IFN-β therapy modulates B-cell and monocyte crosstalk via TLR7 in multiple sclerosis patients

Authors


Full correspondence: Dr. Eliana Marina Coccia, Department of Infectious, Parasitic and Immune-mediated Diseases, Istituto Superiore di Sanità, Viale Regina Elena 299, 00161 Rome, Italy

Fax: +39-06-49903638

E-mail: eliana.coccia@iss.it

Abstract

The implication of B lymphocytes in the immunopathology of multiple sclerosis (MS) is increasingly recognized. Here we investigated the response of B cells to IFN-β, a first-line therapy for relapsing-remitting MS patients, upon stimulation with TLR. IFN-β restored the frequency of TLR7-induced IgM and IgG-secreting cells in MS patients to the levels found in healthy donors, showing a specific deficiency in the TLR7 pathway. However, no difference was observed in the TLR9 response. Furthermore, in MS-derived PBMCs, TLR7-mediated production of IL-6 and the ex vivo expression of B-cell-activating factor of the TNF family, two crucial cytokines for B-cell differentiation and survival, were induced by IFN-β. Depletion of monocytes, which are key producers of both IL-6 and B-cell-activating factor of the TNF family, showed that TLR7-mediated B-cell differentiation into Ig-secreting cells is strongly dependent on the cross-talk between B cells and monocytes. Accordingly, impaired expression of TLR7 mRNA was observed in PBMCs and monocytes isolated from MS-affected individuals as compared with those from healthy donors, which was rescued by IFN-β therapy. Collectively, our data unveil a novel TLR7-regulated mechanism in in vivo IFN-β-stimulated whole leukocytes that could be exploited to define new TLR7-based strategies for the treatment of MS.

Introduction

Growing evidence is accumulating on the central role that B lymphocytes play in the immunopathology of multiple sclerosis (MS) [1, 2]. For example, oligoclonal IgG bands, found in the cerebrospinal fluid of more than 90% of patients with MS, indicate an intrathecal Ab production [3]. The presence of clonally expanded B cells, plasma cells, complement and myelin-specific Abs in chronic MS lesions also suggested an intrathecal, Ag-driven humoral immune response in the central nervous system of MS sufferers [4-6]. In addition, B-cell follicle-like structures are detectable in the meninges of MS patients [7, 8]. More recently, B-cell depleting therapies, such as Rituximab (that targets the B lymphocyte surface antigen CD20 [9-11]), together with Ocrelizumab and Ofatumumab (two other humanized anti-CD20 monoclonal Abs), are proving their efficacy at various stages of clinical development [12]. All together these findings contribute to the compelling evidence that B cells and the humoral immune response are implicated in the pathogenesis of MS and suggest the therapeutic implications that all this may have for the treatment of this disease.

B lymphocytes play an essential role in bridging innate and adaptive immunity. To differentiate into specialized cells capable of communicating with helper T cells and undergo Ab diversification, clonal expansion, and Ig secretion, B lymphocytes need the support of a coordinated network of cytokines, growth factors, adhesion, and ligand-receptor signals [13]. Among B-cell receptors, the TLRs and their natural agonists have raised interest since they elicit direct effects on human B cells [14]. TLRs are germ-line encoded pattern recognition receptors that can detect conserved molecular patterns either expressed on microorganisms or of self-origin. Targeting them or modulating their functions may have therapeutic potential in autoimmune diseases, including MS [15]. B lymphocytes selectively express TLR7 and TLR9 and are activated by their specific ligands [16, 17]. At variance with other TLRs, TLR7 and TLR9 share relevant properties. Indeed, they both recognize microbial and endogenous nucleic acids; in particular, TLR7 specifically binds guanosine- and uridine-rich ssRNAs while TLR9 senses hypomethylated CpG-rich dsDNAs. Furthermore, they both reside in the endosomal compartments, unlike the other TLRs present on the cell surface.

In this study, we characterized the TLR-induced modulation of B-cell responses and differentiation in MS patients undergoing IFN-β therapy, a first-line drug used in the treatment of relapsing remitting MS (RRMS), whose mechanism of action remains poorly understood. IFN-β-mediated immunomodulatory functions may differentially operate depending on the responding cell subset acting on T- or B-cell proliferation, modulation of cytokine production, and regulation of adhesion molecules involved in lymphocyte migration across the blood-brain barrier [18]. For these reasons, investigating the action of IFN-β therapy on B cells might be of great relevance to understand their pathogenic role in the development and regulation of autoimmune inflammatory response in MS.

