Failure of regulation results in an amplified oxidation burst by neutrophils in children with primary nephrotic syndrome

Abstract

The mechanism responsible for proteinuria in non-genetic idiopathic nephrotic syndrome (iNS) is unknown. Animal models suggest an effect of free radicals on podocytes, and indirect evidence in humans confirm this implication. We determined the oxidative burst by blood CD15⁺ polymorphonucleates (PMN) utilizing the 5-(and-6)-carboxy-2′,7′-dichlorofluorescin diacetate (DCF-DA) fluorescence assay in 38 children with iNS. Results were compared with PMN from normal subjects and patients with renal pathologies considered traditionally to be models of oxidative stress [six anti-neutrophil cytoplasmic autoantibody (ANCA) vasculitis, seven post-infectious glomerulonephritis]. Radicals of oxygen (ROS) production was finally determined in a patient with immunodeficiency, polyendocrinopathy, enteropathy X-linked (IPEX) and in seven iNS children after treatment with Rituximab. Results demonstrated a 10-fold increase of ROS production by resting PMN in iNS compared to normal PMN. When PMN were separated from other cells, ROS increased significantly in all conditions while a near-normal production was restored by adding autologous cells and/or supernatants in controls, vasculitis and post-infectious glomerulonephritis but not in iNS. Results indicated that the oxidative burst was regulated by soluble factors and that this regulatory circuit was altered in iNS. PMN obtained from a child with IPEX produced 100 times more ROS during exacerbation of clinical symptoms and restored to a near normal-level in remission. Rituximab decreased ROS production by 60%. In conclusion, our study shows that oxidant production is increased in iNS for an imbalance between PMN and other blood cells. Regulatory T cells (T_regs) and CD20 are probably involved in this regulation. Overall, our observations reinforce the concept that oxidants deriving from PMN are implicated in iNS.

Introduction

Idiopathic nephrotic syndrome (iNS) is the most frequent cause of proteinuria in children. The term identifies a panel of heterogeneous conditions supported by different genetic backgrounds and pathology variants [1,2] that influence the final clinical outcome. The pathogenesis of primary forms is still hypothetical and related therapies appear circumstantial [3–5].

Animal models that mimic focal segmental glomerulosclerosis, the most serious variant of iNS in children, are induced by an infusion of chemicals which play a direct role in oxidative stress in glomeruli. The most popular are adriamycin [6–11] and puromycin [12], which have been utilized widely in the past, but as models of oxidative damage they appear to be too strong and their relation to human disease is unclear. In fact, a definitive demonstration that radicals of oxygen (ROS) generation are implicated in iNS has not been reported thus far, probably because ROS have a very short time span and react rapidly with cells. For the above reason, we have a unique opportunity to detect traces of their short existence [13]. A recent study indicating that the oxido-redox status of plasma albumin is amplified in children with iNS points in this direction [14]. In fact, the analysis of serum albumin in children with iNS performed by means of liquid chromatography electrospray ionization-mass spectrometry (LC-ESI-MS/MS) demonstrated chemical modifications of the unique free ³⁴Cys of the protein sequence to a sulphonic group (SO₃H), which is the end-product of its oxidation and therefore represents a surrogate biomarker of oxidative stress [15,16]. This finding suggests indirectly that unstable oxygen radicals (ROS) production is increased in iNS and underscores the potential participation of oxidative stress in the pathogenesis of proteinuria.

Little is known about the source of ROS in iNS, but it seems reasonable to suspect that they derive from circulating polymorphonuclear neutrophils (PMN). This work addresses the key point of the oxidative burst in PMN and its regulation in iNS utilizing the 5-(and-6)-carboxy-2′,7′-dichlorofluorescin diacetate (DCF-DA) fluorescence assay [17], allowing reliable monitoring of ROS generation by PMN.

