Neutrophil Traps Linked to Disease Activity in IgA Vasculitis

This study was approved by the ethics committee of the First Affiliated Hospital of Guangxi Medical University (NO. 2019 KY-E-137), and written informed consent was obtained from all the parents or guardians of the patients involved.

A case–control study was performed to evaluate levels of NETs in IgAV and whether circulating NETs could predict the activity of IgAV. This study included 193 cases IgAV in children and 192 healthy control children who were recruited from February 2019 to May 2020 in the First Affiliated Hospital of Guangxi Medical University. The IgAV patients were divided into four sub-groups: 61 patients were allocated to the onset group, 64 patients were in the remission group, 43 patients were allocated to the active group (which included some subjects undergoing relapse), and 25 patients were in the drug withdrawal group (withdrawal group).

The inclusion criteria were as follows: All the patients met the clinical diagnosis of IgAV after use of the EULAR/PRINTO/PRES criteria (10). Renal involvement (IgAV nephritis, IgAVN) was defined as the presence of hematuria (red cell count >5/mm³ at high magnification) and/or a proteinuria and/or an eGFR <60 ml/min in children (11). The onset group included the patient diagnosed with IgAV treatment without glucocorticoids and/or immuno-suppressor or treatment with glucocorticoids for less than 3 days for the first time, and the untreated active patients were those without glucocorticoids and/or immuno-suppressor treatment. The active/relapse group was defined as presence one or more of the following symptoms: new rash, arthralgia, abdominal pain, bloody stools, gastrointestinal bleeding, hematuria, and proteinuria, during the maintenance immunosuppressive therapy or after withdrawal of this treatment (i.e., an immunosuppressive therapy-free period). Remission was defined as patients without active/relapse of the disease as described above during their stable maintenance through immunosuppressive therapy periods. The drug withdrawal group was defined as "clinical cured" patients who were neither undergoing current drug treatment nor having regular follow-ups during their drug-free periods.

The exclusion criteria were the following: (1) absence of skin lesions, (2) thrombocytopenia (platelet <100,000/mm³), and (3) presence of any other immune and infection diseases such as respiratory and urinary tract infections, sepsis and systemic inflammatory response syndrome.

Peripheral venous blood samples from the IgAV patients and healthy control were collected into tubes containing ethylene diamine-tetra-acetic acid (EDTA) anticoagulating agent and centrifuged at 1000g for 10 min at 4°C. After centrifugation, the supernatant plasma was separated and then stored at -80°C until use. Fifteen IgAVN patients' renal tissue samples and 9 IgAV patients' (with gastrointestinal involvement and without IgAVN) gastrointestinal tissue samples were formalin-fixed. These were then paraffin-embedded into blocks, and these were used to evaluate the status of NETs. All the patients who were subjected to renal puncture and gastrointestinal endoscopy had to undergo these procedures in order to verify their medical evaluation before the initiation of treatment. The biopsies pertinent to the study were taken during these procedures after consent of the parties involved.

Controls renal biopsies were taken from the normal tissues adjacent to the presence of tumors of 5 patients who underwent surgery for nephrocytoma. Controls gastrointestinal biopsies were taken from five samples of that anatomical area of healthy children with normal endoscopy and histology who were undergoing routine treatment for other reasons.

It is known that NETs are composed of cf-DNA, MPO, NE, and histones, including citrullinated histone H3 (cit-H3). NETs are also the major source of circulating cf-DNA. To identify NETs, we quantified the cf-DNA, cit-H3, MPO-DNA, and NE in plasma samples as was done in a previous study (12). The cf-DNA was quantified by fluorescent assays using ds-DNA Quant-iT PicoGreen quantification kit (Invitrogen). The tests were performed according to the manufacturer's instructions. Briefly, a DNA standard curve of DNA was constructed by using a range of dilutions from 1 to 1000 ng/ml and these samples were incubated for 5 min with Quant-iT PicoGreen reagent at room temperature. The plasma samples to be tested were diluted to 1:5, and the fluorescence signals obtained were measured using a microplate fluorescence reader (Thermo Fisher Spectrophotometer 1510) with filter settings set at wavelengths of 480 and 520nm for excitation and emission, respectively.

