Sjögren’s Syndrome Antigen B Acts as an Endogenous Danger Molecule to Induce Interleukin-8 Gene Expression in Polymorphonuclear Neutrophils

Background Sjögren’s syndrome antigen B is expressed in the nucleus and surface membrane of human polymorphonuclear neutrophils and is released after cell death. However, its biological role is not clear. This study is aimed to investigate the effect of Sjögren’s syndrome antigen B on human polymorphonuclear neutrophils. Methods Human recombinant Sjögren’s syndrome antigen B (rSSB) purified from E. coli was incubated with human polymorphonuclear neutrophils as well as retinoid acid-induced granulocytic differentiated HL-60 cells, HL-60 (RA). Interleukin (IL)-8 protein production and mRNA expressions were measured by enzyme-linked immunosorbent assay and quantitative-polymerase chain reaction, respectively. Uptake of fluorescein isothiocyanate (FITC)-rSSB was assessed by flow cytometry and fluorescence microscopy. Moreover, mitogen-activated protein kinase (MAPK) pathways and nuclear factor-kappaB activation were investigated. Results Human rSSB stimulated IL-8 production from normal human neutrophils and HL-60 (RA) cells in a time- and dose-dependent manner. This IL-8-stimulated activity was blocked by chloroquine and NH4Cl, indicating that endosomal acidification is important for this effect. We found rSSB activated both MAPK pathway and nuclear factor-kappaB signaling to transcribe the IL-8 gene expression of cells. Furthermore, tumor necrosis factor-α exerted an additive effect and rSSB-anti-SSB immune complex exhibited a synergistic effect on rSSB-induced IL-8 production. Conclusions Sjögren’s syndrome antigen B might act as an endogenous danger molecule to enhance IL-8 gene expression in human polymorphonuclear neutrophils.


Patients and controls
Patients with SLE [31] and normal individuals were included in this study. Human PMNs were isolated as described in our previous study [30]. This study was approved by the Research Ethics Committee of National Taiwan University Hospital, Taipei, Taiwan. Written informed consent was obtained from each participant.

Expression, purification, and identification of human full-length recombinant SSB/La protein
Expression and purification of histidine-tagged rSSB protein from E. coli was performed as described in our previous report [30]. Immobilized polymyxin B gel (Thermo Scientific, Rockford, IL, USA) was used to remove any potentially contaminating lipopolysaccharide (LPS) in the preparations, with an efficiency of greater than 99.95%. The concentrations of endotoxin and protein in the purified rSSB preparations were determined using a Limulus amebocyte lysate assay (Lonza, Walkersville, MD, USA) and a bicinchoninic acid assay (Thermo Scientific, Rockford, IL, USA), respectively. The rSSB molecule was separated in 12% SDS-PAGE, followed by reactions with Coomassie blue, purified human polyclonal anti-SSB antibodies, mouse monoclonal antibodies against histidine tag (anti-HisTag), or mouse monoclonal anti-SSB antibodies.

Purification of human polyclonal anti-SSB autoantibodies
Human polyclonal anti-SSB antibodies were purified as described in our previous study [30]. The preparations were stored at -80°C until use.

Cell stimulation
Cells were treated with rSSB at 37°C at the indicated doses and times. LPS (100 ng/ml, Sigma-Aldrich) or TNF-α (20 ng/ml, ProSpec, NJ, USA) was used as the PMN stimulant for the positive controls, and bovine serum albumin (BSA, Sigma-Aldrich) as a negative protein control. In some experiments, PMNs were preincubated with polymyxin B (100 μg/ml, Sigma-Aldrich) to inactivate LPS before stimulation with rSSB or LPS. In other experiments, rSSB was pretreated with proteinase-K (Sigma-Aldrich), RNase A (Sigma-Aldrich) or boiling and freezing for 1 hour before incubation with the cells. In TNF-α priming experiments, PMNs were pretreated with/without TNF-α (20 ng/ml) for 30 minutes and then stimulated with/without rSSB (10 μg/ml) for 2 hours. In immune complex (IC) experiments, the cells were treated with preformed ICs made up of rSSB and purified anti-SSB antibodies for 2 hours. In endosome inhibition experiments, the cells were pretreated with different concentrations (5-50 μg/ml) of chloroquine (CQ, Sigma-Aldrich) or 30 mM NH 4 Cl (Sigma-Aldrich) for 1 hour, followed by stimulation with rSSB at the indicated times.

IL-8 quantitation in cultured supernatants by ELISA
Human IL-8 protein levels in different cultured supernatants were measured using a commercially available ELISA kit obtained from R&D Systems (Minneapolis, MN, USA) according to the manufacturer's instructions.

