1 Introduction
Systemic lupus erythematosus (SLE) is a chronic multiorgan autoimmune disorder that disproportionately affects women of fertile age. It creates a challenging dilemma for pregnant cases because a relatively stable immune environment has been proven to be a prerequisite for a successful pregnancy outcome. Thus far, studies have revealed that patients with SLE are subject to an increased risk for unfavorable maternal and fetal outcomes, such as hypertensive disorders, fetal loss, intrauterine growth restriction, small-for-gestational-age (SGA) infants, and preterm delivery [
1,
2]. Although substantial progress has been made in the management of lupus pregnancy during the past 20 years, keeping lupus under control and securing an uneventful pregnancy safe from lupus flares remain challenging. Plasmacytoid dendritic cells (pDCs) have been recognized as critical drivers in lupus pathogenesis by producing large amounts of type I interferon (IFN-I) in response to Toll-like receptor (TLR) activation [
3]. The increased expression of IFN-I genes in peripheral blood cells is associated with lupus activity [
4] and elevated titers of antinuclear autoantibodies [
5]. Patients with SLE exposed to high levels of IFN-α during pregnancy are more prone to poor pregnancy outcomes than other patients [
6]. The longitudinal profiling of the blood transcriptomes of patients with SLE and complicated pregnancies revealed the incomplete downregulation of the IFN signature, which is remarkably downregulated in noncomplicated lupus pregnancies [
7]. Given that pDCs are the main source of IFN-I, their central role in SLE pathogenesis has been brought to the forefront in the last decade. As a result, several therapeutic strategies that highlight targeting the IFN-I pathway or pDCs have gained remarkable progress [
8,
9].
Antibodies have been proposed to be crucial drivers in the pathogenesis of SLE, as well as many other autoimmune diseases, and the glycosylation pattern of IgG could be decidedly critical for its anti-inflammatory functionality [
10,
11]. IgG is composed of two light chains and two heavy chains [
12]. Each heavy chain harbors an
N-glycosylation site at the conservative residue Asn297 [
13]. The core glycan structure contains two consecutive
N-acetylglucosamine (GlcNAc) molecules attached by a mannose that are further extended by two mannose antennae with each antenna linked by an additional GlcNAc molecule [
14]. This basic structure could be further modified by adding fucose, galactose, or sialic acid and bisecting GlcNAc [
14], resulting in the diversity of glycoforms. The IgG glycome has been reported to regulate antibody effector functions, including antibody-dependent cellular cytotoxicity, antibody-dependent cellular phagocytosis, and complement-dependent cytotoxicity [
15]. The alteration of the IgG glycome is closely associated with many autoimmune conditions, such as SLE, rheumatoid arthritis (RA) [
11], and human immunodeficiency virus (HIV) infection [
15,
16]. Emerging evidence suggests that SLE is associated with a shift toward a highly proinflammatory glycan profile characterized by a decreased level of sialylation [
10]. Moreover, sialylation has been demonstrated to be critical in regulating immune homeostasis at the fetal–maternal interface, and in mice, the depletion of sialylation leads to embryonic lethality [
17], suggesting the indispensable role of IgG sialylation in immunoregulatory activities and fetal development. However, thus far, the profile of IgG sialylation in pregnant females with SLE has not been clearly elucidated. Herein, by analyzing the IgG glycome in females with SLE, we identified the sialylation of IgG as a potential modulator of the IgG inflammatory profile that in turn influences the functions of pDCs. Our work thus affords insight into lupus flares and future therapeutic targets.
