Introduction

Immune thrombocytopenia (ITP) is an immune-mediated bleeding disorder in which platelets are opsonized by autoantibodies and prematurely destroyed by phagocytic cells in the reticuloendothelial system [1, 2]. The autoantibodies produced by autoreactive B lymphocytes against self-antigens are considered to play a crucial role. B-cell activating factor belonging to the tumor necrosis factor (TNF) family (BAFF) is critical for the maintenance of normal B-cell development and homeostasis [3, 4], T-cell costimulation [5], and certain. Th1-associated inflammatory responses [6]. It binds to three TNF receptor superfamily members: B-cell maturation antigen (BCMA), transmembrane activator, and CAML calcium-modulating cyclophilin ligand interactor (TACI) and BAFF receptor (BAFF-R/BR3). BR3 is the key receptor that triggers BAFF-mediated survival, as mice deficient in BR3 display a phenotype similar to that of BAFF-null mice [7]. BCMA−/− mice are born with no major immune defect [7]. TACI, by contrast, emerged as a negative regulator of B-cell activation and expansion, as numbers of B cells are increased in TACI−/− mice [7]. Excess BAFF results in the rescue of self-reactive B cells from anergy, thus implicating an important role in the development of autoimmunity [8]. Overexpression of BAFF in mice leads to autoimmunity with SLE-like symptoms, while mature B cells are lacking in BAFF-deficient mice [9]. Some studies demonstrated elevated BAFF levels in patients with systemic autoimmune diseases such as SLE, rheumatoid arthritis (RA), and Sjogren syndrome [10–12]. Elevated expression of BAFF has been reported in ITP patients [13, 14].

The treatment regimens for ITP include glucorticosteroids (GCs), intravenous immunoglobulin (IVIg), intravenous anti-D immunoglobulin, splenectomy, the thrombopoietin receptor agonists, danazol, and other immunosuppressive drugs in which GCs have been widely recognized as the most appropriate first-line treatment [15]. Recently, high-dose dexamethasone (HD-DXM) has been used as the first-line therapy for adult patients with ITP, which has a higher response rate than conventional prednisone doses [16–18]. However, the impact of HD-DXM on BAFF in ITP patients is still unclear. The objective of the study was to estimate the levels of BAFF and BAFF mRNA in untreated adult patients with ITP and to investigate the effect of HD-DXM on BAFF expression in patients and in vitro.

Materials and Methods

Patients and Controls

Twenty ITP patients with active disease (13 women and seven men, age range 19–66 years, median 41 years) were enrolled in this study (Table 1). The platelet counts of patients at pretherapy (median 17 × 109/l, range 5–36 × 109/l) were significantly lower than that of patients at posttreatment (median 159 × 109/l, range 23–332 × 109/l; P < 0.001). Patient 11 to patient 20 before HD-DXM were also used for cell culture (six women and four men, age range 19–66 years, median 44 years; platelets range 8–34 × 109/l, median 18 × 109/l; Table 1).

Table 1 Clinical Characteristics of Active ITP Patients
Full size table

The normal control group consisted of 24 adult healthy volunteers (15 women and nine men, age range 23–73 years, median 39 years). Platelet counts ranged from 159 to 289 × 109/l, with the median count of 197 × 109/l. Eleven of 24 healthy controls were also used for cell culture (six women and five men, age range 21–70 years, median 42 years).

Enrollment took place between November 2007 and January 2009 at the Hematology Department of Qilu Hospital affiliated to Shandong University. All of the cases met the diagnosis criteria of ITP as previously described [19]. Samples were collected both before and 2 weeks after HD-DXM treatment. Patients complicated with diabetes, hypertension, cardiovascular diseases, pregnancy, active or chronic infection, or connective tissue diseases such as systemic lupus erythematosus were excluded. The study was approved by the Medical Ethical Committee of Qilu Hospital and the Second Hospital of Shandong University. Informed consent was obtained from each patient before being included in the study.