Materials and methods

Patients

A total of 38 children with iNS, six with vasculitis, seven with acute post-streptococcal glomerulonephritis and five with other glomerulonephritis, were enrolled into this study. For definition of the normal ranges of ROS production by PMN, we utilized 31 normal blood samples as controls obtained from the Gaslini staff (aged 22–45 years) and 10 obtained from children with hypercalciuria (aged 2–16 years) who were followed at our department. Clinical data for all the patients in the study are given in Table 1. One major criterion for eligibility of iNS patients was the absence of familiarity and/or of relevant mutations of slit-diaphragm genes (NPHS2, CD2AP2, WT1). Blood samples from children with iNS for ROS determination were collected and analysed repeatedly during different phases of the disease, i.e. recurrence and remission of proteinuria and partial remission. Nephrotic syndrome was defined by the presence of florid proteinuria > 40 mg/h/m² that corresponded to a urinary protein : urinary creatinine (Pu/Cu) ratio > 4. Different values of Pu/Cu (between 0·5 and 4) identified intermediate phases of the disease in which proteinuria had been lowered, but not normalized, by therapies and were defined as partial remission. Renal biopsies were obtained in 24 patients of the nephrotic group and in all the 11 patients with other renal diseases (acute post-streptococcal glomerulonephritis did not require a renal biopsy because it was diagnosed on the basis of specific biological markers; see below). Tissue samples were processed by standard procedures for light microscopy and immunofluorescence; in a few cases, electron microscopy was also performed. Focal segmental glomerulosclerosis (FSGS) was observed in 15 cases, mesangial proliferation with immunoglobulin (IgM) deposition and minimal lesions were found, respectively, in three patients and minimal change nephropathy (MCN) in one patient. Following our internal protocol, the other 19 patients did not undergo a renal biopsy. With a few exceptions, most patients were receiving treatment with steroids alone (at variable doses) and/or in association with cyclosporin and with angiotensin-converting enzyme (ACE) inhibitors. Steroids were given at a starting dose of prednisolone 2 mg/kg followed by gradual tapering [18,19]. Cyclosporin was administered at a 5 mg/kg starting dose, followed by dose adjustment to maintain cyclosporin serum levels between 50 and 100 ng/ml. Five patients with idiopathic nephrotic syndrome who did not respond to therapies or were dependent on steroids and/or cyclosporin were finally treated with Rituximab, Hoffmann-La Roche, Basel, Switzerland (Mabthera 375 mg-m², two doses at a distance of 2 weeks) and blood for ROS analysis was tested after 30 and 60 days from the second dose (Table 2).

Table 1.

Clinical data relative to patient categories involved in the study.

Pathology	n		n	Sex M/F	Age (years)	Age at onset (years)	Activity +/−	Renal function grades 1/2/3	C3 serum levels	Auto-antibodies	Therapy St/Csa/ACEi/other
Nephrotic syndrome	38	FSGS	15	11/4	9	6	15/0	12/2/1	Normal	No	8/12/15/0
		IgM	3	3/0	6	3	3/0	3/0/0	Normal	No	2/3/3/0
		MCN	1	1/0	5	2	0/1	1/0/0	Normal	No	1/0/0/0
		no biopsy	19	9/10	8	6	11/8	19/0/0	Normal	No	16/2/10/0
Vasculitis	6	pANCA	2	1/1	6	5	0/2	5/0/0	Normal	Yes	1/0/0/2
		cANCA	1	1/0	6	6	0/1	2/0/0	Normal	Yes	0/0/1/1
		Schonlein–Henoch	2	1/1	12	12	2/0	2/0/0	Normal	No	2/0/0
		anti-phospholipid	1	1/0	12	12	1/0	0/0/1	Low	Yes	1/0/0
Post-streptococcal glomerulonephritis	7			5/2	14	14	7/0	4/1/0	Low 7/7	No	0/0/7/0
Other glomerulonephritis	5	IgA	2	2/0	8	6	4/0	5/0/0	Normal	No	0/0/5/0
		MPGN	2	2/0	4–10	3–7			Low 2/2	No
		MGN	1	1/0					Normal	Possible

Renal function was graduated according to creatinine levels as 1 (serum creatinine < 1mg%), 2 (serum creatinine between 1 and 2 mg%) and 3 (serum creatinine > 2 mg%). Activity (+) indicates the presence of an increased proteinuria (Pu/Cu > 1). ACEi: angiotensin converting enzyme inhibitors; ANCA: anti-neutrophil antibodies; csa: cyclosporin; FSGS: focal segmental glomerulosclerosis; IgM: mesangial proliferation with IgM deposition; MGN: membranous glomerulonephritis; MPGN: membrano-proliferative glomerulonephritis; St: steroids; M/F: male/female.

Table 2.

Clinical data relative to one child affected by IPEX and nephrotic syndrome and seven patients with idiopathic nephrotic syndrome who had been treated with Rituximab.