Plasma levels of MPO-DNA, NE, and DNase I were determined by enzyme-linked immunosorbent assays (ELISAs). The human MPO-DNA Quantikine ELISA Kit, human PMN elastase ELISA kit, and human deoxyribonuclease I, ELISA Kit used were purchased from Cusabio, Wuhan, 430071. Briefly, after preparing the standards and diluted samples (1:5), the plates were covered with an adhesive strip and incubated for 60 min at 37°C. Horseradish peroxidase-conjugated antibody specific for MPO-DNA (100 µl) was added and incubated for 60 min at 37°C. After washing five times with 200 μl of wash buffer, 90 µl of 3,3',5,5'-tetramethylbenzidene substrate was added and incubated for a further 20 min at 37°C in dark. The O.D. absorbance at 540 nm was read in a microplate reader (Thermo Fisher spectrophotometer 3020) within 20 min after adding 50 µl of stop solution. In order to measure NE and DNase I, the methods were used in accordance with the manufacturer's instructions.

Neutrophils were isolated from healthy children which were then degraded to generate NETs in vitro according to a previously published method (13, 14). Neutrophils were freshly isolated from healthy children and seeded into a 96-well plate. Phorbol-12-myristate-13-acetate (20 nM) (Sigma) was added to the plate and incubated for 4 h at 37°C in an atmosphere of 5% CO₂ so as to generate NETs. The sera from patients and control subjects were diluted in 10 mM Tris-HCl, pH 7.5, 50 mM NaCl, 10 mM MgCl₂, and 2 mM CaCl₂ and added to NETs, and these were incubated at 37°C for 60 min. Aliquots of the solution were then transferred to PBS to form a final concentration of 2 mM EDTA order to stop further degradation. The DNA concentration was quantified using Quant-iT PicoGreen quantification kit (Invitrogen) in accordance with the manufacturer's instructions. A sample of pooled normal healthy children's serum samples was used as an internal control. All samples were compared to the mean of the internal control, and they were measured in duplicate. The amount of degradation calculated for the positive and negative internal controls was set to 100 and 0%, respectively. The amount of degradation of NETs observed in serum samples was expressed as a percentage.

Tissue samples for histology, immunohistochemistry, and immunofluorescence were taken from formalin-fixed, paraffin-embedded blocks. For general morphology and immuno-histochemical analysis, 5–7 μm thick sections were stained with hematoxylin and eosin (H&E) and Masson or hexamine silver stains. For immunofluorescence analysis studies, paraffin sections were de-paraffinized in xylene and dehydrated in graded ethanol solutions and then incubated for 15 min at 60–70°C in EDTA antigen-retrieval buffer (pH8.0). Slides were cooled to room temperature, and then rinsed in PBS (pH 7.4) three times. The endogenous peroxidase was blocked with 3% hydrogen peroxide solution and the sections were incubated at room temperature in the dark for 25 min. Samples were then washed with PBS three times and then blocked with 2% bovine serum albumin solution and normal goat serum for 30 min at room temperature. Samples were then incubated with primary antibodies dissolved in PBS and incubated at 4°C overnight. The following primary antibodies were used: rabbit anti-human NE (1:3000; Abcam, Cambridge, UK), rabbit anti-human MPO (1:1000; Abcam), and rabbit anti-human citH3 (1:100; Abcam). Slides were washed three times with PBS and incubated with CY3-TSA (Servicebio, Technology Co. Ltd., Wuhan, China; diluted 1:2000), Cy5 AffiniPure Goat Anti-Rabbit IgG (1:400, Servicebio), and FITC-TSA (1:1000, Servicebio) in PBS and incubated at 4°C for 50 min. Coverslips were mounted on glass slides using prolonged gold anti-fade reagent with DAPI (Servicebio) for 10 min to counterstain the DNA. Slides were washed a further three times for 5 min each time with PBS and covered with a water soluble fluorescence mounting medium (Servicebio). Images were acquired using 3 DHISTECH Digital slice scanning system (Pannoramic, 3DHISTECH Ltd., Hungary). Image analysis was performed by using the Image J software version 1.51a (NIH, Bethesda, MD). The percentage area occupied by NETs in the tissue was calculated as follows: area of NETs as determined by cit-H3 that co-localized with MPO/NE within the tissue/per high magnification 40× field. When multiple lesions were present in a specimen, the mean value were calculated.