Treatment Regimen

All patients received HD-DXM 40 mg/day for four consecutive days. Initial response evaluation was made at the end of the second week after treatment initiation. The response was evaluated according to the following criteria [19]: complete response (CR) was defined as platelet count ≥100 × 109/l and absence of bleeding; response (R) was defined as any platelet between 30 and 100 × 109/l and at least doubling of the baseline counts and absence of bleeding; no response (NR) was defined as any platelet count less than 30 × 109/l or less than doubling of the baseline counts or bleeding.

Preparation of Plasma and Peripheral Blood Mononuclear Cells

Plasma obtained from all subjects by centrifugation of heparinized peripheral blood samples was stored at −80°C until determination of cytokines. Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized blood using 1.077 g/ml Ficoll-Hypaque (Invitrogen, Carlsbad, CA, USA) gradient centrifugation (2,000 rpm for 20 min, 20°C). The isolated PBMCs were washed twice with 0.9% NaCl then resuspended and adjusted to 1 × 106 PBMCs/ml for cell culture, and 1 × 106 PBMCs were stored at −80°C for use.

PBMCs isolated from ten untreated ITP patients and 11 healthy controls with active disease (patient no. 11–20, Table 1) were adjusted to 1 × 106/ml in RPMI-1640 culture medium (Invitrogen), cultured at a density of 1 × 106 cells/well in a 24-well culture plate and incubated in humidified air in 5% CO2 at 37°C with different concentrations of DXM. For enzyme-linked immunosorbent assay (ELISA), the cells were also treated with 10 μg/ml phytohemagglutinin (PHA; Sigma, USA). After 72 h, PBMCs and cell supernatants were collected for quantifying BAFF expression and secreted protein of IFN-γ and IL-4 using ELISA method.

BAFF, IFN-γ, and IL-4 Enzyme-Linked Immunosorbent Assay

Plasma BAFF was measured by a commercial ELISA according to manufacturer's (R&D Systems, Minneapolis, MN, USA) instructions. The lower detection limit of this assay was 62.5 pg/ml. IFN-γ and IL-4 in supernatants of cultures were determined using a commercial ELISA (Bender MedSystems, Burlingame, CA, USA) according to manufacturer’s instructions.

Determination of the Expression of BAFF mRNA

The TRIzol reagent (Invitrogen, Carlsbad, CA, USA) was used to isolate total RNA. RNA was converted into cDNA using the PrimeScript™ RT Reagent Kit (Perfect Real Time; Takara, Japan) according to the manufacturer's instructions. Multiplex Real-time PCR was performed for BAFF and the endogenous control (β-actin) on an ABI PRISM_7500 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) by using SYBR Green (Toyobo, Japan) as a double-strand DNA-specific binding dye. The primers for all mRNA assays were intron spanning. The PCR reactions were cycled 40 times after initial denaturation (95°C, 5 min) with the following parameters: denaturation at 95°C for 15 s; annealing at 60°C for 15 s; extension at 72°C for 35 s, with temperature transition rates of 20°C/s. The primers for BAFF and β-actin are as follows: BAFF-F, AAGACCTACGCCATGGGACATC; BAFF-R, TCTTGGTATTGCAAGTTGGAGTTCA; β-actin-F, TTGCCGACAGGATGCAGAA; β-actin-R, GCCGATCCACACGGAGTACT.

We used the comparative Ct method (using arithmetic formulae) for relative quantification of cytokine mRNA according to relative expression software tool (REST©) [20]. The amplification efficiency between the target (BAFF) and the reference control (β-actin) were compared in order to use the delta delta Ct (ΔΔCt) calculation.

The Proliferation Assay

2 × 105 PBMCs/well were cultured in RPMI 1640 medium supplemented with 10% FCS in 96-well flat-bottomed plate (0.2 ml final vol) with different concentrations of DXM with at 37°C with 5% CO2. Proliferation was assessed after 72 h by cell counting kit-8 (CCK-8).

Statistical Analysis

Data were expressed as mean ± SD. Statistical significance was determined by ANOVA, and difference between two groups was determined by Newman–Keuls multiple comparison test (q test). Pearson correlation was used for correlation analysis. All tests were performed by SPSS 13.0 system. P value less than 0.05 was considered statistically significant.

Results

Clinical Therapeutic Effect of HD-DXM

Responses were reached in patients: CR in 16 (80%), R in three (15%), and NR in one (5%) The mean platelet count was 165 × 109/l (range 23 to 332 × 109/l) 2 weeks after the initiation of treatment.

Elevated Expression of BAFF in ITP Patients

Figure 1a shows the serum BAFF levels of different groups. Serum BAFF levels in untreated ITP patients (mean ± SD, 597 ± 197 pg/ml) were significantly higher than that of healthy controls (454 ± 132 pg/ml, P < 0.05).

Fig. 1
figure 1

Levels of plasma BAFF and BAFF mRNA in patients and controls. a Analysis of plasma BAFF from 20 active ITP patients both before and after HD-DXM treatment and 24 healthy controls. Plasma BAFF was elevated in untreated patients (mean 597 pg/ml) compared to patients received HD-DXM treatment (mean 297 pg/ml, P < 0.001) and healthy controls (mean 454 pg/ml, P < 0.05). b The relative expression of BAFF mRNA. Levels of BAFF mRNA in untreated patients were increased compared to healthy controls (P < 0.001), and after administration of HD-DXM, it reduced to lower than normal (P < 0.001). Bars represent SD. *P < 0.05, **P < 0.001

Full size image

Using the REST software, the data are presented as the fold change in gene expression normalized to an endogenous reference gene and relative to healthy controls. The relative amount of BAFF mRNA in untreated patients was 2.5-fold of that of healthy controls (P < 0.001; Fig. 1b).

Decreased Expression of BAFF in Active ITP Patients with HD-DXM Treatment

After administration of HD-DXM, plasma BAFF levels in ITP patients were significantly decreased (297 ± 120 pg/ml) compared with pretherapy (597 ± 197 pg/ml, P < 0.001) and lower than that of the normal controls (454 ± 132 pg/ml, P < 0.001; Fig. 1a).

Similar results were found on BAFF mRNA levels. After administration of HD-DXM, the relative amount of BAFF mRNA was 0.32- and 0.13-fold of that of healthy controls (P < 0.001) and untreated patients, respectively (P < 0.001; Fig. 1b).

Changes of BAFF Correlated with Clinical Responses

After administration of HD-DXM, 18 of 20 patients had reduced plasma BAFF level, with two patients who relapsed shortly after discontinuation of the therapy and later received splenectomy had similar or even higher expression. Changes of plasma BAFF in patients before and after HD-DXM treatment in active ITP patients were shown in Fig. 2.

Fig. 2
figure 2

Changes of plasma BAFF in patients before and after HD-DXM treatment from 20 active ITP patients. After administration of HD-DXM, 18 of 20 patients had reduced plasma BAFF level, with two of them having similar or even higher expression compared to pretreatment. Asterisks Patients who had similar or even higher plasma BAFF between pre- and posttreatment

Full size image

Correlation of Plasma BAFF and its mRNA Levels in Active ITP Patients

We observed that plasma BAFF correlated with its mRNA levels in active ITP patients (r = 0.45, P < 0.01, Fig. 3).

Fig. 3
figure 3

Positive correlation between plasma BAFF levels and its mRNA level in active ITP patients. Plasma BAFF and its mRNA levels were measured by ELISA and real-time PCR. There was a positive correlation between them (P < 0.01)

Full size image

Expression of BAFF Is Inhibited by DXM In Vitro

Because BAFF is elevated in untreated ITP patients and after administration of HD-DXM it reduced significantly, we investigated whether DXM, a most used glucocorticoids for ITP therapy, is able to inhibit the expression of BAFF. We cultured PBMCs from ten untreated ITP patients with active disease (patient no. 11–20, see Table 1), and 11 controls were adjusted to 1 × 106/ml in RPMI-1640 culture medium (Invitrogen), cultured at a density of 1 × 106 cells/well in a 24-well culture plate, and incubated with different concentrations of DXM at 37°C with 5% CO2; after 72 h, cells were harvested for real-time PCR. The results showed that DXM suppressed the expression of BAFF in a dose-dependent manner. Figure 4 shows the dose-dependent inhibition of BAFF mRNA of DXM in active ITP patients.

Fig. 4
figure 4

DXM inhibits expression of BAFF mRNA in active ITP patients. BAFF mRNA expression was determined by real-time PCR. β-Actin mRNA was used as control. Relative expression represents the BAFF expression normalized by β-actin (BAFF/β-actin) in each cell culture with respect to the expression of BAFF/β-actin measured in untreated cells

Full size image

Effects of DXM on the Expression of IFN-γ and IL-4

PBMCs from active ITP patients and healthy controls were cultured with different concentrations of DXM (0, 7.8, 15.6, 31.3, 62.5, 125, 250 nM) at 37°C with 5% CO2, with 10 μg/ml PHA (Sigma). After 72 h, supernatants were collected, and levels of IFN-γ and IL-4 were measured. When PHA was added, levels of IFN-γ were available in all samples. A dose-dependent inhibition of IFN-γ by DXM was obvious. Figure 5a shows the dose-dependent inhibition of IFN-γ of DXM in active ITP patients. The levels of plasma IL-4 were lower than the detection limits.

Fig. 5
figure 5

Effects of DXM on IFN-γ and proliferation of lymphocytes. a Inhibition of IFN-γ by DXM in a dose-dependent manner. Regarding the level of IFN-γ with no DXM as 100%, the levels of IFN-γ was inhibited to 82%, 58%, 43%, 28%, 15%, and 10% with DXM at 7.8, 15.6, 31.3, 62.5, 125, and 250 nM, respectively. b Inhibition of proliferation of lymphocytes by DXM. Regarding proliferation of lymphocytes with no DXM as 100%, the levels of proliferation were inhibited to 73%, 61%, 48%, 45%, 29%, and 25% with DXM at 0.125, 0.25, 0.5, 1, 2, and 4 μM, respectively. Results are expressed as mean ±SD

Full size image

Effects of DXM on the Proliferation of PBMCs

After cultured with different concentrations of DXM (0, 0.125, 0.25, 0.5, 1, 2, and 4 μM) for 72 h, proliferation in active ITP patients and controls was assayed by CCK-8. The results showed that DXM inhibited the proliferation of PBMCs of both patients and controls, reaching about 50% of inhibition with 0.5 μM DXM and about 75% inhibition with 2 μM DXM in active ITP patients (Fig. 5b). Similar results in healthy controls were observed (data not shown).

Correlation of BAFF with Clinical and Laboratory Parameters in ITP Patients

Correlations between levels of BAFF and platelet counts were analyzed in ITP patients, and no significant associations were found (P > 0.05). There was no significant correlation between changes of BAFF and platelets.

Discussion

In this study, HD-DXM showed a good initial response in ITP patients, which was in accordance with previous reports [16–18]. The mechanism of effects of GCs on immunosuppression has been studied in details. Some studies have showed that the treatment with GCs induces a marked decrease in BAFF levels in patients with SLE and bullous pemphigoid [10, 21]. Recently, a study about the inhibition of DXM on BAFF in fibroblast-like synoviocytes from patients with RA has been reported, which showed that DXM inhibited the expression of BAFF in a dose- and time-dependent manner [22]. Our data demonstrated the plasma BAFF and BAFF mRNA were remarkably decreased in active patients after HD-DXM treatment, and in in vitro study, DXM significantly inhibited the expression of BAFF mRNA in a dose-dependent manner.

In our study, we found that plasma BAFF and BAFF mRNA were significantly increased in active ITP patients at pretherapy. After 4-day treatment with HD-DXM, the plasma BAFF and BAFF mRNA were remarkably decreased and lower than that of the normal controls. These results indicated that BAFF may also play a critical role in ITP. As a member of the TNF superfamily, BAFF not only regulates B-cell immunity by inducing the survival [23] and maturation of B cells [24], decreasing activation-induced cell death in stimulated B cells [25], but also stimulates T cells by delivering costimulation to TCR-dependent signals [5]. It is mainly expressed by cells of the myeloid lineage such as monocytes, macrophages, and dendritic cells [3]. The expression of BAFF can be up-regulated in myeloid cells by pro-inflammatory cytokines such as IFN-γ, IFN-α, and TNF-α [26]. Oppositely, IFN-γ and TNF-α were also reported to be induced by recombinant BAFF (rBAFF) [5].

ITP is an autoimmune disorder manifested by antibody-mediated platelet clearance and dysmegakaryocytopoiesis, disturbed apoptosis of T-lymphocyte, imbalance of Th1/Th2 ratio, and CTL-mediated platelet lysis [27–30]. It is well known that GCs is considered the most largely used therapeutic approach as first-line treatment for ITP patients. Recently, a single HD-DXM course was administered as first-line therapy in adult patients with ITP [16–18]. It not only could increase the number of CD4+Foxp3+ Treg cells and myeloid DCs but also could correct the Th1 cytokine dominance in ITP patients [31, 32]. In addition, it also could inhibit the expression of BAFF.

Several possible mechanisms may contribute to the effects of DXM to BAFF from our study. First, DXM may influence BAFF expression partly by regulating the production of IFN-γ. GCs were known to affect cytokine synthesis in T cells by binding to and activating cytoplasmic GC receptors. Several studies indicate ITP patients were Th1 dominant with elevated IFN-γ or other th1 cytokines [30, 32]. Level of plasma IFN-γ in ITP patients reduced significantly in ITP patients after HD-DXM treatment compared to that before treatment [32]. In in vitro study, our data showed that DXM significantly inhibited the expression of IFN-γ in a dose-dependent manner. DXM may inhibit BAFF expression partly by reducing expression of IFN-γ. However, the expression of BAFF can be up-regulated by IFN-γ, and IFN-γ also can be induced by BAFF [5, 26]. The reduction of IFN-γ may also be a result of reduced levels of BAFF. Our preliminary study indicated that rBAFF promoted the expression of IFN-γ in PBMCs from ITP patients in vitro; however, the data are limited, and further studies are needed to be done to clarify it. Second, DXM may influence BAFF expression partly by inhibiting proliferation of lymphocytes and/or monocytes. BAFF is mainly expressed by cells of the myeloid lineage such as monocytes, macrophages, neutrophils, and dendritic cells [3, 26], and its three receptors are primarily expressed on B cells, with BAFF-R expressed on all peripheral blood B cells [33]. In our study, DXM significantly inhibited proliferation of PBMCs, which may lead to reduced production of BAFF or abnormal signaling pathway of BAFF/BAFF receptors. In addition, the function and signaling pathway of BAFF/BAFF receptors is complex, especially with negative regulation by TACI, so more work needs to be done to clarify the exact mechanism of DXM on BAFF in ITP patients.

In summary, elevated BAFF in untreated active ITP patients indicate its possible role in the pathogenesis of ITP; DXM inhibited the expression of BAFF possibly by its influence on cytokines such as IFN-γ that regulating BAFF or by its effects on myeloid cells such as lymphocytes and/or monocytes that express BAFF and its receptors. The elevated BAFF could be corrected by HD-DXM.