Autoantibody production in patients treated with anti-TNF-α

Patients with rheumatoid arthritis, Crohn’s disease or spondyloarthritis who are treated with selective TNF-α inhibitors may develop autoantibodies, such as antinuclear antibodies (ANAs) and anti-dsDNA antibodies. Various methods have shown that infliximab led to ANAs in 29–76.7% and anti-dsDNA antibodies in 10–29% of rheumatoid arthritis patients participating in clinical trials. Furthermore, ANAs and anti-dsDNA antibodies have appeared in 11–36 and 5–15% of rheumatoid arthritis patients treated with etanercept and 12.9% and 5.3% of those treated with adalimumab, respectively. Antiphospholipid antibodies, which are mainly detected by means of anticardiolipin assays, have also been found in rheumatoid arthritis patients receiving TNF-α blockers. There have been a number of reports of the development of antidrug antibodies, of which those against infliximab lead to infusion reactions and shorter responses to treatment. This has led some authors to conclude that it is necessary to add methotrexate to infliximab in order to reduce the risk of the appearance of anti-idiotype autoantibodies     

The effect of rituximab on anti‐platelet autoantibody levels in patients with immune thrombocytopenia

Rituximab is an effective therapy resulting in a platelet count improvement in 60% of patients with immune thrombocytopenia (ITP). Rituximab depletes B cells; thus, a reduction in platelet autoantibody levels would be anticipated in patients who achieve a clinical response to this treatment. The objectives of this study were to determine whether rituximab was associated with a reduction in platelet autoantibody levels, and to correlate the loss of autoantibodies with the achievement of a treatment response. We performed a case‐control study nested within a previous randomized controlled trial of standard therapy plus adjuvant rituximab or placebo. We measured platelet‐bound anti‐glycoprotein (GP) IIbIIIa and anti‐GPIbIX using the antigen capture test. Of 55 evaluable patients, 25 (45%) had a detectable platelet autoantibody at baseline. Rituximab was associated with a significant reduction in anti‐GPIIbIIIa levels (P = 0·02) but not anti‐GPIbIX levels (P = 0·51) compared with placebo. Neither the presence of an autoantibody at baseline nor the loss of the autoantibody after treatment was associated with a response to rituximab. The subset of patients with persistent autoantibodies after treatment failed to achieve a platelet count response, suggesting that persistence of platelet autoantibodies can be a marker of disease severity.     

Efficacy of acitretin in the treatment of reactive neutrophilic dermatoses in adult-onset immunodeficiency due to interferon-gamma autoantibody

Reactive neutrophilic dermatoses in adult-onset immunodeficiency due to interferon-γ autoantibody (AOID) are usually associated with concomitant active opportunistic infections. Data focusing on the treatment of these dermatoses with non-immunosuppressive drugs are still lacking. The aim of this study was to assess the efficacy and safety of acitretin treatment of reactive neutrophilic dermatoses in AOID. We conducted a retrospective review of all patients with AOID who had reactive neutrophilic dermatoses and had been treated with acitretin from January 2008 to December 2018. In total, 23 patients had been diagnosed with AOID, with 27 episodes of reactive neutrophilic dermatoses (20 episodes of Sweet syndrome and seven episodes of generalized pustular eruption) and treated with acitretin. The median effective dose of acitretin was 10 mg/day. The mean initial response was 5.6 ± 2.3 days. The rash had almost or completely cleared within 2 weeks in 70.4% of patients. One case had developed a reversible acitretin-induced liver injury with hepatocellular pattern. The median total duration of treatment was 3 months. In conclusion, this study demonstrates the potential role of acitretin as one of the treatments of choice for reactive neutrophilic dermatoses in AOID, attributable to its favorable response and good tolerability.     

Determination of plasma antiglycan autoantibodies in patients with IgA nephropathy and the correlation with clinical characteristics