Statistical analysis was performed with the Statistical Package for the Social Sciences software, release 16.0 for Windows (SPSS 16.0, USA). The Shapiro-Wilk test was used to determine the normality of the data which were expressed as the means ± standard deviations. Categorical variables were expressed as percentages. Differences of categorical data were assessed using the chi-square test. Differences of quantitative data were performed using Student's t test, or one-way ANOVA between more than two groups and followed by a post-hoc Dunnett's T3 test and the mean difference (MD) and standard error of the mean are given (SEM). Pearson's or Spearman's tests were used for correlation analysis as appropriate. A P value of <0.05 was considered as statistically significant.

Among the IgAV patients, there were 108 males and 85 females. The mean age of the patients at disease onset was 5.66 + 2.82 years. Among of the healthy controls, which consisted of 98 males and 93 females, the average age was 4.3 ± 1.2 years. There was no difference in gender and age between the IgAV patients and healthy control group. The baseline clinical characteristics of the patients in the different subgroups are shown in Table 1.

One-way ANOVA was used to analyze differences between the IgAV subgroups and the control group. The results indicated that circulating cf-DNA levels were significantly different among the IgAV subgroups and control group (F = 122.29, P = 0.000). The post-hoc multiple comparison tests using Dunnett T3 test were conducted to evaluate the significance of the differences in any two groups. The post-hoc tests showed higher circulating cf-DNA levels in the onset group (327.52 ± 64.31 ng/ml) compared with the withdrawal group (208.8 ± 62.4 ng/ml). The MD and SEM between the two groups was 129 ± 12.8 ng/ml, P = 0.000. The results were similar when comparing the onset group (327.52 ± 64.31 ng/ml) with the remission group (200.47 ± 54.42 ng/ml, the MD and SEM were 127.1 ± 10.68 ng/ml, P = 0.000) or control group (187.91 ± 36.38 ng/ml, the MD and SEM were 139.6 ± 8.6 ng/ml, P = 0.000). No difference was found between the onset group and active group (315.15 ± 73.43 ng/ml), and the MD and SEM were 12.4 ± 13.9 ng/ml, P = 0.990.

The data also showed higher circulating cf-DNA levels in the active group compared with the withdrawal group (the MD and SEM were 116.8 ± 14.9 ng/ml, P = 0.000) or control group (the MD and SEM were 127.2 ± 11.5 ng/ml, P = 0.000). The results were similar when comparing the active group with remission group (the MD and SEM were 114.7 ± 13.1 ng/ml, P = 0.000). Analysis of the circulating cf-DNA levels of the withdrawal group and the remission group showed no significant difference (the MD and SEM were 2.1 ± 11.9 ng/ml, P = 1.0). Neither were there any differences observed between the two groups and the control group (Figure 1).

MPO-DNA and NE levels were significantly different among the IgAV patients' subgroups and the control group when analyzed by one-way ANOVA. The post-hoc multiple comparison tests showed higher MPO-DNA and NE levels in the onset group (MPO-DNA and NE were 613.54 ± 318.7 and 134.83 ± 76.27 ng/ml, respectively) compared with the withdrawal group (MPO-DNA and NE were 306.19 ± 80.72 and 54.54 ± 28.67 ng/ml, respectively; the MD and SEM were 306.64 ± 43.88 and 80.3 ± 11.3 ng/ml, respectively, P = 0.000 in both cases), the remission group (MPO-DNA and NE were 330.89 ± 98.93 and 72.45 ± 49.78 ng/ml, respectively; the MD and SEM were 282.66 ± 42.64 and 62.4 ± 11.6 ng/ml, respectively, P = 0.000 in both cases), or the control group (MPO-DNA and NE were 298.96 ± 89.77 and 60.25 ± 37.49 ng/ml, respectively; the MD and SEM were 314.58 ± 41.31 ng/ml and 74.6 ± 10.1 ng/ml, respectively, P = 0.000 in both cases).