Patholgy	n	Sex M/F	Age (years)	Age at onset (years)	Activity +/−	Renal function grades 1/2/3	Therapy St/Csa/ACEi/other
IPEX	2	1/1	13–40	10/−	1/1	2/0/0	1/0/1/1
iNS treated with Rituximab	7	5/0	2–12	2–8	7/0	7/0/0	2/4/1/0

Renal function was graduated according to creatinine levels as 1 (serum creatinine < 1 mg%), 2 (serum creatinine between 1 and 2 mg%) and 3 (serum creatinine > 2 mg%). Activity (+) indicates the presence of presence of an increased proteinuria (Pu/Cu > 1). ACEi: angiotensin converting enzyme inhibitors; Csa: cyclosporin; St: steroids; M/F: male/female; IPEX: immunodeficiency, polyendocrinopathy, enteropathy X-linked syndrome; iNS: idiopathic nephrotic syndrome.

In the group with active vasculitis six children [three with p/c anti-neutrophil cytoplasmic autoantibody (ANCA⁺), two with Schonlein–Henoch purpura, one with catastrophic anti-phospholipid syndrome] were enrolled. Finally, seven patients presented with acute post-infectious glomerulonephritis or other glomerulonephritis and one child was affected by immunodeficiency, polyendocrinopathy, enteropathy X-linked syndrome (IPEX) with L242P mutation in forkhead box P3 (FoxP3) (Table 2). Acute post-streptococcal glomerulonephritis was defined on the basis of low C3 associated with acute urinary aspects more related to haematuria. The child with IPEX developed overt proteinuria with nephrotic syndrome 2 years from the onset of acute bowel disease and was treated in a first step with a combination of steroids (prednisolone 2 mg/kg/day) and cyclosporin (5 mg/kg/day). He had stable negative autoantibodies, normal serum C3 levels and his proteinuria was highly selective (albumin > 90%). After 2 months, cyclosporin was stopped and tacrolimus was started at an initial dose of 0·1 mg/kg/twice a day followed by modification on the basis of serum levels (target 5–10 ng/ml). He did not receive a renal biopsy because in a former surgical procedure he presented a severe allergic reaction to anaesthetic drugs that required resuscitation.

Cell sorting and fractionation

To obtain the whole blood cell population, peripheral blood was layered onto a Histopaque 1119 (Sigma-Aldrich, St Louis, MO, USA) density gradient and centrifuged at 290 g for 30 min. Cells collected at the interfacies were washed in Hank's balanced salt solution (HBSS) and resuspended at a concentration of 10⁶ cells/ml.

To fractionate PMNs and monucleated cells, respectively, 10 ml of whole blood were mixed with 8 ml of dextran solution (Plander 70 000; 30 g/500 ml; Fresenius Kabi Italia s.p.a, Isola della Scala, Verona, Italy) and left to sediment by gravity for 45 min at room temperature. The resulting yellow supernatant was layered onto Ficoll-Histopaque 1077 and centrifuged at 290 g for 30 min at room temperature; the monucleated cells (lymphocytes and monocytes) were collected at the interfacies and washed with HBSS; the pellet containing PMNs was resuspended in hypotonic solution to eliminate contaminating red blood cells and then washed in HBSS. The purity of the neutrophil fraction was ≥95%, as assessed by CD15 monoclonal antibody. Cells were then resuspended at a concentration of 10⁶ cells/ml for DCF-DA fluorescence assay.

Supernatant preparation

Mononucleated cells obtained as described above were resuspended at a concentration of 10⁶ cells/ml in HBSS and incubated for 30 min at 37°C, then spun down briefly and the free cell supernatant was added to PMNs, before incubation with DCF-DA.

ROS production

Cells were incubated in Eppendorf tubes (whole blood cell population) or 24-well plates (fractionated populations) with 2 µM CM-H₂-DCF-DA (Molecular Probes, Eugene, OR, USA) for 50 min at 37°C. When using fractionated populations, PMNs were incubated alone and/or with autologous mononucleated cells (1:1 ratio) and/or the free cell supernatants; in the latter case heterologous supernatant was also used. Intracellular fluorescence (excitation 492 nm, emission 527 nm) was measured using a Becton Dickinson FACSCalibur instrument (Franklin Lakes, NJ, USA) equipped with Cellquest software. When analysing the whole blood cell population, fluorescence was detected only in the gated neutrophil fraction, based on evaluation of forward-scatter (FS) and side-scatter (SS) parameters.