Q-PCR for IL-8 mRNA expression
Total cell RNAs were extracted using the RNeasy mini kit (Qiagen, Hilden, Germany) and then reversely transcribed into cDNA. Q-PCR was conducted using a StepOne Real-Time PCR System (Applied Biosystems, Foster, CA, USA). The primer pair for human IL-8 (forward 5'-atg act tcc aag ctg gcc gtg gct-3'; reverse 5'-tct cag ccc tct tca aaa act tc-3') were purchased from Bio-Basic (Markham, Canada). The relative abundance of IL-8 mRNA was calculated using the ΔΔCt method and normalized to cyclophilin.

Detection of FITC-rSSB uptake by flow cytometry and fluorescence microscopy
Fluorescein isothiocyanate (FITC) was purchased from Sigma-Aldrich and conjugated to rSSB or BSA according to the manufacturer's protocols. Subsequently, the cells (1x10 6 /ml) were incubated with 10 μg/ml of FITC-rSSB or FITC-BSA at 4°C or 37°C for the indicated times. After washing, 0.4% trypan blue was added to quench the extracellular FITC fluorescence. The percentage of positive cells was analyzed using a BD FACSCalibur Flow Cytometer (San Jose, CA, USA). We used propidium iodide to stain the dead cells. To study the effect of different molecules on the uptake of FITC-rSSB, HL-60 (RA) cells were incubated with medium alone or with FITC-rSSB (40 μg/ml) in the presence of medium, LPS, non-specific IgG or purified human polyclonal anti-SSB antibodies for 15 minutes, followed by flow cytometric analysis. The FITC-rSSB uptake by HL-60 cell lines was also observed under fluorescence microscopy.

Detection of signaling pathways in PMNs and HL-60 (RA) cells
For detection of MAPK activation, whole cell lysate was subjected to 12% SDS-PAGE and transferred to a PVDF membrane. Rabbit polyclonal antibodies as the primary antibodies against signaling molecules including ERK (p42/p44), phospho-p42/p44, p38, phospho-p38, JNK, and phospho-JNK, were purchased from Cell Signaling Technology (Danvers, MA). Mouse polyclonal antibodies against GAPDH were purchased from Sigma-Aldrich. After incubation with the appropriate primary and horseradish peroxidase-conjugated secondary antibodies, blots were developed using SuperSignal West Pico reagents (Thermo Scientific, Logan, Utah, USA).

NF-κB binding activity in nuclear extracts
Nuclear extracts were isolated using Thermo Scientific NE-PER Nuclear and Cytoplasmic Extraction Kits. The binding activity of NF-κB subunits p65 and p50 in the nuclear extract were detected using TransAM NF-κB Kits (Tokyo, Japan) according to the manufacturer's instructions.

Statistical analysis
Results are presented as the mean ± S.D. The Student's t-test was used for comparisons of continuous variables between two groups. One-way analysis of variance (ANOVA) with Bonferroni correction for multiple comparisons was applied for more than two groups. A p-value of 0.05 or lower was considered to indicate statistical significance.

Identification of recombinant human SSB
To identify the prepared rSSB, the preparations (0.5 to 1 μg) were subjected to 12% SDS-PAGE. As shown in Fig 1A, a prominent band with a molecular weight around 50 kDa was found in Coomassie blue staining. The higher molecular weight than human SSB (48 kDa) was due to the inserted polyhistidine tag in the molecule. The purified molecule was proven to be histidine-tagged recombinant human SSB by Western blotting using mouse monoclonal anti-His-Tag antibodies, human polyclonal anti-SSB antibodies, and mouse monoclonal anti-SSB antibodies as probes ( Fig 1B).

Recombinant SSB stimulated IL-8 production from normal human PMNs
Incubation of normal human PMNs (1x10 6 cells/ml) with rSSB resulted in a dose-(1-20 μg/mL) and time-(1-4 hours) dependent increase in IL-8 production (Fig 2A). We found that 10 μg/ml of rSSB (219.4±152.5 pg/ml) was as effective as 100 ng/ml of LPS (160.8±100.6 pg/ml) in stimulating IL-8 production from normal PMNs ( Fig 2B). In contrast, BSA as a protein control failed to enhance IL-8 production in 3 normal control, even at the dose of 80 μg/ml (54.14 ±75.35 pg/ ml in medium group vs. 47.56±66.62 pg/ml in BSA group). This IL-8 stimulating activity of rSSB was not due to endotoxin contamination during preparation since polymyxin B (100 μg/ml) had no effect on rSSB-induced IL-8 production from PMNs ( Fig 2C). In addition, by Limulus amebocyte lysate (LAL) detection method, the average endotoxin level in the rSSB preparations were 0.0032 EU/10g/ml.  To further characterize the IL-8 stimulating activity of rSSB, we performed enzyme-digestion or heat-denaturation of rSSB. The results showed that an intact protein structure of rSSB mediated the activity as proteinase K treatment and heat-denaturation, but not RNase A treatment of rSSB reduced the rSSB activity to produce IL-8 ( Fig 2D). In brief, the rSSB protein molecule per se, but not contaminating LPS or RNA, was responsible for the IL-8 stimulating activity.