The results were similar when comparing patients of the active group (MPO-DNA and NE were 561.65 ± 299.57 and 148.78 ± 61.39 ng/ml, respectively) with the withdrawal group (the MD and SEM were 254.74 ± 45.12and 94.2 ± 10.9 ng/ml, respectively, P = 0.000 in both cases) and the remission group (the MD and SEM were 230.75 ± 35.38 and 76.3 ± 11.2 ng/ml, respectively, P = 0.000 in both cases). However, no difference was found when compared to the remission, withdrawal, and control groups (P > 0.05 in all cases; Figure 1).

The results of correlation analysis showed that MPO-DNA and NE were highly correlated with circulating cf-DNA. The cf-DNA levels were positively correlated with MPO-DNA (r² = 0.766, P < 0.001). The results were similar between the cf-DNA and NE levels (r² = 0.687, P < 0.001; Figure 2).

The DNase I levels were significantly decreased in the IgAV patients' subgroups when compared with the control group. The post-hoc multiple comparison analysis of the IgAV patients showed lower DNase I levels in the onset (637.59 ± 169.17 pg/ml) and active groups (671.64 ± 202.01 pg/ml) when compared with those in the withdrawal (1130.85 ± 409.93 pg/ml) and remission groups (1165.71 ± 465.83 pg/ml) as well as in healthy subjects (1253.25 ± 427.03 pg/ml; P < 0.001 in all cases; Figure 1). The cf-DNA levels in NETs were negative correlated DNase I (r² = 0.433, P < 0.001; Figure 2).

The ability of serum from IgAV patients to degrade NETs was decreased when compared with healthy control. Sera from the onset and active groups had decreased abilities to degrade NETs. Analysis of the onset (49.88 ± 15.21%) and active groups (58.05 ± 14.14%) showed significantly lower ratios of degradation of NETs compared with the control group (95.31 ± 4.07%), P < 0.001 in both cases. Although the remission group (86.48 ± 10.0%) showed improvement in degradation ability, the levels observed in these patients did not return to normal when compared to the healthy control group (P = 0.035). The ability to degrade NETs was not significantly different between the withdrawal and the control groups (P > 0.05; Figure 3).

NETs complexes were detected in renal specimens of 15 cases of IgAV at both the onset or active stages and in gastrointestinal specimens taken from 9 cases of IgAV with gastrointestinal involvement (i.e., without IgAVN) by multiple immunofluorescence analysis. The immunohistochemistry results from histological classification showed that 4 of the 15 IgAVN patients had minor degenerative nephritis, 9 expressed mesangial proliferation, and 2 had focal proliferative nephritis (Figure 4A). In all the IgAVN patients, IgA and C3 complex deposition were observed in the mesangial area of the glomeruli.

NETs were also found in all the onset and active IgAV patients in both renal and gastrointestinal tissues. NETs were characterized by extracellular DNA and cit-H3 as localized with either MPO or NE in the involved mesangial areas of the glomeruli as well as gastrointestinal mucosal and submucosal tissues (Figures 4A, 5A, and 6A). These were the typical characteristics for NETs seen in IgAV patients. This was confirmed by the overlap with diffused DNA, and NETs deposition was evident in the onset and active IgAV patients. There was no evidence for NETs formation and deposition in control healthy children, who had normal renal and gastrointestinal histology (Figures 4A, 5A, and 6A controls). The area occupied by NETs within the lesion in all the tissue was significantly positive in quantitative analysis compared to the control (Figures 4B, 5B, and 6B, all P < 0.01).