Statistical analysis

The one-way analysis of variance was used for comparison of ROS levels in various conditions. Results are given as mean ± standard error of the mean (s.e.m.).

Results

ROS are generated by resting PMN in iNS

ROS generation was first evaluated in resting PMN utilizing the DCF-DA fluorescence assay. DCF-DA is a non-fluorescent molecule which becomes fluorescent in the presence of a wide variety of ROS, including superoxide anion and hydroxyl radicals [17]. ROS generation was evaluated in 38 children with iNS (a few cases were evaluated in different phases) and results were compared to six patients with vasculitis, seven with acute post-infectious glomerulonephritis, five with other glomerulonephritis and 41 normal samples (Table 1).

In a first set of experiments, the oxidative burst by CD15⁺ cells (PMN) was evaluated in the presence of other nucleated cells in order to maintain potential intracellular interactive regulatory pathways (Fig. 1). Compared to normal cells, ROS generation by iNS PMN was increased markedly (+400–600%), this increment being particularly evident in those patients with Pu/Cu > 4 (P < 0·01). ROS generation was higher in patients with active iNS compared to other glomerulonephritis (+300%) patients and the difference with vasculitis was still evident (+50%).

Fig. 1.

Radicals of oxygen (ROS) production by resting polymorphonucleates (PMN). 5-(and-6)-carboxy-2′,7′-dichlorofluorescin diacetate (DCF-DA) fluorescence assay was performed with whole blood in which PMN (CD15⁺ cells) were analysed as separated gates by fluorescence activated cell sorter (FACS). Thirty-one normal subjects and 38 children with idiopathic nephrotic syndrome (iNS) were considered in this first set of experiments. Patients were evaluated in different clinical phases as defined by their urinary protein/urinary creatinine ratio (Pu/Cu) (remission Pu/Cu < 0·5; relapse Pu/Cu > 4; intermediate Pu/Cu 0·5–4). Six patients with renal vasculitis and five with other glomerulonephritis were evaluated in the same experimental conditions. Statistical analysis performed with parametric tests (one-way analysis of variance) showed significant difference in ROS production *P < 0·05; **P < 0·01.

ROS generation by PMNs is regulated by supernatants deriving from other cells

In order to verify whether ROS generation by PMN was influenced by other mononucleated blood cells, CD15⁺ cells were separated from other cells (CD3⁺, CD20⁺, CD11⁺) by dextran sedimentation, followed by gradient-density centrifugation. While ROS generation was modified only partially in CD15⁺ cells purified from iNS (+90%), the difference (six-fold increment) was marked in the case of normal PMN (Fig. 2a). In order to exclude that the purification procedure had activated normal PMN, the normal blood cell composition was reconstituted by putting autologous CD3⁺CD20⁺CD11⁺ cells together with normal CD15⁺ cells (Fig. 2b). In this case, ROS production decreased significantly (−36%, P < 0·01), and the same effect was observed by incubating autologous supernatants (−45%, P < 0·01). Supernatants from iNS PMN produced no effect on normal PMN (−6%) (Fig. 2b), suggesting that a regulatory factor is produced by normal non-PMN but not by PMN of nephrotic patients. Finally, ROS production by iNS was reduced only slightly by autologous supernatants (−25%), whereas normal supernatants produced a more evident effect (−54%, P < 0·01) (Fig. 2b), suggesting that in iNS there is a defect in a regulatory factor implicated in ROS generation.

Fig. 2.

Radicals of oxygen (ROS) production by polymorphonucleates (PMN) isolated from other cells. (a) 5-(and-6)-carboxy-2′,7′-dichlorofluorescin diacetate (DCF-DA) fluorescence assay performed utilizing whole blood as in Fig. 1. In other experiments, PMN (CD15⁺ cells) were first isolated by dextran sedimentation in various experimental conditions: comparison of PMN obtained from normal donors and from children with active idiopathic nephrotic syndrome (iNS); (b) comparison of isolated PMN obtained from normal subjects and from patients with iNS. DCF-DA analysis was performed with isolated PMN alone and/or in presence of autologous–heterologous cells and supernatants deriving from CD3⁺, CD11⁺ and CD20⁺ cells; (c) patients with vasculitis and post-streptococcal glomerulonephritis. Statistical analysis performed with parametric tests (one-way analysis of variance) showed significant difference in ROS production *P < 0·01.