Results

IFN-β therapy selectively promotes TLR7-mediated B-cell maturation and Ig production

There is increasing recognition that TLRs and TLR-driven responses can play a key role in the pathogenesis of several autoimmune diseases, including MS. TLR7 and TLR9 are selectively expressed by B cells, and when activated by specific ligands, lead to their proliferation and differentiation into Ig-secreting cells. Given the key importance of B lymphocytes in MS disease, we investigated whether IFN-β therapy would modulate Ig synthesis in MS patients by performing a longitudinal study conducted with unseparated PBMCs isolated from 15 MS patients before (T0) and 1 month after (T1) the beginning of IFN-β therapy. Moreover, PBMCs isolated from 10 healthy donors (HDs) were also included in this study as comparative control.

To this end, PBMCs were cultured in vitro with either a specific TLR7 (the synthetic small molecule 3M001) or TLR9 (a type B CpG, 2006) agonist for 7 days and then IgM (Fig. 1A) and IgG production were evaluated by Elispot (Fig. 1B) and Elisa assay (Supporting Information Fig. 1). The TLR9-mediated B-cell stimulation led to a similar frequency of IgM- and IgG-secreting cells in both HD- and MS-affected individuals and this Ab release was not modified in response to IFN-β treatment. On the other hand, it was very interesting to find that the basal level of TLR7-induced Ig production was significantly lower in MS patients as compared with that in HD, showing a specific defect in TLR7 responses in B cells from MS sufferers.

Figure 1.

TLR7-mediated Ig production is selectively promoted by IFN-β therapy. PBMCs from 10 HDs or 15 MS patients before (T0) and 1 month (T1) after IFN-β therapy were treated with TLR7 (3M001) or TLR9 (CpG) agonist for 7 days. (A) IgM- or (B) IgG-secreting cells were detected and enumerated by isotype-specific ELISPOT. The values shown represent the means +SEM of results pooled from the indicated number of donors. p-values were calculated by two-tailed Student's t-test. For IgM: *p = 0.0353, **p = 0.0138. For IgG: *p = 0.0332, **p = 0.0169.

Surprisingly, 1 month of IFN-β therapy was able to partially restore this deficiency and selectively increase the production of IgM and IgG upon TLR7 triggering, re-establishing the level of Ab release found in HDs. The analysis of Ig content by Elisa confirmed the results obtained by Elispot assay (Supporting Information Fig. 1). IFN-β-mediated effect was long-lasting since it was still observed after 6 months of IFN-β treatment (data not shown). However, IFN-β did not enhance auto-Ab production as demonstrated by measurement of both homogeneous and speckled patterns of anti-ANA Abs on sera of MS patients before and after therapy (data not shown) [19].

In accordance with the Ig production, flow cytometry analysis of the co-stimulatory marker CD86 also highlighted an increased maturation status in in vivo IFN-β-treated CD19+ B cells from MS patients stimulated with TLR7, whereas no major differences were found upon TLR9 stimulation (Supporting Information Fig. 2A). In this experimental setting, we also observed a significant increase in the expression of the activation marker CD38 on B-cell surface after IFN-β treatment (Supporting Information Fig. 2B). Given that this protein is notoriously type I IFN inducible [20], this result clearly shows that B lymphocytes are target of the IFN-β therapy confirming previous study by Zula et al. [21] who described a rapid activation of IFN signal transduction pathways in B cells present in unseparated blood from RRMS patients soon after IFN-β injection.

TLR7 expression is lower in PBMCs and monocytes of MS patients compared with HDs

In the past, we dissected the regulation of TLR7 in maturing monocyte-derived DCs and observed that its transcription was dependent on the endogenous IFN-β release [22]. Thus, to evaluate whether IFN-β therapy would modulate TLR7 expression in MS patients, we first monitored by real-time RT-PCR TLR7 level of transcription, together with that of TLR9, in MS patients versus HDs. It was of great interest to find that PBMCs obtained from MS patients display a clear defect, as compared with those of HDs, in TLR7 expression that was statistically significant (25 HDs and 45 MS patients analyzed) (Fig. 2A). This difference was not observed in the transcription of TLR9 gene (Fig. 2B), demonstrating that in MS patients, the defective TLR7 expression is specific. Furthermore, we observed that in PBMCs isolated from the same MS patients following 1 month of IFN-β therapy, the level of TLR7 mRNA was restored to the level observed in HDs, while that of TLR9 was not modulated (Fig. 2A and B).