Objective To establish the measurement of IgA1 O-glycan-specific antiglycan autoantibodies in patients with IgA nephropathy (IgAN), and evaluate their role in the development and progression of IgAN. Methods In the IgAN regular follow-up cohort of Peking University Institute of Nephrology from January 2006 to December 2015, 170 patients drawn by stratified randomization were enrolled in this study. Enzyme-linked immunosorbent assay (ELISA) was used to determine the levels of plasma galactose-deficient IgA1 (Gd-IgA1) and antiglycan autoantibody (IgG and IgA1). The correlation between antiglycan autoantibodies and clinicopathological parameters was analyzed by linear correlation and multiple linear regression analysis. The receiver operating characteristic curve (ROC) was used to evaluate the value of plasma anti glycide antibodies in the diagnosis of IgAN. Results IgG and IgA1 antiglycan autoantibodies that specifically recognized Fab-hinge region (Fab-HR) antigens could be detected in both IgAN and healthy control group. Agglutinin inhibition test showed that the specific antigen epitope was N-acetylgalactosamine (GalNAc) residue exposed to galactose deficiency in IgA1 hinged region. There was no significant difference in the absolute levels of plasma IgG antiglycan autoantibodies between IgAN and healthy controls (P=0.963). After adjustment of the plasma level of IgG, the normalized antiglycan autoantibody (ln[IgG antiglycan antibody/IgG]) in patients with IgAN was significantly higher than that in healthy controls (0.58±0.31 vs 0.37±0.11, P﹤0.01). The normalized level of IgG antiglycan autoantibody in IgAN patients was positively correlated with 24 h urine protein level during renal biopsy (Spearman r=0.183, P﹤0.05), and was also significantly correlated with 24 h urinary protein level after adjusting for baseline clinical and pathological factors (β=0.713, 95%CI 0.323-1.102, P﹤0.01). The area under ROC curve (AUC) of normalized IgG antiglycan autoantibody in the diagnosis of IgAN was 0.764 (95% CI 0.682-0.845, P﹤0.05). Using the cut-off value of 0.396, the sensitivity and specificity of normalized IgG antiglycan autoantibody for IgAN were 0.729 and 0.700 respectively. There was no significant difference in the absolute or normalized levels of IgA1 antiglycan autoantibodies between IgAN patients and healthy controls. Conclusions Gd-IgA1-specific antiglycan autoantibodies can be detected both in IgAN patients and healthy controls. They are elevated in some patients with IgAN and possibly involved in the development of IgAN.     

Exosomes expressing the self‐antigens myosin and vimentin play an important role in syngeneic cardiac transplant rejection induced by antibodies to cardiac myosin

Long‐term success of heart transplantation is hindered by humoral and cell‐mediated immune responses. We studied preexisting antibodies to cardiac self‐antigens, myosin and vimentin, and exosomes induced by antibodies to self‐antigens in eliciting immune responses to cardiac grafts. After syngeneic heterotopic murine heart transplantation, rabbit anti‐myosin or normal rabbit immunoglobulin was administered at day 0 or 7. Sera were collected after heartbeat cessation, cellular infiltration was analyzed, and exosomes were isolated from sera. Histopathologic examination of the controls' transplanted hearts demonstrated normal architecture, and their sera demonstrated neither antibodies to self‐antigens nor exosomes expressing self‐antigens. Administration of antibodies to cardiac myosin immediately posttransplantation (day 0) but not on day 7 triggered graft failure on day 7, and histopathologic examination revealed marked cellular infiltration with neutrophils and lymphocytes. Histopathologic examination of rejected hearts also demonstrated myocyte damage as sera had increased antibodies to myosin and vimentin and development of exosomes expressing self‐antigens. Administration of exosomes isolated from failed grafts containing self‐antigens induced graft dysfunction; exosomes isolated from stable mice did not induce graft failure. Antibodies to self‐antigens can induce exosomes containing self‐antigens, initiating an immune response and causing graft failure after cardiac transplantation.     

De novo development of antibodies to kidney‐associated self‐antigens angiotensin II receptor type I,collagen IV,and fibronectin occurs at early time points after kidney transplantation in children

Chronic rejection is the leading cause of graft loss following pediatric kidney transplantation. Our group and others have demonstrated an association between the development of Abs to self‐antigens and chronic rejection following adult lung and heart transplantation. The goal of this study was to determine whether Abs to kidney‐associated self‐antigens develop following pediatric renal transplantation. We investigated post‐transplant development of Abs to kidney‐associated self‐antigens angiotensin II receptor type I, Fn, and collagen IV in a pediatric cohort. Using ELISA, we measured Abs to kidney‐associated self‐antigens in serum. Our cohort included 29 subjects with samples collected pretransplant and for 12 months post‐transplant. No samples had Abs to kidney‐associated self‐antigen pretransplant. In contrast, 50% (10/20) of subjects developed Abs to one or more kidney‐associated self‐antigen post‐transplantation. The median time to antibody appearance and duration of persistence were 103 and 61 days, respectively. Development of Abs did not correlate with graft function. Half of subjects developed Abs to kidney‐associated self‐antigens angiotensin II receptor type I, Fn, or collagen IV in the first year after kidney transplantation—a higher rate of early antibody development than expected. In this small study, Abs did not correlate with worse clinical outcomes.     