2 Materials and methods
2.1 Study population
A total of 156 pregnant patients with SLE and 90 pregnant healthy females at different trimesters of pregnancy were recruited as the SLE and control cohorts, respectively, during prenatal care visits between June 1, 2017 and November 15, 2019. Another 90 nonpregnant healthy females without previous adverse obstetric histories and 36 nonpregnant SLE female patients who attended pre-pregnancy counselling were also included in the control and SLE cohorts, respectively. The two cohorts only included females aged 18–45 years old and consisted exclusively of Han individuals. Only singleton pregnancies were included in the study to rule out the perplexing influence of twins or multiple pregnancies. All patients who were diagnosed with SLE fulfilled the revised criteria of the American College of Rheumatology [
18] and were not afflicted with other autoimmune diseases. Moreover, individuals who were diagnosed with cancer; metabolic diseases; or other severe diseases irrelevant to SLE involving the cardiovascular, respiratory, digestive, endocrine, genitourinary, or nervous system were excluded.
2.2 Assessment of disease activity
Disease activity in nonpregnant patients with SLE was scored by using the Systemic Lupus Erythematosus Disease Activity Index 2000 [
19]. Disease activity in pregnant patients with SLE was assessed by applying the Systemic Lupus Erythematosus Pregnancy Disease Activity Index (SLEPDAI) [
20].
2.3 Sample collection and follow-up
Medical history and blood samples were obtained from every participant. For pregnant participants, serum samples were collected during the first trimester ( < 14 weeks of gestation), second trimester (14–28 weeks of gestation), and third trimester (> 28 weeks of gestation). None of the participants had clinically detectable infections during blood sample collection. All pregnant patients were followed up during the timeframe of our study, and major obstetric and medical events were documented as they occurred. Adverse pregnancy outcomes (APOs) were registered, and the major events are described as follows:
● SGA: birthweight below the 10th percentile for gestational age [
21].
● Preterm birth: live birth that occurred prior to 37 weeks of gestation.
● Fetal loss: stillbirth [
22] and spontaneous abortion [
23].
● Pre-eclampsia: hypertension (blood pressure > 140/90 mmHg taken at two time points 6 h apart) and presence of proteinuria (urinary protein ≥ 300 mg/24 h) after the 20th gestational week [
24].
● Preterm premature rupture of the membranes: rupture of the amniotic sac prior to the start of labor before 37th week of pregnancy [
25].
2.4 IgG isolation
IgG was isolated from the serum of all participants via affinity chromatography by using a 96-deep-well protein G monolithic plate equilibrated with four column volumes (CV) of binding buffer (20 mmol/L Na3PO4, pH 7.0) two times before use to retain IgG stability. A total of 70 μL of each serum sample was diluted with 100 μL of binding buffer (PBS), applied to the abovementioned protein G plate, and incubated at room temperature for 45 min. The plate was then washed three times with binding buffer to remove excess unbound proteins. IgG was eluted with 2 CV of elution buffer (0.1 mol/L glycine, pH 2.8) and further collected into a shallow plate containing 0.075 CV neutralization buffer (Tris-HCl, pH 9.0). The concentration of IgG was determined by using a BCA kit.
2.5 Enzymatic N-glycan release and purification
Purified isolated IgG was incubated with 10 μL of diluted PNGaseF at 37 °C overnight. For the complete release and absorption of IgG N-glycans, the enzymatic solution was repeatedly added to the preactivated and balanced PGC plates three times. After being cleaned of residual reagents and solvent, the released IgG N-glycans were evaporated to dryness and prepared for fluorescent labeling.
2.6 Glycan fluorescence labeling
Samples and the blank control were subsequently labeled with 2-aminobenzamide, a fluorescent dye. After incubation at 60 °C for 2 h, each sample was diluted with 50 μL of ultrapure water then filtered with a 0.22 μmol/L filter membrane before ultraperformance liquid chromatography (UPLC) analysis.
2.7 Hydrophilic interaction chromatography of labeled glycans
Fluorescently labeled IgG
N-glycans were sequentially analyzed via hydrophilic interaction chromatography (HILIC) on a Waters Acquity UPLC instrument that consisted of a quaternary solvent manager, sample manager, and FLR fluorescence detector as previously described [
26]. An automatic processing method was used for data processing, and each HILIC-UPLC chromatogram was manually corrected by using a traditional integration algorithm. The released
N-glycans separated into 24 chromatographic peaks.