Figure 2.

PBMCs and monocytes derived from MS patients display a low level of TLR7 gene expression. Cells were isolated from freshly drawn blood of HDs and MS patients before (T0) and after 1 month (T1) of IFN-β therapy. (A, B) Total RNA was prepared and (A) TLR7 or (B) TLR9 gene expression was evaluated by quantitative real-time RT-PCR. Data obtained from all enrolled HDs and MS individuals are shown as mean ± SEM. (C–E) Twenty-five HDs and 45 MS were enrolled for PBMC isolation; (C, D) 7 HDs and 13 MS for B cells purification; (E) 8 HDs, 8 MS patients without treatment (T0); and additional 5 MS patients before (T0) and after 1 month (T1) of IFN-β therapy for monocyte isolation. The values shown represent the means +SEM of results pooled from the indicated number of donors. *p = 0.03965 (two-tailed Student's t-test for unpaired data); **p = 0.029 (two-tailed Student's t-test for paired data); ***p = 0.0466 (two-tailed Student's t-test for unpaired data); ****p = 0.0014 (two-tailed Student's t-test for paired data).

In the attempt to investigate which TLR7-expressing cell types in the peripheral blood might be responsible for this defect in MS patients, B cells and monocytes were purified from both HDs and MS patients at baseline and 1 month after the beginning of IFN-β therapy, since these two leukocyte populations express TLR7. Data on TLR7 expression in B cells isolated from HDs or MS (7 and 13 individuals, respectively) did not mirror the impairment observed in the context of the mixed cell population of PBMCs (Fig. 2C and D), although a slightly enhanced level of TLR7 transcription in response to IFN-β occurred also in this experimental setting. As observed in unseparated PBMCs, TLR9 levels of B cells did not differ in HDs and MS patients irrespective of IFN-β treatment. Interestingly, when the expression of TLR7 was analyzed in monocytes of MS patients (13 individuals), a different picture appeared. Indeed, a lower TLR7 mRNA level was highlighted in monocytes from MS patients than that obtained from HD (8 individuals) and, moreover, also a robust induction was observed in response to IFN-β therapy (longitudinal analysis of 5 patients at baseline and 1 month after IFN-β treatment) (Fig. 2E). TLR9 expression was absent in monocytes (data not shown). These data for the first time indicated a defect in TLR7 signaling in monocytes of MS patients.

IFN-β enhances the expression of soluble factors involved in B-cell differentiation

To mature and differentiate into Ig-secreting cells, B lymphocytes need the support of a coordinated action of many cytokines, growth factors, and adhesion-dependent signals. Among them, IL-6 plays a pivotal and not redundant role acting on the survival and on the Ig secretion capacity of B cells [23].

Thus, to characterize the mechanisms underlying the differential responsiveness of B cells from MS patients to TLR7 and TLR9 stimulation, we measured by ELISA the production of IL-6 in PBMCs isolated from 10 HDs and from 15 MS patients before and after IFN-β therapy (Fig. 3A). PBMCs were treated for 24 h with the TLR7 and TLR9 ligands, 3M001 and CpG, respectively, and cytokine release was quantified. In line with the data obtained for Ig production, HD PBMCs secreted a robust level of IL-6 following TLR7 triggering that was significantly higher than the production seen in PBMCs of therapy-free MS patients. The lower cytokine release observed in MS patients was almost completely restored following IFN-β administration. TLR9 stimulation induced low amounts of this cytokine in HDs and the level of production was not differently modulated in MS individuals before and after IFN-β therapy.

Figure 3.

IL-6 and BAFF expression is strongly induced in MS patients undergoing IFN-β therapy. Whole PBMCs obtained from 10 HDs and 15 MS were cultured for 24 h with a TLR7 (3M001) or a TLR9 (CpG) ligand, respectively 3M001 and CpG. (A) IL-6 was measured in culture supernatants by ELISA. Data are shown as mean +SEM of results pooled from 10 HDs and 15 MS individuals before (T0) and after 1 month (T1) of IFN-β therapy. (B, C) BAFF expression was evaluated in HDs or MS patients before and after IFN-β therapy either (B) at the mRNA level by quantitative real-time PCR in freshly isolated PBMCs (5 HDs and 7 MS patients) or (C) at protein level by BAFF ELISA in sera (6 HDs or 12 MS patients). The values shown represent the means +SEM of results pooled from the indicated number of donors. For IL-6 *p = 0.005 (two-tailed Student's t-test for paired data).

