Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Arthritis Rheum. Author manuscript; available in PMC Sep 3, 2009.
Published in final edited form as:
PMCID: PMC2737264

Augmented Interferon-α Pathway Activation in Patients With Sjögren's Syndrome Treated With Etanercept



Recent clinical trials suggest that etanercept is ineffective in controlling Sjögren's syndrome (SS). To address the hypothesis that tumor necrosis factor blockade can result in increased levels of interferon-α (IFNα) and BAFF, we quantified those mediators in plasma from etanercept- and placebo-treated SS patients.


We studied plasma samples from 20 patients with SS treated with etanercept (25 mg twice weekly) or placebo in a 12-week, randomized, double-blind clinical trial. In addition, we studied plasma samples from 29 healthy controls. IFNα activity was determined by reporter cell assay, and BAFF levels were determined by enzyme-linked immunosorbent assay.


Baseline IFNα plasma activity and BAFF levels were increased in SS patients compared with healthy controls (mean ± SD IFNα plasma activity score 4.43 ± 2.60 versus 2.08 ± 0.91; P < 0.0001) (mean ± SD BAFF level 0.83 ± 0.27 ng/ml versus 0.60 ± 0.15 ng/ml; P = 0.008). A significant increase in IFNα activity was detected after 12 weeks of treatment in the etanercept group, but not in the placebo group (P = 0.04 and P = 0.58, respectively). Furthermore, a statistically significant increase in BAFF levels was noted in patients receiving etanercept, but not in those receiving placebo (P = 0.01 and P = 0.56, respectively). In vitro culture of control peripheral blood mononuclear cells with etanercept resulted in a dose-dependent increase in the expression of IFNα and the IFNα-inducible genes IFN-induced protein with tetratricopeptide repeats 1 and BAFF.


IFNα activity and BAFF levels are elevated in the plasma of patients with SS compared with healthy controls. Etanercept treatment exacerbates IFNα and BAFF overexpression, providing a possible explanation for the lack of efficacy of this agent in SS.

Sjögren's syndrome (SS) is a chronic autoimmune disorder characterized by lymphocytic infiltration and destruction of the exocrine glands resulting in oral and ocular dryness. B cell hyperactivity is the hallmark of the syndrome, manifested as hypergammaglobulinemia and production of rheumatoid factor (RF) and autoantibodies specific for Ro/SSA and La/SSB autoantigens (1). Several cytokines have been demonstrated to mediate B cell survival and function in SS. BAFF, a tumor necrosis factor (TNF) family ligand produced mainly by myeloid cells, is overexpressed in patients with SS and other autoimmune disorders, is a critical B cell survival factor, and is associated with the production of autoantibodies (2,3). In addition, activation of the interferon-α (IFNα) pathway has been implicated in the pathogenesis of both systemic lupus erythematosus (SLE) (48) and SS (9,10). In SLE, expression of IFN-inducible genes and plasma IFN activity are associated with the presence of anti-Ro/SSA antibodies (11,12). Of interest, IFNα has been associated with BAFF production, and SS salivary gland epithelial cells are particularly responsive to IFNα, resulting in expression of BAFF messenger RNA (mRNA) (13).

Treatment of SS is mainly empiric, with the goal of relieving symptoms (14). In efforts to identify a more effective therapeutic approach, and based on the demonstration of overexpression of TNF in plasma and in salivary gland biopsy samples from patients with primary SS and in animal models (15,16), pilot and randomized controlled trials of anti-TNF agents have been performed (1721). While an initial uncontrolled trial of infliximab therapy in primary SS showed promising results (17), 2 randomized placebo-controlled clinical trials demonstrated that anti-TNF therapy is without efficacy in this disease (19,20). The reason for the failure of TNF blockade in SS is not clear. Recently, in a small study of patients with systemic-onset juvenile idiopathic arthritis, anti-TNF therapy was associated with increased expression of IFNα-inducible genes (22). On the basis of these collective data, we hypothesized that failure of anti-TNF therapy in SS might be related to augmented IFNα pathway activation along with a subsequent increase in BAFF production. To investigate this possibility, we determined IFNα plasma activity and BAFF levels in patients with SS before and after treatment with etanercept or placebo. The effect of etanercept on the induction of IFNα, TNFα, and IFNα-inducible genes was also investigated in an in vitro system.

Patients and Methods


The study population included 20 patients with SS diagnosed according to the American-European Consensus Group criteria (23) who were assigned to treatment with etanercept (25 mg twice a week) or placebo in the setting of a 12-week, randomized, double-blind, placebo-controlled clinical trial (20). Clinical characteristics and detailed evaluation of sicca symptomatology of the study participants have been previously reported (20) (a summary of medical therapy is available at http://www.hss.edu/research-staff_crow-mary.asp). Briefly, no differences were detected in the clinical parameters tested, including subjective measures of oral or ocular sicca symptoms (by visual analog scale), results of the Schirmer I test, the van Bijsterveld score, or salivary flow (20). Three patients had secondary SS (2 with rheumatoid arthritis [RA] and 1 with CREST syndrome [calcinosis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly, telangiectasias]). Plasma samples that were available at baseline and at 12-week followup were studied. The healthy donor group consisted of 29 individuals, without any known illness, with ages similar to those of the patients.