2.8 Reverse transcription polymerase chain reaction
Total RNA was extracted from the peripheral blood mononuclear cells (PBMCs) of pregnant patients with SLE by using Trizol reagent (Invitrogen) and reverse transcribed into cDNA by using a PrimeScript Reagent Kit (TaKaRa) then amplified on a 7900 Real-Time PCR System (Applied Biosystems) with the following settings: 30 s at 95 °C followed by 40 cycles of 5 s at 95 °C and 34 s at 60 °C. The paired primers were as follows:
MX1: 5′-GGGTAGCCACTGGACTGA-3′ (Forward),
5′-AGGTGGAGCGATTCTGAG-3′ (Reverse);
IRF7: 5′-TGAAGCTGGAACCCTGG-3′ (Forward),
5′-GATGTCGTCATAGAGGCTGTT-3′ (Reverse);
OAS1: 5′-GAAGGCAGCTCACGAAAC-3′ (Forward),
5′-TTCTTAAAGCATGGGTAATTC-3′ (Reverse);
IFIT1: 5′-GCCTCCTTGGGTTCGTCTACAA-3′ (Forward),
5′- TCAAAGTCAGCAGCCAGTCTCA-3′ (Reverse);
IFI44: 5′-TCCCCAACTAATTTCCAGAT-3′ (Forward),
5′-TCCAGTGAATCTTCGCATC-3′ (Reverse);
RPL13A: 5′-CCTGGAGGAGAAGAGGAAAGAGA-3′ (Forward),
5′-TTGAGGACCTCTGTGTATTTGTCAA-3′ (Reverse).
2.9 Isolation of pDCs from peripheral venous blood
The peripheral venous blood of healthy females and pregnant patients with SLE was collected in anticoagulation tubes, and PBMCs were further isolated via density gradient centrifugation by using Ficoll-Paque™ PLUS. Human primary pDCs were isolated and enriched through negative selection by using a human PDC Isolation Kit II (Miltenyi Biotec). Then, the pDCs were further purified into Lineage–HLADR+CD123+BDCA-2+ cells through fluorescence-activated cell sorting (FACSAria Cell Sorting System, BD Biosciences).
2.10 Immunoglobulin preparation and cleavage of terminal sialic acids
Purified human IgG (Equitech-Bio) was treated with α2-3,6,8,9 neuraminidase (Rhinozyme) to cleave terminal sialic acids and obtain desialylated IgG (deSia-IgG) in accordance with the manufacturer’s protocol. IgG and deSia-IgG were tested for endotoxin levels, and the endotoxin levels of both preparations were less than 2 EU/mg.
2.11 Cell stimulation and culture
Purified pDCs pooled from healthy donors (200 mL of peripheral blood per donor) were seeded at the density of 1 × 105 cells/100 μL of medium in round-bottomed 96-well plates, whereas pDCs pooled from patients with SLE (5–8 mL peripheral blood per donor) were cultured at the density of 5 × 103 cells/50 μL of medium in round-bottomed 96-well plates. PDCs were pretreated with IgG or deSia-IgG in accordance with the experimental design then stimulated with the TLR-9 agonist ODN 2216 (0.5 μmol/L, Invivogen) and TLR-7 agonist R848 (10 μg/mL, InvivoGen), respectively.
