Deglycosylation of myeloperoxidase uncovers its novel antigenicity.

Myeloperoxidase (MPO) is a common target antigen of anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis and is recognized in one-third of patients with anti-glomerular basement membrane (GBM) disease. Our previous study identified over 60% of patients with anti-GBM disease recognizing linear peptides of MPO heavy chain. Here we tested whether aberrant glycosylation alters MPO antigenicity through exposure of neo-epitopes on MPO molecules. Atypical glycosylated MPO molecules, including all possible glycosylation types, were prepared by exoglycosidase and endoglycosidase treatments. Antibodies were detected from the sera of 40 patients with anti-GBM disease without the coexistence of MPO-ANCA. Circulating antibodies against aberrant glycosylated MPO existed in 21 of these patients. Non-glycan MPO and MPO with only N-acetylglucosamine had high frequencies of recognition (16 and 15 patients, respectively). Antibodies binding to aberrant glycosylated MPO could not be inhibited by intact MPO or GBM antigen. When applied to ethanol-fixed neutrophils from normal individuals, these antibodies yielded a typical cytoplasmic staining pattern (c-ANCA). Antigen specificity was detected in 90% of the antibodies using five peptides containing one of the five N-glycosylation sites each, mostly on N323, N355, and N391. The antibodies were restricted to IgG1 subclass, could activate complement, and induce neutrophil degranulation in vitro. Thus, aberrant glycosylated MPO exposed neo-epitopes and was recognized by half of the patients with anti-GBM disease. Their antibodies possessed pathogenic characteristics and may be associated with kidney injury.

New lymphatic cell formation is associated with damaged brain tissue clearance after penetrating traumatic brain injury.

We characterized the tissue repair response after penetrating traumatic brain injury (pTBI) in this study. Seventy specific pathogen-free Kunming mice were randomly divided into the following groups: normal control, 1, 3, 7, 15, 21, and 30 days after pTBI. Hematoxylin and eosin (H&E) staining, immunohistochemistry, and immunofluorescence were performed to examine and monitor brain tissue morphology, and the distribution and expression of lymphatic-specific markers lymphatic vessel endothelial receptor-1 (LYVE-1), hematopoietic precursor cluster of differentiation 34 (CD34) antigen, and Prospero-related homeobox-1 (PROX1) protein. H&E staining revealed that damaged and necrotic tissues observed on day 1 at and around the injury site disappeared on day 7, and there was gradual shrinkage and disappearance of the lesion on day 30, suggesting a clearance mechanism. We explored the possibility of lymphangiogenesis causing this clearance as part of the post-injury response. Notably, expression of lymphangiogenesis markers LYVE-1, CD34, and PROX1 was detected in damaged mouse brain tissue but not in normal tissue. Moreover, new lymphatic cells and colocalization of LYVE-1/CD34 and LYVE-1/PROX1 were also observed. Our findings of the formation of new lymphatic cells following pTBI provide preliminary insights into a post-injury clearance mechanism in the brain. Although we showed that lymphatic cells are implicated in brain tissue repair, further research is required to clarify the origin of these cells.

A sparse deep learning model for privacy attack on remote sensing images.

Deep learning tools have been a new way for privacy attacks on remote sensing images. However, since labeled data of privacy objects in remote sensing images are less, the samples for training are commonly small. Besides, traditional deep neural networks have a huge amount of parameters which leads to over complexity of models and have a great heavy of computation. They are not suitable for small sample image classification task. A sparse method for deep neural network is proposed to reduce the complexity of deep learning model with small samples. A singular value decomposition algorithm is employed to reduce the dimensions of the output feature map of the upper convolution layers, which can alleviate the input burden of the current convolution layer, and decrease a large number of parameters of the deep neural networks, and then restrain the number of redundant or similar feature maps generated by the over-complete schemes in deep learning. Experiments with two remote sensing image data sets UCMLU and WHURS show that the image classification accuracy with our sparse model is better than the plain model,which is improving the accuracy by 3%,besides, its convergence speed is faster.

Antibodies against linear epitopes on Goodpasture autoantigen in patients with anti-neutrophil cytoplasmic antibody-associated vasculitis.

In a substantial number of patients with crescentic glomerulonephritis, both anti-glomerular basement membrane (GBM) antibodies and anti-neutrophil cytoplasmic antibodies (ANCA) are detected simultaneously. ANCA is presumed to be the initial event but the mechanism is unknown. In the present study, we investigated the antibodies against linear epitopes on Goodpasture autoantigen in sera from patients with ANCA-associated vasculitis, aiming to reveal the mechanisms of the coexistence of the two kinds of autoantibodies. Thirty-one patients with ANCA-associated vasculitis were enrolled in this study. Twenty-four overlapping linear peptides were synthesized across the whole sequence of Goodpasture autoantigen. Serum antibodies against linear peptides were detected by ELISA and their associations with clinical features were further analyzed. Twenty-five out of the thirty-one (80.6%) sera from patients with ANCA-associated vasculitis possessed antibodies against linear peptides on Goodpasture autoantigen. These antibodies could be detected in 50% of patients with normal renal function (Scr ≤ 133 µmol/L), 70% of patients with moderate renal dysfunction (133 µmol/L < Scr ≤ 600 µmol/L), and 94% of patients with renal failure (Scr > 600 µmol/L) (P = 0.032). The highest recognition frequencies were found for peptides P4 (51.6%), P14 (54.8%), and P24 (54.8%), which contained the sequences that constitute the conformational epitopes of EA (P4) and EB (P14) recognized by anti-GBM antibodies. The level of anti-P4 antibodies was positively correlated with the percentage of crescents in glomeruli (r = 0.764, P = 0.027). Patients with anti-P24 antibodies had a significantly higher prevalence of renal dysfunction on diagnosis (88.2 vs. 42.9%, P = 0.018). Antibodies against linear epitopes on Goodpasture autoantigen could be detected in sera of patients with ANCA-associated vasculitis, which might mediate the production of antibodies towards the conformational epitopes on Goodpasture autoantigen, namely, the anti-GBM antibodies.