key: cord-0274987-0kcnl5xp
authors: Bradford, H. F.; Menon, M.; Haljasmagi, L.; Vanker, M.; Peterson, P.; Wincup, C.; Abida, R.; Fernandez Gonzalez, R.; Bondet, V.; Duffy, D.; Isenberg, D.; Kisand, K.; Mauri, C.
title: Inactive disease in lupus patients is linked to autoantibodies to type-I interferons that normalize blood IFNα and B cell subsets
date: 2021-04-07
journal: nan
DOI: 10.1101/2021.04.07.21255049
sha: ce2f09845cbe85caec58d47583351db5261d4240
doc_id: 274987
cord_uid: 0kcnl5xp

Systemic Lupus Erythematosus (SLE) is characterized by a prominent increase in expression of type-I interferon (IFN)-regulated genes in 50-75% of patients. Here we investigate the presence of autoantibodies (auto-Abs) against type I IFN in SLE patients and their possible role in controlling disease severity. We report that out of 491 SLE patients, 66 had detectable anti-IFN-auto-Abs. The presence of neutralizing anti-IFN- auto-Abs correlates with lower levels of circulating IFN protein, inhibition of IFN down- stream signalling molecules and gene signatures and with an inactive global disease score. Previously reported B cell frequency abnormalities, found to be involved in SLE pathogenesis, including increased levels of immature, double negative and plasmablast B cell populations were partially normalized in patients with neutralising anti-IFN-auto- Abs compared to other patient groups. We also show that sera from SLE patients with neutralising anti-IFN-auto-Abs biases in vitro B cell differentiation towards classical memory phenotype, while sera from patients without anti-IFN-Abs drives plasmablasts differentiation. Our findings support a role for neutralising anti-IFN-auto-Abs in controlling SLE pathogenesis and highlight their potential efficacy as novel therapy.

To evaluate whether SLE patients develop endogenous auto-Abs to cytokines, we tested sera from 474 SLE patients and 312 healthy controls for auto-Abs against various cytokines with the Luciferase Immunoprecipitation system (LIPS) assay (clinical characteristic, genders, ethnicity are reported in Table 1 ). The auto-Abs to cytokines were measured in groups that included an IFNa pool (IFNa1, IFNa2, IFNa8, IFNa21), IFNw, Table 2 ). The majority of the auto-Abs to cytokines were either undetectable or were produced at a very low concentration in patients or control sera. However, we detected a significant increase in auto-Abs to IFNa (66 out of 474 patients) and IFNw (59 out 474 patients) in SLE patients compared to healthy volunteers (Fig. 1b, c) . Reactivity towards IFN-I subtypes was partially overlapping as 12% (n=43) of patients have auto-Abs to both IFNa and IFNw, whereas anti-IFNa or -IFNw single positive patients comprise 4% each, moreover we show that compared to anti-IFNa-auto-Ab negative and to those with low anti-IFNa-auto-Abs titres (Fig. 1f) . The capacity of anti-IFNa-auto-Abs to neutralise IFNa was assessed using a reporter cell line based neutralization assay as previously described 26 . Serum levels of IFNa negatively correlated with anti-IFNa-auto-Abs titres in patients where the neutralizing capacity is IC50>100 (Fig. 1g) . We observed that serum samples with high anti-IFNa-auto-Abs levels were more efficient in blocking all tested subtypes (IFNa2/5/6/8) of IFNa bioactivity in vitro (Fig. 1h, Supplementary Fig. 1b) .

To gain mechanistic insight into the capacity of neutralizing anti-IFNa-auto-Abs to reduce downstream IFN-I signalling, we compared the IFN composite score 27 , a cumulative measure of mRNA expression of four individual ISGs: MX1, MCL1, IRF9 and STAT1 genes (see methods), in SLE patients with and without anti-IFNa-auto-Abs and healthy controls. The IFN-I score was significantly higher in anti-IFNa-auto-Ab negative patients compared to healthy controls; anti-IFNa-auto-Ab positive patients displayed an IFN-I score comparable to healthy controls (Fig. 1i, Supplementary Fig. 1c ).

To measure the stability of the neutralizing anti-IFNa-auto-Abs, we selected five auto-Abs high patients from the cross-sectional group and measured the titre and neutralization capacity of these auto-Abs longitudinally over an average of 10 years from the first sample collection. All patients tested have auto-Abs against 12 subtypes of IFNa (IFNa1, IFNa2, IFNa4, IFNa5, IFNa6, IFNa7, IFNa8, IFNa10, IFNa14, IFNa16, IFNa17 and IFNa21) at high titres. Interestingly, two of the patients developed anti-IFNa-auto-Abs during followup analysis; one patient initially developed auto-Abs against IFNa8, and subsequently developed cross-reactivity to all the subtypes during following months ( Fig. 1j and Supplementary Fig. 1d ).