In the same way, another key component of the milieu responsible for B-cell proliferation and differentiation into plasma cells is the B-cell-activating factor of the TNF family (BAFF), whose mRNA expression was found to be comparable between PBMCs of HDs and untreated MS patients but strongly induced upon IFN-β therapy (Fig. 3B). Accordingly, similar levels of BAFF were present in the sera of HDs and MS patients and were induced in response to IFN-β therapy (Fig. 3C), confirming previous data from Krumbholz et al. [24]. All together these results show that in MS patients, the lower humoral immune response upon TLR7 triggering is replenished by IFN-β treatment likely through the release of factors, such as IL-6 and BAFF, that mediate B-cell differentiation into Ig-secreting cells.

Monocytes have a key role in TLR7-stimulated B-cell differentiation and Ig production in MS patients

Having found that in MS patients monocytes display an impaired expression of the TLR7 gene (Fig. 2E), we hypothesized that this cell type might have a role in the defective TLR7-induced Ab response of MS patients through the release of cytokines involved in B-cell differentiation and activation, such as IL-6 and BAFF [25-27].

To test our hypothesis, we depleted PBMCs from 4 IFN-β-treated MS patients of monocytes and cultured total or monocyte-depleted PBMCs with the TLR7- or TLR9-specific ligands. While the poor induction of IL-6 was not affected by the depletion of monocytes upon TLR9 stimulation, a strong dependence on monocytes was observed for IL-6 release in response to TLR7 (Fig. 4A). In a similar manner, BAFF mRNA, strongly expressed in freshly drawn total PBMCs upon IFN-β therapy, displayed a clear reduction in the level of expression in IFN-β-treated monocyte-depleted PBMCs (Fig. 4B).

Figure 4.

Monocyte depletion completely abrogates IFN-β-induced IL-6 and BAFF expression in PBMCs from MS patients. (A) Supernatants were harvested from cultures of TLR7- or TLR9-stimulated total or monocyte-depleted PBMCs obtained from 4 IFN-β-treated MS patients. IL-6 levels were measured by ELISA. Data are shown as mean +SEM of results obtained from the four IFN-β-treated MS patients analyzed. (B) RNA was isolated from total or monocyte-depleted PBMCs obtained from 4 IFN-β-treated MS patients and BAFF expression calculated by real-time RT-PCR as specified above. The values shown represent the means +SEM of results pooled from the indicated number of donors.

In this scenario, it was not surprising to find that the TLR7-driven IgM and IgG synthesis, observed in PBMCs from IFN-β-treated MS patients, was dramatically decreased when monocytes were depleted (Fig. 5A and B). Similarly, when BAFF activity was prevented by the addition of a specific BAFF neutralizing Ab to PBMC cultures, a reduction in the TLR7-stimulated IgM and IgG production was obtained (Supporting Information Fig. 3). A different picture was found when Ig release was measured upon TLR9 triggering in either monocyte-depleted PBMCs or whole PBMCs treated with anti-BAFF Ab. Indeed, an enhanced release of both IgM and IgG was observed in response to TLR9 stimulation in the absence of monocytes while the neutralization of BAFF poorly affected Ig production (Fig. 5A and B and Supporting Information Fig. 3, respectively). This result was not obvious and, at this stage, it is difficult to explain but it suggests that monocytes could be associated to a negative feedback loop on TLR9-driven B-cell differentiation while they positively act on the TLR7 responsiveness of Ig-producing B cells. Thus, we can envisage that changes in the basal and/or TLR-induced cytokine milieu of in vivo IFN-β-conditioned PBMCs could profoundly impact on Ig production from B cells in response to TLR7 or TLR9 stimulation.

Figure 5.

In the absence of monocytes TLR7-treated B cells do not differentiate into Ig-secreting cells. Whole or monocyte-depleted PBMCs obtained from IFN-β-treated MS patients were treated with TLR7 (3M001) or TLR9 (CpG) agonists for 7 days. (A) IgM- or (B) IgG-secreting cells were detected and enumerated by isotype-specific ELISPOT. The values shown represent the means +SEM of results pooled from 4 MS patients. For IgM *p = 0.0325 and for IgG **p = 0.039 (two-tailed Student's t-test for paired data).

Collectively, these findings demonstrate that the cross-talk between monocytes and B cells is essential for the release of an effective humoral immune response in the context of TLR7 stimulation affecting the maturation and differentiation status of B lymphocytes into Ig-secreting cells.

Discussion

Over the past decade, there has been growing understanding and acceptance of the pathological involvement of B cells and humoral response in MS [1, 2]. The demonstration that peripheral B-cell depletion leads to a rapid decline in disease activity in MS is the strongest evidence of the central role of these cells in MS autoimmunity [9, 11]. However, the key question that still remains unsolved is when and how in the life of an individual B cell does provide immunopathogenic support or arise as a disease-relevant cell type in MS.

In this study, we investigated whether IFN-β targets B lymphocytes and modulates their functions contributing to the protective effects of this treatment. Only a few studies have thus far addressed this point and most have investigated the ability of highly purified B cells from MS patients to present antigens and subsequently regulate T-cell responses [28, 29]. In contrast, we studied whether IFN-β therapy would regulate the maturation and differentiation of B cells into Ig-secreting cells in response to TLR7 or TLR9 stimulation. Indeed, it has been shown that TLR triggering is necessary for extensive human naïve B-cell proliferation, isotypic switching, and production of Abs providing the third signal upon BCR cross-linking by antigen and interaction with T helper cells [30]. However, compared with naïve B cells, memory B cells have less stringent triggering requirements and can be readily activated also in the absence of BCR stimulation by polyclonal stimuli such as bystander T-cell help and TLR triggering, which modifies also the cytokine milieu, accelerating Ab production in secondary responses and maintaining the serological memory [31, 32]. Thus, in our experimental setting, the simultaneous presence of different immune populations in total PBMCs assured the presence of all the required signals for B-cell differentiation and offered a faithful representation of what is actually happening in vivo in the peripheral blood of MS patients.

Our results demonstrate a fundamental difference in the outcome of either TLR7 or TLR9 stimulation of B cells in the context of PBMCs isolated from HDs or MS patients. Indeed, while the treatment with a TLR9 ligand induced a comparable production of both IgG and IgM in control or MS-affected individuals, we highlighted for the first time a clear deficiency in TLR7-mediated B-cell differentiation into Ig-secreting cells in MS patients.

In vivo administered IFN-β is able to replenish in MS patients the low TLR7-induced Ig production to the level observed in HDs. In line with this evidence and consistent with previous findings [33], TLR7 expression was also upregulated by IFN-β both in whole PBMCs, purified B cells, and monocytes. Furthermore, three studies reported with different experimental approaches how IFN-α, another subtype of the type I IFN family to which IFN-β belongs, exogenously provided or in situ produced by plasmacytoid DC, enhances B-cell differentiation into IgM- and IgG-producing cells only in response to TLR7, but not TLR9, triggering [34-36]. We believe that in our settings in vivo IFN-β therapy might have similar activity to what is described in vitro for IFN-α.

IFN-β treatment enhances TLR7-induced B-cell responses in MS patients acting at different steps: not only on the regulation of TLR7 gene expression but also on the secretion of soluble factors of key importance for B-cell differentiation, namely IL-6 and BAFF. IL-6 promotes terminal differentiation of B cells to plasma cells [23, 37] and exerts also a pronounced effect on the survival and/or Ig secretion [38]. BAFF regulates, in tandem with APRIL (a proliferation-inducing ligand), B-cell survival, differentiation and class switching, determines the size of the peripheral B-cell pool and is essential for maintenance of the peripheral B-cell repertoire and initiation of T-cell independent B-cell responses [39]. BAFF has been implicated in the development of autoimmunity in experimental settings and in several human B-cell-related autoimmune diseases, including MS [39]. Interestingly, Serafini and Aloisi in collaboration with our team also found that BAFF is expressed in EBV-infected B cells in acute MS lesions and ectopic B-cell follicles [40], highlighting the key role of this factor in B-cell activation also in the MS brain.

In our experimental setting, we observed that IL-6 was strongly induced in TLR7-treated PBMCs derived from IFN-β-treated MS patients and this result likely correlates with the replenishment of TLR7 expression and responsiveness in in vivo IFN-β-conditioned monocytes. Indeed, when PBMCs derived from IFN-β-treated patients were depleted of monocytes, the strong induction of IL-6 observed in total PBMCs was completely lost. In addition, a strong reduction of BAFF expression was observed in in vivo IFN-β-conditioned PBMCs after the depletion of monocytes. In a similar fashion, in the absence of monocytes, there was no induction of TLR7-driven IgM and IgG production, indicating that IFN-β treatment could exert its therapeutic effects by fine-tuning monocyte functions, in the context of TLR7 stimulation, that act through bystander mechanisms on the differentiation of B lymphocytes. Taking into account that TLR7 is crucial for type I IFN release from pDC [41] and is, at the same time, an IFN-inducible gene [22], we can envisage the existence of a tight relation between IFN-β response and TLR7 responsiveness of MS monocytes, whose full comprehension deserves further investigation.

In line with this view, recent data obtained by Molnarfi and collaborators showed that monocytes from RRMS patients exhibited a reduced ability to produce HGF, a neuroprotective and neuroinflammation-suppressive mediator, when compared with HD [42]. Treatment with IFN-β significantly enhanced HGF synthesis and secretion by blood monocytes, contributing to the clinical benefit of IFN-β in RRMS via the combined HGF-mediated neuroprotective and anti-inflammatory mechanisms.

In this context, it is also important to remind that monocytes are abundant in inflammatory MS brain lesions and displayed also altered functions and an activated innate immune signature in MS patients with clinically more severe course [43]. In particular, the type I IFN pathway is dysregulated in these monocytes, which may contribute to more active disease. In addition to that, conditional genetic knockout of IFNAR1 in monocytes, but not in T cells, B cells, or central nervous system cells, leads to enhanced disease severity in the animal model of MS [44]. All these evidences indicate that perturbations of the type I IFN signaling pathway and response in monocytes could represent crucial events in MS immunopathology and, at the same time, a key target of IFN-β therapy. On the other hand, we cannot exclude that the replenished TLR7 responsiveness in PBMCs and monocytes of IFN-β-treated MS patients could be related to the rescue or prevention of TLR7 tolerance, that is generally induced by specific ligands of this receptor and leads to a reduced cytokine and Ig production [45]. Indeed, Poovassery and Bishop [45] recently demonstrated that IFN-β controls TLR7 tolerance and activation through the PI3K/Akt/mammalian target of rapamycin signaling pathway but also enhancing TLR7 expression in human B cells. These data allow us to envisage that a chronic stimulation of TLR7 by a not-yet identified microbial or self-derived agonist could lead to a reduced sensitivity to TLR7 stimulation in MS patients, which might result in altered B-cell functions.

To date, the enhancement of Ab synthesis mediated by IFN-β treatment is not resulting in an excessive Ig production or in an induction of auto-Abs (data not shown and [46]). Rather, this therapy restores via monocyte-mediated bystander mechanisms the correct TLR7 responsiveness of MS-derived B cells, which in this way fully acquire the capacity to mature into Ig-producing cells, similar to HDs. In this scenario, the study from Warrington et al. [47] is of great interest that demonstrates how naturally occurring polyclonal human Abs (in particular IgM) can strongly promote remyelination inducing a transient Ca2+ influx in myelin-forming cells. Thus, the ability of IFN-β therapy to induce polyclonal Abs (and in particular IgM) with potential remyelinating activity reveals another mechanism of protection possibly mediated by this drug, that could lead to amelioration of neurological symptoms in MS patients.

An additional aspect to take into account from our findings is that the deficient TLR7-induced IgM and IgG production observed in MS patients might correlate with worsening of disease or impaired immune responses against infections with TLR7-recognized RNA viruses, such as influenza, or upon vaccination. Many studies have been conducted in this regard. Different groups have reported that the risk of relapse is increased in individuals with MS bacterial or viral infections [48, 49]. In the case of influenza, it was shown that the reduction of infection episodes leads to a lower number of exacerbations in MS sufferers. In a study with 180 RRMS patients, 33% of individuals, who became infected with this virus, developed an acute relapse within 6 weeks [50]. However, randomized, double-blind, placebo-controlled studies during the past decade have shown that influenza vaccination of MS patients neither increases the relapse rate nor worsens the course of disease [51]. Indeed, the administration of standard vaccines in MS patients is considered safe worldwide, it follows the same recommendations as in healthy adults and actually should be recommended to MS patients in order to avoid attacks of the disease [52]. Having all this in mind, it cannot be excluded that our data on the reduced level of secreted Abs in response to TLR7 stimulation can have a role in the exacerbation of relapses observed in MS-affected individuals along episodes of influenza infection. The increasing recognition that viruses, and in particular EBV, can be etiological factors driving the development of MS or other autoimmune diseases in genetically susceptible individuals further strengthens the potential of administering anti-viral therapies to people affected by these disorders [12]. In line with this view, the increased TLR7 gene expression observed upon IFN-β might be part of a specific antiviral program induced by this cytokine that could counteract dysregulated responses to viral infection in MS patients.

Targeting TLRs and modulating their functions are increasingly recognized as potential therapeutics for different autoimmune diseases, including rheumatoid arthritis, systemic lupus erythematosus, and MS [15, 53]. Indeed, a common trait of most auto-immune disorders is a chronic inflammation occurring at specific sites within the body or as a systemic complication suggested to be sustained also by TLR activation. In our study, we highlighted a defect in TLR7 gene expression in PBMCs of MS patients as compared with HDs. TLR7 is a member of the TLR family that has been implicated at different levels in autoimmunity. Polymorphisms in the TLR7 gene were shown to have a role in time to disease progression in individuals affected by MS [54] but also in predisposition for systemic lupus erythematosus in Asian population [55].

All together, the above evidence suggests how a tight regulation of both TLR expression and TLR-induced responses, in particular those driven by TLR7 triggering, is necessary to maintain a healthy and tolerant immune environment. Having found that TLR7 responsiveness was clearly rescued by IFN-β treatment, we can envisage that IFN-β therapy creates a new microenvironment in PBMCs and, likely, in other anatomical sites, where novel interactions among leukocyte subsets are established and might influence the outcome of the immune process. These new insights in MS immunopathology and in the therapeutic effects of IFN-β could help to improve existing therapies or define new therapeutic strategies for MS targeting TLR expression or TLR-induced responses.

Materials and methods

Patients

Sixty patients with definite RRMS according to McDonald's criteria [56] (age, 36.8 ± 7.4 years (mean ± SD)) and 35 age- and sex-matched healthy subjects (40 ± 6.3 years) were enrolled at the S. Andrea Hospital MS Center. Patients were longitudinally studied right before (T0) and 1 month (T1) after the beginning of IFN-β treatment (recombinant IFN-β1b in the formulation of Betaferon, Bayer, 250 μg subcutaneously, every other day). Mean Expanded Disability Status Scale was 1.5 (range 0–6), disease duration was from 1 to 26 years. Patients had neither taken steroids during the 3 months preceding enrollment, nor had received other disease modifying therapies before. The study was approved by the Ethics Committee of S. Andrea Hospital and all the subjects involved in the study gave written informed consent.

Cell isolation and stimulation

Peripheral blood (20–50 mL) was collected from MS patients and HDs and PBMCs isolated by density gradient centrifugation using Lympholyte-H (Cedarlane Laboratories, Hornby, Ontario, Canada). B cells and monocytes were obtained by positive sorting by using anti-CD19 and anti-CD14 conjugated magnetic microbeads (Miltenyi Biotec, Bergish Gladbach, Germany), respectively. The recovered cells were >90–95% pure as determined by flow cytometry using anti-CD19 and anti-CD14 Ab (BD Pharmingen, San Diego, CA, USA).

PBMCs (1 × 106/mL) were cultured in RPMI 1640 (BioWhittaker Europe, Verviers, Belgium) supplemented with 2 mM l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin (Gibco, Grand Island, NY, USA), and 10% FBS (BioWhittaker Europe).

PBMCs were subjected to positive sorting using anti-CD14 conjugated magnetic microbeads (Miltenyi Biotec) to remove monocytes from whole PBMCs.

Whole or monocyte-depleted PBMCs were stimulated with optimal doses of TLR7 and TLR9 agonists: 3M001 (25 μM, a kind gift of Dr. Mark Tomai, 3M pharmaceuticals) and type B phosphorothioate-CpG 2006 oligodeoxynucleotides (3 μg/mL, synthetized by Eurofins MWG Operon), respectively.

Monoclonal anti-human BAFF Ab (20 μg/mL; R&D Systems, Minneapolis, MN, USA) was used to block BAFF biological activity, where indicated.

Flow cytometry analysis

Monoclonal Abs for CD19, CD38, CD86 as well as IgG1, IgG2a control Abs (BD Pharmingen), conjugated with FITC, PE, or PERcP as needed, were used for flow cytometry analysis. Briefly, cells (1 × 105) were collected and washed once in PBS containing 2% FBS, then incubated with Abs at 4°C for 30 min. After staining, cells were fixed with 2% paraformaldehyde before analysis on an FACSCan (BD Pharmingen). CD38 and CD86 expression was evaluated in the CD19+/SSC gate.

Elispot assay

PBMCs from HD or MS patients before and after IFN-β therapy were treated with the TLR7 or TLR9 agonist for 7 days as specified. For Elispot assay, cells were then recovered and incubated for 3 h at 37°C in IgM- or IgG-coated 96-well flat-bottomed microtiter plates. Wells were subsequently washed and then incubated overnight at 4°C with alkaline phosphatase-conjugated goat anti-human IgM or IgG (Sigma). After extensive washings with PBS-Tween, the alkaline phosphatase substrate 5-bromo-4-chloro-3-indolyl phosphate (BCIP; Sigma) was added to each well. After rinsing and drying, the spots were enumerated under a stereomicroscope with 40-fold magnification. The ratio between the number of Ig-secreting cells and the number of CD19+ cells present in each culture was evaluated in 10 HDs and 15 MS patients analyzed before and after IFN-β therapy. The values represent the means ± SEM.

ELISA

Supernatants from PBMC cultures were prepared as described in the text, harvested, and stored at −80°C. ELISA kit for IL-6 was purchased from Bender MedSystems (Burlingame, CA, USA). The values shown represent the means ± SEM of the cytokine concentrations detected in the supernatants of cultures collected from independent experiments.

IgM and IgG content present in the supernatants of PBMCs obtained from 6 MS patients and 5 HDs was evaluated by Elisa kit (Bethyl Laboratories, Inc.). The values represent the means ± SEM of Ig concentration.

Sera from 6 HDs and 12 MS patients were also collected and BAFF level was evaluated by Quantikine BAFF immunoassay (R&D Systems) according to the manufacturers’ instruction.

RNA purification and real-time RT-PCR

DNase-I-treated total RNA was purified from MS patient- or HD-derived PBMCs using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) or B cells and monocytes using the high pure RNA isolation kit (Roche Diagnostic GmbH, Mannheim, Germany). Reverse transcription was primed with oligo (dT) and random hexamers by using the Murine Leukaemia Virus Reverse Transcriptase (Invitrogen Life Technologies, Carlsbad, CA, USA). Quantitative PCR assays for GAPDH, TLR7, TLR9, and BAFF were done at least in duplicates by using the Light Cycler Fast Start DNA SYBR Green I Master Mix in the presence of 3 mM MgCl2 on a LightCycler Instrument (Roche Diagnostics) as previously described [22].

Sample values were normalized by calculating the relative quantity of each mRNA to that of GAPDH using the formula 2−ΔCt and expressed as mean ± SD.

Primer pairs for GAPDH and TLR7 was as previously described [22]. TLR9 and BAFF primers used in this study were as follows:

  • TLR9_forward: 5′-TGAAGACTTCAGGCCCAACTG-3′
  • TLR9_reverse: 5′-TGCACGGTCACCAGGTTGT-3′
  • BAFF_forward: 5′-TGAAACACCAACTATACAAAAG-3′
  • BAFF_reverse: 5′-TCAATTCATCCCCAAAGACAT-3′

Statistical analysis

Statistical significance of differences was determined by Student's t-test for paired or unpaired data (p < 0.05 was considered significant) from JAVA Applets & Servlets for Biostatistics software.

Acknowledgements

This work was supported by the Italian Multiple Sclerosis Foundation # 2009/R/7 (to E.M.C.). We thank Dr. Mark Tomai (3M pharmaceuticals) and Francesca Aloisi (Department of Cell Biology and Neurosciences, Istituto Superiore di Sanità, Rome, Italy) for their helpful discussion. We acknowledge Dr. Silvia Romano, Dr. Giulia Coarelli, and Dr. Arianna Fornasiero, who took care of patients and helped with sampling. Furthermore, we thank Eugenio Morassi (Division Service for Data Management, Documentation, Library and Publishing Activities, Istituto Superiore di Sanità, Rome, Italy) for preparing drawings.

Conflict of interest

Marco Salvetti received lecture fees from Biogen-Dompé and received research support from Bayer-Schering, Biogen-Dompé, Merck-Serono, and Sanofi-Aventis.

Abbreviation
BAFF

B-cell-activating factor of the TNF family

HD

healthy donor

RRMS

relapsing remitting MS

Ancillary