PMN purified from patients with vasculitis and post-infectious glomerulonephritis with the technique described above produced very high levels of ROS when separated from other blood cells (Fig. 2c). The increment in respect to non-separated cells was relevant (+290% and +600%, respectively, P < 0·01 for both), suggesting that regulatory factors produced by non-PMN cells were active in these patients. The experiments performed with autologous and heterologous supernatants (Fig. 2c) confirmed that in the presence of soluble factor(s) deriving from other nucleated cells, PMN activation in patients with vasculitis and post-infectious disease was decreased markedly (up to −60% to –67% in both cases, P < 0·01 for both).

The case of IPEX and anti-CD20 treated children

Two cell models characterized by spontaneous and/or drug-induced reduction of specific cell sets were utilized in order to define which cell type among the different panels of nucleated blood cells regulates ROS generation by PMN. In a first approach, we utilized cells deriving from a patient with a mild form of IPEX due to a missense mutation (p.Leu242Pro) of FoxP3 that is associated with a normal forkhead box protein expression [20]. This patient had developed severe enteropathy associated with steroid-resistant nephtoric syndrome and was treated with the calcineurin inhibitor tacrolimus, which induced stable remission (Table 2). During the proteinuric phase he had stable negativity of circulating autoantibodies, C3 was normal and proteinuria was highly selective. As shown in Fig. 3, ROS generation by CD15⁺ isolated during a clinically active phase of the disease was impressive (200× that of resting normal PMN) and the same parameter was not modified by co-incubation with autologous cells. By comparison, ROS production was reduced markedly in this patient during remission of clinical signs, the same as for the mother who carried the same mutation in a heterozygous state. This observation supported the concept that regulatory T cells (T_regs) play a relevant regulatory effect on ROS production by PMN.

Fig. 3.

Radicals of oxygen (ROS) production by polymorphonucleates (PMN) in an immunodeficiency, polyendocrinopathy, enteropathy X-linked (IPEX) child. 5-(and-6)-carboxy-2′,7′-dichlorofluorescin diacetate (DCF-DA) fluorescence assay performed with whole blood containing non-isolated PMN (CD15⁺ cells) in a child with immunodeficiency, polyendocrinopathy, enteropathy X-linked (IPEX) and his normal mother, carrier of the same mutation (p.Leu242Pro) of forkhead box P3 (FoxP3).

The second model consisted in determining ROS generation in seven patients who had been treated with anti-CD20 humanized monoclonal antibodies (Rituximab) for clinical purposes [21], and had their circulating CD20 almost eliminated. ROS production by PMN was reduced significantly (−59%), suggesting that CD20 cells also play a regulatory effect in this context (Fig. 4).

Fig. 4.

Radicals of oxygen (ROS) production after Rituximab. 5-(and-6)-carboxy-2′,7′-dichlorofluorescin diacetate (DCF-DA) fluorescence assay performed with isolated polymorphonucleates (PMN) (CD15⁺ cells) in patients with idiopathic nephrotic syndrome (iNS) after an in vivo treatment with anti-CD20 antibodies (Rituximab). Isolated PMN were studied alone and in the presence of autologous–heterologous cells and supernatants deriving from CD3⁺, CD11⁺ and CD20⁺ cells.

Discussion

Oxidants are toxic for the kidney, and whenever their production overcomes the intra- and extracellular defence, renal damage is generated [22,23]. This may occur when the quota of metabolic ROS produced in mitochondria is altered or not buffered by specific enzymatic pathways (such as in the presence of inherited defect of co-enzyme Q synthesis) [24] or, more frequently, when leucocytes are activated by immunological triggers such as complement, immune complexes or ANCA [25,26]. Experimental models have provided ample evidence for the role of oxidants in leucocyte-dependent glomerulonephritis (for a review see [27]), which occurs when activated cells infiltrate the glomeruli. The rapid degradation of ROS (in the order of milliseconds) implies that the majority of ROS are toxic when produced in contact with cells [28]. Long-living ROS metabolites, such as chloramines, could instead be active when produced outside the kidney, as they have a half-life in the order of minutes and can reach the glomeruli if not buffered by specific systems [29]. In animals, the infusion of stable oxidants such as adriamycin [6,7] and puromycin [30] produces renal lesions and nephrotic syndrome that mimic minimal change disease and/or focal glomerulosclerosis, but as models of oxidative damage they appear to be too strong and cannot be translated readily to human pathology.