We next investigated the effect of neutralizing anti-IFNa-auto-Abs on disease severity.

We observed that patients with at least one neutralizing antibody against an IFNa subtype displayed significantly lower disease activity (as measured by the global British Isles All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. ; https://doi.org/10.1101/2021.04.07.21255049 doi: medRxiv preprint Lupus Assessment Group (BILAG) score) compared to patients without Abs in circulation or to those displaying Abs with negligible neutralizing capacity (Fig. 2a) .

Principal component analysis (PCA) performed on 54 patients with detectable anti-IFNaauto-Abs confirmed that the neutralizing capacity correlates with antibody titres, and that patients with high titres of neutralizing Ab clustered separately from those with active disease and high levels of serum IFNa (Fig. 2b) . No specific organ involvement was identified amongst the different clusters of active patients. No significant differences in disease duration are found.

The results so far suggest that the presence of neutralizing anti-IFNa-auto-Abs is associated with a better clinical outcome. To gain more understanding into the kinetics of antibody production, putative reduction in IFNa levels and the effect on disease progression, we measured the levels of IFNa and anti-IFNa-auto-Ab titres in a longitudinal cohort of SLE patients. In 4 out of 5 patients (depicted here), high titres of neutralizing anti-IFNa-auto-Abs were stable over time, neutralized all the IFNa subtypes (with some variation with IFNa1) and reduced pan-IFNa protein levels in circulation down to undetectable levels (Fig. 2c, Supplementary Figure 2a-b) . The prolonged presence of neutralizing anti-IFNa-auto-Abs together with a consistently low concentration of IFNa, mirrored a persistent inactive clinical score (Fig. 2c, Supplementary Figure 2a) . Patients with non-neutralizing anti-IFNa-auto-Abs in circulation were characterized by high levels of IFNa and were generally associated with a more severe disease activity (Fig. 2d) .

These results suggest that these non-neutralising auto-Abs may stabilize circulating IFNa levels as previously suggested for other cytokines 28-30 . The association of nonneutralizing auto-Abs with high IFNa concentration is intriguing. It has been previously suggested that in certain cases, including more recently in COVID-19 patients 31 , circulating auto-antibodies can increase the half-life of the molecule they bind, possibly through the uptake and release of immune complexes by the neonatal Fc receptor on endothelial cells 32,33 .

It is plausible that IFNa bound to auto-Abs can still be detected by the Simoa assay (Bondet et al, submitted). Depending on the epitope bound, IFNa in immune complexes can be neutralized or remain bioactive. To assess whether IFNa in patients containing non-neutralizing auto-Abs is functionally active, we tested SLE patient sera on isolated monocytes using a pSTAT1 induction assay. pSTAT1 was upregulated by SLE sera containing both IFNa and low-titre anti-IFNa auto-Abs, in comparison to healthy control sera or patients without measurable IFNα or anti-IFNa auto-Abs. Excessive pSTAT1 upregulation was partially abrogated by preincubation with a monoclonal anti-IFNa antibody (26B9) that neutralizes all subtypes of IFNa (Fig. 2e) . Taken together, our results suggest that anti-IFNa-auto-Abs influence disease outcome of SLE depending on the titres of antibodies and their neutralization capacity. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. In response to viral infections, plasmacytoid dendritic cells (pDCs) rapidly produce IFNa that drives B cell maturation into plasma cells producing antibody against viral antigens 6 .

In view of the recent findings showing the detrimental effect of neutralizing anti-IFNaauto-Abs in patients with COVID-19, it is important to understand the impact of neutralizing anti-IFNa-auto-Abs on "nascent" IFNa produced by challenged pDCs, and how this affects healthy B cell differentiation. PBMCs from healthy donors were stimulated with CpGC and cultured respectively with serum from SLE patients with no antibodies or with neutralizing anti-IFNa-auto-Abs; healthy allogeneic serum was used as a control (as depicted in Fig. 4a ). Sera from patients containing neutralizing anti-IFNa-auto-Abs significantly downregulated ISG expression in cultured healthy PBMCs confirming their ability to inhibit IFNa-down-stream signalling (Fig. 4b, Supplementary Figure 3d ).

Although sera from both groups of SLE patients decreased to a certain extent the frequency of immature B cells and plasmablasts and increased memory and mature B All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. ; https://doi.org/10.1101/2021.04.07.21255049 doi: medRxiv preprint cells, the sera from anti-IFNa-auto-Ab positive patients profoundly reduced healthy immature B cell and plasma cell differentiation confirming the ex vivo neutralizing capacity of anti-IFNa-auto-Abs (Fig. 4c) .

The frequencies of IL-10 + B cells is significantly decreased after culture with both groups of SLE sera compared to control sera (Fig. 4d ). There is no difference observed in the frequency of TNFa + B cells (Fig. 4e) .