Uptake of rSSB by PMN
To investigate whether SSB autoantigens enter into the cell interior or merely bind to the cell surface of PMNs to activate IL-8 production, normal PMNs were incubated with FITC-rSSB for different time periods. Uptake, but not mere attachment, of FITC-rSSB by PMN began at 5 minutes and reached a maximal level after 15 minutes of incubation ( Fig 3A). As shown in Fig 3B, the uptake of rSSB was specific and irrelevant to cell death, and appeared higher at 37°t han that at 4°. However, the difference in our small sample size did not reach a statistically significant level (p = 0.21). More samples are needed to determine whether uptake of rSSB is energy-dependent. Likely, although it seemed that human polyclonal anti-SSB autoantibodies enhanced FITC-rSSB uptake at 15 minutes compared to medium, LPS or IgG (Fig 3C), more experiments are needed to draw any conclusion (p = 0.55, compared with medium control). The uptake of FITC-rSSB by HL-60 cells was also observed by fluorescence microscopy (Fig 3D).

Enhanced PMN IL-8 production by SSB-anti-SSB immune complex
As anti-SSB antibody seemed enhanced FITC-rSSB uptake by HL-60 (RA) cells (Fig 3C), we next investigated whether anti-SSB autoantibody or preformed SSB-anti-SSB ICs influenced IL-8 production from human PMNs and HL-60 (RA) cells. As shown in Fig 2F, anti-SSB antibody (10 μg/ml) exerted a negligible effect on IL-8 production from normal PMNs. However, rSSB-anti-SSB ICs markedly enhanced IL-8 production. The synergic effect of SSB-anti-SSB ICs on IL-8 production was also observed in HL-60 (RA) cells (S1 Fig). We speculate that anti-SSB antibody at this concentration might merely facilitate the uptake of SSB molecules by these cells (Fig 3C), but have no effect on IL-8 production (Fig 2F). Instead, uptake of SSB-anti-SSB ICs by phagocytes through Fc-gamma receptors (FcγRs) exerts a synergistic effect on IL-8 production from these cells (Fig 2F).

Chloroquine and NH 4 Cl decreased SSB-induced IL-8 gene expression in HL-60 (RA) cells and human PMNs
As endosomes contain internalized molecules, we compared the effect of endosome acidification inhibitors on rSSB-induced IL-8 gene expression in HL-60 (RA) cells (Fig 4A). Pretreatment with different concentrations of CQ (5-50 μg/ml) and NH 4 Cl (30 mM) displayed a doseresponsive inhibitory effect on rSSB-induced IL-8 gene expression. Likewise, the inhibitory effect of CQ on IL-8 gene expression was observed in normal human PMNs (Fig 4B). Since hydroxychloroquine (HCQ) is frequently used in patients with SLE, we evaluated the effect of CQ on rSSB-induced IL-8 production from the PMNs of lupus patients. Pretreatment of lupus PMNs with 5 μg/ml of CQ reduced the rSSB-induced IL-8 production by 22.7% compared to the medium pretreatment group (Fig 4C).

Discussion
There are several original findings in this study. First, rSSB enhances IL-8 gene expression in normal human PMNs in a time-and dose-dependent manner. Second, rSSB transduces p38, ERK 1/2 and JNK MAPK signaling and NF-κB subunit p65 and p50 nuclear translocation that is responsible for rSSB-induced IL-8 gene expression in human PMNs. Third, TNF-α exerts an additive effect whereas the SSB-anti-SSB IC exerts a synergistic effect on rSSB-induced IL-8 production. Finally, CQ reduced the rSSB-induced IL-8 production from neutrophils in both normal controls and patients with systemic lupus erythematous.
Cell death and delayed clearance of intracellular antigens are involved in the development of autoimmunity [16,17]. Many endogenous adjuvant molecules and nuclear autoantigens are released from excessive tissue damage and cell death [19][20][21]. As TNF-α and IL-8 play a critical role in amplifying the inflammatory response and sustaining tissue damage and cell death, our results suggest that released SSB from dead cells might act as a danger molecule to induce IL-8 production from PMNs.
Gallucci et al. [18], Shi et al. [22], and Matzinger et al. [34] reported that danger signals from tissue damage and cell death can alert the immune system to responses. Currently, intracellular molecules such as ATP [21], UTP [20], uric acid [22], and heat-shock proteins [35,36] are recognized to be endogenous danger signals that activate immune cells by themselves or in conjunction with TNF-α. Moreover, some autoantigens have been reported to mediate pathologic immune responses in systemic autoimmune diseases [23][24][25][26][27][28][29]. For example, the carboxy-terminal domain of tyrosyl-tRNA synthetase (TyrRS, 1 nM) has been shown to exhibit chemotactic activity for monocytes and PMNs and to stimulate the production of myeloperoxidase, TNF-α, and tissue factor. The amino-terminal domain of TyrRS (1 nM) has been reported to induce PMN migration [27], and human myelin basic protein (10-30 μg/ml) has been reported to induce a dose-dependent release of interferon-γ, TNF-α and IL-10 from the mononuclear cells of patients with multiple sclerosis [28]. In addition, HSP60 (10 μg/ml) has been reported to induce significant amounts of TNF-α, IL-1 and IL-10 from peripheral blood mononuclear cells of patients with juvenile dermatomyositis [29]. As rSSB (10 μg/ml), alone or in combination of TNF-α, induces IL-8 production from normal neutrophils, it is suggested that SSB might be a new endogenous danger molecule that can activate PMNs to produce IL-8. Moreover, SSB-anti-SSB ICs induce a much higher level of IL-8 production from PMNs.
In our study, the concentration of rSSB (10 μg/ml) we used to stimulate cells was similar to that of other autoantigens mediating immune responses [28,29]. However, data on the During infections or autoimmune responses, SSB molecules are released from damaged tissues and dead cells. The free SSB molecules activate PMN via MAPK pathways and NF-κB nuclear translocation to transcribe the IL-8 gene. TNF-α and SSB-anti-SSB immune complexes, which may utilize TNF-α receptors (TNFR) and Fc-gamma receptors (FcγR), respectively, augment IL-8 production probably through a final common pathway. concentration of SSB in vivo was not addressed in the literature we reviewed. As SSB has been very abundant in the cell (20 million copies per cell) [3], the estimated intracellular concentration of SSB is around 400 μg/ml assuming that average cell volume to be 4 billionths of a cubic centimeter [37]. Moreover, SSB expression may be upregulated in inflamed tissues [38]. Hence, the possible local concentration of released SSB after cell death was estimated to be high enough (in the order of μg/ml) to induce biologic effects. It is also postulated that the high expression of SSB in labial salivary gland ductal cell might contributed to the local immune response in pSS. [38].
Neutrophils have been suggested to play a central role in initiation and perpetuation of aberrant immune responses in systemic autoimmune diseases, including SLE and rheumatoid arthritis [39]. Although the role of neutrophils in the pathogenesis of pSS remained elusive, there are several findings that support neutrophils play a part in the pathogenesis of pSS. First, we demonstrated human SSB as well as SSB-anti-SSB IC induced IL-8 production from normal neutrophils. Second, neutropenia is frequently noted in pSS (33%) and is closely associated with anti-SSB antibody [40]. As SSB can be expressed in the surface membrane of human neutrophils and purified anti-SSB antibody increased apoptosis and IL-8 production of these cells [30], the abnormal SSB antigen expression in neutrophil membranes may induce the synthesis of autoantibodies against these cells and cause their lysis [40]. This also supports a possible role of neutrophil in the pathogenesis of pSS. Finally, neutrophils are activated in pSS [41].
A schematic diagram illustrating the interaction of SSB with human PMN is shown in Fig 6. After uptake of free rSSB, MAPK signaling (p38, ERK1/2, and JNK) and NF-κB subunit p65 and p50 nuclear translocation were activated to transcribe IL-8 gene expression in human PMNs. In addition, TNF-α and SSB-anti-SSB ICs may utilize TNF-α receptors and FcγRs, respectively, to augment the production of IL-8. These findings are compatible with the wellknown intracellular mechanisms for IL-8 production [42][43][44]. As many autoantigens and endogenous adjuvants activate Toll-like receptors [45], nucleotide-binding oligomerization-like receptors [46] and chemokine receptors [24], it would be interesting to investigate the role of these receptors in the rSSB-induced activation of PMNs. The basic mechanisms by which rSSB stimulates IL-8 production and correlations of serum cytokines levels with concentrations of SSB as well as SSB-anti-SSB IC levels in patients with pSS are now under investigation. Further larger studies are also needed to explore the role of CQ/HCQ in the rSSB-induced activation of PMNs from patients with SLE.

Conclusions
SSB might act as an endogenous danger molecule to enhance IL-8 gene expression of PMNs. This not only expands the biological function of SSB, but supports the concept that autoantigens per se contributes to the generation of autoimmunity in genetically susceptible subjects.
Supporting Information S1 Dataset. Individual data points in this study.