Plasma type I IFN activity

Type I IFN activity was detected using a reporter cell assay (12). Cells of the WISH epithelial cell line (no. CCL-25; American Type Culture Collection, Manassas, VA) express the type I IFN receptor and are highly responsive to type I IFN. WISH cells were plated at a density of 5 × 105/ml in 96-well plates in minimum essential medium (Cellgro, Herndon, VA) with 10% fetal calf serum (FCS) and then were cultured with 50% patient plasma for 6 hours. Recombinant human IFNα (IFN-αA; BioSource International, Camarillo, CA) and media alone were used as positive and negative controls, respectively. In some cultures a monoclonal anti-IFNα antibody (10 μg/ml) (Chemicon, Temecula, CA) was added.

The WISH cells do not express significant levels of Toll-like receptors (data not shown), so immune complexes containing DNA or RNA in plasma should not confound the assay by causing IFNα generation in the WISH cells. In addition, preincubation of WISH cells with cycloheximide does not decrease the IFNα-induced gene response (12), suggesting that IFNα that is already present in the samples is driving the IFNα-induced gene expression by the reporter cells. The ability of plasma to cause IFN-induced gene expression is largely abrogated by the addition of monoclonal anti-IFNα antibody, confirming that IFNα is the major active type I IFN causing the IFN-induced gene expression (12). In addition, stimulation of WISH cells with other proinflammatory cytokines, such as interleukin-1 (IL-1), IL-6, and TNF, did not result in increased expression of IFNα-inducible genes, suggesting the specific regulation of these genes by type I IFN stimuli (data not shown).

In a large cohort of 142 individuals tested, the relative expression of IFN-induced protein with tetratricopeptide repeats 1 (IFIT-1) and that of RNA-dependent protein kinase (PKR) were highly correlated with each other (Spearman's r = 0.39, P < 0.0001), suggesting a coordinate expression of these genes (data not shown). Values for each of the 2 genes individually showed a strong correlation with the combined sum (Spearman's r = 0.79 for IFIT-1, Spearman's r = 0.84 for PKR, both P < 0.0001), confirming coordinate regulation of the 2 transcripts. The mean ± SD IFN plasma activity score in a group of 50 SLE patients was 6.76 ± 9.18, while the corresponding value for healthy controls was 2.08 ± 0.91.

Preparation of complementary DNA (cDNA)

Total cellular mRNA was purified from stimulated cells at the end of the culture period using the Qiagen Turbocapture oligo(dT)-coated 96-well plate system according to the manufacturer's protocol (Qiagen, Valencia, CA). Briefly, the cells were washed once with phosphate buffered saline and then lysed in lysis buffer. The lysates were applied to the oligo(dT)-coated wells and washed to remove genomic DNA and protein. The mRNA was then eluted from the oligo(dT)-coated wells by incubating the plate at 65°C for 5 minutes. Total cellular mRNA was reverse-transcribed to cDNA immediately following purification using the SuperScript III reverse transcriptase system from Invitrogen (Carlsbad, CA). Oligo(dT) primer was used to amplify mRNA, specifically, and an RNase inhibitor was included to prevent degradation.

Quantitative real-time polymerase chain reaction (PCR)

Quantitative real-time PCR was used to quantify specific cDNA using the Bio-Rad SYBR Green intercalating fluorophore system with a Bio-Rad iCycler thermocycler and fluorescence detector (Bio-Rad, Hercules, CA). The following primers for genes that are highly induced by type I IFN signaling (IFIT-1 and PKR) were used in the PCR on WISH cell–derived cDNA: for IFIT-1, 5′-CTCCTTGGGTTCGTCTATAAATTG-3′ (forward) and 5′-AGTCAGCAGCCAGTCTCAG-3′ (reverse); for PKR, 5′-CTTCCATCTGACTCAGGTTT-3′ (forward) and 5′-TGCTTCTGACGGTATGTATTA-3′ (reverse) (Operon, Huntsville, AL). The housekeeping gene GAPDH (5′-CAACGGATTTGGTCGTATT-3′ [forward primer] and 5′-GATGGCAACAATATCCACTT-3′ [reverse primer]) was also quantified in the cDNA samples to control for background gene expression.

Threshold values were recorded for each sample at the logarithmic portion of the amplification. Melting curve analysis was used to ensure the specificity of the PCR product. Standard curves using known dilutions of cDNA were generated to control for differing efficiency of the PCR at different substrate concentrations. Expression of the type I IFN–induced genes was compared with housekeeping gene expression to determine relative expression. The relative expression was then normalized to the relative expression of the respective genes in unstimulated cells from the same population. The IFNα plasma activity score is expressed as the sum of the relative expression of the genes tested.