Amongst different B cell subsets that produce IL-10, immature B cells and plasma cells are the two subsets that produce the most of IL-10. It is interesting, to note that the remaining immature B cells present in the culture stimulated with the serum from the anti-IFNa-auto-Ab positive patients, produced similar levels of IL-10 as B cells stimulated with control sera (Fig. 4f) . Expression of IL-10 by the plasmablast subset was significantly reduced by both anti-IFNa-auto-Ab positive and negative sera compared to control (Fig.   4g ). The overall reduced frequencies of IL-10 + B cells in PBMCs treated with sera containing neutralising anti-IFNa-auto-Abs may be, on one hand, a reflection of the contraction of the immature B cell compartment, since IL-10 + B cells have been shown to be enriched within this subset following IFNa stimulation 35 . However, it should be also acknowledged that sera from anti-IFNa auto-Ab negative patients significantly inhibited IL-10 + B cell expansion, suggesting that other unknown factors, present in the sera of SLE patients, contribute to the overall observed inhibition of IL-10 + B cells.

In summary, we report that a subset of SLE patients harbour neutralizing anti-IFNa-auto-Abs that can modulate B cell responses and are associated to a better disease outcome. This is in contrast with patients with non-neutralizing low titres of anti-IFNa-auto-Abs, which appear to stabilize IFNa in the blood and expand circulating frequencies of DN (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 14/SC/1200. Sample storage complied with requirements of the Data Protection Act 1998.

A total of 50ml whole peripheral blood was collected from an individual patient or healthy donor for PBMC isolation using Ficoll-based density gradient centrifugation. A total of 10ml whole peripheral blood was collected into serumseparator (SST) tubes, centrifuged for 10 minutes at 1200g at RT and serum decanted. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. ; https://doi.org/10.1101/2021.04.07.21255049 doi: medRxiv preprint negative discrimination level was calculated as mean plus 3 SD from 1% trimmed values of healthy controls.

For the detection of IgG subclass-specificity, serum samples were incubated with fusion protein solutions (10 6 LU per well) overnight at +4 C. Next day, agarose beads bound with streptavidin (Novagen, USA) were incubated with biotin-conjugated human subclassspecific antibodies (anti-IgG1, anti-IgG2, anti-IgG4 from BD Pharmingen, USA; anti-IgG3 from Sigma-Aldrich, USA) in microfilter plates for 1 h at room temperature. Overnight incubated serum samples with fusion protein solutions were added to microfilter plate and incubated at room temperature for 2h. Microfilter plates were washed and luminescence intensity measured as above. The results were expressed as luminescence units (LU). Three-fold serially diluted serum samples were co-incubated with interferons for 2h at 37 C, 5% CO2. 10 5 IFN-a-HEK-Blue cells were added to microtiter plate wells and incubated 20-24hours at 37 C, 5% CO2. QUANTI-Blue (InvivoGen, USA) colorimetric enzyme assay was used to determine AP activity in overnight supernatants. Optical density (OD) was measured at 620 nm with Multiscan MCC/340 ELISA reader (Labsystems, USA). Neutralization activity was expressed as IC50, which was calculated from the dose-response curves and represents the serum dilution at which the IFN bioactivity was reduced to half of its maximum. pSTAT1 induction assay. Monocytes were stimulated with IFNa (dilution series between 12.5 U/ml to 400 U/ml) (Miltenyi Biotec), with two SLE patient serums with IFNa All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. ; ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p<0.0001.

No data sets were generated or analysed during the current study. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 7, 2021. ; https://doi.org/10.1101/2021.04.07.21255049 doi: medRxiv preprint

A New Population of Cells Lacking Expression of CD27 Represents a Notable Component of the B Cell Memory Compartment in Systemic Lupus Erythematosus

Activated memory B cell subsets correlate with disease activity in systemic lupus erythematosus: Delineation by expression of CD27, IgD, and CD95

Correlation between circulating CD27high plasma cells and disease activity in patients with systemic lupus erythematosus

Activation of the interferon-α pathway identifies a subgroup of systemic lupus erythematosus patients with distinct serologic features and active disease

Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus

Plasmacytoid dendritic cells induce plasma cell differentiation

Each organ system was given a grade A, B, C, D or E, where A was the most active and E as least active. In this study the grades were converted into numerical scores using the BILAG-2004 index, where A=12, B=8, C=1, D=0 and E=0. Following this, global BILAG scores were calculated by adding the sum of the values from all organ systems. Patients with a global score higher than 6 were considered active. The following abbreviations are used: African-Caribbean (AC), Caucasian (C), South Asian (SA), East Asian (EA), Other (O), hydroxychloroquine (HCQ), prednisolone (Pred), methotrexate (MTX), mycophenolate mofetil (MMF), azathioprine (Aza), general (Gen) central nervous system (CNS)

This work is funded by Versus Arthritis UK program grant (21140) and Research Award