Dot enzyme‐linked immunosorbent assay strip as a screening tool for detection of autoantibody to interferon gamma in sera of suspected cases of adult‐onset immunodeficiency

Background

Being able to detect the presence of autoantibodies to interferon (IFN)‐γ in serum is essential for evaluating patients with suspected adult‐onset immunodeficiency (AOID) with unusual intracellular infections. Most reported patients with AOID have been Asian, although the exact prevalence of this illness is unknown. To date, no standard assay exists to detect autoantibodies to IFN‐γ. An easy‐to‐use, low‐cost assay that can be performed in any laboratory would be a valuable tool for clinical management of AOID, as well as better reveal its prevalence.

Methods

Our experimental study exploited a dot enzyme‐linked immunosorbent assay (Dot‐ELISA) strip to detect autoantibodies to IFN‐γ. Sera from 66 HIV‐negative patients having autoantibodies to IFN‐γ as determined by indirect ELISA were tested.

Results

Dot enzyme‐linked immunosorbent assay was sensitive (100%) and specific (94.5%), with a positive predictive value of 97.6% and a negative predictive value of 100%.

Conclusion

This simple method provides prompt qualitative results that can be read visually and used in facilities with limited testing capabilities.

     

Many alphaIIbbeta3 autoepitopes in chronic immune thrombocytopenic purpura are localized to alphaIIb between amino acids L1 and Q459

In chronic immune thrombocytopenic purpura (ITP), autoantibodies bind to platelet surface proteins, particularly alphaIIb, resulting in platelet destruction by the reticulo-endothelial system. In order to better localize the autoepitopes on alphaIIb, we studied the binding of antibodies to Chinese hamster ovary (CHO) cells expressing either alphaIIbbeta3 or alphaIIb-alphavbeta3 chimaeras in which a segment of alphaIIb (either amino acids L1-Q459, L1-F223 or F223-Q459) was substituted for that portion of alphav. We evaluated platelet-associated autoantibodies from 14 ITP patients with alphaIIb-dependent antibodies. Ten of 14 bound to alphaIIb (L1-Q459)-alphavbeta3, showing that autoepitopes were often localized to this region of alphaIIb. In addition, each of the autoantibodies binding to alphaIIb (L1-Q459)-alphavbeta3, also bound to CHO cells expressing either alphaIIb(L1-F223)-alphavbeta3 or alphaIIb(F223-Q459)-alphavbeta3). In two of the three eluates tested, > 95% of the autoantibody binding to alphaIIb could be adsorbed using CHO cells expressing any of the three chimaeras, showing that the epitope(s) have contact points on either side of amino acid F223; in the third eluate, only a portion ( approximately 40%) could be adsorbed by the chimaeric cell lines showing that, in this patient, an additional antibody was also present, directed to a site distal to amino acid Q459. The remaining four eluates bound to CHO cells expressing alphaIIbbeta3 but to none of the chimaeras, suggesting that these epitopes are also distal to amino acid Q459. We conclude that the binding of many anti-alphaIIbbeta3 autoantibodies is dependent on the presence of alphaIIb amino acids L1-Q459.     

Localization and immunogenicity of tubulin in the filarial nematodes Brugia malayi and B. pahangi

Tubulin was identified in the filarial nematodes Brugia malayi and B. pahangi by several approaches. Initially, a monoclonal antibody (6D8) was selected for its unusual binding to B. malayi microfilariae in indirect immunofluorescence assays: 6D8 showed granular, heterogeneously dispersed fluorescence on fixed parasites but did not bind to unfixed microfilariae. The microfilarial sheath did not bind 6D8, although it did bind fluoresceinated wheatgerm agglutinin. By Western blotting against microfilarial sonicate, 6D8 reacted with a 50,000-55,000 mol. wt protein, and also bound to purified chicken brain beta-tubulin. Additionally, this monoclonal antibody reacted with a recombinant fusion protein expressed by a clone (Bpa-7) originally isolated from an adult B. pahangi cDNA expression library by its reaction with chronic human filariasis serum. This clone encodes a small 40 amino acid C-terminal segment corresponding to residues 409-449 of beta-tubulin, and shows complete amino acid sequence homology with vertebrate beta-tubulin from 409 to 430 but 55% divergence (six amino acid substitutions, four insertions and one deletion) from human and chicken beta-tubulin over positions 431-449 at the C terminus. Antibody to both parasite and vertebrate (chicken) tubulin was found in filarial infection sera, with higher levels of autoreactive antibody apparent in amicrofilaraemic individuals. Immunogold electron microscopy was then used to localize beta-tubulin in B. malayi microfilariae and adult worms. Tubulin was shown not to be exposed on the microfilarial sheath or in the cuticle of either stage, but was found to be abundant in the somatic tissues. In microfilariae, 6D8 bound myofibril structures under the hypodermal layer, and also bound within cell nuclei. In the adult stage, tubulin was associated with muscle blocks, as well as the intestinal brush border and the embryonic uterine microfilariae.