2.12 Enzyme-linked immunosorbent assay and flow cytometry
Cell supernatant and pDCs were harvested separately from culture wells after 24 h of stimulation. IFN-I, including IFN-α and IFN-β, was tested by using enzyme-linked immunosorbent assay (ELISA) kits (PBL Assay Science) as per the manufacturers’ protocols. Other inflammatory cytokines, including interleukin-6 (IL-6) and tumor necrosis factor α (TNF-α), were detected by using a Cytometric Bead Array (CBA) Human Soluble Protein Master Buffer Kit (BD Biosciences) and quantified through flow cytometry. The expression levels of the costimulatory molecules CD40 and HLADR were determined through flow cytometry. For the measurement of the phosphorylation level of SYK (pY348) and BLNK (pY84), cells were fixed with prewarmed buffer at 37 °C for 10 min, permeabilized at room temperature for 10 min, then stained with PE Mouse anti-SYK (pY348) and anti-BLNK (pY84). For the measurement of the phosphorylation of IRF-7 (pS477/pS479), cells were fixed for 10 min at 37 °C then permeabilized on ice for 30 min or more. Then, the cells were stained with PE Mouse anti-IRF-7. Flow cytometry data were analyzed with FlowJo software (Tree Star, Ashland, OR).
2.13 Inhibition of SYK in pDCs
Human primary pDCs pooled from five healthy female donors were pretreated with 0.005 μmol/L SYK inhibitor R406 (Selleck Chemicals, Houston, TX, USA) for 1 h, treated with 10 mg/mL IgG or deSia-IgG for another 1 h, then stimulated with CpG for 20 h.
2.14 RNA sequencing
Total RNA was extracted from the PBMCs of pregnant SLE females by using Trizol reagent (Invitrogen) then determined by using Bioanalyzer 4200 (Agilent). A VAHTS mRNA-Seq v2 Library Prep Kit for Illumina (Vazyme) was applied to construct next-generation libraries of mRNA as per the manufacturer’s instructions. mRNA libraries were sequenced on an Illumina platform (HiSeq X 10 system) by using a 150-bp paired-end run. Genes that were differentially expressed were identified on the basis of the log2 fold change of ≥0.5 or ≤−0.5 and adjusted P value < 0.05. Differentially expressed genes were mapped to biological pathways to determine whether alterations in the expression levels of genes affected the patterns of biological pathways.
2.15 Statistical analysis
Statistical analyses were carried out by using SPSS (version 22.0; IBM Corporation, Armonk, NY, USA) and GraphPad Prism (version 8.0.1; GraphPad Software; La Jolla, CA, USA). Normalization by total area was performed, and the peak area of each of 24 basic glycan peaks was divided by the total integrated area to eliminate experimental variations and make measurements comparable across all samples. The differences in terms of clinical and biochemical variables between two different groups were compared by using unpaired t-test if the data were normally distributed. Otherwise, the Mann–Whitney U test was employed. Meanwhile, categorical variables were compared by using Fisher’s exact test. For cell experiments, unpaired t-test was performed to evaluate the statistical significance of differences between two groups. The receiver operating characteristic (ROC) curve was constructed, and the area under the curve (AUC) was applied to examine the diagnostic value of glycan traits. Pearson’s correlation analysis was used to assess the correlation of clinical parameters and ISG expression with glycan traits. Two-tailed P value < 0.05 was considered statistically significant.
3 Results
3.1 IgG sialylation is associated with disease activity and fetal loss in pregnant females with SLE
Of the 488 participants screened for eligibility, only 372 were retained for the final analysis (Fig.1 and Table S1). Their IgG N-glycomes were profiled, and 24 directly measured chromatographic peaks were calculated and expressed as the percentage (%) of the total integrated area (Fig.2). Derived traits S total and S1 total representing groups of glycans with sialic acids were quantified by summing up the corresponding peak areas (Table S2).
We assigned all participants into four stages, namely, preconception and the first, second, and third trimesters, to investigate how IgG sialylation differed between the SLE and control groups. We observed a significant decrease in sialylation in the SLE group relative to that in the control group irrespective of stage (Fig.2). We further explored whether the level of sialylation was associated with lupus activity during pregnancy. By performing Pearson’s correlation analysis, we found significant correlations between sialylation and disease activity as assessed by SLEPDAI (Fig.2). Other laboratory indicators of lupus activity, such as anti-dsDNA antibody and C3/C4 titers, were also correlated with sialylated IgG glycans. Specifically, S total and S1 total showed a moderate positive correlation with C3 and a negative correlation with anti-dsDNA antibody levels (Fig.2). These results strongly suggested that increased lupus activity is associated with the downregulation of IgG sialylation.
