body fluids, appears to be highly useful for quantification of the activity of those IFNs in patients and their relationship to serologic and clinical features of disease. PBMCs may prove more useful as target cells for study of those components of plasma or serum, including cell debris, nucleic acids, and immune complexes, that contribute to the induction of type I IFN production.

In summary, we utilized a functional assay of IFN activity to demonstrate that IFNα is the IFN isoform most responsible for IFIG expression in SLE and that plasma IFN activity is highly correlated with anti-RBP autoantibody titer. Additional investigation is required to identify the upstream mediators responsible for the initial induction and maintenance of autoimmunity in SLE.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors thank all of the patients who participated in this study, as well as their physicians, who provided important clinical data and facilitated collection of blood samples. The authors also thank Dr. Jane Salmon and the coordinators of the Hospital for Special Surgery Autoimmune Disease Registry, who assisted with identification of candidate study patients and sample accession.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES