The winner of our Bright Sparks postdoc session at the BSI/NVVI Joint Congress in December was Dr Madhvi Menon from University College London with her talk entitled ‘Abnormal crosstalk between plasmacytoid dendritic cells and regulatory B cells contributes to the pathogenesis of systemic lupus erythematosus’. Here, Madhvi tells us more about her research on the pathogenesis of this inflammatory autoimmune disease.
Systemic lupus erythematosus (SLE) is an inflammatory autoimmune disease characterised by diverse clinical features including arthritis, rashes, kidney disease, seizures, heart and lung disease, and mental impairment.1 The heterogeneity of clinical manifestations in SLE patients reflects the multisystem inflammation caused by diverse immunological abnormalities.
B cells are considered central to the pathogenesis of SLE, contributing to the disease by both antibody-dependent and antibody-independent mechanisms. B cell defects in SLE include hyperproduction of autoantibodies, dysregulated cytokine production, and abnormal expression and function of key signalling molecules.2 Because of the multiple B cell abnormalities in SLE and their role in SLE pathogenesis, patients are often treated with rituximab (B cell depletion; anti-CD20) therapy.3,4 In addition to aberrant B cell responses, a gene expression signature characterised by an upregulation of interferon (IFN)-I induced gene transcripts (IFN-I signature) has been identified in 60–80% of SLE patients.5 IFNα (a type-I IFN) is primarily secreted by plasmacytoid dendritic cells (pDCs) upon toll-like receptor (TLR) activation.6 In SLE patients, chronic TLR activation results in excessive IFNα production by pDCs. The elevated IFNα levels drive T cell-dependent inflammation, impair apoptotic cell clearance, induce plasma cell differentiation and promote autoantibody production.7
Unriddling the Breg abnormalities in SLE
Claudia Mauri’s lab was the first to discover that regulatory B cells (Bregs), which are important contributors to the maintenance of tolerance, are numerically and functionally impaired in patients with SLE.8 The defect was associated with the inability of immature (CD19+CD24hiCD38hi) B cells to differentiate into IL-10-producing Bregs that exhibit immunosuppressive capacity. When I joined the Mauri lab as a PhD student, the key questions to address were: what are the signals controlling Breg differentiation and why do patients with SLE display a reduced Breg frequency?
As increased IFNα and its associated gene signature is a prominent feature of SLE, we hypothesised that pDCs, via IFNα production, inhibit Breg differentiation whilst promoting plasma cell expansion. To test our hypothesis, we co-cultured healthy B cells and pDCs with CpGC, a TLR9 agonist that activates both cell types. Much to our surprise, we discovered that, in addition to expanding plasmablasts (precursors to antibodyproducing plasma cells), the co-culture of pDCs with B cells results in a dramatic expansion of immunosuppressive IL-10producing CD24+CD38hiBregs. Importantly, this expansion was dependent on the production of IFNα and the expression of CD40L by TLR9-activated pDCs. We further demonstrated that Bregs, in turn, could suppress IFNα production by pDCs, via the production of IL-10. Thus, we established the existence of a novel immune-regulatory feedback loop between pDCs and Bregs in healthy individuals.
"We have established the existence of a novel pDC-Breg crosstalk in healthy individuals, which is aberrant in patients with SLE"
Next, we investigated the outcome of the pDC–Breg interaction in patients with SLE. The conundrum we faced was the following: if pDCs induce Bregs in an IFNα-dependent manner, then why are Bregs reduced in SLE patients despite the presence of chronically activated pDCs and a type-I IFN signature? Using ImageStream technology and in vitro assays, we learned that the pDCs in SLE patients promote plasmablast expansion and antibody production, but fail to induce Breg differentiation. As hyperactivated pDCs in SLE patients produce increased levels of IFNα, we tested the concentration-dependent effect of IFNα on B cell responses. In doing so we discovered that the level of exposure to IFNα is critical in determining the fate of B cell responses. Indeed, exposing healthy B cells to high concentrations of IFNα reduced IL-10 production by B cells to levels similar to that observed in SLE B cells. In contrast, there was a continued expansion of plasmablasts with increasing concentrations of IFNα.
Further analysis revealed that the Breg defects in SLE were also associated with alterations in phosphorylation of STAT1 and STAT3, signals downstream of the IFNα/β receptor. This is most likely due to chronic exposure of B cells to IFNα in vivo. As a consequence, functionally impaired Bregs from SLE patients failed to restrain IFNα production by hyperactivated pDCs. This dysfunctional pDC–Breg crosstalk in SLE patients is likely to increase production of autoantibodies and inflammatory cytokines such as IFNα, creating a ‘vicious cycle’ that contributes to the pathogenesis of disease (see Figure).
Targeting the compromised pDC–Breg crosstalk
Our in vitro results revealed a role for the altered pDC–Breg crosstalk in the pathogenesis of SLE. As we have access to SLE patients undergoing rituximab therapy, we were able to investigate the effect that in vivo B cell depletion and repopulation has on the pDC–Breg crosstalk. We utilised clinically well-characterised cohorts of rituximab-treated SLE patients upon B cell repopulation: those that respond to rituximab and those that fail to respond to rituximab. We found that SLE patients responding to rituximab therapy display a low IFN-I gene signature, and following B cell repopulation, show a restoration in Breg number and function. Moreover, the restored Breg response corresponded to normalised activation of pDCs. In contrast, SLE patients not responding to rituximab therapy have a numerical deficiency in Bregs (following B cell repopulation), hyperactivated pDCs and display an ‘elevated’ IFN-I gene signature. This suggests that the rituximab therapy in SLE might work, in part, by normalising the pDC–Breg crosstalk.
We have established the existence of a novel pDC–Breg crosstalk in healthy individuals, which is aberrant in patients with SLE. This work has been published in the journal Immunity.9
It is plausible that therapeutic strategies targeting the pDC–Breg crosstalk may provide ways of resetting the immune system for better management of SLE. This would be particularly relevant for SLE patients that display an elevated IFN-I signature and fail to respond to B cell depletion therapies. Our current work is focussed on harnessing the therapeutic potential of the pDC–Breg crosstalk for long-term remission from SLE. If successful, this would be an important step towards personalised medicine for the treatment of SLE.
Postdoctoral Research Associate, University College London
- Rahman & Isenberg 2008 New England Journal of Medicine 358 929–939
- Dorner et al. 2011 Arthritis Research & Therapy 13 243
- Anolik et al. 2004 Arthritis & Rheumatology 50 3580–3590
- Aguiar et al. 2017 Arthritis Care and Research 69 257–262
- Bennett et al. 2003 Journal of Experimental Medicine 197 711–723
- Hoffmann et al. 2015 Trends in Immunology 36 124–138
- Pascual et al. 2006 Current Opinion in Immunology 18 676–682
- Blair et al. 2010 Immunity 32 129–140
- Menon et al. 2016 Immunity 44 683–697