Peripheral blood mononuclear cell (PBMC) cultures

PBMCs were isolated from healthy donors using Ficoll-Hypaque (Pharmacia, Uppsala, Sweden). The cells were plated at a density of 2 × 105/well (100 μl/well) in 96-well plates in RPMI 1640 and 10% FCS and cultured for 6 hours with either recombinant IFNα (100–400 IU) or the soluble TNFα receptor etanercept (Enbrel; Wyeth, Collegeville, PA) at varying concentrations (range 0.125–0.50 ng/liter). These concentrations are consistent with the serum concentrations of etanercept-treated patients, as previously reported (24). Unstimulated PBMCs from the same pool of cells served as the negative control. After 6 hours, the cells were lysed, mRNA was purified, and cDNA was transcribed using the methods described above. Quantitative real-time PCR was performed on the cDNA samples using primers for IFIT-1 (see above), IFNα (IFNα2) (5′-CTTGAAGGACAGACATGAC-3′ [forward] and 5′-TGTGCTGAAGAGATTGAAG-3′ [reverse]), TNFα (5′-AGGTCTACTTTGGGATCATTG-3′ [forward] and 5′-GGGGTAATAAAGGGATTGGG-3′ [reverse]), and BAFF (5′-AGTTCAAGTAGTGATATGGATG-3′ [forward] and 5′-GGAGGATGGAAACACAC-3′ [reverse]).

BAFF, RF, and immunoglobulin assays

BAFF levels were measured using a polyclonal human BAFF enzyme-linked immunosorbent assay (ELISA) kit according to the manufacturer's protocol (Bender MedSystems, Burlingame, CA). Briefly, patient samples were loaded into polyclonal anti-BAFF antibody–coated wells, incubated for 3 hours, and then washed with a wash buffer. Biotin-conjugated polyclonal anti-BAFF was then added, followed by streptavidin–horseradish peroxidase to detect bound BAFF. Each sample was tested in duplicate, and standard wells provided by the manufacturer were used to generate a standard curve to estimate the BAFF concentration. According to the manufacturer's instructions, no cross-reactivity between other proteins and BAFF has been reported. Total serum IgG, IgM, and RF levels were also detected and quantified using nephelometry (Beckman Image; Beckman Coulter, Fullerton, CA). Anti-Ro/SSA antibodies were determined by commercial ELISA. The presence of anti–double-stranded DNA (anti-dsDNA) antibody was determined by Crithidia luciliae assay (Zeus Scientific, Raritan, NJ).

Statistical analysis

Mean values of IFNα plasma activity scores and plasma BAFF levels of SS patients and healthy controls were compared using the Mann-Whitney test. Mean values between etanercept and placebo groups prior to and after therapy were compared using paired t-tests. Correlation between changes in IFNα plasma activity and BAFF levels after 3 months of treatment was determined using the nonparametric Spearman's test. For the in vitro experiments, mean values of relative expression between unstimulated and stimulated cells were compared using paired t-tests.


Increased IFNα plasma activity in patients with SS

To establish the baseline level of type I IFN activity in the plasma of patients with SS, the WISH epithelial cell line assay was used to determine the capacity of plasma to induce expression of IFN-regulated genes. As shown in Figure 1, type I IFN plasma activity in 20 SS patients was increased compared with that in 29 healthy controls (mean ± SD baseline value 4.43 ± 2.60 versus 2.08 ± 0.91; P < 0.0001). After addition of anti-IFNα antibody (10 μg/ml) to the samples tested in the WISH assay, type I IFN plasma activity was nearly abolished (mean ± SD 4.43 ± 2.60 at baseline versus 1.76 ± 0.54 after addition of anti-IFNα; P < 0.0001), indicating that the ability of SS plasma to stimulate expression of type I IFN–inducible genes is predominantly due to the presence of IFNα.

Figure 1
Type I interferon (IFN) plasma activity in patients with primary Sjögren's syndrome (SS) compared with that in healthy controls (HD). Plasma type I IFN activity was quantified using the WISH epithelial cell line as described in Patients and Methods. ...

IFNα plasma activity in SS patients before and after treatment with etanercept or placebo

Figure 2 demonstrates the IFNα plasma activity in the etanercept- and placebo-treated SS patient groups before and after 3 months of therapy. A statistically significant increase in IFNα plasma activity was seen in patients who received etanercept treatment (mean ± SD 4.12 ± 1.77 at baseline versus 7.46 ± 5.34 at 3 months; P = 0.04). No difference in IFNα plasma activity after 3 months was observed in patients treated with placebo (mean ± SD 4.75 ± 3.30 at baseline versus 3.95 ± 3.59 at 3 months; P = 0.58).

Figure 2
Plasma type I IFN activity in patients with primary SS treated with etanercept or placebo. Plasma was assayed for type I IFN activity in the WISH assay. A statistically significant increase in IFNα plasma activity was observed at the 3-month time ...

BAFF levels in plasma of SS patients before and after treatment with etanercept or placebo

Since IFNα levels were elevated in SS patients before treatment and were further increased by etanercept, we next determined levels of an IFN-inducible gene product, BAFF. To demonstrate a relationship between TNF blockade and BAFF expression in SS patients treated with etanercept, plasma BAFF was measured by ELISA at baseline and at the 3-month time point. Figure 3 shows increased serum concentrations of BAFF in 20 SS patients at baseline compared with 17 healthy controls (mean ± SD 0.83 ± 0.27 ng/ml versus 0.60 ± 0.15 ng/ml; P = 0.008).

Figure 3
Plasma levels of BAFF in patients with primary SS compared with those in healthy controls. BAFF levels in plasma samples from patients with primary SS and healthy controls were assayed by enzyme-linked immunosorbent assay. Horizontal bars show the means. ...

Figure 4 illustrates BAFF concentrations before and after treatment with etanercept or placebo. A statistically significant increase in BAFF levels was observed in patients after 3 months of etanercept treatment (mean ± SD 0.81 ± 0.30 ng/ml at baseline versus 1.10 ± 0.23 ng/ml at 3 months; P = 0.01), while BAFF levels were not significantly changed by placebo treatment (mean ± SD 0.84 ± 0.25 ng/ml at baseline versus 0.92 ± 0.31 ng/ml at 3 months; P = 0.56).

Figure 4
Plasma levels of BAFF in patients with primary Sjögren's syndrome (SS) treated with etanercept. A statistically significant increase in serum levels of BAFF was noted in patients with primary SS after 3 months of treatment with etanercept compared ...

Increased expression of BAFF has been documented in SS salivary glands, and both IFNα and IFNγ have been shown to induce BAFF in salivary gland epithelial cells (13,25). Together with these reports, our data suggest a functional relationship between the increased levels of IFNα and BAFF observed after TNF inhibition, although additional factors are likely to be involved in BAFF regulation. This point is supported by the significant correlation between the change in IFNα plasma activity and the change in BAFF levels over 3 months of etanercept or placebo treatment (r = 0.58, P = 0.008) (Figure 5A).

Figure 5
Correlation between change in plasma type I interferon (IFN) activity and BAFF levels in all study subjects. A, A statistically significant correlation was detected between change in type I IFN plasma activity and BAFF levels in plasma of study subjects ...

Effect of TNF inhibition on total serum IgG and IgM levels and plasma autoantibodies

Both IFNα and BAFF are known to promote B cell survival and Ig class switching (2528). To determine whether IgG levels were preferentially increased in the etanercept-treated patients compared with the placebo-treated patients, total plasma IgG levels were determined by ELISA in baseline and 3-month samples. IgG levels increased at least 10% from baseline values in 6 of 10 etanercept-treated patients compared with 2 of 10 placebo-treated patients, although the difference did not reach statistical significance (median change 25 mg/dl in the etanercept group versus −30 mg/dl in the placebo group; P = 0.35). However, a statistically significant correlation was observed between changes in type I IFN plasma activity and IgG levels over the 3-month period in all study participants (r = 0.50, P = 0.02) (Figure 5B), while no statistically significant correlation was observed between changes in BAFF and IgG levels (r = 0.20, P = 0.39) or between BAFF and RF levels (r = −0.29, P = 0.08). In addition, no statistically significant differences in IgM or RF levels were observed between the etanercept and placebo groups (data not shown).

New anti-dsDNA antibodies developed in 1 patient in each of the treatment groups during the 3-month study. Salivary gland tissue antibody levels were not evaluated, but the level of systemic anti-Ro/SSA antibodies was not different at baseline in the 2 groups and did not change significantly during the brief 3-month study (illustrations of these findings are available online at http://www.hss.edu/research-staff_crow-mary.asp).

Effect of IFNα on BAFF production in vitro

In light of the observation that changes in plasma IFNα activity and BAFF levels were significantly correlated in the SS patients (Figure 5A), we wished to confirm the capacity of IFNα to directly mediate increased expression of BAFF mRNA. When PBMCs were isolated from healthy individuals and cultured with medium alone or with increasing concentrations of IFNα, a dose-related induction of BAFF mRNA expression was noted after 6-hour culture (Figure 6A). These in vitro studies support our in vivo evidence that increased expression of IFNα in SS patients, as well as the heightened levels of this cytokine after etanercept treatment, are linked to the elevated BAFF levels that we detected in the plasma of the study patients.

Figure 6
Etanercept (ETN) induces interferon-α (IFNα) and IFN-inducible genes and suppresses tumor necrosis factor α (TNFα) production in vitro. A, Peripheral blood mononuclear cells (PBMCs) were cultured for 6 hours with varying ...

Effect of TNF inhibition on type I IFN pathway activation and on BAFF and TNF production in vitro

In order to provide support for the hypothesis of a functional relationship between TNF inhibition and increased IFN pathway activation and BAFF production, we performed in vitro experiments in which healthy control PBMCs were cultured with etanercept to inhibit the activity of any endogenous TNFα produced during the culture period. Consistent with the in vivo data, and as shown in Figure 6B, a dose-related increase of IFNα mRNA transcripts was noted after 6-hour culture of PBMCs from healthy controls with varying concentrations of etanercept. In contrast, TNFα mRNA production was inhibited (Figure 6B). Figure 6C shows a dose-related induction of the IFN-inducible genes BAFF and IFIT-1 under the same culture conditions. When varying doses of recombinant TNFα (0.36, 0.72, and 3.6 ng/ml) were included in the cultures containing PBMCs and etanercept (0.25 ng/liter), IFIT-1 expression was inhibited by 52.7%, 55.5%, and 71.8%, respectively, and BAFF expression was inhibited by 69.6%, 83.7%, and 49.8%, respectively (data not shown).

It should be noted that in our in vitro experiments, further increases in IFNα and etanercept dose were associated with a decrease in the magnitude of response, resulting in a bell-shaped dose-response curve, a common pattern in biologic systems. This could be related to cell toxicity, a receptor saturation process, or the presence of a negative feedback loop. Taken together, these data suggest that endogenous TNFα present in the PBMC cultures contributes to inhibition of IFNα expression, and that blockade of that TNFα reverses that effect, resulting in increased IFNα mRNA expression along with increased expression of the IFN-inducible genes IFIT-1 and BAFF.


TNFα is an important proinflammatory cytokine, and its blockade by monoclonal antibodies or soluble TNF receptor has resulted in significant and clinically meaningful responses in RA, psoriatic arthritis, ankylosing spondylitis, and Crohn's disease. In contrast to the success of this therapeutic approach in those diseases, and in spite of the documented presence of TNFα in labial salivary gland biopsy samples from SS patients (15), a pilot study of etanercept and 2 randomized placebo-controlled trials failed to show a significant effect of anti-TNF agents on clinical disease manifestations (1820).

Mariette et al (19) reported results of a multi-center study of 103 SS patients who were randomly assigned to receive 3 infusions of infliximab or placebo and were followed up for 22 weeks. There was no difference between the 2 groups in favorable overall response (≥30% improvement in 2 of 3 domains—oral, ocular, skin, or vaginal dryness; joint pain; or fatigue) at 10 weeks or 22 weeks. Of interest, gamma globulin levels and serum IgM titers increased significantly in the infliximab-treated patients compared with the placebo group. Results of a 12-week, randomized, placebo-controlled trial of etanercept (25 mg twice weekly) in 28 patients were reported by Sankar et al (20). Although the primary outcome measure was more lenient in the etanercept study than in the infliximab study (20% improvement from baseline in 2 of 3 clinical domains), no significant clinical efficacy was observed, although a significant decrease in erythrocyte sedimentation rates was seen in the etanercept group.

We hypothesized that a potential mechanism for this lack of efficacy might be impaired control of type I IFN, a cytokine that has been associated with autoimmunity and increased disease activity in SLE and is a potential target of negative regulation by TNFα (4,5,11,22,29,30). We predicted that in the setting of therapeutic TNF inhibition, production of IFNα might be augmented. To address this hypothesis, we studied plasma samples from SS patients treated with etanercept or placebo for 3 months in the setting of the randomized controlled trial reported by Sankar et al (20). We observed a significant increase in plasma IFNα activity in the etanercept-treated group, but not in the control patients, using a sensitive reporter cell assay (12).

Detection of IFN in SS sera was first reported in 1979 (31). Since then, increasing evidence suggests that activation of the type I IFN pathway plays a role in the pathogenesis of SS, consistent with a possible viral etiology of the syndrome and similar recent observations in SLE (32). Microarray data from salivary gland biopsy samples from SS patients revealed increased expression of IFNα-inducible genes (33), and salivary gland biopsy samples from SS patients were shown to have a large number of IFNα-producing plasmacytoid dendritic cells (PDCs) compared with salivary gland biopsy samples from controls (9,10). Expression of TRAIL, a highly IFNα-responsive member of the TNF molecular family, has been demonstrated on infiltrating cells in SS salivary glands and may provide a mechanistic link between IFNα production and glandular epithelial cell damage (34).

In the present study, we demonstrated a statistically significant increase in type I IFNα plasma activity in SS patients treated for 3 months with etanercept, but not in those who received placebo. A similar effect was confirmed in an in vitro culture system in which healthy control PBMCs cultured with etanercept showed a dose-dependent increase in IFNα, IFIT-1, and BAFF mRNA; at the same time, a decrease of TNFα mRNA expression was noted in the etanercept-treated cells. Unfortunately, accurate quantification of TNFα protein in the etanercept-treated culture supernatants is complicated by formation of immune complexes and was not performed (35).

The presence of TNFα has previously been shown to inhibit IFNα release by PDCs in response to influenza virus, implying that in both in vivo and in vitro systems, blocking the IFN-attenuating effect of TNFα might permit increased endogenous IFNα production (22). The mechanisms that account for regulation of IFNα production by TNF have not been fully elucidated. TNFα has been shown to inhibit differentiation of hematopoietic progenitor cells, including those that generate IFN-producing PDCs (36). The cytokine receptor flt-3 and activation of its downstream signaling mediator STAT-3 are important for PDC development. A STAT-3 inhibitor, suppressor of cytokine signaling 3 (SOCS-3), is inducible by TNFα and may contribute to reduced production of PDCs (30). In addition, TNFα may promote the maturation of PDCs toward a phenotype that is more effective at antigen presentation but less capable of IFNα synthesis or that may induce PDC apoptosis (37).

These mechanisms likely contribute to in vivo control by TNF of the number of effective IFN-producing cells. However, our in vitro data suggest that TNF may also act on existing IFN-producing cells to dampen their capacity to transcribe new IFNα or inhibit induction by IFNα of its target genes, since short-term culture of PBMCs with soluble TNF receptor increased IFNα, IFIT-1, and BAFF mRNA over a 6-hour period. In contrast, etanercept reduced the expression of TNFα mRNA. The mechanisms responsible for these short-term effects are not known, but they may involve the induction of SOCS proteins or intracellular cross-talk between signaling pathways, as has been shown for the inhibitory effect of IFNα on TNFα gene transcription in macrophages (3840). The adaptors and kinases that are implicated in gene regulation induced by TNF intersect the pathways that induce transcription of type I IFN and may abrogate or redirect their activity (38).

BAFF levels were also increased in SS patients compared with healthy controls, in accord with previous reports (2). Interestingly, after 3 months of treatment with etanercept, all but 1 of the patients receiving etanercept showed an increase of at least 15% in the plasma BAFF level compared with fewer than half of the patients receiving placebo. IFNα has been reported to stimulate the production of BAFF in epithelial cells from SS salivary glands, suggesting that the increase in IFNα expression in the etanercept-treated patients may subsequently increase BAFF production (13). Consistent with this interpretation, we observed that changes in IFNα plasma activity correlated significantly with serum BAFF levels, suggesting coordinate expression of these cytokines. Cumulative evidence identifies BAFF as an important mediator in the pathogenesis of SS. BAFF-mediated survival signals provided to autoreactive B cells may play a role in ectopic germinal center formation, promotion of autoantibody production, and class switching to IgG isotypes (2,25,26,4144). It is also of interest that a condition reminiscent of human SS develops in a BAFF-transgenic murine model characterized by a lupus-like disease, with lymphocytic infiltration of salivary glands as the mice age (45).

Taken together, these data suggest an explanation for the failure of etanercept to alleviate sicca symptoms in patients with SS. In view of the recent implication of IFNα pathway activation and BAFF overexpression in SS disease pathogenesis, we propose that the inefficacy of anti-TNF agents could be attributed to exacerbated activation of these already-activated pathways.

Although the cross-regulation of TNFα and IFNα has been previously described (22,29,37,39,46), the present study is the first that directly measures IFNα and BAFF expression in patients who have experienced TNF blockade in vivo. In conclusion, the present study provides a possible explanation for the lack of efficacy of etanercept in SS and points to a different therapeutic approach in this disease, such as targeting the type I IFN or BAFF pathways.


We would like to thank Jing Hua, MD, PhD, for her essential contribution of developing the WISH assay.

Supported in part by grants from the NIH (AI-059893 and AR-050829), the Alliance for Lupus Research, the Lupus Research Institute, and the Intramural Research Program of the NIH, National Institute of Dental and Craniofacial Research. Dr. Mavragani is recipient of a fellowship through the Stavros S. Niarchos International Fellowship Exchange Program of the Stavros S. Niarchos Foundation and the Arthritis Foundation, New York Chapter. Dr. Niewold's work was supported by NIH grant T32-AR-07517. Dr. Crow's work was supported by a grant from the Mary Kirkland Center for Lupus Research, the Alliance for Lupus Research, the Lupus Research Institute, and the NIH.


Dr. Crow owns stock in Johnson & Johnson and has applied for a patent for an interferon assay.

Author Contributions: Dr. Crow had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Mavragani, Pillemer, Crow.

Acquisition of data. Mavragani, Niewold, Moutsopoulos, Pillemer.

Analysis and interpretation of data. Mavragani, Niewold, Moutsopoulos, Pillemer, Wahl, Crow.

Manuscript preparation. Mavragani, Pillemer, Wahl, Crow.

Statistical analysis. Mavragani.

Contributor Information

Clio P. Mavragani, Mary Kirkland Center for Lupus Research, Hospital for Special Surgery, and Weill Medical College of Cornell University, New York, New York.

Timothy B. Niewold, Mary Kirkland Center for Lupus Research, Hospital for Special Surgery, and Weill Medical College of Cornell University, New York, New York.

Niki M. Moutsopoulos, National Institute of Dental and Craniofacial Research, NIH, Bethesda, Maryland.

Stanley R. Pillemer, National Institute of Dental and Craniofacial Research, NIH, Bethesda, Maryland.

Sharon M. Wahl, National Institute of Dental and Craniofacial Research, NIH, Bethesda, Maryland.

Mary K. Crow, Mary Kirkland Center for Lupus Research, Hospital for Special Surgery, and Weill Medical College of Cornell University, New York, New York.


1. Kassan SS, Moutsopoulos HM. Clinical manifestations and early diagnosis of Sjögren syndrome. Arch Intern Med. 2004;164:1275–84. [PubMed]
2. Szodoray P, Jonsson R. The BAFF/APRIL system in systemic autoimmune diseases with a special emphasis on Sjögren's syndrome. Scand J Immunol. 2005;62:421–8. [PubMed]
3. Mariette X, Roux S, Zhang J, Bengoufa D, Lavie F, Zhou T, et al. The level of BLyS (BAFF) correlates with the titre of autoantibodies in human Sjögren's syndrome. Ann Rheum Dis. 2003;62:168–71. [PMC free article] [PubMed]
4. Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann WA, Espe KJ, et al. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc Natl Acad Sci U S A. 2003;100:2610–5. [PMC free article] [PubMed]
5. Bennett L, Palucka AK, Arce E, Cantrell V, Borvak J, Banchereau J, et al. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood. J Exp Med. 2003;197:711–23. [PMC free article] [PubMed]
6. Kirou KA, Lee C, George S, Louca K, Papagiannis IG, Peterson MG, et al. Coordinate overexpression of interferon-α–induced genes in systemic lupus erythematosus. Arthritis Rheum. 2004;50:3958–67. [PubMed]
7. Crow MK, Kirou KA. Interferon-α in systemic lupus erythematosus. Curr Opin Rheumatol. 2004;16:541–7. [PubMed]
8. Ronnblom L, Alm GV. A pivotal role for the natural interferon α-producing cells (plasmacytoid dendritic cells) in the pathogenesis of lupus. J Exp Med. 2001;194:F59–63. [PMC free article] [PubMed]
9. Bave U, Nordmark G, Lovgren T, Ronnelid J, Cajander S, Eloranta ML, et al. Activation of the type I interferon system in primary Sjögren's syndrome: a possible etiopathogenic mechanism. Arthritis Rheum. 2005;52:1185–95. [PubMed]
10. Gottenberg JE, Cagnard N, Lucchesi C, Letourneur F, Mistou S, Lazure T, et al. Activation of IFN pathways and plasmacytoid dendritic cell recruitment in target organs of primary Sjögren's syndrome [published erratum appears in Proc Natl Acad Sci U S A 2006;103:5242] Proc Natl Acad Sci U S A. 2006;103:2770–5. [PMC free article] [PubMed]
11. Kirou KA, Lee C, George S, Louca K, Peterson MG, Crow MK. Activation of the interferon-α pathway identifies a subgroup of systemic lupus erythematosus patients with distinct serologic features and active disease. Arthritis Rheum. 2005;52:1491–503. [PubMed]
12. Hua J, Kirou K, Lee C, Crow MK. Functional assay of type I interferon in systemic lupus erythematosus plasma and association with anti–RNA binding protein autoantibodies. Arthritis Rheum. 2006;54:1906–16. [PubMed]
13. Ittah M, Miceli-Richard C, Eric Gottenberg J, Lavie F, Lazure T, Ba N, et al. B cell-activating factor of the tumor necrosis factor family (BAFF) is expressed under stimulation by interferon in salivary gland epithelial cells in primary Sjögren's syndrome. Arthritis Res Ther. 2006;8:R51. [PMC free article] [PubMed]
14. Mavragani CP, Moutsopoulos NM, Moutsopoulos HM. The management of Sjögren's syndrome. Nat Clin Pract Rheumatol. 2006;2:252–61. [PubMed]
15. Boumba D, Skopouli FN, Moutsopoulos HM. Cytokine mRNA expression in the labial salivary gland tissues from patients with primary Sjögren's syndrome. Br J Rheumatol. 1995;34:326–33. [PubMed]
16. Hunger RE, Muller S, Laissue JA, Hess MW, Carnaud C, Garcia I, et al. Inhibition of submandibular and lacrimal gland infiltration in nonobese diabetic mice by transgenic expression of soluble TNF-receptor p55. J Clin Invest. 1996;98:954–61. [PMC free article] [PubMed]
17. Steinfeld SD, Demols P, Salmon I, Kiss R, Appelboom T. Infliximab in patients with primary Sjögren's syndrome: a pilot study. Arthritis Rheum. 2001;44:2371–5. [PubMed]
18. Zandbelt MM, de Wilde P, van Damme P, Hoyng CB, van de Putte L, van den Hoogen F. Etanercept in the treatment of patients with primary Sjögren's syndrome: a pilot study. J Rheumatol. 2004;31:96–101. [PubMed]
19. Mariette X, Ravaud P, Steinfeld S, Baron G, Goetz J, Hachulla E, et al. Inefficacy of infliximab in primary Sjögren's syndrome: results of the randomized, controlled Trial of Remicade in Primary Sjögren's Syndrome (TRIPSS) Arthritis Rheum. 2004;50:1270–6. [PubMed]
20. Sankar V, Brennan MT, Kok MR, Leakan RA, Smith JA, Manny J, et al. Etanercept in Sjögren's syndrome: a twelve-week randomized, double-blind, placebo-controlled pilot clinical trial. Arthritis Rheum. 2004;50:2240–5. [PubMed]
21. Moutsopoulos NM, Angelov N, Sankar V, Leakan RA, Pillemer S, Wahl SM. Immunological consequences of thalidomide treatment in Sjögren's syndrome. Ann Rheum Dis. 2006;65:112–4. [PMC free article] [PubMed]
22. Palucka AK, Blanck JP, Bennett L, Pascual V, Banchereau J. Cross-regulation of TNF and IFN-α in autoimmune diseases. Proc Natl Acad Sci U S A. 2005;102:3372–7. [PMC free article] [PubMed]
23. Vitali C, Bombardieri S, Jonsson R, Moutsopoulos HM, Alexander EL, Carsons SE, et al. European Study Group on Classification Criteria for Sjögren's Syndrome. Classification criteria for Sjögren's syndrome: a revised version of the European criteria proposed by the American-European Consensus Group. Ann Rheum Dis. 2002;61:554–8. [PMC free article] [PubMed]
24. Kawai S, Sekino H, Yamashita N, Tsuchiwata S, Liu H, Korth-Bradley JM. The comparability of etanercept pharmacokinetics in healthy Japanese and American subjects. J Clin Pharmacol. 2006;46:418–23. [PubMed]
25. Lavie F, Miceli-Richard C, Quillard J, Roux S, Leclerc P, Mariette X. Expression of BAFF (BLyS) in T cells infiltrating labial salivary glands from patients with Sjögren's syndrome. J Pathol. 2004;202:496–502. [PubMed]
26. Le Bon A, Schiavoni G, D'Agostino G, Gresser I, Belardelli F, Tough DF. Type I interferons potently enhance humoral immunity and can promote isotype switching by stimulating dendritic cells in vivo. Immunity. 2001;14:461–70. [PubMed]
27. Jego G, Palucka AK, Blanck JP, Chalouni C, Pascual V, Banchereau J. Plasmacytoid dendritic cells induce plasma cell differentiation through type I interferon and interleukin 6. Immunity. 2003;19:225–34. [PubMed]
28. Litinskiy MB, Nardelli B, Hilbert DM, He B, Schaffer A, Casali P, et al. DCs induce CD40-independent immunoglobulin class switching through BLyS and APRIL. Nat Immunol. 2002;3:822–9. [PubMed]
29. Gary-Gouy H, Lebon P, Dalloul AH. Type I interferon production by plasmacytoid dendritic cells and monocytes is triggered by viruses, but the level of production is controlled by distinct cytokines. J Interferon Cytokine Res. 2002;22:653–9. [PubMed]
30. Campbell JS, Prichard L, Schaper F, Schmitz J, Stephenson-Famy A, Rosenfeld ME, et al. Expression of suppressors of cytokine signaling during liver regeneration. J Clin Invest. 2001;107:1285–92. [PMC free article] [PubMed]
31. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med. 1979;301:5–8. [PubMed]
32. James JA, Harley JB, Scofield RH. Role of viruses in systemic lupus erythematosus and Sjögren syndrome. Curr Opin Rheumatol. 2001;13:370–6. [PubMed]
33. Hjelmervik TO, Petersen K, Jonassen I, Jonsson R, Bolstad AI. Gene expression profiling of minor salivary glands clearly distinguishes primary Sjögren's syndrome patients from healthy control subjects. Arthritis Rheum. 2005;52:1534–44. [PubMed]
34. Matsumura R, Umemiya K, Kagami M, Tomioka H, Tanabe E, Sugiyama T, et al. Expression of TNF-related apoptosis inducing ligand (TRAIL) on infiltrating cells and of TRAIL receptors on salivary glands in patients with Sjögren's syndrome. Clin Exp Rheumatol. 2002;20:791–8. [PubMed]
35. Moreland LW, Baumgartner SW, Schiff MH, Tindall EA, Fleischmann RM, Weaver AL, et al. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N Engl J Med. 1997;337:141–7. [PubMed]
36. Ramsfjell V, Borge OJ, Cui L, Jacobsen SE. Thrombopoietin directly and potently stimulates multilineage growth and progenitor cell expansion from primitive (CD34+ CD38-) human bone marrow progenitor cells: distinct and key interactions with the ligands for c-kit and flt3, and inhibitory effects of TGF-β and TNF-α J Immunol. 1997;158:5169–77. [PubMed]
37. Dolganiuc A, Chang S, Kodys K, Mandrekar P, Bakis G, Cormier M, et al. Hepatitis C virus (HCV) core protein-induced, monocyte-mediated mechanisms of reduced IFN-α and plasmacytoid dendritic cell loss in chronic HCV infection. J Immunol. 2006;177:6758–68. [PubMed]
38. Baetz A, Frey M, Heeg K, Dalpke AH. Suppressor of cytokine signaling (SOCS) proteins indirectly regulate Toll-like receptor signaling in innate immune cells. J Biol Chem. 2004;279:54708–15. [PubMed]
39. Vlotides G, Sorensen AS, Kopp F, Zitzmann K, Cengic N, Brand S, et al. SOCS-1 and SOCS-3 inhibit IFN-α-induced expression of the antiviral proteins 2,5-OAS and MxA. Biochem Biophys Res Commun. 2004;320:1007–14. [PubMed]
40. Ivashkiv LB. Type I interferon modulation of cellular responses to cytokines and infectious pathogens: potential role in SLE pathogenesis. Autoimmunity. 2003;36:473–9. [PubMed]
41. Szodoray P, Alex P, Jonsson MV, Knowlton N, Dozmorov I, Nakken B, et al. Distinct profiles of Sjögren's syndrome patients with ectopic salivary gland germinal centers revealed by serum cytokines and BAFF. Clin Immunol. 2005;117:168–76. [PubMed]
42. Batten M, Groom J, Cachero TG, Qian F, Schneider P, Tschopp J, et al. BAFF mediates survival of peripheral immature B lymphocytes. J Exp Med. 2000;192:1453–66. [PMC free article] [PubMed]
43. Melchers F. Actions of BAFF in B cell maturation and its effects on the development of autoimmune disease. Ann Rheum Dis. 2003;62 2:ii, 25–7. [PMC free article] [PubMed]
44. Szodoray P, Jellestad S, Teague MO, Jonsson R. Attenuated apoptosis of B cell activating factor-expressing cells in primary Sjögren's syndrome. Lab Invest. 2003;83:357–65. [PubMed]
45. Groom J, Kalled SL, Cutler AH, Olson C, Woodcock SA, Schneider P, et al. Association of BAFF/BLyS overexpression and altered B cell differentiation with Sjögren's syndrome. J Clin Invest. 2002;109:59–68. [PMC free article] [PubMed]
46. Banchereau J, Pascual V, Palucka AK. Autoimmunity through cytokine-induced dendritic cell activation. Immunity. 2004;20:539–50. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...