Previous studies have linked the elevated risk for fetal loss with increased SLE activity [
27,
28]. We also found that patients with SLE and failed pregnancy had higher SLEPDAI scores (mean ± SD, 8.00 ± 4.35 vs. 2.96 ± 1.99,
P < 0.0001), lower C3 levels (mean ± SD, 0.70 ± 0.23 vs. 1.01 ± 0.18,
P = 0.0002), and C4 levels (mean ± SD, 0.11 ± 0.06 vs. 0.21 ± 0.08,
P = 0.0014) and higher anti-dsDNA titers (mean ± SD, 55.36 ± 31.48 vs. 23.34 ± 16.21,
P = 0.0026) than those without fetal loss (Table S3). A significant decrease in the level of sialylation was documented in mothers with SLE and fetal loss (
n = 12) compared with those without (
n = 32). This decrease was represented as the lower level of S and S1 totals in mothers with fetal loss than those without (Fig.2). The ROC curve showed the fair performance of S total (AUC 0.78, 95% CI 0.63 to 0.94) and S1 total (AUC 0.79, 95% CI 0.64 to 0.95) in discriminating patients with SLE with fetal loss from those without. The cut-off value for each derived trait was 15.59% and 14.06% (Fig.2).
3.2 IFN-I signature is negatively correlated with IgG sialylation level
The dysregulation of IFN-I is associated with a series of autoimmune diseases, including SLE [
29]. The increased expression of ISGs, which has been deemed as the IFN-I signature, is positively correlated with lupus activity [
30]. Moreover, dysregulated IFN-I signaling could be pathogenic and detrimental to fetal survival [
31,
32] and could be even more so under the condition of SLE because the disease per se is distinguished by the overactivation of the IFN-I signaling pathway [
4]. We examined the expression of five ISGs (MX1, IRF7, OAS1, IFIT1, and IFI44) in PBMCs from 13 pregnant patients with SLE to investigate whether ISG expression is associated with sialylated IgG glycoforms (Fig.2). We further introduced two derived traits (FG1S1/(FG1 + FG1S1) and FG2S1/(FG2 + FG2S1 + FG2S2)) to represent the level of terminal sialic acid on IgG glycan while ruling out the influence of other glyco-modifications. Our results showed that the expression levels of the five ISGs were negatively correlated with the level of IgG sialylation, indicating the probable effect of sialylated IgG glycoforms on IFN-I production.
3.3 Lack of sialylation dampens the ability of IgG to suppress the function of activated pDCs
PDCs are now commonly acknowledged to be a major source of IFN-I, which promotes a myriad of clinical manifestations in SLE [
33]. We used α2-3,6,8,9 neuraminidase to cleave IgG terminal sialic acid and obtained deSia-IgG to study whether modifications in the level of IgG sialylation regulated the production of IFN-I via affecting pDC functions. The desialylation of IgG was confirmed through HILIC-UPLC, and the intact structure of deSia-IgG was determined through Western blot analysis (Fig. S1). The effects of IgG and deSia-IgG were investigated by using purified human pDCs stimulated by the TLR7 ligand R848 and TLR9 ligand ODN2216. PDCs were activated upon TLR7 and TLR9 stimulation with the upregulation of the expression of the costimulatory molecules CD40 and HLADR on the surfaces of pDCs and the significantly increased secretion of type I interferon (IFN-I), IFN-α, and IFN-β (Fig.3). However, IgG inhibited the activated state of pDC by reducing surface CD40 and HLADR expression and IFN-I in the supernatant and demonstrated a stronger inhibitory effect at the concentration of 10 mg/mL than at 5 mg/mL (Fig.3 and 3B). Surprisingly, the anti-inflammatory effect of deSia-IgG on activated pDCs was markedly attenuated compared with that of untreated IgG at the same concentration. The results of flow cytometry showed that in addition to IFN-I, IgG inhibited the production of IL-6 and TNF by pDC under TLR7 and TLR9 stimulation, whereas the inhibitory effect of deSia-IgG significantly weakened or disappeared (Fig.3).
3.4 IgG transduces signals via the SYK pathway
We performed RNA sequencing on ODN 2216-stimulated pDCs treated with IgG or deSia-IgG to obtain a comprehensive view of the effect of IgG and deSia-IgG on activated pDCs. The RNA-seq results indicated that the gene expression profile of pDCs treated with CpG + IgG and CpG + deSia-IgG markedly differed in terms of the expression of genes associated with TLR signaling and IFN-I-related pathways (Fig.4). Furthermore, the PI3K–AKT and Fc receptor pathways were found to be differentially regulated by IgG and deSia-IgG. The former is critical for pDC IFN-I production, and the latter is downstream of IgG. The results of gene set enrichment analysis demonstrated that importantly, IgG suppressed the TLR pathway in a considerably more efficient manner than deSia-IgG (Fig.4). Furthermore, we observed that the IFN-I genes in IgG-treated pDCs were downregulated relative to those in deSia-IgG treated pDCs, though the latter also showed a moderate decrease compared with those in controls (Fig.4). Remarkably, kinase enrichment analysis found that the expression of genes associated with the SYK signaling pathway significantly differed between IgG and deSia-IgG treated pDCs (Fig.4 and 4E). Specifically, genes that could be induced by SYK signaling were upregulated by IgG treatment, and genes that were reported to be inhibited by SYK were suppressed by IgG engagement.
Under normal conditions, IFN-I production is kept under control to assure a balanced TLR-mediated response, which is attributed to certain regulatory receptors on the pDC surface [
34,
35]. ITAM-containing adaptor signaling, which involves the phosphorylation of SYK and downstream BLNK, exerts a potent negative regulatory effect on IFN production by pDCs [
36]. A significant increase in the phosphorylation of SYK and BLNK was observed in IgG-treated pDCs, whereas the ability to phosphorylate SYK and its downstream molecule BLNK was impaired in deSia-IgG-treated pDCs (Fig.5), suggesting that sialylation is necessary for IgG to enhance the activation of the SYK pathway. Meanwhile, the mean fluorescence intensity (MFI) of phosphorylated IRF-7 (pIRF-7) was suppressed in the IgG group relative to that in the control group, whereas deSia-IgG-treated pDCs demonstrated a similar pIRF-7 MFI as the control (Fig.5). Furthermore, blocking SYK in pDCs eliminated the inhibitory effect of IgG on the production of IFN-I (Fig. S2). Taken together, these results demonstrated that IgG regulates pDC function via the SYK pathway in a sialylation-dependent manner.
3.5 IgG controls IFN-I produced by pDCs from pregnant patients with SLE in a sialylation-dependent manner
Given that IgG exerted an inhibitory effect on pDCs from healthy donors and exhibited impaired capacity after desialylation, we further tested the effects of IgG and deSia-IgG on pDCs from pregnant patients with SLE. We used ODN2216 to stimulate pDCs from seven pregnant patients with SLE. Treatment with IgG at a concentration similar to normal serum levels markedly reduced the production of IFN-α, IL-6, and TNF-α (Fig.6). However, with the abrogation of the terminal sialic acid, deSia-IgG was observed to only mildly suppress IFN-α production while exerting no significant effect on the levels of IL-6 and TNF-α, indicating a compromised anti-inflammatory function.
4 Discussion
Our study provides a preliminary description of the change in IgG sialylation in lupus pregnancy and how the sialylation of IgG interferes with disease activity and pregnancy outcomes in patients with SLE.
The glycosylation of IgG at the conserved site at Asn297 has been proven to be an indispensable element influencing IgG effector functions. Thus far, however, the mechanisms regulating IgG glycosylation, especially the α2,6-sialylation of the
N-glycan of IgG, remain poorly understood. Although data on the relationship between IgG sialylation and B-cell ST6Gal1 expression have been published [
37], they are contradicted by the results of another study demonstrating that B-cell ST6Gal1 is not a prerequisite in IgG sialylation [
38]. Moreover, Jones
et al. [
39] elucidated the possibility that IgG sialylation can be modified extracellularly. However, Oswald
et al. [
40] arrived at a different conclusion that neither ST6Gal1 in B cells nor in plasma induces the sialylation of IgG. Thus, how the change in IgG sialylation occurs under different physiological and pathological circumstances has not been conclusively determined. The removal of the terminal sialic acids attached to IgG Fc renders IgG incapable of suppressing inflammation [
41]. Recent studies have suggested that terminal sialylation has a critical role in suppressing autoimmunity through the upregulation of the inhibitory effect of FCγRIIb on macrophages [
42] and the reduction in the affinity of IgG for type I FcγRs [
43]. Moreover, in RA, sialylated IgGs mediate anti-inflammatory effects by inhibiting macrophages through binding to C-type lectin receptor SIGN-R1 [
44]. Furthermore, T cell-independent immune responses induce suppressive sialylated IgGs, which inhibit B-cell activation and pathogenic immune reactions [
45]. In the present study, we observed that consistent with the inflammatory profile of the disease itself, sialylation in the SLE cohort was downregulated compared with that control cohort. Moreover, hyposialylation was associated with high disease activity as measured by SLEPDAI. Patients with SLE and higher disease activity are more prone to severe APOs than other patients [
46–
48]. In our study, we observed the remarkable downregulation of sialylated IgG glycans in patients with SLE and lupus flares who experienced fetal loss in the second trimester. The highly proinflammatory glycan profiles of patients with active disease suggests the potential pathogenic activity of hyposialylated IgG, which might pose a threat to fetal survival during early pregnancy. Furthermore, derived traits associated with sialic acid showed potential diagnostic performance for fetal loss in lupus pregnancy, highlighting a probable biomarker for severe adverse fetal outcome in mothers with lupus.
The sialylation-dependent anti-inflammatory properties of IgG have been addressed in a K/N arthritis model [
41]. Intriguingly, Wiedeman
et al. [
49] previously proposed that sialylation is not a requisite for the anti-inflammatory effect of IgG on pDCs. Notably, the concentration of IgG was relatively low (500 and 5000 μg/mL) in their study. However, by adopting a concentration of IgG/deSia-IgG similar to that in serum (10 mg/mL), we introduced a different mechanism by which the sialylation of IgG indeed regulated the effect of IgG on pDC functions. These results indicated a concentration-dependent mechanism in sialylation-regulated IgG function.
IFN-I plays a crucial role in the pathogenesis of SLE and has become one of the promising targets of immunotherapy [
50]. The persistent activation of the IFN-I/IFNAR axis renders pregnancies susceptible to inflammation-driven complications [
51]. By performing RNA-seq analysis, we confirmed the differential regulation of IFN-I production by IgG and deSia-IgG as demonstrated by the expression profiles of genes in the TLR signaling and SYK pathways. In such a context, our present study provided a novel mechanism for the regulation of IFN-I that functions via the modulation of terminal sialic acids on IgG.
Collectively, our study demonstrated that the inflammatory ability of IgG in controlling pDC functions acts in a sialic acid-dependent manner, suggesting that IgG sialylation may serve as a pivotal modulator in lupus flares during pregnancy. Sialic acid modulates the effect of IgG on pDCs via regulating the phosphorylation of downstream molecules of SYK pathway, indicating a potential axis for future therapeutic strategies for pDC- or IFN-I-related diseases.
PDF
(5322KB)