However, with the exception of mutations of genes coding for enzymes deputed to co-enzyme Q synthesis (see above) [24,31], no clear evidence that oxidants are implicated in human renal pathologies is available. Studies in humans are particularly difficult, mainly because of the short-lived ROS, and the only opportunity we had in the recent past was the possibility of determining the products of their interaction with membrane and circulating proteins [13]. Musante and co-workers [15] described the presence of oxidation products of serum albumin in patients with focal glomerulosclerosis (sulphonic-³⁴ Cys albumin), suggesting that ROS had been produced and buffered by protein sulphydril groups. Technology evolution based on specific fluorochromes that are activated by oxidants and are detected by fluorescence activated cell sorter (FACS) analysis [17] now allows determination of the oxidative burst in circulating neutrophils and represents an opportunity to address the main problem of ROS production directly in patients by nucleated circulating cells.

This study utilized the technique described above to determine the oxidative burst in children with iNS and/or in other pathologies that are considered traditionally to be models of oxidative stress, such ANCA vasculitis. The results indicated clearly a 10-fold increase of ROS production by PMN in children with iNS, correlating with the proteinuria level. The second finding was that the oxidative burst by PMN was regulated highly by T lymphocytes, mainly T_regs, by means of soluble factors and that this regulatory circuit was altered in iNS. Despite some pitfalls (i.e. most patients were receiving treatment at the time of the study, no clear-cut association with the presence of proteinuria in single patients), this study is the first to have performed direct analysis of ROS in blood and create a basis for considering a regulatory defect in cell–cell cross-talk.

The demonstration of a marked increase of oxidative burst in iNS was first obtained in resting PMN isolated by simple gradient density centrifugation with no other manipulation. In this case, the difference between iNS and controls was impressive. When the same cells were, instead, separated from other cells, the oxidative burst increased in all other samples but not in iNS. Finally, when the composition of blood nucleated cells was reconstituted by gathering PMN and other homologous cells together, then the same marked difference between iNS and normal or other renal conditions was restored. This observation suggested to us that a regulatory mechanism deriving from other constituents of the nucleated blood cell panel was active in maintaining a low-oxidative status in normal conditions, and that this mechanism was altered in iNS. The definition of the cell regulating the oxidative burst in PMN and the soluble factor that mediates this effect was a primary topic for the second part of this study. We obtained indirect evidence that CD20 and T_regs are implicated in the regulation of the PMN oxidative burst, and other studies are clearly needed to reinforce our hypothesis. The implication of T_regs is supported mainly by the observation of markedly increased ROS production in a patient affected by a mild form of IPEX who presented with an associated nephrotic syndrome. IPEX is a rare and severe disorder of the immune regulation system due to mutation of FoxP3, which leads to life-threatening autoimmune phenomena [32,33]. The child described here suffered from a mild form of IPEX produced by a missense mutation of FoxP3 [20] and developed iNS, which disappeared after several months of therapy with the calcineurin inhibitor tacrolimus. IPEX patients may present with iNS that can be substained by different renal lesions, i.e. minimal changes [34] and/or membranous deposits [35,36]. The child described here did not undergo a renal biopsy because of a life-threatening episode of allergy to anaesthetics during a preceding surgical procedure. Therefore, we are left to consider surrogate biomarkers such as selectivity of proteinuria to propose iNS with minimal lesions, which remains a speculative, albeit most probable, diagnosis in this case. The concept that T_regs inhibit ROS production by PMN is not new, as it has been demonstrated clearly that these cells modulate PMN function and lifespan [37]. Moreover, in animal models of nephritic syndrome induced by oxidants [38] and in Buffalo/Mna rats [39], infusion of activated T_regs protects against renal injury and slow down the progression to renal failure.

A second clue on the cell network regulating PMN burst derived from the analysis of patients who had been treated with anti-CD20 antibodies (Rituximab) for iNS. Indeed, it is known from recently published case reports and a small series of treated patients [21] that these antibodies reduce proteinuria successfully in iNS patients. Besides showing that CD20 cells are implicated in regulation of the oxidative burst by PMN, this finding offers an adjunctive explanation of the successful effect of Rituximab in iNS.

In conclusion, our study shows for the first time that oxidant production by PMN is increased markedly in children with iNS as the result of an imbalance between PMN and other nucleated blood cells. T_regs are probably involved in this regulation, and CD20 cells also have a role. Overall, our observations reinforce the general concept that oxidants are implicated in iNS and provide a rational link to considering T_reg and oxidants within a common mechanism.

Disclosure

None.